è<r^'

1^ IC Sê «•

THE

JOURNAL

OF THE

COLLEGE OF SCIENCE,

IMPERIAL UNIVERSITY OF TOKYO,

VOL. XX.

m M ^ 19 :;^ ^ W It
published by the university.

t5ky5, japan.

1904—1905.
MEIJI XXXVII.— xxxviir.

Publishing Committee.
^4.

Prof. K. IVIitSUkuri, Ph. D., Rigakuhakushi, Director of the College,

(ex officio).
Prof. J. Sakurai, LL. D., Rigakuhakushi.
Prof. I, Ijima, Ph. D., Rigakuhakushi.
Prof. F. Ömori, Rigakuhakushi.
Prof. S. Watase, Ph. D., Rigakuhakushi.

NÖTIGE.

Beginning with Vol. XVI., each article has a paging of its oion ;
and the position of an article in a volume is indicated by the number
placed at its head.

It is hoped that this arrangement, which enables us to print papers
independently of one another, loill ensure a more rapid publication of the
material than has been possible heretofoi'e.

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

2

2/3f

CONTENTS.

Art. 1. — On the Formation of the Germinal Layers in Gastropoda.

( IVith 3 plates.) By T. Fujita.— Publ. Oct. 10th, 1094.
Art. 2. — Cretaceous Cephalopoda from the Hokkaido. Part. II.

(With 6 plates.) By H. Yabe.— Pub). Oct. 15th, 1904.

Art. 3.— Revisio Euphorbiacearum et Buxacearum Japonicarum.

{With 6 plates.) By B. Hayata.— Publ. Dec. 28th, 1904.

Art. 4.— The Gephyrea of Japan. ( JVith 4 plates.) By K. Ikeda.—
Publ. Dec. 30th, 1904.

Art. 5. — Mesozic Plants from Nagato and Bitchu. (With 3 plates.)
By M. YoKoYAMA.— Publ. April 18th, 1905.

Art. 6. — On the Magnetization and the Magnetic Change of Length
in Ferromagnetic Metals and Alloys at Temperatures rang-
ing from - 186°C to + 1200°C. {With 4 plates.) By K. Honda
and S. Shimizu.— Publ. April 18th, 1905.

Art. 7. —On a New Cestode Larva Parasitic in Man. {With 1
plate.) By I. Ijima.— Publ. May 25th, 1905.

Art. 8.— Mesozoic Plants from Korea. {With 4: plates.) By H. Yabe.
—Publ. August 15th, 1905.

Art. 9.— Modulus of Elasticity of Rocks and Some Inferences
relating to Seismology. {With 2 plates.) By S. Kusakabe. —
Publ. August 5th, 1905.

Art. 10. — Kinetic Measurement of the Modulus of Elasticity for
158 Specimens of Rocks and a Note on the Relation be-
tween the Static and the Kinetic Values of the Same. ( With
1 plate.) By S. Kusakabe. — Publ. September, 29th, 1905.

Art. 11.— On Two New Species of Chimaera. {With 2 plates.) By
S. Tanaka.— Publ. December 18th, 1905.

Art. 12.— A Revised List of Corallinae. Pty K. Yendo.— Publ. Decem-
ber 25th, 1905.

PRINTED AT THE "TOKYO TSUKIJI TYPE FOUNDRY."

^^- ^

JOURNAL OF THE COLLEGE OP SCIENCE, IMPERIAL UNIVERSITï,
^ TOKYO, JAPAN.

VOL. XX., ARTICLE 1.

On the Formation of the Germinal Layers
in Gastropoda.

Bv

T. Fujita, R{^nhushi. NogaJuishi.

With 3 plates.

In spite of tlie foct that within recent years the morphogenesis
of the Mollusca has received a. great deal of attention from
embryologists, there is still much disagreement among authors
with regard even to some of the fundamental points in the early
phases of development. This is no doubt due to actual differences
in developmental processes found in the various animals employed
for investigation, and this makes it highly desirable to extend
our researches to as many different forms as possible, if we are
to arrive at any comprehensive results in these vexed problems.
With this aim in view I first paid a great deal of attention to
the early development of the Gastoropoda, Siphonaria lepida
Gould. A preliminary note on that subject was published in the
Tokyo Zoological Magazine, Vol. VIL, No. 79, 1895. A few
years later I undertook a similar investigation on another form,
Ajjlysia sp. The conclusions reached in the latter study were

2 AET. 1. T. FUJITA : ON THE FORMATION

almost the same as those obtained from Siphoncma, there being
only slight differences in minor details.

The present paper embodies the more essential points brought
out in the investigation on these two animals.

THE EGGS.

The eggs of both species here to be considered are deposited
on rocks between the tide-marks from the early pai't of March
till late in June. Each form presents so many characteristic
differences in egg masses, that it is convenient to describe them
under separate headings.

Eggs of Sipho7iaria : — The eggs of Siphonaina are almost
always found in masses, the shape of which may be described as
an elliptical loop (Fig. 1, Plate I), the last laid portion usually
resting on the portion laid first. When oviposition is completed,
the head of the animal is j^ointed in the direction of the arrow
in Fig. 1. PI. I. When magnified the mass is seen to be formed
of an innumerable number of oblong vesicles, which are linked
one to another by a fine thread, the whole being cemented
together by a gelatinous coating. Each vesicle is enveloped in
two membranes, very thin and transparent, which are closely
applied to each other, and separated only at the poles of the
longer axis (Fig. 4). The egg (Figs. 2 and 4 e) found in each of
these vesicles which must be regarded as the chorion lies rather
eccentrically in the portion furthest removed from the genital
opening. It is perfectly spherical in form, and is 0.1 mm. in
diameter. When freshly deposited it is very slightly tinged with
yellow, but in course of development it gradually becomes white.
In far advanced stages especially in the larval form its color

OF THE GERMINAL LAYERS IN GASTROPODA. à

becomes completely changed to dusky brown, owing to the pre-
sence of the shell. Besides the egg there is, within the chorion,
an abundant supply of albuminous material uniformly distributed
in the form of rather coarse spherules. Fig. 2 shows the general
view of the egg in the chorion, which is seen to be enveloped
in a thin layer of gelatine not yet coagulated. The spherules
just cited are homogenous and translucent ; but at about the time
of the expulsion of the second polar body, vacuoles are produced
in them (Fig. 3). Subsequently, all of the spherules gradually
disintegrate, leaving behind only very fine almost imperceptible
granules. They, however, can be easily coagulated, and therefore
may be brought into view by the application of hardening reagents.
By the time this coagulation stage is reached, it is found that
the distribution of the albuminous material is no longer uniform
throughout as was the case before disintegration. There is
almost always a smaller quantity towards the side, which is upper-
most in Fig. 4, i.e. the future dorso-anterior end of the egg, than
on the opposite side.

JEffgs of Aplysia : — In tliis species too, the eggs are deposited
in a mass ; but here it is in the form of a very long and intricately
and loosely entangled cord which is about 1-1 è vim. in diameter.
In this cord there is recognizable, even with tlie naked eye an
innumerable number of yellow capsules, which are, as in the case
of the former species, connected by a thin thread and enveloped
in gelatine. The capsule represents the chorion, and enclosed
therein we find eggs usually more than thirty in number (Fig.
37, PI. II). The egg has a spherical contour, and is about 0.12
m7)i. in diameter. As is seen in Fig. 38 coarse granules of
deutoplasm are conspicuous in that part of the egg which becomes
the anterior portion of the future embryo. Their color is very

4 ART. 1. T. FUJITA : ON THE FOEMATION

bright, particularly when fresh, varying from Hght red to yeUow.
In advanced stages, however, the color changes completely to
dull brown just as in the case of Siphonaria.

THE ELASTOMERE STAGE.

Siphonaria : — In three or four hours after deposition, the
egg gradually begins to elongate in the antero-posterior dii-ection,
i.e. almost parallel to the long axis of the chorion. Subsequently
a portion becomes somewhat flattened or even depressed, and tlie
cytoplasm thereabout becomes translucent in the form of an
inverted cone. This portion begins to bulge out, and fifteen
minutes later it finally assumes a spherical shape. This eventually
separates from the mother egg, forming the first polar body.
After the complete detachment of this body, the egg returns to
its normal spherical form, and loses its transparency throughout
its entire mass. Then ensues an interval of about thirty minutes ;
after which the egg again behaves as above described. The new
spherical structure or the second polar body is thrown out liere
very close to the first. After the completion of the maturation
processes, the restoration of the egg to the normal spherical form
is again effected. Generally fifteen minutes are required for each
of the two processes : — the formation of the second polar body,
and the restoration of the spherical form. Then follows a pause
of one or more hours, after which the egg elongates for the third
time. This is the preparation for the first cleavage. If we
carefully observe the exact contour of the egg at this stage,
we shall find that it is not strictly elliptical in shape. The
portion which becomes the future posterior end is almost always
slightly pointed. Hence, using an exaggerated expression, it may

OF THE GERMINAL LAYERS IN GASTROPODA. 5

be said to have an ovoid shape. The same state of affairs has
been observed in Limax by JMark^^''. Thus, before the cleavage
process commences, it already shows a tendency to present
inequalities in its two lialves. After the elongation, a depression
begins to appear in the animal pole near the polar body. It
gradually extends downwards until it almost reaches the center
of the egg. At about the same time a similar groove with a
much sharper angle appears in tlie vegetative pole which proceeds
upwards towai'ds the center of the egg. (Fig. o). The depression
from the animal pole, however, proceeds much the more rapidly
of the two until it reaches about the center of the egg. There
it seems to stop, and to meet the approaching groove from the
vegetative pole. Thus these two grooves encircle the egg leaving
a bridge which for the time being connects the two halves. In
the section, this bridge is proved to be the body of the spindle.
This first cleavage plane is often described as moving directly
from the animal pole to the vegetative in other species of Mollusca,
for example, in Umbrella (Heymons)^"^', Unio (Lillie)^''^\ Suecinea
(Schmidt) *''^', Limax (Meisenheimer)^^^\ etc. In either case the
cleavage results in the total division of the egg into two
halves or blastomeres. The plane itself in Siphonaria stands
almost transverse to the longitudinal axis of the future adult.
The blastomeres thus formed seem at first sight similar in size ;
but a careful observation shows that in reality that carrying the
polar hodj is a little larger than its complement. This difierence
in the size of the two halves is the consequence of the original
inequality existing in the natural configuration of the egg before
the cleavage. When the final constriction of the blastomeres
takes place each of them has a spherical outline, not departing
from it even at the point where they come in contact with each

6 ART. 1. T. FUJITA : ON THE FORMATION

Other. When we observe a fresh specimen in such a stage, we
notice the existence of a period, although very short, in which
the two blastomeres are completely separated from each other by
a narrow space. In Limax, Kofoid^'^*^^ claims to have found at the
spot " a rather vague, transparent, and unstained protoplasmic
connection." In my specimens, both fresh and preserved, I could
not finds any instance of such connection, although Kofoid's
statement seems to be a highly reasonable one. The large blas-
tomere thus formed contributes mostly to the formation of the
anterior portion of the adult, while the smaller contributes to
the posterior. Therefore, they will hereafter be designated re-
spectively as the anterior and the posterior, blastomeres. The
eggs of various jMolluscan species which have been thus far studied
are not alike in the size of the two blastomeres. For example,
in Unio the posterior blastomere is larger, while in Neritina
(Blochmann)*^^ Umbrella, etc. it is smaller than the anterior one.
Soon, the two blastomeres again come gradually in contact
with each other, the surfaces by which they touch, becoming
flattened. Simultaneously their general configuration begins to
alter, and becomes somewhat elongated in the direction parallel
to the plane of contact. During this period of apposition the
anterior blastomere slightly shifts its position obliquely upwards
in the direction of the animal pole. In about one hour, the
blastomeres begin to stretch further in the direction already
elongated and finally divide in a plane which is at right angles
to the plane of contact or that of the first cleavage. This
phenomenon appears usually first in the anterior one (Fig. 6),
and is soon followed l)y the posterior complement. In the
blastomeres thus transformed, the furrow causing the second
cleavage commences almost simultaneously at the animal and the

OF THE C.EKMINAL LAYERS IN GASTROPODA. 7

vegetative poles. It stands almost at right angles to tlie first
cleavage plane and coincides with the future median axis of the
body. As a consequense the egg is subdivided into the right,
and the left halves (Fig. 7). Concerning the priority of the
cleavage in the first two blastomeres just described, there is much
controversy. In Umbrella, Aplysia, and others the posterior
blastomere divides first. On the contrary, in Neritma, Unio and
others the anterior precedes the other. Thus it is easy to see
that in Mollusca n eggs the priority of cleavage is quite independent
of the amount of the deutoplasm and its distribution in the egg.
The four cells formed by the second cleavage take a spherical
form just as i'l the case of the first blastomeres, but soon become
flnttcr.^d on one side as the result of mutual contact. They, as
a whole, constitute the first quartet and are commonly designated
as A, B, C, and D, (Fig. 8) of which A and B come from the
anterior, and C and D from the posterior, blastomeres of the
two cell stage. During the formation, and especially during the
accommodation of this quartet the cell C shifts its position
obliquely upwards. At length, it comes to stand in the same
level with the two anterior blastomeres leaving its complement
cell J) in the original level lower than the others. By this shifting
it also happens that the two diagonally opposite cells B and D
are brought in direct contact at the animal, as well as at the
vegetative, poles, forming what has been called the cross furrow.
The cells forming this furrow differ in different species, and
sometimes the furrow is formed by different cells at the two poles.
Moreover, there are many cases, in w^hich the length and the
direction of the furrow presents marked differences in these two
poles. However, in the species under consideration, the furrow
is formed by the juxtaposition of the same cells both in the

8 ART. 1. — T. FUJITA : ON THE FORMATION

(lorsiil and the ventral views, and it has also the same length on
both snrfaces.

Aplysia : — The egg of this species when freshly deposited is
at almost the same stage as that of Sijihonaria. It has not yet
discharged the polar bodies. The process of fertilization, however,
shows no material difference from that of the preceding species.
It is worthy of note that the polar bodies appear in the small
cytoplasmic portion of the egg. After the polar bodies are com-
pletely formed the egg begins gradually to elongate and at last
assumes an ovoid shape, the pointed end being toward the
cytoplasmic portion. The first cleavage plane appears just as
in Siiihonaria from both the animal and the vegetative poles
(Fig. 39). The result of this division is the formation of two
blastomeres totally different in size as well as in quality (Fig.
40). The larger blastomere is heavily laden with deutoplasm ;
while the smaller one is almost wholly devoid of it. Such a
striking difference in the two blastomeres is observed in only a
few cases in Gastropoda even in the so called yolk-laden eggs,
of Umbrella, Planorbis, etc. Of these two blastomeres the larger
half represents the future anterior portion of the body, and the
smaller half its posterior portion, these corresponding to the
smaller and the larger blastomeres of Siphonaria. The further
changes accompanying the first cleavage, such as the gradual and
mutual accommodation of the blastomeres take the same course
as in Siphonaria already fully described above. The second
cleavage in this species takes place first in the posterior half (Fig.
41, PL II). The new cells thus generated are, as in the case
of Siphonaria, designated as A, B, 0, and I) (Fig. 42). Their
i-elation to the future axis of the body coincides with that of the
former species and therefore requires no further explanation.

OF THE GERMINAL LAYERS IN GASTROPODA. 9

After complete accommodation it will be seen that here again
the cross furrow lies between the cells B, and D, ])Oth at the
animal and the vegetative poles. The same phase is observable in
A. limacina,^-^ and also with only a slight difference in Umbrella.
The hypothesis advanced by Kofoid/-^^^ as to the relation existing
between the cross furrow and the presence of an abundant deu-
toplasm can not he accepted in these cases.

THE FIRST QUARTET OF ECTOMERES.

Sipkonaria (cci, b^, Ci, d^, Figs. 9-11, PL I) : — About two
hours ai'ter the complete formation of the original quartet, the
spindles appear simultaneously in all the blastomeres ; but actual
cleavage begins always at the left posterior one D. The axes of
the spindles in all the cells are directed obliquely upwards (Fig.
10), and when seen from the animal pole they seem to converge
towards one another (Fig. 9). The cleavage proceeds in one
cell after anotlier in a certain regular course. In this species
the course of this cleavege is in a right-handed spiral or in the
direction of the motion of clock-hands as shown in Fig. 9, the
apical view, and also in Fig. 10, the posterior side-view. In the
former figure we see that the blastomere D precedes all others in
development, presenting already its small daughter-cell. This
priority of cleavage and the further course of development is
made clear by the next figure. Here the first ectomere cell, di
from the blastomere D is seen to be already well formed, while
the formation of a^ from the blastomere A is not as much advanced.
In the other two blastomeres B and (7, the division is indicated
only by a spindle. It is stated by Kofoid, that in Limax the

10 ART. 1. — T. FUJITA : ON THE FORMATION

spindles gradually shift their positions during this cleavage. Such
does not seem to be the case in Sii^lionaria : the axis of the spindle,
as may be partly gathered from the figures cited above, remains
unchanged in direction throughout the whole of this phase. The
newly formed cells, however, moves lightly when they come to adjust
themselves to one anotlier. Four daughter cells a■^, bi, Ci and di
(Fig. 11) thus propagated form the first quartet of ectomere
cells, lie, when completely detached, on the dorsal side of the
mother-blastomeres. They are small in size and are at first
merely an aggregation of spherical bodies, which however soon
begin to be applied one against another. When finally settled,
they seem to have been shoved somewhat towards the anterior
end (Fig. 11), Now in this new quartet the cross furrow is
again formed by a juxtaposition of the ectomere-cells bi and c/j
just as it was in the preceding stage by their mother-blastomeres
B, and D (Fig. 11). Thus the furrow of the ectomere-cells and
that of the blastomeres are formed by the cells genetically related
to one another. The lines of these two furrows form a small
angle with each other, i.e., neither runs parallel with, nor stands
at right angles to, the other as is the case in many other species.
Aplysia (a^, h^, c^, d^, Figs. 43-44, PL III) : — The form
tion of the first quartet of ectomeres here also commences
first in the posterior blastomere (Fig. 44), most usually from its
left component D, and is followed immediately by the cells C, B,
and A in the order named (Fig. 43). Thus the cleavage proceeds
in a left-handed spiral, i. e. in the direction just the reverse of
that of /Siphonaria. This direction is contrary to that ascertained
by Blochmann in A. limaoina, and also in other forms of yolk-
laden egg noted by various authors. The first quartet of ectomeres
formed are much smaller than the mother blastomeres ; and when

OF THE GERMINAL LAYERS IN GASTROPODA. 11

finally settled, the position they occupy is on the posterior dorsal
portion of the egg, instead of on the anterior dorsal as in Sl-
phonaria. On the other hand the cross furrow keeps its relative
position exactly as in the case of the preceding species (Fig. 45).

THE SECOND QUARTET OF ECTOMERES.

Siphonaria [a.^, bo, c^, do, Fig. 12-13, PI. I) : — After a pause
of about one hour, the second generation of ectomeres originates
again from the original blastomeres. They arise first in the
posterior blastomeres as in the case of the first genera-
tion, but this time from the right cell C. Although spindles
often appear in all the blastomeres at the same time as shown
in Fig. 13, yet the cell C is always ahead of all the others in
actual cleavage. The order of division here is just the reverse
of that in the first quartet as shown in Fig. 12, in which we
notice that of the second quartet cells Co, b-,, and Oo have already
been produced, while D is just preparing to bud off cL. These
ectomere-cells are of the same size as those of the first quartet,
and roughly speaking occupy positions alternate with these latter.

Aplysia (a.,, bo, Co, do, Figs. 45-47, PI. III.) : — In this species
too, the second quartet of ectomere-cells arise from the blasto-
meres, the process begining as in Siphonaria from the right
posterior one C. Cleavage, however, proceeds in a right-handed
spiral, the reverse of the order in which the first generation was
budded off. This point as well as the relative position of the
newly formed generation are clearly illustrated in Figs. 45, and
46, which give respectively the a pical and the left lateral, views
of the egg. In Fig. 45 the spindles are seen in all the blasto-
meres except B, and that of C is most advanced. In the next

12 ART. 1. T. FUJITA : ON THE FOEMATION

figure also, the order of cleavage is clearly indicated inasmuch as
the blastomere C has already almost detached its new daughter
cell Co ; and D is about to divide. In this species the second
quartet of ectomere-cells are much larger than the first set and
occupy positions alternate with these as in Sijihonaria.

THE THIRD QUARTET OF ECTOMERES.

Siphonaria («1.1, ^1.1, c'l.i, dx,i, Figs. 14-15, PL I) : — In
a majority of the Molluscan eggs direct segmentation from the
blastomeres stops temporarily with the second quartet of ectomeres,
the third being derived mostly from the first quartet of ectomere-
cells. This is also the case in the present species. In my pre-
liminary note on the development of the present species com-
municated to the Zoological Magazine already referred to, I noted
that this generation also originates from the blastomeres. A
renewed investigation has shown that this view was a mistaken
one, having been caused by the abnormal condition of the speci-
mens then accessible. In reality, this new generation arises from
the first quartet of ectomere-cells. Cleavage begins from the first
descendant {d-^ of the left posterior blastomere and proceeds in
a right-handed spiral. In Fig. 14, the apical view of this stage,
di.i alone has been budded off; while all the others show the
spindle in a more or less similar state. These ectomere-cells of
the third generation or trochoblast cells, as they are usually
called, are smaller than any of the preceding generations.
Nevertheless, they are rather conspicuous, coloring deeply when
stained. They lie in the same level and in alternation with the
ectomere-cells of the first (piartet, ])ushuig the cells of the second
quartet {a->, h etc) to the left, in the order which indicates a

OF THE GERMINAL LAYERS IN GASTROPODA. 13

riglit-handed spiral. Such a displacement of the cells is recog-
nizable when we compare Figs. 12 and 15.

Aplysia {((.,.1, bo.i, Co.u d-^.u Figs. 48-50, PI. Ill) : — In this
species as well as in A. limacina the ectomere-cells of the third
generation are not the descendants of the first quartet of ectomeres.
Here it is the ectomere-cells of the second quartet, which present
the spindles in a left-handed spiral, the right posterior cell C2
taking the lead in cleavage. Fig. 48, the apical view, shows
the priority of cleavage in the cell Co, although sometimes we
meet with abnormal cases as represented in Fig. 49, in which
the cell «2 precedes the others in cleavage. The new daughter
cells are similar in size to any of the preceding generations, and
do not take a bright stain like those of Siphonaria. Again as
they are the descendants of the second generation, they stand in
alternation with the latter, and their positions with regard to the
first generation are différent from what is seen in the preceding-
species.

THE FOURTH QUARTET OF ECTOMERES.

Siphonaria («y, 4» <^'oj d-^, Figs. 15-16, PI. I) : — An hour
or so after the last stage, the blastomeres again exhibit the
spindles. This time the spindle appears first in the right posterior
blastomere C\ and the cleavage proceeds in a right-handed spiral.
The period in which this generation arises seems to differ with
species, as will be seen directly in the case of Aplysia. In the
present species its formation commences somewhat later so that
it appears concomitantly with that of the next generation. This
is easily proved by referring to Fig. 15, the apical view, and Fig.
16, the sagittal section seen from the left side. In the former,

14 AET. 1. — T. FUJITA : ON THE FORMATION

all of the blastomeres except D are shown to have budded off
their daughter cells c^ etc, while in the latter which represents
a slightly earlier stage the blastomeres of which A and D are seen
show their spindles in an early phase. It is moreover evident in
both the figures, that the formation of this generation is im-
mediately followed by that of the fifth quartet [a^.x, etc). The
cells of the fourth generation are, as a rule, rather larger than
any of the foregoing ones. They occupy a position on the ventral
aspect of the egg, just under the third generation («i.i, etc).
By the formation of the fifth generation they, however, shift
their position more to the left side ; and eventually come to in-
terpose themselves together with the fifth quartet between the
cells of the second {a.2, etc) generation (Fig. 15).

After throwing ofi* three series of ectomeres, i.e., at the forma-
tion of this fourth generation of ectomeres the blastomeres are
completely differentiated, and make no further contribution to the
formation of the ectomere-cells. The cleavage which occurs later
in the blastomeres oives rise to the Q-enerations of entomere-cells.
Hence, the blastomeres receive the name of entomeres with the
single exception of the left posterior one D. This latter ultimately
gives birth, beside the entomere, to the mesomere, so that it is
generally designated as the ento-mesoniere.

Ajjlysia («„ h, c„ d„ Figs. 48-50, PI. Ill) :— The
fourth generation arises as in Siphonaria from the left posterior
blastomere, and proceeds in a right-handed spiral. In this
species too, as was mentioned in Siphonaria, the formation
of this generation very often takes place simultaneously with
that of another quartet, but here it is with the preceding (co,
etc), and not with the following, generation. This fact is well
illustrated in Figs. 48 and 50, respectively showing the apical

OF THE GERMINAL LAYERS IN GASTROPODA. 15

view, and the sagittal section seen from the right side. The
ectoniere-cells of this generation are a little larger than any of
the preceding. When fully adjusted (Fig. 51), they take their
positions, as in the first species interposing themselves between
the cells of the second {a-i, etc) and the third («2.1, etc) quartets.
In this species also, this generation is the final ectomere product
from the blastomeres. Henceforth, the blastomeres are called
entomeres or ento-mesomere according to their characters just as
in Siphonaria.

At about this time a noticeable change occurs with regard
to the relative positions of the anterior and the posterior ento-
meres. By the repeated cleavage already gone through, it is seen
that the posterior half is gradually brought upwards toward the
apical pole. This shifting increases with development and is
combined with a slight movement toward the right side. If we
refer to Fig. 44 a right lateral view of the blastomere stage, and
to Figs. 46 and 49, representing the posterior views of the dif-
ferent stages, one before the formation of the second generation,
and the other after the formation of the third generation, we
can easily comprehend the above mentioned change in the cell
arrangement. At the outset both the anterior and the posterior,
halves of the egg stand ahnost on the same level (Fig. 44). By
two consecutive cleavages of the first and the second quartets of
ectomeres, tlie posterior half has shifted its position almost half
way up the anterior components (Fig. 46). Finally the former
seems to stand high up, and to surmount, the latter. That there
is at the same time a slight shifting toward the right is seen by
comparing Figs. 46 and 49. In the former the plane of contact
of A and B is seen through the right portion of D and near the
contact plane of D and Ö. In Fig. 49 the latter contact plane

16 ART. ]. — T. FUJITA I ON THÉ FORMATION

is much further to the right. Moreover, it is to be observed that
the difference in the amount of cytoplasm contained in the
anterior and the posterior halves of the egg causes the gradual
displacement of the ectomeres in general toward the posterior end.
This seems to be due partly to the fact that during all the cleavage
processes the axes of the spindles in the anterior blastomeres always
have a tendency to be inclined toward the posterior end, so that
an ectomere-cell budded off generally lies dorsally and posteriorly
to its mother cell. The posterior position of ectomeres is marked
even from the stage of the first ectomere generation ; thus in
Fig, 45 which shows the first quartet of ectomeres in their
proper position we see how they lie more on the posterior blasto-
meres. The same arrangement is also continued up to quite late
stages. Although this tendency is noticeable in other yolk-laden
Molluscan eggs it does not seem as pronouned as in Aplysia, for
their blastomeres shows no such marked difference in nature as
there is in this species.

THE FIFTH QUARTET OF ECTOMERES.

Slphonaria («2.1, /^.u g-ia, fh.u Figs. 15-16, PL I) : — As
I hiive already stated the formation of this generation takes place
in conjunction with the preceding one. It arises from the second
quartet of ectomeres. As is seen in Fig. 15 the cleavage makes
its first start from the right posterior cell Co and its course is in
a right-handed spiral. The daughter cells Co.u etc thus formed
are as large as those of the fourth generation. They are situated
more veiitrally than the mother cell and finally interpose them-
selves between the cells of tiie preceding generation, pushing the
latter more to the left along the right spiral. From this time

OF THE GEKMINAL LAYERS TN OJASTROPODA. 17

onwni'd it <;'r:i(lii;illy hocomes evi(l(Mit tluit :ill tlic generations of the
oetonioro-cell, as a whole, make but a sliglit displaceiiient compared
witli wliat is generally seen in other s^iecies. They do not strictly
ov(M'lay tlie eiitonieres ; bnt lie toward the ant(M'ior, and not
toward the postei'ior, end as was the case in Aplysi(( already
allnded to.

Aplysia (/^.,, /;,.,, r,.,, r/^ ,, Figs. 51-'")2, PL IIT) : — Tn
A. limncrna this generation is said to l)e propagated from tlie
tliird quartet of eetomercs ; l)nt in tliis S})ecies it is descended
from the first set of ectomeres. Cleavage begins with the cell r,
and its course is in a left-handed spiral. Tn Fig. »jl, represent-
ing the apical view of such a stage, the new ectornere-cell (',.1 is
seen to have been completely l)ndded off from its mother-cell c^
and to have already assumed the normal form ; while in all of the
other quadrants the spindle remains in the amphiaster stage. The
same phase in the j^osterior two cells of the first ectoraere-quartet
is îdso recognizable in tlu; o]^tical transverse section seen from
the jiosterior side (Fig. /)2). The new ectomere-cell or tlie trocho-
blast is very much smaller than any of the foregoing ccioineres
as in the case oïSiphonaria. However in its general outline and in
its reaction toward the staining fluids it bears a great similarity to
the third quartet of ectomeres in Siphonaria, so that its recogni-
tion is a matter of considerable ease. The resemblance of these
two generations of ectomeres in the two species is intelligible
when we remember that they are both the first direct descend-
ants of the first generation of ectomeres and thus have the same
genealogical history. The fifth quartet of ectomeres occupies a
position alw^ays ventral to its mother-cell and lies almost exactly
on the ectomere-cells of the third generation. It also interposes
itself between the ectomere-cells of the second generation.

18 AKT. 1. T. FU.TITA : ON THE FORMATION

THE SIXTH QUARTET OF ECTOMERES.

Siphonaria (r/,,„ />,,,, rv„ ^Z,.,, Fig'^. 17-19, PI. T) :— Tlio
origin of this goiiorjitioii prosents miicli v;iri;itioii aeeording to
the species studied. In Siphonaria it again arises from the
second quartet of ectomeres. Here the cleaA^age takes place first
in the right jiosterior cell c. followed immediateh^ by the cell bo,
üo, etc,, thus taking the course of a left-handed spiral. Fig. 17
illustrates a horizontal section of this stage, in which all the ectomere-
cells of the second generation show their spindles. By the seg-
mentation of this stage, the ectomere-cells of the second quartet
are divided into two cells of unequal sizes. The newly formed
daughter-cell a2,2, etc. is smaller than its mother-cell ; but it comes
to occupy the position originall}^ occupied by the latter. As a
consequence, the mother-cell shifts its position ventrally toward
the rio'lit side, and at lensjth resis on the cells of the fourth and
the fifth generations. In this stage or sometimes somewhat later,
there takes place a noteworthy event, viz., the cleavage of the ento-
raesomere D (Fig. 18). The spindle lies in the low^er anterior ])art
of the cell so that it is easily seen in the horizontal and the
sagittal sections, Figs. 18 and 19. The cleavage is soon followed
l)y the division into two cells, the entomere-cell Di, and the
mesomere-cell M (Fig. 20). They present considerable ditferences
not merely in size, but also in general character. The entomere-
cell Di is of about the same size as the ectomere-cell of the third
quartet, and moreover like the latter has a comparatively large
nucleus. It is situated anteriorly, ventrally to the mesomere-cell
M, (Fig. 21) and maintains its direct contact with the right
antci'ior entomci'c />, with the cross furrow between as before

OF THE GERMINAL LAYERS IN GASTROPODA. 19

(Fig. 20). The inesomere-cell, uii the other hand, is large, and
transkicent, and now conies to occupy a position almost in the
median axis of the egg.

Aplysia {((■>.■>, b^., c.>.->, d.,;,, Figs. 03-55, PI. Ill) : — Yolk-laden
eggs of Mollusca present nuicli difference in the origin of the
present (piartet of ectomeres. In Unibrella it arises from the third
generation and in Neritina from the fifth. In the species of
Aplysia other than the one studied in the present investigation,
this generation is described as arising from the fourth quartet of
ectomeres. In the present species, however, it is the outcome of
the second generation just as in Siphonaria ; and the course of
cleavage is likewise quite identical in both cases. The spindle in
the ectomere-cells of the second quartet lies almost horizontal, so
that the daughter-cells «...j, etc. stand at the same level with
their mother-cells ; and when fully accomodated they push the
eetomere-cells of the fifth generation to the left, finally interpos-
ing themselves between these latter and the mother-cells. Fig.
53, the optical horizontal section from the animal pole shows
that the right posterior cell c-i of the second generation of ecto-
mere-cells has already propagated its new daughter-cell t'2.2 ;
while its left component f/^ still presents a spindle. From this the
course of cleavage and the relative position of this new series of
daughter-cells will be easily comprehended. In this figure it will
also be seen at a glance that the formation of the ectomere genera-
tions no longer follows with a strict regularity as in earlier stages,
and spindles are present simultaneously in cells other than the
second quartet of ectomeres. Such an irregularity in cleavage is
also seen in Fig. 54, the optical sagittal section near the median
line, the division of two consecutive sets of ectomeres here arising
almost at the same time.

20 AKT. 1. T. FUJITA : ON THE FORMATION

THE SEVENTH QUARTET OF ECTOMERES AND THE CHANGE
FROM THE RADIAL TO THE BILATERAL, SYMMETRY.

Siphonaria (a,.i.i, ^»,.1.1, Co.i.i, (/2.1.1, Fig. 20, PL I, and Fig. 21,
PI. II) : — Tlic seventh quartet of ectomeres is the lirst clescedaut of
the fifth (|uartet (^/,,.i, etc.). In Fig. 20, a horizontal section near
the vegetative pole, the first cell (L.,., of the new generation is
quite detached from the left posterior cell (/,-.i of the fifth
generation, which, however, is not seen in this figure being
situated more dorsally. Cleavage ]n-oceeds in a right-handed
spiral. The daughter-cell is quite similar in its general character
to the cells of the third, and the sixth generations. It lies ventral
to the mother-cell, and ultimately assumes a position alternate with
the cells of the fourth generation «;;, etc. (Fig. 24). From this
time on until the differentiation of the three germinal layers none
of the cells at the vegetative pole of the egg presents any marked
change. The vegetative pole is occupied hy the entomeres A, i/,
C, i>i, and the mesomere M surrounded l)y the ectomere-cells of
the fourth {a.;, etc.) and the seventh (c/j.i.ij titc.) generations (Figs.
20 and 24).

Up to the [)resent stage each of the successive generations
of ectomeres has invariably been formed of a quartet, the cells
of which are produced in a certain regular way, and keep their
relative positions in a spiral form. Such a disposition of cells is
generally known as the spiral symmetry. This form of symmetry
is, however, generally not retained after the formation of the
mesomere-cell M, and is eventually changed into a bilateral sym-
metry. This, so far as 1 am aware, is said by investigators to be
caused, in most of the Molkiscan eggs simply by the new posi-

OF THE GERMINAL LAYERS IN GASTKOrODA. 21

tioiis taken up l)y tlint quartet of ectomeres which are ])roduced
as tlie first descendant of the fourth generation. Subsequently,
all the cells come by degrees to shift their positions so as to
accommodate themselves to this new generation. Thus the egg
has ))een said to ultimately assume a bilateral synnnetry, and as
this takes place only very gradually, it is said that it is some-
times quite im])Ossible to point out with exactness the transitional
stage. In Siphonaria tliis phenomenon happens at, or a little
after, the stage in which the seventh generation of ectomere-cells
has l)een budded oft'. The alteration of the body-form is in this
species also induced by the same generation of ectomeres that
causes the change in the other species alluded to above {viz.
the first descendant of the fourth generation) ; but what I make
out respecting its formation and disposition is totally difterent
from what has hitherto been stated. The two poderior quadrants
of the fourtli cctomere-quartet (c^ and (?;•), ((djüiitiii(j laterally
the mesoine re-cell M begin to present the spindles [Figs. 20 ami 21),
tJie left component {d-.') taking the lead. Unlike all others Uicsc
spindles lie in a strictly radial and horizontal direction. Con-
sequently the daughter-cells, Ic and re when divided, come
each to occapy a position more tnedian than th(d of the mother-
cell, close to the sides of the mesomerc-cell JI and also they are
in exact bilateral symmetry with regard to each, other (Figs.
24 and 25). They are quite like the ectomere-cells of the
seventh quartet in size as well as in general appearance. The
two anterior (piadi'ants a.^ and b.; of this fourth generation,
which are so often tlescribed as dividing almost synchronously
with, or a little later than, their posterior component cells, never
present in this species any trace of a spindle until after the
complete differentiation of the germinal layers. On account of

22 AKT. 1. T. FUJITA : ON THE FORMATION

tliits liiglily specialized method of the formation and charaeteristic
disposition of the cells, such as has not yet been desciibed in
other species, I think it is better to denominate these as the
bilateral cells, thus distinguishing them from other ectomere-
cells. With the ])irth of the bilateral cells the spiral symmetry
is materially and abrui)tly disturbed ; and the configuration of the
egg is at length completely transformed into a bilateral synnnetry.
Thus it is seen that the change of the symmetry takes ])lace in
this species at a })eriod more definite than in any others tlius I'ar
known to us.

Aplysia (''^.1.1, *^2.i.i> ^-i.i.u d'l.i.u Fig. 54-55, PL III) : — In
Aplysia the seventh quartet arises from one diH'erent in order, but
of strictly the same genealogical derivation as the corresponding
one of the foregoing species. Here it arises from the third
generation, and indeed first from its left posterior-cell iL^^ thence
following the left-handed spiral. This state of things will be
fairly understood when we compare Figs. 54 and bïS, illustrating
the sagittal, and the horizontal, optical sections respectively. In
these two figures the ectomere-cell d-i.i alone is represented. It
is noteworthy that while the spindles of the cells heretofore
described as well as of the other quadrants of this same (juartet
{a-,,i^ etc.) lie almost horizontally the spindle in this cell is
seen to take a somewhat vertical position. Consequently, when
the new daughter-cell (/'o.i.i is budded off, it lies more ventrally
than the other (juadrants «2.1.1, />lm.i, etc. The daughter-cells thus
g(;nerated have a great öimilarity in general outline to those of
either the rilUi, or the sixth generations. They generally border
the ventral side of the egg, lying almost uutler the cells of the
second generation.

Concomittantly with this atage or a little earlier (Figs. 00

OF THi: CERiNriNAL LAYER« TN Ci AlS'l IM )PODA. 2o

and ")4), tluTC takes place the differentiation of the ento-mesomere
cell D. Tin's often liappens as early as the stage of the sixth
quartet, as is shown in Fig. 54. The details of this event are
qnite like those of Siphonaria, and therefore need no further
explanation. Even the entomere-cell />, itself thus formed, is
like that of the former species. It lies anteriorly and ventrally
to its mother-cell or mesomere jNI, and of conrse has the cross
fni'row in its original relation.

Besides this notahle phenomenon there always ensues still
another i-emarkable change in the process of cleavage, namely
the formation of the bilateral cells similar to those described above
in Slphonaria. The division of these new dnughter-cells is
quite similar to what takes place in the species just cited. The
two posterior ectomere-cells of the fourth generation c-^ and d,; ad-
joining the right and the left sides of the mesomere-cell jM now
begin to divide quite independently of their two anterior com-
panions. As is well shown in Fig. 53 an apical view of the
animnl pole, the spindles in these two cells lie almost horizontally
and are directed, unlike all the preceding ones, toward the
median axis, that is, in a somewhat radial direction. The spindle
usually appears first in the right cells c-. (Fig. 55), and produces
a small daughter-cell re (Fig. 50) which resembles an ectomere-
cell of the seventh generation, especially f/2.,.1 in its general
appearance. The division of the left cell c/o soon follows, produc-
ing the cell Ic. Although the two bilateral cells Ic and re are
propagated from the cells of the same generation they behave in
a slightly difterent manner. The left l)ilateral cell Ic at th(^
beginning of its detachment lies very close to, or even in contact
with, the entomere-cell D^ (Fig. 56). Subsequently, by the
cleavage of the mesomere-cell it is at last brought moi-e toward

24 ART. 1. T. FUJITA : ON THE FOKMATTOX

the posterior portion of the l)Ocly so as to lie ahnost on tlie same
level with its complementarv cell re on the opposite side.
By comparing Figs, ^"^ß and ö?, the optical liorizontal sections of
two different stages, we are aide to perceive this change in the
position of the left bilateral cell. Idms it seems that the final
position of the bilateral cells in the two species under considera-
tion coincides in almost every respect. I am not in position to
contradict the views expressed by previous authors, concerning
the transition from the radial to the Inlateral symmetry ; l)ut at
least in the species that T h.ave studied I am convinced that the
Itilateral cells are the main, if not the sole, cause of the trans-
formation of the l)ody symmetry. I should like to make a
reniîirk here on the same stage in A. limacina Blochmanx
makes no allusion to the phenomenon mentioned above. But a
close inspection of his figures shows us the occurrence of the same
events in that species. His Fig. 13 has an intimate relation to
the stage of the seventh ectomere generation, although it is
introduced to show an earlier sta2;e. There it is seen that c-, has
given rise to a small cell. This does not seem to form a quad-
rant of a quartet, lait corresponds in its general character to
the left bilateral cell. Again in his Figs. 14 and 15 the cell
r.y as well as the first entomere-cells are seen to occupy the same
position with those noted above by me. Although it is to
be gathered from the author's interpretation that the change of the
body symmetry occurs as early as the stage of about 24 cells,
yet his ligures incline me to believe that this phenomenon takes
place nuu'h latei-, just as in the case of the present species.

Innnediately after the formation of the bilateral cells or
sometimes simultaneously with it there takes place the cleavage
of the iiiesomere M. A spindle appears lying in the transverse

OF HIE GERMINAL LAYEKS IN GASTROPODA. 25

<liroclioii (Fig. ")()), and siibscqiiciitly divides the cell into two
nlinost equal halves. The dauglitev-eells (y>/, Figs. ,>7 and ;>S) ai-e
ovate in form and are characterized byt heir less granular contents.
Their position is in a strict bilateral symmetry with respect to
each other, and the plane in which they come in contact almost
coincides with the future median axis of the body.

Previous to this stao'e the nuclei of the entomere-cells A,
and 7> lie usually near the posterior end in the middle line
uf each cell as is shown in Fig. 5ö. They, however, have
commenced to shift their position toward the right (Fig. ''){]).
This goes so far in A that the nucleus reaches the ventral side
of the anterior end as will l)ecome intelligible l)y a reference
to Figs. ÖG and Ö8. It would seem that such a shifting of the
nucleus in the two entomeres could have no other purpose than
to prepare for the formation of the future entoderm-cell. This
latter event arises shortly after the bilateral symmetry becomes
well pronounced by the clejp'age of the mesomere-cell M. The
formation of the entoderm-cells has its begining in the posterior
cells ; and indeed in the right component cell C (Fig. ")?). In
the formation of this generation the sj^indle lies almost horizon-
tally, and the daughter-cells detached which are known as the
entoderm-cells (A^, B^, ({, Figs. o9 and GO) are rather small
and situated alternately with, and outside of their mother-cells
{A, II 0).

THE EIGHTH QUARTET.

Siphonaria (ry, ., />,.,>, ^,.,, ^/,.,, Figs. 22-24, PI. IT) :— With
the difterentiation of the entomere-cell D^ the behaviour of
individual cells becomes hardlv ascertainable in a surface view.

2(\ ART. 1. — T. FUJITA : ON THE FOE^FATION

Tlio formation of this generation was therefore made ont in
Siphonaria from sections. In this species the first ectomere
generation gives birth to the eighth quartet of ectomeres (Fig. 22),
the riglit posterior cell r, taking precedence over tlie others. The
spindles lie in rather a horizontal plane (Fig. 23), and the
course of the division is in a left-handed spindle. The newly
formed cells r^i 9, bi.o, etc., which are shown in the next Fig. 24,
are much larger than their mother-cells «1, b^, etc., with which
they lie on the same level and the same radii. After this cleavage
the first ectomere generation takes the shape of a rosette, the cross
fin-row l)eing almost obliterated. At this stage the normal and
typical form of the egg is rather oval in a horizontal view, nar-
rowing toward the posterior side (Fig. 27). In a side view, how-
ever, it takes a wedge shape with the thicker end turned also
toward the posterior side. About this time the ectomere-cells
lose their spherical contour, and losing their individuality become
more or less flattened, altogether presenting the appearance of a
layer (Fig. 20). Hence they will hereafter l)e called the ectoderm-
layer. The ectoderm is, moreover, disposed more or less in dif-
ferent zones. Fig. 24 illustrates the egg in this stage, which is
seen from the apical pole showing clearly the exact positions as
well as the mutual relations of the cells. The highest or dorsal
zone is occupied l>y the first {a^, h^, etc.), the third («u, /;, 1, etc.),
and the eighth (r/,,., /-»i.o, etc.) generations of ectoderm-cells, and the
ventral zone (Fig. 25) is formed of the fourth (o^o, b., etc.), and the
seventh (r^a.i.i, b.,_^.u etc.), generations of ectoderm-cells and the
entomere-cells {A, B, C, D^), together with the bilateral cells (re
and Ic) and the mesomere-cell M ; while the lateral zone is
])ordcred with the ectoderm-cells of the second (a.,, h.,, etc.), the
fifth (^2.1» Ku etc.), and the sixth (r^,„, /;.,.,, etc.), generations.

OF THE GERMINAL LAYERS IN GASTROPODA. 27

A simihir but Bliglitly advanced stage in Aplysia is also
shown in Fig. ö9. In this it is seen that unhke ^Siphonaria
the tii'st (ai, A,, etc.), the second {a.,, do, etc.), the fifth («ji,
(î»i.i, etc.), and the sixth (r/o..,, 00,2, etc.), generations of ectoderm-
cells cover the dorsal apex of the egg. Futhermore the
ectoderm- cells, as a whole, are aggregated in this species in
a more posterior portion of the egg than in Siphonaria. Again,
owing to the large size of the anterior entomeres the egg of
Aplysia is generally thicker in that portion than is the case in
the preceding species (see Fig. -54). Although not as marked
as in other species, the first Anlage of the trochoblast-cells can
be distinctly made out in the two species under consideration.
As will be seen clearly in Figs. 24 and 09 the apical quartet of
the first ectomeres is encircled by a girdle of eight cells, of
which four («1.1, Aj.i, etc.), i.e. the first descendants of the first
ectomere-quartet form the trochoblast. While the apical quartet
gives rise to the " arms of the cross " its further development does
not take place in these two species until after three germinal layers
are firmly established.

In Siphoîiaria, after such a disposition of cells has been
completed, the cleavage of the mesomere M and of the entomeres
^4, B, and G occurs synchronously (Fig. 25). Usually, however,
the mesomere- cell M precedes the other three, as shown in the
horizontal section (Fig. 25). The spindle in this cell is directed
almost exactly horizontally (Fig. 26) and the cleavage plane
thus formed conies to correspond with the median axis of the
egg. The resultant cells ^iii) are in their nature (piite identical with
those of Aplysia to which allusion has already made (p. 25). It
suffices here only to call attention to figures mentioned above,
and to Fig. 27, which shows the horizontal section through the

28 ART. 1. ï. FUJIÏA : ON THE FORMATION

middle portion of the egg. This latter is intended to illus-
trate the mesomere after its complete division into two daughter-
cells and after their full accommodation. The three entomere-cells
A, B, and G now commence to divide as in Fig. 25. In this case
their nuclei do not show any shifting, such as was seen in Aplysla.
The entomere-cells are divided into two lialves almost equal in size ;
and all are arranged on the same level. From this time onward
the entomere-cells dispose themselves in a layer and may now l)e
called the entoderm -layer.

The further developmental course of the germ-layers has
been studied with some degree of exactness only in Siphonaria.
The ectoderm-cells gradually begin to divide indiscriminately ; and
hence it is wholly impossible to trace them in surface views beyond
the tenth generation. The ninth quartet (^<2.2.b '^2.'2.i5 etc.), arises
from the sixth generation of ectoderm-cells (^i;^ h,2, etc.), and
the tenth («2.1.1.1, ^^2a.\a, etc.), from the seventh («2.1.1, '^2. 1.1, etc.), (Fig.
30). In what quadrant they first appear, and how the courses of
cleavage run are quite uncertain. The new daughter-cells are
always as small as their mother-cells.

The entoderm-cells iro through their second cleavage, at
the same time with the second division of the mesoderm-cells. This
process begins with the cell A^^ and proceeds in a right-handed
spiral, as seen in Fig. ^34, a horizontal section of this stage, in
which A^ and i)i have given off respectridy Ai,^, and i)i.i, and
C\ shows a spindle. The small daughter-cells thus produced
exactly alternate in position with their original mother-cells.
The subsequent growth of the entoderm-cells seems very slow, and
I have not traced it Iteyond this stage. I will only add here that
the first entoderm-cells retain their original relative positions' for
a long time as indicated l)y the persistence of the cross furrow.

OF THE GERMINAL LAYERS IN GASTRORODA. 29

Til fact it remains iincliaiiged until tlio cells enter into the i)er-
maneiit portion of the alimentary canal. Hence it in of great
use in determining the orientation of the l)ody.

As to the mesoderm, its formation may he followed u]) to a cer-
tain stage with some exactness. ^Viter the eighth ectomere stage each
of the two daughter mesomere-cells m come to present the spindle.
It is directed obliquely forward and upward (Fig. 28), and appears
first most frequently in the left component (Fig. 29). The daughter-
cell or the first mesoderm-cell m^ is very small ; but its presence
is easily recognizable owing to its large nucleus (Fig. 30). The
second mesoderm-cell originates from the mesomere-cell as did the
first. The fact is shown in the horizontal section Fig. 32, and
the sagittal section Fig. 33. This, however, happens much later
ill time when there has already been much increase in the
numljer of the ectoderm-cells. The new mesoderm-cell in~z is
much like the first one but it takes a position more ventral than
the latter. The exact seat of these cells is clearly indicated in
Fig. 36, a sagittal section of the egg. The third mesoderm-cell
nh i« now derived from in^ beginning with the right component.
It is the most dorsally situated of all as is shown in Fig. 35, a
transverse section through the posterior side of the egg. The
first mesoderm-cell soon again subdivides, thus giving rise to the
formation of the fourth one ni^. Fig. 36 just mentioned above
also shows this fact, the S])indle in the first mesoderm-cell ii^
being in the amphiaster stage. The mesoderm-cells, henceforth,
seem to segment rather rapidly with no apparent regular order.
At first they all aggregate as a mass in the posterior median
line of the 1)ody. Gradually, how^ever, they l)egin to spread toward
the lateral parts, where they soon arrange themselves in distinct
layers. In some other species of Mollusca the so called " larval

30 AKT. 1. — T. FIJJJTA : ON THE FORMATION

or Hecoudarv mesoblaHt " is often ck'sciibed tis boiiii;; foi'incd iVoiii
one or otlier of the ectoderm-cells, and it is sm)posed to contril)ute
toward the foruiation of the niesencliynie. In the present species
I was not able to lind any trace of it so far as the investigation
extended.

Several years have elapsed since I carried ont the alnjve
investigation, and during that period, tlie investigations by
WiEKZEjsKr, Meisenheimer, Heath, Holmes, Cauazzi, Geor-
GEWiTCH etc. on similar subjects have been published. Generally
s})eaking, they have tended to confirm and verify the pro2)Osition«
that had already been made known by previous authors. I con-
sider it advisable to refer to some of these works which bear
directly on what I have described above.

In January 1900 there appeared a paper on the development
of Aplijsla liiiiacina by Carazzi, and again eight months after
another paper by George witch on A. dep'dans.

Carazzi has carefully traced out all the developmental
processes up to the formation of the mesoderm-layer noting the
time exactly. When we compare his results with those given
above, it will be seen that they agree in the main. Nevertheless,
there are discrepancies on some important points which cannot
l)e ])assed without a word. The difterentiation of the germ-layers
in general is reached a little earlier in our species than in that
studied by the Italian author, excepting the formation of the
(iitomei'e />,, whicli is far earlier in his sj^ecies. In his Fig.
2 we see the illustration of an egg from its apical pole,
whicli coincides with Fig. 53 of the present pa[)er. But one of

OF THE GERMINAL LAYERS TN rjARTROPODA. 31

tlie (Iniwing^ in his Fig. o (illustrating the vegetative pole of the
egg) which is like my Fig. -57, is interpreted in a different \\<\y.
The hilateral cell, to which I have called attention, is in our species
pi'odnced at nhout this time from the posterior half r.. and d,. of
the fourth generation of ectomere-cells. Such a cleavage really
occur also in A. limacma ; but it is said to be accompanied or-
dinarily with the simultaneous division in its anterior components.
Hence, in that species the new generation is formed of a quartet
like all the preceding ectomeres and does not influence in any
way the transformation of the cell-arrangement. Nevertheless
the bilateral form of cleavage occurs also in the ectomere-cells
o6'^ and of?\ as may be gathered from the author's own words as
"subito dopo, alia 28" ora oc^ e 3(/^ si dividond con fusi transversali,
cioè con divisione bilaterale, mentre oa} e ?)b^ rimango in riposo."
However, this is the second division of those ectomeres, and indeed
after the l)ilateral symmetry is fairly established by the formation
of the mesomere-cell. Thus the transformation of the body sym-
metry here seems to be effected l^y the gradual and renewed dis-
2:)Osition of already formed ectomeres rather than by a single
cleavage of the posterio]* half of the fourth ectomei-es. Such a
method of transformation is seen in various species as T have
already noted.

Concerning Georgewitcii's paper on A. depUans the author,
it appears to me, has fallen into some confusion on important
points. At all events his ideas on the orientation of the egg are
exastly contrary to those usually held, tlie portion called by him
the posterior, being really the anterior, so that his " Ur-
mesodermzell " is in fact derived fi'om the anterior blastomere.
And it is very strange that he describes the posterior (really
anterior) half of the second quartet of ectomeres as the " Ur-

32 T. ART. 1. FUJTTA : OX THE FORMATION

mesomere." By my investigation given above as well as by those
of others it is known that this generation is the most active in
its growth, at least before the establishment of a bilateral sym-
metry. They ought to have given rise to daughter-cells before
the true mesoderm "Kleine Zelle???," was first perceived by the
author. By a careful study, however, T have at last found that
his description and figures do not harmonize. In his description
not a' word is said as to the fate of his anterior blastomeres A
and B i. e. really the posterior C and D of authoi'S. But we
can ascertain from his Fig. 20 that these cells segment almost
:it the same time. In sliort, according to his paper we nnist
finally arrive at the very embarrassing conclusion that he has put
three totally different kinds of cells under one and the same name
of "the mesoderm" viz., (1) the descendants of the second ectomere-
cells or according to the author 2e and Id (Fig. 10), and (2)
the descendants of the entomere C or according to the autho]" rt>,
in addition to (3) the descendants of the proper mesomere, which
seems to have been entirely overlooked by the author. Yet these
c(dls, as they represent the ectomere-, the entomere-, nnd the meso-
mei-e-cells ought not of course to be confounded. This and some
other conclusions of the author, which I can not help considering as
too hasty are derived from his assumption of the homology
existing in the formation of the original mesomere-cells in the
species in question and in Polyclads. The fact is clearly seen
in his own tei-m — "Der Ursprung und die Lage dieser 2 Zelle
m und Wx stimmt so aufiallend mit denjenigen bei Polycladen

überein, "

In the same year a paper entitled " The Early Development
of Planorl)is " was made public by Holmes. By a careful study
of this author new liglit has been thrown on many points,

OF THE GERMINAL LAYERS IN GASTEOPADA. Öi>

especially concerning the fate of the cells. Kevertheless the
alteration of the body-form in this species seems to be not so
pronounced as in those I have studied. Holmes seems to think
that the ultimate cause of such a transformation lies in the dif-
ferent behaviour of the posterior cells of the third ectomere
generation. Thus he goes on to say:- "we may view the earlier
division of the cell oa and Sb as the first foreshadowing of
bilateral cleavage." • By this expression it is quite evident that
there is no special cell formation as in the cases described in the
present paper. Such a gradual modification as is elucidated by
the author is of a wide occurrence in the Molluscan egg and the
change of symmetry by the cell formation near the posterior end
of the body is demonstrated at present only in the species studied
by myself. Nevertheless I believe that the same fact will here-
after be confirmed in other forms.

It may not be useless to summarize here the main points
brought out in the present investigation.

1. Throughout the whole process of cleavage it is observa-
ble, that there is no fixed regularity in the course such as is
expressed in the so-called law of alternation of spirals as stated
by Wilson, Kofoid et al. Even the corresponding daughter- cells
from the same blastomeres or ectomere-cells are propagated dif-
ferently in different forms. The first generation of ectomere-cells
is produced in S'qjhonaria and in Aplysia in spirals of opposite
directions and even in the same species the cleavage sometimes
takes place consecutively in the same direction, and not alternately
to rio'ht and to left as has been observed in other forms.

:34 ART. 1. — T. FUJITA ! ON THE FORMATION

^Nevertheless, it should be borne in mind, that cleavage commences
invariably from one or the other of the posterior blastomeres or
its descendants.

2. After the second cleavage the opposite quadrants of the
blastomeres usually come in jaxtaposition forming the cross furrow
between. Kofoid has made the statement that the quadrants
forming the furrow at the two opposite poles differ with species
according to the amount of the deutoplasm in the egg. This
does not hold good at least in the present cases. Siphonaria and
Aplysia perhaps represent two extremes with regard to the
quantity of the deutoplasm. Nevertheless, as has been seen, the
cross furrow in these two species is formed at both poles by the
same quadrants.

3. During the cleavage of the egg, the daughter-cells are
disposed in a spiral form, that is, the individual cells shift
their respective positions either toward the right or the left of
their original positions. Such a spiral arrangement or symmetry
is retained for some time. It is then abruptly transformed into
a bilateral symmetry. This important phenomenon, so tar as I
am aware, has been interpreted as due to the rearrangement of
the component cells. It is true that in an advanced stage the
disposition of the cells becomes altered ; but there is another im-
portant factor which necessitates such a transformation of the
body symmetry. The factor is the existence of certain new cells
propagated from the fourth generation of ectomere-cells. Of the
«iuartet which forms this generation, the two posterior cells give
rise in a peculiar way to the daughter cells which I have called
the bilateral cells. These play an important role in bringing
about the change of the -body symmetry, which seems to take
place in a comparatively short space of time. The period of

OF THE GERMINAL LAYERS IN GASïOROPODA. OO

his change is, as Wilson remarks, after the entoiiiere-cell
differentiates itself from the left posterior ento-mesomere-cell, and
takes its seat along the median axis. In other words, the bilateral
symmetry appears just after the cells of the three germ-layers
are distinguished. This corresponds with the stage when the sixth
or seventh quartet of ectomeres is formed and the egg is com-
posed of 29-33 cells as is shown in the tabulated form below.
The exact period of the occurrence could not be expressed as it
varies with the species, and even in the same species with the
environment.

4. The cleavage of the mesomere-cell takes place after the
bilateral symmetry is completely established. It is then subdivided
into two equal halves, which stand in a strict bilateral symmetry
with regard to the median axis. Shortly afterwards from each of
these cells mesoderm- cells are propagated, which by the further
division eventually form a layer. Hence, it is clear in these
species that the mesoderm-cells take their first origin from the
posterior quadrants of the blastomere as in many other species.
The differentiation of the three germ-layers strictly speaking
dates from the formation of the mesoderm-cells. It is at about
the stage of the eighth generation of ectomere-cells in Siphoîiaria
and of about the tenth generation in Ajjiysia. It is therefore
the 42-cells stage in the former, while it is the 50-cells stage in
the latter. Thus, the differentiation of the mesoderm-cells varies
in time not only in the species named above, but also in all
those forms before thoroughly investigated. The results of the
present investigation as well as of those of other authors are
compiled below^ in the form of tables to facilitate a comparative
study of the Molluscan development : —

36

ART. 1. T. FUJITA : ON THE FORMATION

Tables showing the cell-lineage in Molluscan Egg.

b.c bilateral-cell ni mesomere-oell segmented

d entomere-cell m,, ni,, etc raesuderm-cells

ent entomere-cells excepted n generation <jf ectonieie-cell-

M mesome re-cell

I Siphon A RIA

7 hours

Order ot" (Jiiartet

I II III IV V VI VII VIII I XI X

9-10

II Aplysia

Order of Quartet

X

1

I II

III

IV 1 V VI VII

VIII

IX

Egg

Elastomere

1

!/'

/

^h^

U 1^1

^

~^~-— ^

~-^_

L__ T'~^2.2

^^

^^

b::::^-

-(ent.)

~im,)

Ko. of Cells

4

s 12

16

20

24 i ;îi

;i9

4;;

47

51

OF THE GERMINAL LAYEBS IX GASTOROPODA.

37

III Neritina (Bloclimaim)

Order of Quartet

]

11

HI

IV

\'

VI

VII

VIII

IX

X

Egg

Elastomere

r

/

~H 2.1.1 -

/"--

-

"'•'■\

/ y

"s--

^--

""""2.1.2

^^^

^«2. 1.1.1

"^^

— —

-------

____

---

"""•" 3. 1

-^1.2

---

""~'~

----_.

J (MX
1(d) N

\

^^

^--^

-(ent.)

\m)

Ni. of Cells

4

S

12

Ki

L'l)

•_'4

2S

;;■)

;î6

44

30

IV U31BRELLA (Heynions)

Order of Quartet

I

II

111

IV

V

VI

VII

VIII

XI

X

Egg

Elastomere

f-

- — — __

1.1

/

___

/">--.

^^_

'l;.i

— "j--.

^^

^^^

~~~^ll.2

^^^^^

-.

--n,.,

~~^"2.2

^;::;^

-<m).^

^-^

^'lll)

-(ent.)

Xo. of CelK

4

8

12

1()

2(1

24

29

.j.'>

40

47

Time

."ÎO hours

4S

0(1

72

72

70

DC)

96

120

12(»

ART. 1. — T. FUJITA.

Y LiMAX (Meisenheimer)

Order of Quartet

I

II

III

lY

^'

VI

VII

VIII

IX

X

Egg

Elastomere

/ '"""

/

n

■

—

— 11,.:

y —

--___

-•'^

/x

2^

--

~-Jl,,,_,

--^

■-----,

~— -

~--- —

^^^^^^

^-«3.1

~ ^

--n^..

^^^

-^^

"~

i.(M)

--^

^eiit.) ~-(niO

No. of Cells

4

s

12

IG

20

24

28

;î2

.".6

45

51

These tables show, that in most cases the mesoderm forma-
tion arises in the stage of the tenth generation of ectomere-
cells and at about the period when the cells number 50 or
more without any regard to the duration of time consumed in
their development. Further we learn that the genealogy of the
ectomere-cells up to the stage of the sixth generation follows
almost the same course, although the order of development some-
times varies with to species. And at the stage named, the diöer-
entiation of the germ-layers is first attained in some species. By
referring to the table it is furthermore evident that the amount
of deutoplasm present in the egg seems not to have any causal
effect on the rate of differentiation. In this respect, Ker'itina, an
egg rich in deutoplasm, forms the germ-layers in the same stage as
Li max, and S'q^honarla, which are scantily supplied with food-yolk.

Before closing I wish to express here my deepest obligations
to Professors K. Mitsukuri and S. Watnsé for their kind super-
vision of the present work.

BIBLIOGRAPHY. 39

Bibliography.

1. Blochmann, F .Ueber die Entwicklung der Neritina fluviati?.

Zeitschr. f. w. Zool. Bd. XXXVI. 1882.
"2. „ Beiträge znni Kenntnis der Entwicklung der

Gastropoden. Zeitschr. f. w. Zool. Bd.

XXXVIII. 1883.

3. BoBRETZKY, N Studien über die embryonale Entwicklung der

Gastropoden. Arch. f. mikr. Anatomie Bd.

XIII. 1877.

4. BÜTSCHLi, 0 Entwicklungsgeschichtliche Beiträge. Zeitschr.

f. w. Zool. Bd. XXIX. 1877.

5. Carazzü, D L'embriologie dell' Aplysia limacina. Anat.

Anz. Bd. XVII. 1900.

-6. CoNKLi^^, E. G The Cleavage of the Ovum in Crepidula for-

nicata. Zool. Anz. 1892.

7. „ The Embryology of Crepidula. Journ. Mor-

phol. Vol. XIII. 1897.

8. Crampton, H. E Reversal of Cleavage in a Sinistral Gasteropod,

Annals N. Y. Acad. Sc. Vol. VIII. 1894.

9. „ Experimental Studies on Gasteropod Develop-

ment. Arch. f. Entvvicklungsmechanik d. Or-
ganismen. Bd. III. 1896.

10. „ Observations upon Fertilization in Gasteropod.

Zool. Anz. 1897.

11. Erlanger, R. v Zur Entwicklung von Paludina vivipara. Mor-

phol. Jahrb. Bd. XVII. 1891.

12. „ Beiträge zur Entwicklungsgeschichte der Gastro-

poden, I. Bithynia tentaculata. Mittheil, aus
d. Zool. Stat. zu Neapel Bd. X. 1892.

13. „ Mittheilungen über Bau und Entwicklungeiniger

marinen Prosobranchien. Zool. Anz. 1892.

14. ,, Zur Bildung des Mesoderms bei der Paludina

vivipara. Morphol. Jahrb. Bd. XXII. 1894.

15. Fol, H Sur le développement des Gastéropodes Pul-

monés. Arch, de Zool. exp. et génér. Tome
VIII. 1880.

40 BIBLIOGRAPHY.

IG. FujiTA, T Preliminary Note on the Mesoderm Formation

of Pulmonata. Zool. Mag. Vol. YII. 1895..

17. „ Notes of 8ome Experiments on Mollnscan Eggs.

Zool. Mag. Vol. VIII. 1896.

18. Georg EWiTCH, P. M..,.Zur Entwicklungsgeschichte von Aplysia depilans..

Anat. Anz. Bd. XVII. 1900.

19. GoETTE, A Bemerkungen über die Erabryonalentwicklung

der Anadon ta piscinalis. Zeitschr. f. w. Zool.
Bd. LII. 1891.

20. Haddon, a. C Notes on the development of Mollusca. Quart.

Journ. of Micr. Sc. Vol. XXII. 1882.

21. Hatscher, B Ueber die Entwicklungsgeschichte von Teredo.

Arb. aus d. Zool. Inst, zu Wien. Bd. III. 1880.

22. Heath, H The Development of Ischnochiton. Zool..

Jahrb. Bd. XII. 1899.

23. Heymons, E Zur Entwicklungsgeschichte von Umbrella me-

diterranea. Zeitschr. f.w. Zool. Bd. LVI. 1893..

24. Holmes, S. J The Early Development of Planorbis. Journ.

of Morph. Vol. XVI. 1900.

25. Horst, Pi On the Development of the European Oyster.

Quart. Journ. of Micr. Sc. Vol. XXII. 1882.

26. KiNPOWiTSCH, N Zur Entwicklungsgeschichte von Clione limacina.

Biolog. Centralbl. Bd. XI. 1891.

27. KoroiD, C. A On Some Laws of Cleavage in Limax. Proc.

of the Amer. Acad, of Arts and Sc. 1894.

28. „ On the Early Development of Limax. Bull, of

the Mus. of Comp. Zool. at Harv. Coll. Vol.,
XXVII. 1895.

29. KowALEVSKY, A Etude sur l'embryogénie du Dentale. Annal

du JMus. d'hist. Nat. de Marseille. Tome I. 1883.

30. „ Embryogénie du Chiton i)olii. Annal du Mus.

d'hist. nat. de Marseille Tome I. 1893.

31. LiLLiE, F. Pi The Embryology of the Unionidae. Juurn.

of Mor])!.. Vol. X. 1895.

32. Mark, E. L Maturation, Fecundation and Segmentation of

Limax cami)estris. Bull, of the Mus. of

Comp. Zool. at Harv. Coll. Vol. VI. 1881.

33. Meisenheiner, J Entwicklungsgeschichte von Limax maximiis.

Zeitschr. f. w. Zool. Bd. LXII. 1896.

42 EXPLANATION OF FIGUlîRS.

Explanation of Figures.

AD the figures are drawn with a camera hioida. and are. unless otlier-
wise designated, magnified with Zeiss I) x 4.

Reference letters used.

a. anterior ])lastomerp. M mesomere-cell.

<". egg. p. posterior hlastomere.

ÏC. left l)ilat(n'aj cell. ph. polar body.

in. mesomere-cell segmented. re. right l)ilateral cell.
The arrow under the ectomere-cell indicates its course of cleavage and
tJK' Roman numeral shows the order of the quartet to be formed.
Figs. 1~?)6 illustrate the development of Siphonnrla.
Figs. 37-00 illustrate tliat of Aplysia.

BIBLTOGRAPHY. 41

;Î4. Meisenheimer. J Ent\vicldnno;sp;osohichte von DreisPonHia p<>ly-

morplia Pall. Zeitpchr. f. w. Zo(^l. Pxl. TjXIX.
1901.

?,~). ]\[etcai,f, m Contributions to the Embryology of Chiton.

Stud, from the biolog. Labor, of the «lolin Mop-
kins Univ. Vol. V. 1893.

?>C). Pattens, W The EmV)yology of Patella. Arb. ans dem

Zool. Inst, der Univ. Wien. Bd. VI. 1S8G.

o7. Sakasin, P Entwicklungsgeschichte der Bithynia tentaculata.

Arb. aus dem zool. zootom. Inst, zu Würzburg
Bd. VI. 1883.

38. S<'iiMH>T, F Beiträge zur Kenntniss der Entwicklungsgeschich-

te der tStylommatophoren. Zool. Jahrb. Bd.
VIII. 189.5.

39. Stai'FFACHer, H Eibildung und Furchung bei Cyclas cornea.

Jen. Zeitschr. f. Naturw. Bd. XXVIII. 1893.

40. TöNNiGES, C Die Bildung des Mesoderm bei Paludina vivi-

para. Zeitschr. f. w. Zool. Bd. LXI. 189G.

41. Wterzejski, a Ueber die P^ntwicklung des Mesoderm bei Physa

fontinalis. BioL Centralbl. Bd. XVII. 1897.

42. Wilson, E. B The Cell-Lineage of Nereis. Journ. of Morph.

Vol. VI. 1892.

43. ZiE(ir,ER, E Entwicklung von Cyclas cornea. Zeitschr. f.

w. Zool. Bd. XLI. 188.>.

"TS'sfe'sr^-

T. I- 1. 1 ITA.

ON THE FORMATION OF THE GERMINAL LAYERS IN GASTROPODA.

PLATE 1.

Plate I.

Fig. J. Egg mass of Slphoitarla tleposiletl uu ruck. Natural size. Arrow
indicates the orientation of the animal during egg-de})osition.

Fig. 2. Three eggs magnified to show their manner of connection. ]<"'resh.
Ax 2.

Fig. 3. Yolk splierules with vacuoles before disintegration. Fresh. E x 2.

Fig. 4. Egg after total disintegration of yolk spherules.

Fig. 5. First cleavage stage, two blastomeres being connected only by a
narrow bridge. Fresh.

Fig. 6. {Second cleavage stage. Its commencement, from above. Fresh.

Fig. 7. Same stage, further advanced.

Fig. 8. Completion of the same stage. J31astomeres are now designated as
A, the left anterior, B, tlie right anterior, C, the right [losterior,
and ]), the left posterior.

Fig. 9. Cleavage of the first quartet of ectomere-cells. From ai)ical [)ole.

Fig. 10. »Same stage. From posterior side.

Fig. 11. Completion of the same stage, to show the position of cells. From
a[)ical pole.

Fig. 12. Second ectomcre-quartet stage, showing its formation. A[)ical ^'i('w.

Fig. 13. Same stage. Posterior view.

Fig. 14. Commencement of third ectomere-quartet. Ajncal view.

Fig. 15. Cleavage of fourth and fifth ectomere-quartets. Horizontal section.

Fig. 16. Fifth ectomere-quartet stage with formation of entomere-cell Dj.
Sagittal section.

Fig,' 17. Sixth ectomere-(|ua.rtet stage. Horizontal section.

Fig. 18. Formation of entomere-cell Dj. Horizontal section through ve-
getative pole.

Fig. 10, Same stage. Sagittal section.

Fig. 2U. Cleavage of seventh ectomere-cpiartef. Jbjri/.ontal section through
vegetative pole.

Jour. Sei. Coll. Vol. XX., Art. /.. PI. I.

Ulk é Imp Th* Toki« Printinn Co

I. IUI I A.

THE FORMATION OF THE GERMINAL LAYERS IN OASTROPODA.

PLATE 11.

Plate II.

Fig. 21. Furmatioii of bilateral cells. Transverse section of a stii^e sliglitl}'

more advanced than in Fig. 20.
Fig. 22. Formation of eighth ectomere-qnartet. Horizontal ^-ection thnnigh

a[)ical pole.
Fig. 23. »Same stage. Transverse section.
Fi<i'. 24. Stajie before cleavage of mesomere-cell ; a)nstrncted from .M'clions

so as to show the mutual relative position of cells. Cells furming

the ventral side of the egg are colored blue.
Fig. 25. Cleavage of mesomere-. and entoraere-cells. Horizontal section

through vegetative pole.
Fig. 2G. Cleavage of mesomere-cell M. Transverse section.
Fig. 27. Completion of same stage, showing the relative [)osition of ecto-

mere-quartets. Horizontal section.
Fig. 28. Formation of tirst mesoderm-cells. Sagittal section.
Fig. 29. Same stage. Transverse section.
Fig. 30. Stage after formation of tenth ectomere-qnartet. Horizontal section

near vegetative pole.
Fig. 31. Horizontal section through vegetative pole, showing complete en-
closure of mesomere-cells l^y epibolic growth of entoderm-cells.
Fig. 32. Cleavage of second mesoderm-cells. Horizontal section.
Fig. 33. Same stage. Sagittal section.

Fig. 34. Formation of second entoderm-cells. Horizontal section.
Fig. 35. Cleavage of third mesoderm-cells. Transverse section of the [»osterior

side.
Fig. 36. Formation of fourth mesoderm-cells. Sagittal section.
J''ig. 37. Eggs of Aplysia. Fresh.
Fig. 38. Same, highly magniüed. Fresh.

Fig. 39. Conuuencement of tirst cleavage stage. Side view. Fresh.
Fig. 40. Formation of blastomere -cells. Side view.

Fig. 41. A slightly more advanced stage of Fig. 40. A}»ical vi(;w. Fresh.
Fig. 42. Com[)letion of same stage, showing the pro[)er positions of blasto-

meres. Apical view. Fresh.

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

iJlk d- Imp The Tokio Pnntim Co.

I. Il JIIA.

ON THE FORMATION OF THE GERMINAL LAYERS IN GASTRfiPODA.

PLATE III.

L'

Plate III.

'^. 43. Formation of first ectoniere-qnartot. Apical view.

g. 44. Samo stage, Eight side view.

g. 45. Formation of second ectomere-qnartet. A[)ica] view.

g. 4G. Same stage. Posterior view.

g. 47. Same stage. Left side view.

g. 4S. Simultaneous formation of third and i'oiirth eetoinere-((nart<'ts sfagxv^.

Apical view,
g. 49. Same stage. Posterior vie\v.
g. 50. Same stage. Optical sagittal section,
g. .51. Formation of rifth ectomere-qnartet stage. A[»ical view,
g. 52. Same stage. ( )ptical transverse section,
g. 53. Cleavage oi sixth ectomere-qnartet, and formation of bilateral cells.

Optical horizontal section.
g. 54. Same stage, with formation of seventh ectomere-qnartet, and

entomere-cell D,. Sagittal section.
g. ~)5. Same stage, with formation of hilateral cells. Horizontal section.
g. .~)G. Cleavage of mesomere-cell. ( )ptical horizontal section,
g. 57. Formation of first entoderm-cells. 0|itical horizontal section,
g. 58. Completion of same stage. Optical transverse sectitm.

59. Stage liefore formation of mesoderm-cell »?, showing the arrangement

of cells. Cells on ventral side are colored l)lne.
(iO. Formation of first mesod(M'm-cell stage. 0[itical horizontal section.

Jour. Sel. Coll. Vol. XX., Art. I. PI. III.

/m

''1 ho

h'i'i. ih

'2.1 ^ r.

Fi g. 51

1 ":' f^ c,.

J-'io. 5U

L^ r

rf3 -'j;li^/^

rf.,

Fut. 'A

Wim

^i 6i, ^1.1

Fi«. 53

ri "i'^ ''» '"' ''i.ï

I i , ' '■5^.

/^,'

Fij? 55

2 ''ïi^^li'!^-' ^<^ *^

t"fi 5h

/ï^. 57

t«i. 59

lÀlh & Imp The Toki« Printing Co.

„i 1'

Fig- 58

m, "2.1.1 rr '

Fig. 6('

JOURNAL OF THE COLLEGE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 2.

Cretaceous Cephalopoda from the Hokkaido.

PART II.i)

Tum'lffes, Jlelirorevas, JfcferoceraSf JSipjfonifes,

Olcosteph runts, Dtsiaocrras, Ilnitericpi'fts,

and an atidetei'iinned (Jenas.

By

Hisakatsu Yabe, RigakusM.
University Hall, Imperial Univ. of Tokyo.

IFith 6 plates

TURRILITES Lam.

D'Orbigny l)ronglit nearly all of the tiirreted forms of
Ammonites under one of tlie three genera, Turrilites, Helicoceras,
and Heteroccras, distinguishing at the same time two groups in
the first genus, namely Turrilites roiundati and T. angulati.

1) Since the publication of the first part of this memoir, a number of valuable papers
relating to foreign Cretaceous Cephalopod fauna have appeared both in Europe and America.
First of all, mention must be made of the works of Dr. F. M. Anderson (Cretaceous De-
posits of the Pacific Coast, 1902) and of Dr. J. F. Whiikaves (On some Additional Fossils
from the Vancouver Cretaceous, 1903), to botli of whom we are indebted for mucli accurate
and important knowledge concerning the Cretaceous deposits of the region along the Pacific
coast of the United States and Canada. A comparison of the results of these papers with
those reached by Professors Yokoyama and Jimbo and the present writer, will at once
bring out some remarkable faunistic resemblances between the deposits of California and

2 ART. 2. H. YABE : CRETACEOUS

Soon after D'Orbigny, Quenstedt^- considered the separation
of Helicoceras from Tunnlites quite unnecessary, while Sharpe^'
and Stoliczka'^^ were of opinion that the so-called group of T.
7'otundaü should be brought under the genus Helicoceras, retain-
ing the name Tiirrilites only for the T. cmc/ulaii of D'Orbigky.
Pjctet/^ however, proposed to divide Helicoceras into two groups,
the one having a short spire and widely separated volutions and
the other showing a high s^iire and less Avidely separated volutions.
The first of these groups he called Helicoceras proper, while the
second as well as the group, T. roiundati, were considered by him
to be distinct genera.

Meek'*^ like D'Orbigny took Helicoceras in the most restricted
sense and divided it into Helicoceras proper, Patoceras, and

Canada on the one side, and those of tlie Hokkaido on the other. A discussion of these,
however, the writer intends to put forth in the last part of this work.

As to the new fossil localities in the Hokkaido, no important addition to our know-
ledge has been made during recent years ; but several new forms have been brought back
by geologists who have lately visited the localities already known. Among these, the writer
Avould mention a valuable collection made by Prof. Yokoyajnia at Urakawa, a locality
that had not been visited by geologists for more than twenty years.

During the last four years, the writer has devoted mucli time to the study of- the
morphology of the Hokkaido Ammonites, and has also taken the opportunity of making
himself acquainted with the development of several of tiieir predominating types, for it has
now become clear that a mere morphological study of adult specimens is not sufficient for
the establishment of their real affinities, and that by the study of their development only
can such knowledge attained. Consequently, the result obtained from this side of the be
study is also occasionally embodied in the following descriptions.

In this second part, the following genera are described : —

1) Turrilites, Helicoceran, Uderoccras and Nip-ponitci<.

2) Olcostcplianus.

'.'>) Desnto<;er((s and Jlauericcms.

There is also one species whose generic position is still undetermined.

The order in which tlie different groups are here treated is not due to the jiresence of
any mutual relations existing among them but merely for the sake of convenience and of
avoiding too long a delay in the ])ub]ication of the paper.

1) Quenstedt: Petrefaktenkimde Deutschlands. I. Ceplialopoden. P. 297.

2) Sharpe: Mollusca of tlie Ciuilk. P. 59.

3) Stouczka : I.e. p. 18;{.

4) Pictkt: Paléont. Suiss. Foss. de Ste. Croix. P. ]20.

0) Meek: Invert, ('ret. and Tort. Foss. of the Fjipcr Missouri Comity, p. -IS-").

ci: PII A LU rO DA FROM TRE JIOKKAIDO. ö

/Spirocenfs. Stoliczka" said tluit " tlic poyitioii uf the sipliuncle
iuid the symmetrical or iin symmetrical division of the sutures
ought to be regarded as a much better distinguishing character
than the kind of volutions," while Meek'"^ argued, "it seems not
justified to unite typical Ilelicoceras characterized by very slender,
widely separated volutions, forming so broad a curve as to give
origin to a verv wide umbilical cavity and Turrilites roiuiidatl,
generally provided with much rounded, contiguous whorls."

ZriTEL,"' in his Handbuch, accepted four subgenera of Turri-
Utes, namely Turrilltei^ s. str., Liiidr/ia (? Hell cane i/kcs), Ile-
ieroceras, and Helicocei'as.

Remarkably different from the views above alluded to, are
those held by Neumayr and Hyatt. Neumayr"^' believed that
Hdicoceraa together with Toxoceras, Anisoceras, Plychoceras and
others should be In'ought under the genus H(f)ii,ües (in the wide
sense), while he placed Heteroceras near to Crioceras, a genus which
according to him is tolerably distinct from Turriliies and Ilamiles,

According to a later publication of Hyatt,^^ these secondary
forms have no afHnity with Lytoceras and its allies, and conse-
quently are brought by him under Acanthoceratida and Cos-
moceratida. The genera which are treated as Acanthoceratids are
Helicoceras, Heteroceras, and Lindgla, which form his family of
Hamitidfc. . Of those placed under Cosmoceratida, Bostrycoceras
Hyatt is treated under Nostoceratidse, Helicancylus under An-
cyloceratidœ, Turriliies and Osilingoceras Hyatt under Turrilitidse,
and Hyphantoceras Hyatt under Anisoceratidie.

1) Stui-iczka : i.e. p. Js:).

2) Mekk : I.e. p. 487.

Ö) Zittel: Handhncli tier Palaeontologie, I, 2. p. 44Ô.

4) Xeujiayr; Die Ammoniteii der Kreide n. Systematik; d. Arunioiiitiden, pp. 894, 938.

5) Hyatt: Cephalopoda in Zittkl's Text P>ook of Palaeontology, translated by Eastman.

4 AllT. 2. — H. YA13E : CRETACEOUS

111 the study of the Cretaceous Cephalopodii, one of the most
difficult tasks is to deHue the exact limits of the diifeient genera
established ou the turreted forms of Ammonites. Hyatt's classi-
fication by creation of many new genera may possibly be an
improvement to the previously existing ones but it cannot Ije
followed here, for the generic diagnoses and discussions on the
taxonomy are not yet published. Therefore, the j^resent writer
is obliged to follow the old and well known classification of
D'Orbigny as modified by Sharfe and Stoliczka tliuugh it is
apparently unnatural. Yet, it is quite necessary to create a new
genus, as there is one specimen which entirely diifers frum the
known types in the develojmient of its shell.

The group Turriliies, in its most comprehensive sense, foi'ins
a rather inconspicuous element in our fauna. Notwithstanding
their display of various modifications in the upper Ammonite-
beds, specimens are rare in all the localities, so that the descrip-
tion of several species of them must be left imperfect. Besides
those from the upper Ammonite-beds, two species of the Ceno-
manian ty[)es are also described below. These the writer considers
as specially important, for the fauna of the lower horizons is
very poor.

TURRILITES cfr. BERGERI Jîrongniakt.
PL HI., fig. 1.

1902. Turrilites cfr. 3Iantelli Yabe : List Cret. Amm. Hok-
kaido in the Outline of the (leology of Japan. 2"'^ Edi-
tion. rJ02.

CEl'HALOrODA FROM THE HOKKAIDO. Ö

Compare :

1840. T. Bergen D'Orbigny : Pal. Fr. Terr. Cret. Tome I.,

p. 590, pi CXLIIL, ligs. 3-G.
1847. T. Boyerl Pictet et I^oux : UoW. des Grè.s Verts.,

p. 148, pi. XV., fig. 8.
1847. ,, ,, QuENSTEUï : Pctief. Deutseli., L, p. oOO,

pi. XXL, lig. 2G.
1856. „ „ Shaepe : Moll, of the Clialk, p. 6Ö, pi.

XXVL, figs. 9-11.
18(31. „ ,, PiCTET et Campiche : Saiiite-Croix. T. XL,

p. 134, pi. LVIIL, fig. 1-5.
1866. ,, ,, ÖTOLiczKA : Cret. Cepli. 8outli India, p.

185, pi. LXXXVI., figs. 3-6.
1897. ,, ,, KossMAT : Öüdindisclie Kreidef. p. 45.
1903. ,, ,, Choffat : Lc Cret. de Condiicia, p. 15, pi.

I., figs. 3-5.
Öliell turreted, sinistral, with an apieal angle of about 10°.
Whorls contiguous, slowly increasing in size ; rounded externally,
and slightly concave on the upper side. Umbilicus very narrow.
External surface furnished with four rows of numerous short and
laterally compressed spines, the upper two of which are a little
closer together than the others. Also the luunber and size of
s})ines in the two upper rows (about 32 on the last volution) are
ditlerent from those in the lower ones, those of the latter being-
very slightly larger in size and fewer in number. The upper-
most row of spines is concealed under the succeeding whorl.
There are also feeble traces of ribs on the upper and lower
surfaces of the whorl. The number of these equals that of the
spines on it.

The septum has not been seen.

G ART. -2. — II. YABE : CRETACEOUS

Only a single fragment of this shell having been found, it
is hardly possible to get a general idea of its shape and sculpture.
The writer has long been under the impression that it resembles
more or less clearly 7\ 3IanlclU Shaupe, but the slender, elongate
outline of the shell and the four rows of spines on the whorls
being not nuich différent in size and number, seem to show that
it is more closely related to T. Berger I Bron(;niart. Among
the figures of the said species from foreign Cretaceous deposits,
which are accessible to the writer, that of a fragment with tW'O
contiguous whorls from the Chloritic JMarl of Jîonchurch, Jsle of
Wight (shown by Sharpe in his PL XXVII. , iig. 10) exhibits
the closest resemblances.

Stoliczk.v descril)es the tubercles of T. Berger I as often
s})inose and sharp, but none of his figures shows tubercles so sharp,
and laterally so compiessed as in our specimen which in this
respect comes closer to an immature one figured by Sharfe in
his monograph (PI. XXVI., fig. 11).

In point of the existence of only a slight ditlerence in the
size and number of spines in the upper and the lower rows,
this species resembles 2\ BlanicMl more than T. Bcrgerl, while
the close arrangement of the spines in each row makes it re-
semble T. Bergerl var. 'miliaris (Pictet et Campiche : 1. c.

fig. .-,).

Taking these several points into consideration, it appears to
be quite evident that this form is specifically distinct, not only
from T. Bergerl and T. Mantelli, but also from all others hither-
to described, the nearest ally however being T. Bergerl.

Jjucality — The Popet, near the mouth of the ^Sanushilje.

Horizon : — Lower Ammonite-beds.

X^orcign localities and horizons of T. Berger} and 1\ Matltelll :

CErHALOrODA FROM THE HOKKAIDO. 7

— T. Jiei'gerl Brongniart occurs in tlic deposits of the All)i;in,
Vraconiaii and Cenomanian ages of England (Chalk with green
grains at Chardstock), France (Cenomanian of Bavet, Anxon ;
Albian of Aignlun and Savoy), Austria, Switzerland and Italy.
This species has also been found in Algeria associated with
Turriliies costatus and Acanfhoceras rolo/if/r/e/i.^e, and in Condncia
together with BcJcinnHci< minlnnis Lister, PhyUocevd^ cfr, semi-
xtrhilum. D'OiîB. and Acanihoceras hitichiv'niin var. inoçatnhlquensh
Choffat. Stoliczka calls it a connnon fossil of the Utatur
group of Odium and Moravatur, South India.

T. Mantelli has l)een met with in the Gray Chalk of Sussex
and the Tourtia of Essen.

TURRILITES KOMOTAI m.

Pis. I. and II.
Dimensions : —

Height of th(^. 8i>ire. 18.0 cm.

Diameter of the last volution. 18.0 ,,

Width of the niiihiliciis. 6.0 ,,

Height of the last whorl. G..3 „

Breadth of tlie last whorl. 7.0 „

Shell sinistral, composed of numerous contiguous whorls,^^
rapidly increasing in size, with an apical angle of about 70°.
Whorls rounded externally, slightly convex on the upper and
concave on the lower snrface. Section of the whorl nearly el-
liptical, somewhat broader than high ; rather trapezoidal in the
last volution. Suture of the spire deep. Umbilicus moderately

1) Only four volntinis are present in the specimen, three or four otliers near the apex
havina; been broken ud'.

8 ART. 2. — II. YABE : CRETACEOUS

wide, being about one tliird the diameter of the last vohition.
Surfîice with ribs and spines, the former arranged in two series
and the latter in four rows. The ribs on the upper part of the
whorl are numerous and flat, separated by wide intervals ; they
are oblique, bending strongly forward, distinct near the external
side, ])ut gradually becoming feeble townixl the umbilicus. Each
of these ribs bears tliree short spines, close to the external side ;
they are moderately sharp and compressed in the spiral direction.
The external surface of the whorl shows a second series of ribs
which are much broader than those of the first series and al)Out
a half their numl)er. The ribs of this series are also slightly
oblique and somewhat concave in front, with flat intervals of
almost the same width as the ri])s themselves. The middle portion
of each of these ribs on the body whorl carries a long spine.
The sj^ines are more or less trigonal in a transverse section,
rounded behind, and flattened in front. Their surface is orna-
mented with fine lines of growth, in addition to a delicate median
line on the flat side. Most of the spines on the other whorls
have been cast off, and the remaining basal portions which look
like tubercles are usually covered with a thin calcareous layer.
The umbilical side of the whorl is smooth while the basal sur-
face shows the impression corresponding to the surface marking
of the upper side of the foregoing volution.

The specimen is unfortunately not in a condition to show
the position of the siphuncle, nor all the ramifications of the
lobes and saddles which are only partly thougli distinctly seen
on the whorl near the apex.

T. Komotai is related to T. Cunliffianus Stol.'^ from the
Utatur Group of Odium, South India, although the distinction

1) Stqmczka: Cret. S. Indiii, p. 190, pi. LXXXIX.
%

CEPHALOPODA FPvOî\[ THE HOKKAIDO. 9

18 (|uitc obvious. In tlio Indian species, tlic niinil)er of ribs on
tlie upper side of the wliorl is nearly equal to that of the ribs
on the lateral side, or to that of the tubercles arranged in two
rows between the upper and lower series of ribs, l>ut moi-e
numerous than that of the tubercles in the lowest row.

Locality and horizon : — This majestic form of Turrllitei< is
at present represented only by a single specimen which has been
found by J\lr. Komota in a marly nodule found Itolow a cliff of
the Ikushum])ets directly above the coal mine.

The matrix of this specimen is a sandy marl, dark grey in
color, and compact in texture, agreeing in every respect with that
of the nodules in the sandstone of the cliff above cited. There-
fore there is no dou])t that the specimen came from that locality,
as is also showu by the presence of an impression of Blargarita
funiculclta Yokoyama on the matrix of this specimen, a (lasteropoda
very common in tlie above sandstone. This sandstone cliff is the
chief locality of the typical fossils of the lower Acanthoceras-zone.

The writer has much pleasure in associating with this species
the name of its collector JMr. Komota who w\as kind enough to
present it to the museum of the Geological Department of the
Tokyo University.

HELICOCERAS SCALARE m.

PI. III., figs. 2 and 3.

Shell helicoid, forming a. widely open coil ; either sinistral
or dextral. Whorls circular or subcircnlar in section, gradually
enlarging anteriorly. Surface with numerous transverse ribs.

10 ART. 2. H. Y A BE I CRETACEOUS

separated from one another b}^ a tolerably broad and flat interval.
Kibs sharj), always simple and smooth, becoming somewhat feebler
and more flexuous on the umbilical side.

The suture line has been only partly examined. It shows a
single lateral saddle and lobe, both of which are twice deeply
bifid and finely incised. The siphonal saddle is slightly higher
than the lateral one.

One of the specimens (fig. 2.) is from the Sraphites-heds of the
Opiraushibets. It is nearly a volution of a sinistral shell, less
than 4 mm. in diameter at the smaller end, and about 8 mm. at
the other end. One half of this whorl is furnished with 45 ribs
of which 8 are broader and more elevated than the others,
especially on the outer side. The smaller ribs which are found
between these larger ones number 4 to 10. All the ribs are ai-
ranged transversely, except near the larger end of the whorl
^vhere they are a little oblique.

The second specimen is from the Yril)arigawa. It is a volu-
tion of a dextral shell, a little larger than the one mentioned
above. As in Heieroceras Otsukai,^^ the siphon bends downward
from the median line at a certain point of the external side of
the whorl.

Another example drawn in fig. 3, pi. III., is a small frag-
ment of a larger, dextral shell. These figures show in suflficient
clearness the mode of arrangement of the ribs on the whorl.

Localities : — The middle course of the Opiraushibets, and the
Yflbarigawa. Four specimens from the former locality and the
two from the latter have been examined.

Horizon : — Scaphites-heds.

\) Vide p. 1.").

CEPHALOPODA FEOM THE HOKlvAIDÔ. 11

HELICOCERAS (?) VENUSTÜM m.
ri. III., fig. 4; pi. v., iigs. 1 and 2.

VJO'2. IlamUcs vcnustus Yabe : the Li.st of Aniinoiiites from
the Cretaceous of the Hokkaido, in Outlines of the
Geology of Japan : ])escriptive Text to accompany the
Geological Map of the Empire.

Shell sinistral, forming a helicoid spiral of a very wide coil.
AYliorls slender, very slowly increasing in diameter toward the
anterior end ; subcircular in cross-section when young, transversely
oblong in the aged specimens. Surfece with numerous transverse
stricC and periodic ribs, the latter hardly distinguishable from
the former on the umbilical side, but becoming gradually higher
toward the external side, until they are most elevated on the
upper side, each being here furnished with two tubercles. In-
tercalated striœ numerous, simple, smooth and uniform, becoming-
somewhat feebler on the umbilical surface ; their intervals vary
from 1.0 mm. to O.o mm. Both striae and periodic ribs maintain
a slightly oblique course.

The suture line is only partly visible. It is apparently very
similar to that of the preceding species, being divided into three
bifid saddles and a corresponding number of similarly bifid lobes
on either side of the siphonal line. These saddles and lobes are
nearly uniform in size and are moderately incised. The external
lobe is somewhat shallower than the others. The siphon is on
the median line of the external surface of the whorl.

There are two specimens of this pretty species in the writer's
collection, one of which is from lower part of the Tachyducus-
beds, exposed along a right triljutury of the Saushi-sanushibe.

12 APvï. 2. H. Y A BE : CRETACEOUS

It is a young volution with tlie umbilicus 5.5 cm. wide and the
larger end 1.3 cm. bj-oad. The other one which is from the
/Scaph!tes-h(idH of the Opiraushibels, is a fragment of the septate
portion of a larger shell.

The present species is evidently related to Heleroceras cendopse
Anderson'* from the Smith ranch, east of Pha-nix, Oregon, for
they have many points of resemblance in the general form and
size of the shell and also in the character of the ribs. Indeed,
the resemblance is so close that their separation into two distinct
species may be donl)ted. But according to the description of
Anderson, the American fjrni possesses periodic ribs whose
highest elevation is on the siphonal line and is not so elevated
on the upper side as in the Japanese specimens. Although the
difference seems to be tiitling, yet the writer considers it prudent
at present to regard them as two distinct species.

The discovery of more materials from both sides of the Pacific
wdll doul)tless throw light on this question.

Localities and horizons: — The lower horizon of the Pachydiscus-
beds of the Saushi-sanu.sliibe, Iburi Prov., and the fSc(fp/iik'S-hQds of
the Upiraushibets, Teshio Prov. (Jne specimen from each locality.

HETEROCERAS (?» OSHIMAI m.
IM. III., ligs. 5 and C

Shell tui-reted, sinistral ; vohilions free, especially near the
anterior end. Whorls growing slowly ; obliquely elliptical in
section ; the surface carrying both ribs and tubercles. Kibs

1) Andkksok : Cretaceous Deposits of tin/ Fucllie Coast. Proc. Calif. Aead. Sei. old
Series. Vol. 11., no. 1, p. 'Jl, pi. LII., ligs. K.HMOl.

CEi'HALOrODA FROM THE JIOKKATDÔ. 13

numerous, the last volution liaving 45, separated by intervals wider
than their breadth. l\ibs uniform in size except on the upper
border of the whorl, where the tuberculatud ones become thicker,
while the smooth ones become more faint ; arranged obliquely,
curving backward ou the lower l)order and forward on the upper.
fSome of the ribs bear four tubercles or more coi'rectly rounded
spines which form four horizontal rows, two on the upper border,
one in the middle of the external side and another on the lower
border, and these tul)crculated ribs altei'uate Avith 1-4 smooth
ones. Tubercles subeipial in size. Internal side of the whorl
nearly smooth.

Of tlii^5 species there is a non-septete fragment of moderately
large size composed of two circuits, of which the posterior circuit
and one half of the other are very close together although not
quite in contact, while the remaining half is detached from the
other part. The specimen is represented in pi. III., fig. 5.

The second example, which proba1-)ly belongs to the same
species, is a young shell, very loosely twisted and ornamented
with distinct, narrow^ oblique ribs alternately bearing two short
spines on the upper border (fig. 6).

I have not yet seen the suture line of this species, but the
other characters are enough to distinguish it from any known
species of Turrilites. The nearest approach to the Japanese form
is made by T. Reussianus D'Orbigny, from which it differs in
having very oblique, fiexuous and subequal ribs. Still their
resemblance is so great that they must be considered as closely
allied forms.

T. HeussiauLis is placed by Schlütek^^ under the genus

1) C. SchlCter: Cepluilopoden d. oberen deutsch. Kreide. IST!. P. lO'J, pi. XXXII.,
iigä. lo-21 ; pl. XXXIII., lig. 1. — A. Fiarricii u. U. Sculoknjî.vcu : Cei>li. d. böli m. Kreidet".

14 ART. 2. H. YABE : CRETACEOUS

Heteroceras, while Fritsch and others bring it under Helicoceras.
Recently, however, Hyatt^' raised it to the type of his genus
Hyph aniocerru.

T. Remsianus occurs in the Reuss Ian us -zone of England
and Northwestern Germany, in the zone of J^Jpiasicr brevis of
France, in the Pläner Kalk of 8axony and the Priesen beds of
Bohemia.

The species is named in honour of Dr. R. Oshima of the
Hokkaido Colliery Railway Company in recognition of his courtesy
oiïered to the writer and his kind interest taken in this palaeon-
toloa;ical stud v.

Localities : — The larger specimen was obtained in situ near
the confluence of the Yoshiashizawa with the Ikushumbets, and
the smaller one in the upper course of the Shi-kuruki, a tributary
of the Yübarigawa.

Horizon : — ScophUes- beds.

HETEROCERAS (0 OTSÜKAI m.
PL IV., figs. 1-2., ph VL, lig. 7.

1894. Ti/rrilUes sp. Jimbö : Beiträge z. Kennt, d. Fauna d.

Kreidef. Hokkaido, p. 41, pi. I. (XVII.), ligs. 8, 8a.
Shell turreted, volutions free, sometimes sinistral and some-
times dextral ; the anterior ones making occasionally a narrower

V. 47, ].l. Xrrr, % lO?; pi. XIV., %s. 14-19 C/H-IS); pi. XVI., fig. 9?— a. Fritsch:
Stinlieii ini (.rebiete tier l)oliiii. Kreidef. IV. Die Iberscliiclileii, p. 92, tig. 44. V. Prieseiier
Sc'liiclUen, p. 79, fig. 02. — II. Wood: The Mollusca of the Chalk Rock. Pt. 1. Quart. Jonr.
Gcol. Soc. London. Vol. LIT. 189(>. P. 74, pi. II., figs. 3-5.

1) Hyatt: Cephalopoda in Zittel's Text Book of Palaeontology, translated by East-
man. IS'J'J. V. o76.

CEPlIALOrODA FROM THE HOKKAIDO. 15

coil than the posterior, at times a wider one. Umbilicus as wide
as the breadth of the whorl. Whorls slowly increasing in diameter
with growth, nearly round or somewhat transversely oblong in
section and ornamented with numerous ribs (about 50 on a volu-
tion). Ribs uniform, smooth, usually simple, rarely bifid; arranged
slightl}^ obliquely ; moderately sharp but broad at the l^ase on
the external side, thence becoming gradually narrower toward the
upper border and suddenly toward the lower border, while they
are fine, though still distinct, on the umbilical side. Interspaces
between the ribs flat or slightly concave, being twice as broad
as the ribs.

Suture line partly known. External lobe shallow, with a
high broad siphonal saddle ; first lateral lobe very broad, regularly
bifurcating twice; external saddle symmetrically bipartite, inclining
inward ; first lateral saddle broad and also bifid. All the lobes
and saddles are finely serrated.

Of the three examples of this species hitherto found, one (pi.
IV., fig. 2.) is a partly septate fragment composed of I2 volutions,
measuring 2.5 cm. in diameter at the smaller end. Where the
shell is not preserved the ribs appear broad and round, separated
by very shallow furrows. The suture line and the siphuncle are
also seen in the part above alluded to. The siphuncle is abnormal,
being pushed down anteriorly from the normal median position
on the external side of the whorl. A quite similar cnse has
already been noticed by F. B. Meek on n shell of HeteTOceras
tortum Meek and Hayden.^'

The second specimen (pi. IV., fig. 1.) is a somewhat smaller
one, composed also of le volutions.

1) Meek : Invert. Cret. and Tert. Foss. Upp. Missouri County. Rep. U. S. Geol. Surv.
of the Territories. Vol. TX., ISTO. P. 4SI, pi. XXIT., fi^s. 4a, I., c.

16 AUT. 2. H. YABE 1 CRETACEOUS

Turrilites sp. of Jimbö is a small specimen, of ill preservation,
l)nt its form and surface markings are so like the two individuals
above described tliat it may safely be regarded as belonging to
the same species.

Compare Heterocerns elongcdnm Wptiteaves^^ from the Na-
naimo group of Vancouver. Though decidedly distinct, the two
show certain similarities in their characters.

The writer takes this opportunity of associating with this
species the name of Mr. ( )T8ur\:A of the Imperial Geological
Survey, who once undertook a geological trij^ in the IMesozoic
district along the upper course of the Yubarigawa.

Localities and horizon : — The first specimen was picked up
in the river bed of the Yubarigawa above the confluence of its
right branch, the Wenhorokabets, with it, while the second is
from the Pachydi^nis-heds of the Kikumezawa, a tributary of the
Ikushumbets. Jdebo's specimen is lal)elled " Ponnaibuts, a place
close to the town of Urakawa, Hidaka Prov."

Vae. MULTICOSTATA m.
PI. III., fig. 9; pi. IV., fig. 3; pi. VI., fig. S.

Under this head, the writer brings four fragments of turreted
shells which in their generjd ]ia])it agree fairly well with the
typical species, but differ in having more open coils, slender
whorls, and narrower and more numerous ribs. Consequently, the
ribs of these shells are set more closel}^ and also more obliquely
than in the typical species.

1) WiiiTKAVES; Mosozoic Fossils. Pt. 2, ji. 100, pi. XII. Pt. 5, p. 3;31, pi. XUV., %. 2.

CEPHALOPODA FROM THE HOKKAIDO. 17

All these distinguishing characters, however, vary much
among the specimens. One of these specimens approaching the
typical species in the surfoce marking possesses, at the same
time, whorls not so widely open as in the others, thus undoulttedly
forming a connective link between the typical species and the
above mentioned type of the variety. A specimen representing
the opposite extremity of variation shows a suture line which
does not essentially differ from that of the typical species (fig. 8).
There is another specimen from the Shisanushibe which the Avriter
with some hesitation refers to the same variety.

Localities: — The four specimens above mentioned are from
(1) the Opiraushibets where it was found as a pebble close to
the confluence of the Panke-kenebets (fig. o), (2) the Opiraushibets
in its middle course, (3) the Shisanushibe, a tri1)utary of the
Popet, and (4) the Yilbarigawa (fig. {)). The specimen from the
last locality was found in the collection of Jimp.ö.

Horizon : — All the specimens being obtaii:ed from among the
river gravels, the exact geological horizon in ^vhich they occur is
of course unknown. But the one from the middle course of the
Opiraushibets was found associated with Scaphiles puerculus Jimbö
in a marly nodule ; hence, there is a great probability, that it was
derived from the extensive Seaphiies-he(\^ in the neighborhood of
the above river.

HETEROCERAS (?) JAPONICUM m.
PL III., fig. 8.

1902. Turrililes cfr. Indicus Yabe : List of Ammonites from
the Cretaceous of the Hokkaido, in Outlines of the

18 ART. -2. — H. YABE : CRETACEOUS

Geology of Japan : Descriptive Text accompanying the
Geological ]Map of the Empire.

The writer proposes the following diagnosis by examining a
specimen with two volutions of the anterior portion of the shell
preserved.

Shell dextral, narrowly elongated ; volutions widely apart
longitudinally ; umbilicus moderate in width. Whorls somewhat
obliquely elongated in cross-section. Surface ornamented with
numerous ribs aud also some periodic ones. Kibs strong, simple,
smooth and uniform, transverse and nearly straight on the external
surface of the whorls, curving gently on the inner upper margin
and thinning gradually inside ; intervals wide (about 2 ram. on
the average), smooth and flat. Periodic ribs very remote from
one another, one on two volutions, somewhat broader and much
higher than the other ribs. Septation unknown.

It is by no means certain that the present species is specifically
distinct from T. indicus Stoliczka^^ The Indian species is
provided with volutions, contiguous when young and detached in
the advanced stage of growth with the ribs also nearly similar in
characters, the only difference l)eing in that the deeper periodic
furrows are bounded behind by a strong rib. According to
KossMAT,'-^ Heteroceras indicum is a species resembling H. poly-
plocihm Rœmek but distinguished by its constantly smaller size
and total absence of tubercles which often appear on the surface
of the whorl of //. jwlyplocum.

F. M. Axderson"^^ referred a small form from Oregon, coiled
in a flat spiral, to //. indicum with query. The whorl of the

1) Stoliczka : Cret. S. Ind., p. 184, pi. LXXXVI., figs. 1-2. Kossmat : Untsuch. ii. d.
siidind. Kreidef., \>. 14.'!, pi. XX., figs. 5 and 6.

2) Kossmat: 1. c. p. 143:

o) An-derson : Cretaceous Deposits of the Pacific Coast. 1902. P. 91, pi. III., figs. 96-97.

CEPHALOPODA FROM THE HOKKAIDO. 19

American specimen is described as showing three or four con-
strictions, besides the numerous ribs of the ordinary kind.

Besides the specimen on which the above descrij^tion is based,
there is another from the Opiraushibets which is much more
fragmentary, and which therefore the writer refers with much
hesitation to the same species.

Localities : — The Yubarigawa, Isliikari Prov. (figured), and
the Opiraushibets, Teshio Prov.

Horizon : — Upper Ammonite-beds.

Localities and horizons of the allied species : —

H. indlcum Stol. The uppermost lied of the Trichinopolv
group of South India.

H. indicum Stol. (?). The lower Cliico group of Smith
ranch near Phœnix, Oregon.

HETEROCERAS (?) ORIENTALE m.
PL IIL, fig. 7.

There are two internal casts of a shell which the writer takes
for a new species. The larger and better of these two specimens,
has only one and a half volutions preserved. The diagnosis runs
as follow^s : —

Shell sinistral, very much elongated longitudinally. Vo-
lutions slender, increasing in size gradually toward the anterior
portion, making very oblique and open coils. Whorls oblong in
cross-section with the longer axis oblique to the axis of volution.
Surface marked with very flexuous, transverse ribs except on the
smooth inner side. Kibs slightly irregular in size, usually simple

20 ART. li. H. YABE : CRETACEOUS

and strong l)üt not seldom bifurcating and then weaker ; they
make a l)ackward curve on the lower marsjin and another forward
one near the upper, whei'e sooner or later they disappear. In-
tervals also not quite uniform, the breadth varying from 2 mm.
to o mm. Tubercles rather prominent, arranged in four rows of
which three are on the external side and one on the lower margin.
All the rows are equidistant from one another and parallel to
the suture except on the last portion of the shell in which they
are more or less oblique to the latter. Kibs, except weak and
short ones, bear a set of tul)ercles.

Although this species agrees with T. cfr. Bergeri^^ in having
four rows of tubercles and transverse ribs, yet there seems to Ije
no further resemblance between the two. Helicoceras hystricvluiih
White'-' from the Cretaceous of Sergipe, South America, represents
a species probably belonging to the same type as ours. Excepting
this H. hystricuhuii, the writer is at present unaquainted with any
form which makes even a distant approach to the Japanese shell.

Localities : — The Ikushumbets, Ishikari Pro v., the exact posi-
tion of the bed from which the specimens were derived is unknown.
But the (^tlier jMoUuscan remains found together in the same marly
nodule belong to the species common in the Pachydiscus-hQaa.

NIPPONITES MIRABILIS gen. et sp. ^ov.
PL IV., tigs. 4-7 ; pi. VI., fig. 6.

We have only a single specimen belonging to this entirely
new and extraordinary type of Ammonites.

1) Vide supra.

■1) WinrE: (jjutrihntions to the Paleontology of Brazil. 1888. P. 229, pi. XXII., figs. ?,, 4.

CEPHALOrODA FROM THE HOKKAIDO. 21

At fii'st the shell forms a more or less flat spiral coil wound
siiiistrally îis in Helicoceras, its diameter being about 2.7 cm.
and the cross-section of the whorl at the end of the second vo-
lution about 0.9 cm. The further growth is no longer spiral, 1)ut
it consists in turning to the right and then to the left several
times and thus forming many U-shaped curves which cover the
first part of the shell from six directions, so that the plane of
the curves roughly corresponds to the six sides of a cube. The
limbs of each single curve and of those adjacent come nearly in
contact.

In cross-section, the whorls of the younger, spiral portion
are nearly round, while those of the older are somewhat oblong,
the dorso-ventral diameter being a little narrower than the
lateral one. The body chamber is probably long, as about two
anterior curves seem not to be septated.

As our shell consists of several U-shaped curves, wdien it is
compared with the ordinary Turrilites, one limb of a curve corre-
sponds to a whorl of a sinistrally w^ound Turrilites, and the
other to that of a dextrally wound one.

The ribs are simple, smooth and uniform (except at the
anterior end), sharp and high on the external surface, becoming
somewhat weaker inside ; they are closely and obliquely set,
separated by flat intervals of a moderate width. Only near the
anterior end of the shell, the ribs become dissimilar, some of
them being higher and broader than others, which proves the
growth at these points had been frequently checked. With the
exception of the anterior portion, the surface sculpture of this
species is so much like that of If. Otsuhai as to suggest the
existence of a relationship between the two species. In the
.spirally coiled, posterior portion of this shell, the siphuncle is

22 AET. 2. — H. YABE : CRETACEOUS

suturai by which fact it is distinguished from Helicoceras. On
the portion of the irregular growth, however, the siphuncle i&
always (?) situated in the median line of the external side.

The suture line was partly examined on the whorl which
succeeds the regular spiral growth. It shows two saddles and
corresponding lobes on one side of the siphonal line, and does
not essentially differ from that of H. Otsukal. The two saddles
are comparatively slender and nearly equal in height ; both are
bipartite with bifid subdivisions. The lateral lobe is very broad
and bipartite, slightly exceeding the ventral one in depth. The
siphonal saddle is relatively broad and high, with a few ser-
rations along the margin.

The complicated mode of coiling the whorls in this species
is better seen in the accompanying figures. In Plate IV., figs.
4 and 5 show its external aspect seen from two different direc-
tions, each U-shaped curve is marked with the Koman numerals
II. — VI. and I'. II'. in the order of their succession. The
first carve (I.), immediately succeeding the spiral portion, is not
to be seen in these figures ; the second curve which is only
partly visible is marked II. and the third III. and so on. The
seventh curve, being nearly upon the first one, is marked I'.
and the eighth, for the same reason, II'.

Fig. 6 shows a longitudinal section of the specimen, the
cross-sections of the whorls being hatched. The two volutions
of the spiral portion appear as four round sections marked with
Arabian numerals 1-4. The position of the siphuncle is also
indicated on the sections by S.

Fig. 7 was drawn after a wire model showing the mode of
coiling the whorls from the beginning to the end. Roman
numerals are here also used in the same sense as before.

CEPHALOPODA FROM THE HOKKAIDO. 23

We do not know any morphological equivalents of this
peculiar shell among similar tube-forming animals of the past
■or the present. The following cases of abnormal growth of the
shell of lurrilites and its allied genera may be taken into con-
sideration.

(1.) The Ammonites with a conically spiral shell usually
follow a certain law in their growth, having volutions which are
•either mutually in contact or separate ; but cases often occur in
which they show a portion near the mouth more or less per-
ceptibly deflected in position from the preceding volutions. This
-change may be produced by the animal either pathologically or
as the result of old age. AVe also learn from various sources,
the occurrence of many examples in discoidal species, showing
a slight tendency to become conically spiral in the anterior por-
tion. QuENSTEDT and some others early called attention to this
fact and seem to have believed the probable derivation of second-
ary forms from ancestral individuals of pathologic growth.

(2.) In some of the secondary forms, it is known that at
the very beginning of development they show a normal growth,
that is to say, a discoidal shell with volutions more or
less involute. A remarkable case has recently been described
by Whitefield^^ in Heteroceras sinipUcostatum from the Fort
Benton group of the Black Hills. According to this author,
the species has its younger portion composed of two straight
limbs, close together, like Hamites. This seems to indicate the
derivation of a more abnormal type from a less abnormal one.

(3.) Besides, there are a few cases in which the shell is
coiled in a conical spiral, first to the right and then to the left.

1) ^VHITEFIELD : Observations on and emended Descriptions of Heteroceras simpHcostahim
"WiiiTKFiELD. (By review.)

24 AET. 2. — H. YABE : CRETACEOUS

the latter portion surrounding the earlier one. Such double
coils of a tubular shell around the same axis of volution, but
in opposite directions, may be either accidental, or due to the
animal itself having an inherent power of producing such coils.

(4.) Again, there are some species of the later Cretaceous
which are remarkable in exhibiting a great range of individual
variations on the mode of coiling the shell. The variations,,
however, are due to a loss of power in the animal, which as it
grows no longer maintains its normal mode of coiling.

In the present species, in spite of its being represented bv
a single specimen, the mode of coiling the whorls can neither
be accidental nor due to a loss of power to produce the regular-
ly wounded shell, as it is too regular to admit any such sup-
position. That it is a Turrilites or some allied genus which has
modified its mode of growth on account of its advanced age is
also inconceivable. Therefore, this formation of several U-shaped
curves must be ascribed to the inherent power of the animal.
It is for these reasons that the writer proposes to give a new
generic name to this single specimen.

As above mentioned, this species has both suture and sculp-
ture which indicate its derivation from some species of Hetero-
ceras (?), e.g., Jleteroceras (?) Otsidcai in turreted groups of Am-
monites. From the form of the shell, we can readily see that
the animal led neither a free swimming life nor a sessile one,,
but was a creeper on the sea-bottom as is the generally accepted
view of Turrilites.

QuENSTEDT after describing 2\trrilUes reßexas — an extra-
ordinary form of a doubly coiled shell, says:'*

" Es mag wahrscheinlich Formen gehen, ^vo sich dieses hin-

1) Qüenstkdt: I.e. p. oCG.

CEPHALOrODA FROIM THE HOKKAIDO. Zö

und lieicb-elien noch zum üftern wiederholt, sie sind den Weber-
spulen zu verglichen, über welche die Faden sich hin und her
über einander windet. Was setzt das nicht fiir eine Beweglich-
keit der Organe voraus !"

Of course, Qüenstedt did not know, then, that such a
peculiar form as the present species would occur in Japan.

INCERTiE SEDIS.

There are few forms of Ammonites in our fauna, more
interesting than the single specimen described below under the
name of Ammonites Koioi. Its characters agree in several re-
spects with Olcosiephaaus siiperstes Kossmat" which that author,
together with Ammonites rvdra Stoliczka,"' temporally brought
under the genus Olcostephantis. He, of course, acknowledged
the phyletical relation of these two Ammonites to Vascocevas on
one side and to Acanthoceras on the other, although they much
resemble, not only in the general aspect of their external form,
but also in their suture line, the lower Cretaceous and the
Jurassic species of the genera Olcostephanvs and Stephonoceras —
a good example of heterochronous homoeomorphy, a. phenomena
now known as often occurring in the Cephalopoda.

Vascoceraé"^ is a genus lately proposed by P. Choffat to
include Ammonites intermediate between Olcostephanus and
Acanthoceras {3Iammites). According to this author, Ainin.
syperstes differs from Vascocevas in respect of the higher saddles

1) Kossmat: I.e. p. 1?.3, pi. XVII., figs. 1 a, b, c.

2) Kossmat: I.e. p. 20. — Stoliczka : I.e. p. 122, [il. LX. Further see Xeumaye: I.e.
p. 933.

3) Choffat: Faune Crétacùjue du Portugal, II Série. P. 51.

26 ART. 2. — H. YABE I CRETACEOUS

of the suture line, while Grossouvre^^ considered it to be a
species rather referable to the genus Acanthoceras. Tlius, the
question in regard to the particular generic name that should
l)e retained for the group under consideration seems at present
an open one.

Ammonites rudra occurs in the middle, and Amm. superstes
in the upper division of the Utatur group of South India.
According to Peron'-^ and Choffat, Ammonites closely resem-
bling the latter occur also in the deposits of the Turonian age
of Algeria and Portugal. Ammonites [Buchiceras) Hartii Hyatt
from the upper Cretaceous of Sergipe in Brazil which has been
referred by Kossmat to the same group, together with the above
two species, shows also a, close relation to Vascoceras, but the
specimen in this case seems too imperfect for the decision of its
proper generic position.

AMMONITES KOTOI m.

PI. VI., %s. 3 and 4.

The only specimen here figured is an imperfect cast of the
interior in which only a portion of the outer volution is well
preserved. The suture line, however, is tolerably well preserved.

Shell globose, inflated, consisting of many volutions. Umbi-
licus deep, moderate in breadth. Whorls lunular in cross-
section, l)roader than high ; uniformly rounded on the ventral
and lateral sides. Surface Avith numerous transverse ribs, and a

1) Grossouvrk: Sur I' Animonitoi peramplm, p. 334.

2) Peron: Les Ammonites da Crétacé supérieur de l'Algérie, p. 14.— Kossmat : I.e.
p. ]:;5.

CEPHALOPODA FROM THE HOKKAIDO. Z/

row of tubercles bordering the umbilicus. Kibs round, simple
and uniform, separated by intervals of like breadth, gently
curving forward on the ventral surface of the whorl and
generally bearing a moderately sized tubercle on the umbilical
margin, umbilical wall perpendicular and smooth.

The volution next to the outer one differs from it mainly
in having a relatively wider umbilicus.

The siphonal saddle is broad and high, flat on the top, and
with a few denticles on both sides. The external and the first
lateral saddles are exceedingly high and simple, with a few short
branches, and are tripartite on the top. Each branch terminates
in a round lobule. The first lateral lobe is bifid and somewhat
shorter than the external one.

As already mentioned, the present species is an ally of Amm.
superstes and Amm. rudra ; of these two, it exhibits a greater
resemblance to the first than to the second. However, it differs
from the first by its higher whorl bearing more numerous ribs
and tubercles.

Locality : — The specimen was found in a nodule of marl,
obtained as a pebble in the YCibarigawa. This being the sole
example now at the writer's disposal, its stratigraphical position
cannot be given with an}' certainty.

OLCOSTEPHANUS Neumayr.

Neumayr^^ correctly remarked that an Ammonites from
Urakawa brought under the genus Stephanoceras by Brauns"^

1) Neumayr : Neues Jahrb. für Mineralog. Geolog, u. Palœontolog. 1881 Bd. II., p. 80.

2) Brauns: Vorläufige Notiz über Vorkomnisse der Juraformation in Japan.

28 AKT. 2. H. YABE : CRETACEOUS

might belong to Olcostephanus, while Naumann^^ believed it to
be in reality a sjiecimen, foreign to Japan. In the Jimb~> col-
lection, however, there is a specimen of Ammonites referable, as
the present writer believes, to Olcostephanus, with a label written
" Loc. Takambe, a branch of the Ikushumbets river." Thus,
at present there is no doubt about the occurrence of the Am-
monites of the above named genus in our fauna.

A specimen described l)y Jimbô"^ as a species of Olco-
siephanuSj long ago, is now believed by the present writer to be
merely an imperfect specimen of a new species of Scaphites
which will be described in some detail in the succeeding part
of this memoir.

OLCOSTEPHANUS UNICUS 3i.

PI. VI., figs. 5, a, b.
Dimensions : —

Diameter. 4.2 mm.

Height of the last whorl. 1.9 „

Breadth of the last wliorl. 2.3 „

Width of umbilicus. 1.2 ,,

Shell composed of many volutions, inflated. Umbilicus moder-
ate in size, deep ; the umbilical wall is perpendicular and its
margin rounded. Aperture subcrescentic, broader than high,
uniformly rounded on the lateral as well as on the ventral side.
Involution rather deep. Surface wdtli numerous transverse ribs ;
straight, uniform, narrow l)ut prominent, separated by flat intervals
wider than their breadth. The ribs are closely set and nearly

1) Naumann : Ueber das Vorkommen der Kreidefurmation auf der Insel Jezo.

2) Jimb5 : I.e. p. n:j (179), pi. IX., figs. .-5, a, b.

CEPHALOPODA FROM THE HOKKAIDO. 29

jDarallel to one another, while near the umbilical margin, tlii-ee
or more of them unite to form a bundle. Besides, there are five
constrictions on the hist volution which are iDroader and deeper
than other intervals and are bordered behind Ijy a rib somewhat
more prominent than the usual ribs to which they are arranged
quite parallel.

The suture line which is only partly seen, shows three saddles.
The first lateral saddle is narrow, but high and bipartite ; the
lobes are also bipartite.

Of Anmionites described from the Cretaceous deposits of the
Pacific border, Olcostephanus Logannianus Whiteaves^^ from tlie
■Chorizon of the Queen Charlotte Islands in some respects appears
nearest to this species. The periodic ribs seen on the Japanese
specimen are not present in the American species, and moreover
the whorls of the former increase as the shell grows more rapidly
than the latter.

Locality : — The specimen here figured is from the Kami-
takambe, a right branch of the Ikushumbets, between the Poronai
and the Ikushumbets coal mines.

Horizon : — Unknown.

HAUERICERAS Grossouvre.

The genus Hauericeras was established by A. de Grossouvre,^
in 1893, to include Ammonites which have smooth, strongly
compressed, discoidal shells, with a sharp, ventral keel and having
the suture line composed of two broad saddles followed by 3-7

ft

1) Whiteaves: Mes, Foss. Vol. I., pt. 1, p. 27, text fig. ?>; pi. VIII., fig. 2. Pt. 3, p.
211, pi. XXIII., figs. 1 and la. Pt 4. p. 276.

2) Grossouvke : Recherches sur la Craie supérieure. 1893. P. 219.

30 ART. 2. — H. YABE : CRETACEOUS

smaller ones. The external and lateral saddles are unequally
bipartite and especially so in the former, so that it may more
properly be called tripartite ; the lateral lobe is also broad and
tripartite, and about as long as the external one. The siphonal
saddle is relatively high and broad, being inflated at the top.
Tiie auxiliary saddles and lobes rapidly diminish in size and
hang obliquely toward the umbilical suture.

Karl v. Zittel^^ treats Hauer iceras, Desmoceras, Puzosia
and Pachydiscus under the family of Desmoceratidie, while Gros-
souvRE"^ brings them together with Schlüteria under the family
of Phylloceratidœ. Al. Hyatt''^ includes them, with the excep-
tion of Pachydiscus, in the family of Haploceratidic. However^
the general character of Hauericeras in its developmental stages
differs slightly from Phylloceratida?, being on the contrary closely
allied to the typical Desmoceras, and therefore the writer is at a
loss to see wdiy Grossouvre included it among the Phylloceratidit.

As is generally accepted by palaeontologists, Hauericeras has
a close affinity to the group of Desmoceras Sugata Forbes and
also to the genus Puzosia Bayle, the resemblance to the former
beino- in its ventral keel, and to the latter in the suture line.
The Avriter himself also at first adopted the same view, but a
further examination of the specimens from the Hokkaido led him
to think that the latter belonged rather to the family of Phyllo-
ceratidse than to Desmoceratidie. He, of course, thinks it scarcely
safe to determine their taxonomic position on such an external
resemblance, and this question, consequently, may be treated, in
some detail, in the later part of this memoir.

1) Zittel: Griindziige d. Palteontologie 1895. *

2) Grossouvre: 1. c. p. ]6r>.

.".) Hyatt : Cephalopoda in Zittel's Text Book of Palpeontology translated by Eastman..
1890. P. 569.

CEPHALOPODA FROM THE HOKKAIDO. ol

No matter to what family our genus belongs, it is, undoubtedly
one of the most important groups of Ammonites for strati graphical
purposes, as though much restricted in the number of species its
geographical distribution is very wide.

The type of the genus is H. Gardeni Batly sp.,^^ recognized
in the Cretaceous (horizon f. of Griesbach) of Natal in South
Africa, the Ariyalur and Trichinopoly groups of South India, the
Nanaimo group (Horizon B of Richardson) of Vancouver, and
the Cretaceous of the Hokkaido. To the same species, an Am-
monites from the Cretaceous of Tunis has lately been referred by
Pervinquiere-' although with some doubt, and Ammonites sidcatus
Kner'^ from Galicia is by some palseontologists taken as a syno-
nym of H. Gardeni. If the above determinations are correct,
then it is a cosmopoliton species of the upper Cretaceous Am-
monites.

According to Grossouvre and Kossmat, this genus also in-
cludes Ammonites remhda Forbes, Amm. pseudoffardeni Schlüter,
H, Fayoli Gross, and -H. ^yelsc^li Gross. H. rembdaf^ occurs in
the Valudayur bed of South India, the Cretaceous of Natal
(horizon d) and ]Madagascar^^ while Amm. pseudogardeni^^ is found
in the zone of Scaphites binodosus of the German Cretaceous.
Amm. j)seudogardeni var. nodotum ScHLtJTER'^ from Braunschweig

1) Vide p. 32.

2) L. Pervixquikee: I'n facies particulier du Sénoinen de Tunise. 1895. (Bv review),
o) Favre: Descrip. des Mollusques foss. la craie des environs de Lemberg 1869. P. 12,

pi. IV., %. 7.

4) Stoliczka: Foss. Cepli. Cret. Rock South India. P. C^, pi. XXXIII., fig. 5; pi.
LXXI., fig. 9. Griesbach : Geol. of Natal. P. 63, pi. III., figs. 2, 3. Kossmat : Unters, ii. d.
Südind. Kreidefom. P. 124, pi. XVIII., fig. 9.

•5) Grossouvre: Sur quelques Fossiles Crétacés de Madagascar, p. 378, 1899.

6) Schlüber: Cephal. d. oberen deutsch. Kreide, p. 54, pi. XVI., figs. 3-6.

7) Schlüter : Podocrates im Senon von Braunschweig. Zeitsch. d. deut. geol. Gesell.
Bd. LI., p. 411.

32 ART. 2. H. YABE : CRETACEOUS

lias not yet been fully illustrated and the specific identification'
seems to be uncertain. Of the two species described by Gros«
souvRE from the Cretaceous of France, viz. H. Fayoli^^ and H.
Welschi,"^ the former is in the Campanien and the latter in the
zone of Mortoniceras texaimin.

The present writer proposes to bring two other species
under this genus. Tlie one is Amm. lagarus Redtenbacher^^
from Gossan which shows such agreement in every particular
with the above mentioned typical species that the writer has nO'
hesitation in stating their generic identity. The other is a new
form found in the Cretaceous of the Hokkaido, and is described
below under the name of //. antjiisttim.

HAUERICERAS GARDENI Baily.

1855. Amt/ionites Gardeni Baily : Cret. Foss. South Africa,.

p. 456, pi. XI., fig. 3.
1865. Amm. Gardeni Stoliczka : Foss. Cepli. Cret. Rock

S. India, p. 61, pi. XXXIIT., fig. 4.

1879. Amm. Gardetii Whiteayes : Mes. Foss. vol. I., pt,.
2, p. 102.

1880. Haploceras Gardeni Naumann : Kreidef. auf der
Insel Jezso, p. ID.

1890. Desmoceras Gardeiii Yokoyama : Verst. Jap. Kreide^

p. 184, pi. XX., fig. 10 a-c.
1895. Desmoceras Oardeni Whiteaves : On some Foss.

from the Nanaimi Group of the Vancouver Cret. p. 131.

1) Grosouvre: I.e. p. 221, pi. XXVII., fi,c(. n.

2) Grossouvre: I.e. p. 222, pi. NX XV., tig. 9.

Ö) Redtenbachkr: Cephalopoden Fauna d. Gosauseiiichten,. p. 112, pi. XXV., figs. 3 a, b.

CEPHALOPODA FPvOM THE HOKKAIDU. öO

1895. Dei<iiioceras (ktrdeni A\'hiteaves : On some Foss. from
tlie Nanaimo Group of tlie Vancouver Cret. P. 131.

1895. Hauericeras Gardenl Grossouyre : Les Ammonites
de la Craie supérieure. P. 220.

1897. Hauericeras Gardeni Kossmat : Unters, ü. d. S. Ind.
Kreidef P. 123 pi. XVIII. figs. 7 a, b, 8, 10.

1903. Hauericeras Gardeni Whiteayes : Mes. Foss. vol. I.,
pt. 5, p. 352.

The remarks of Yokoyama on the Japanese specimens given
in the above quoted work need no special additions, and every-
one who sees his figure is at once convinced that the Japanese
form must be identified with that of South Africa.

Localities : — LTrakawa and Ghashikots in Ikandai near LTra-
kawa, Hidaka Pro v.; the Sanushibe and the Shirikuru-pokomanai,
in the Söshibets, both branches of the ]\Iukawa, Iburi Prov.;
the Shi-kuruki, a tributary of the Yrd>arigawa and the Kikume-
zawa, a tributary of the Ikusliuml)ets, Ishikari Prov.; the
Opiraushibets and tlie Abeshinai, a tributary of the Teshiogawa,
both in Teshio Prov.; north of Chietomanai near Cape Soya,
Kitami Prov. Although this species is found everywhere in the
Pa chy discus-beds, the number of its individuals is always limited.

HAUERICERAS ANGUSTÜM m.

PI. v., figs. 5, 6.

In the JiMBö collection, there are two other specimens of
Hauericeras showing an aspect distinct from the preceding species.
Both are small but beautifully preserved, one of them showing

34

ART. 2. — H. YABE I CRETACEOUS

the suture lines in detail up to the end of the outer volution.
In the writer's collection, there are also two specimens, referable
to the same species ; one of them is somewhat larger and flattend
on the sides by pressure, l)ut agrees in point of essential characters
with the above examples.

The dimensions of these three examples measured are as
follows : —

(1)

Ratio.

(2)

Ratio.

(-)

3.65 cm.

100

0.45

CIU.

100

2.60 cm.

1.50 „

41

1..05

J)

39

1.10 „

—

—

0.75

»

22

0.55 „

1.15 „

01

1.10

H

31

0.85 „

Ratio.
100

42

21

Diaiueter.

Height of tlie last whorl.

Breadth of the last wliorl.

Width of umbilicus.

Involution.

1). A specimen from the Kikumezawa.

2). „ „ „ „ Urakawa; (pi. V., fig. 5.)

3). ,, ,, „ ,, Urakawa.

As is seen from the above table, the variation is very slight
in the three specimens except in size.

This species is readily distinguished from H. Gardeni Baily,^^
H. rembda Forbes'-' and H. Fayoli Grossouvre''^ by its narrower
umbilicus and higher whorl. It is, no doubt, much like H.
Welschl Grossouvre,^' with which at first glance it seems to
agree in nearly all its characters. But a closer examination
reveals a marked difference in the form of the whorl. In the
Japanese species the lateral sides converge gradually upward
to the ventral keel, while in the Grossouvre's figure of H.
Welschi, the sides converge very slowly to the ventral margin
whence they suddenly bend to the base of the keel.

1) See p. 32.

2) 1.0.

3) I.e.

4) I.e.

CEPHALOPODA FEOM THE HOKKAIDO. öO

This species agrees tolerably well with //. Gardeni in its
suture line, there being in both four auxiliary saddles. That
H. pseudogardeni differs from the present species notwithstanding
their great resemblance in external shape is easily seen in that
it has more numerous auxiliaries.

Localities : — The Kikumezawa, a tributary of the Ikushum-
bets, Ishikari Proy.; the Popets, a branch of the Mukawa, Iburi
Prov.; Urakawa, Hidaka Proy.

Horizon : — The only case in which the mode of occurrence
of this species has been determined, is in the Kikumezawa, where
it was found in the Pachyd i^cus-heds together with II. Gardeni.

DESMOCERAS Zittel.

DESMOCERAS DAWSONI Whiteaves yar. JAFONICA.

PI. v., figs. 3, 4.

1884. Haploceras Beudanti Whiïeaves : Mes. Foss., yol.

I., pt. 3, p. 205, pi. XXVI., figs. 1 and la.
1900. Desmoceras {Puzosia) Dmvsonl Whiteaves. Mes. Foss.,

vol. I., pt. 4, p. 286, pi. XXXVIL, fig. 3.
The specimens of this species in the writer's collection, ex-
cepting one which has been figured, are more or less distorted
by pressure. The suture line is also quite obscure on account of
the bad state of preservation.

36

AKT. 2. H. Y ABE : CRETACEOUS

Dimensions :

Diameter.

Height of the last wliorl.
Breadth of tlie List wliorl.
AViclth of umbilicus.

(1)

Ilatio.

(2)

Ratio.

{•■->)

Katio

11.3 cm.

100.

11.1 cm.

100.

13.2 cm.

100

5.7 „

50.

5.5 „

49.

6.8 „

51

4.5 „

38.

3.3 „

29.

—

—

2.1 „

16.

—

—

2..35 „

18

9^

The figured specimen from the Iknshumbets.

Uesmoceras Dawsoni Whiïeaves from Cumshewa Inlet;

measured from the figure on PI. XXVI. in Mes. Foss.

vol. I., pt. 3.
3) Desmoceras Dawsoni Whiteaves from Cumshewa Inlet;

measured from the figure on PL XXXVII. in Mes.

Foss. vol., I., pt. 4.
Shell discoidal, compressed on the sides and rounded on the
ventral surface. Umbilicus narrow, with a steep wall and more
or less rectangular margin. Involution very deep, about | of eacli
volution being covered. Section of the whorl elliptical, higher
than broad. Surface with eight distant, very flexuous constrictions,
parallel to the lines of growth. Constrictions narrow, bordered
behind by an elevated rim which becomes more prominent and
bends strongly forward on the ventral surface. On aged speci-
mens, these constrictions become strongly sigmoidal as is shown
in the anterior portion of the figured example, and the lines
of growth, otherwise appearing as delicate striœ, also become
coarser.

The suture line is only partly visible. The first lateral lobe
is about as long as the external lol)e and tripartite, but not so
deep and wide as in the case of Puzosia. The external saddle
as well as the two lateral ones are l)ipartite.

The general outline, surface marking and suture line o^ the
Japanese species much resemble those of Desmoceras Dawsoni

CEPHALOrODA FROM THE HOKKAIDO. 37

AVhiteaves from the Queen Charlotte Islands. But the shell is
somewhat broader and periodic constrictions are less numerous in
the Japanese species, so that the writer feels warranted in pro-
posing a new specific name for it.

According to Whiteaves, Desmoceras Daivsoni is quite
abundant in the Cretaceous rocks on the northern shore of
Cumshewa Inlet, while in Bear Skin Bay a few specimens are
.said to occur which are however distinguished from the typical
form by their rectangular umbilical margin. The latter is con-
sidered by that author as a variety to which the Japanese form
seems to approach more closely.^*

At first, Whiteaves""' identified his specimens with a European
form — Ammonites Beudanti Brongniart — which is also recorded
by STOLiczKa as occurring in the South Indian Cretaceous fauna.^'^
Afterwards, however, Kossmat^^ proved the distinction between
Amm. Beudanti and the allied form from South India. The
latter, as described by him, shows the suture line of the type of
Puzosia, and hence it is a species quite different from Ammonites
Beudanti which no doubt belongs to Desmoceras. He stated also
that Ammonites Beudanti of Whiteaves from the Queen Charlotte
Islands is a true Desmoceras. This view was afterward adopted
by Whiteaves'^ himself who in his later publication made a
new specific name for what he had formerly taken for Ammonites

1) J. C. MERRiA:sn later reported this species from Spanisch Gulch of the John Day
Basin.

2) D'Orbigny: Ter. Crétacés. Ceph. P. 278, pi. XXXIII., XXXIV. 1840.— Pictet :
Moll, des Grès Verts. P. ;», pi. II., lig. n. 1848.— QuE^'STEDT : Ceph. P. 222, pi. XVII.,
fig. 10. 1849.— Pictet et Campiche: S. Croix. P. 277, pi. XL., figs. 1-4. I860.— Pakona
■e BoxARELLi: Fossili albiani d'Escragnolles. P. 8() (34), pi. XI. (II.), fig. 6. 1896.

3) Stoliczka: Foss. Ceph. Cret. Rock South India. P. 142, pi. LXXI., figs. 2-4; pi.
LXXII. 1865.

4) Kossmat: Süd. Ind. Kreidet'. P. 120 (18-5). 1897.

5) Whiteaves: Mes. Foss. Vol. I. pt. 4., P. 286. 1900.

38 AET. 2. — H. Y ABE : CRETACEOUS

Beudanti, giving the Characters which distinguish it from P.
Stoliczhai Kossmat of Southern India. However, he diifers from
KossMAT in one })oint ; viz. in referring his species not tO'
Desmoseras, but to Puzosia, in which respect the present writer
is more inclined to adopt the opinion of Kossmat.

The reason why the present writer hesitates to bring this
species under Democeras is because its suture line is only im-
perfectly known. Kossmat who examined the Canadian specimen^
thinks that it has a typical Desmoceras suture as he says " mit
ganz typischer zur Naht gehender Lo1)enlinie." A part of the
suture line seen on a Japanese specimen gives no clue to dis-
tinguish it from Desmoceras. On the other hand, the suture
line on pi. XXVI in the White ayes' monograph which does
not appear to be very satisfactorily drawn, shows in some respects
a resemblance to that of Cleoniceras,^' a genus proposed by Parona
and Bonarelli to include Amm. Qeo7i. d'ORB., formerly as-
signed to Sonneratia, Amm. Bevdanti and a few other allied
forms. Such being the case, it is quite impossil)le at present to
determine whether this species is Desmoceras or Gleoniceras.

Desmoceras Dawsoni shows also a distant relation to Desmo-
ceras voyi Anderson"'^ from the Horsetown beds of California. The
latter possesses, however, a broader and somewhat inflated shell with
lateral sides gently converging to the ventral surface, and although
the periodic ribs are of the same nature as those of Desmoceras
Dawsoni, yet they are fewer in number. The suture line is described
as agreeing wâth that of Desmoceras lai'idorsatum Michelin."^'

1) Parona e Bonarellt : Fossili albiani d'Escragnolles, p. 85 foS"». 189().

2) Anderson: Cret. J)ep. Pacific coast, p. 100, pi. III., figs. 89, •)(). 1902.

o) D'Orbigny: I.e. p. 270, p1. LXXX.— Pictet : Foss. d. (h-es. Yerts. p. 44, pi. III.,
figs. 4, 5.— SxoLiczkA : I.e. p. 14.s, pi. LXXIV., figs. 1-4.— Kossmat : I.e. p. 107 (172), pi.
XIX (XXY), figs. 2-5.

CEPHALOPODA FROM THE HOKKAIDO. 39

The occurrence of Desmocerai Dmusoni in the Cretaceous of
the Hokkaido is of special importance, as among the few Japa-
nese forms which show some reUition to those of the North
American Pacific species, none is more closely related than the
present one, and it is also a remarkable fact that a bivalve,
very much like Thetis affinis Whiteaves'' from the C horizon
of the Queen Charlotte Islands is found abundantly together
with this Ammonites in the same layers.

Locality and horizon : — Several specimens of this Ammonites,
one of which is here figured, were obtained along the Ikushum-
bets 10 miles east of the Ikushumbets coal mines, Ishikari
Prov. Up to this time it has been met with only at the above
locality where it is quite numerous in a sandstone of the Thetis-
zone, associated with Thetis aff. affinis Whiteaves.

DESMOCERAS PORONAICUM m.

PL VL, figs. 1, 2.

Dimensions : —

Diameter.

2..5 cm.

100.

Height of the last whorl.

1.2 „

48.

Breadth of the last whorl.

1.1 „

44.

Width of umbilicus

0.4 „

16.

Shell small, discoidal, composed of a few compressed whorls,
which are nearly parallel sided and slope gradually to the round-
ed ventral surface. Involution very deep ; umbilicus narrow
with the perpendicular wall, exposing only a very small portion

1) Whiteaves: Mes. Foss. vol. I., pt. 3, p. 226, pi. XXX., tigs. 4, 4a, b; pt. 4. p. 290.

40 AET. 2. H. YABE : CRETACEOUS

of the inner whorls. Aperture somewhat elongated, higher than
broad. Surface of the shell apparently quite smooth, with a
faint groove on a portion of the anterior whorl where the shell
has been taken off'. The groove curves strongly forward in
crossing the ventral surfece where it is more distinct than on
the lateral sides. Suture line with four saddles and four lobes
on each lateral side, and the two on the umbilical Avail. These
saddles and lobes diminish very gradually in size and are all
similar in shape, the former being bipartite and the latter tri-
partite excepting the three or four innermost ones which are
always simple.

There is scarcely any doubt as to the existence of a close
relationship between this species on the one hand and Desmoceras
inane Stoliczka^' from the Utatur Group and 1). diphylloides
Forbes-^ from the Ariyalur and the Voludayur group of S. India
ou the other. D. diphylloides agrees fairly well with this species
in its general outline if we except the slightly narrower umbilicus
and more numerous saddles and lobes. D. inane is thicker than
the Japanese species although it much resembles it in the suture
line. Ammonites Sehvynknuis Whiteaves"" from the Horizon
A of Vancouver Island and 1). pyrenaicam CjROSSOUVRe'" from
the Santonien of France are also our distant allies.

From 1). Bawsoni Whiteaves var., the present species is
distinguished by the higher whorls and nar]-ower umbilicus. But
as the specimens of these two species examined by the writer
are quite different in size, the above distinctions must be taken

1) Stoliczka : I.e. p. 121, 111. LIX., iig. IS (non fig. 14).— Kossmat : I.e. p. 107, pi.
XIX., figs. 6-7.

2) Stoliczka: I.e. p. Ill», \A. LIX., figs. S-9.— Kos.smat : I.e., p. I(i8, pi. XIX., figs. 8-9.

3) Whiteavks: I.e. pt. 2, p. 104, pi. XII 1., fig. 1; pt. 5, p. .'îôl.

4) Gkossouvre: Aniru. <le la Craie supérieure, p. 168, pi. XXXVII., figs. 9 a, b, c.

CEPHALOPODA FEOM THE HOKKAIDO. 41

with some reserve. Also it may not be impossible, as it some-
times happens, that they are really individuals of one and the
same species at different stages of growth. However, for the pre-
sent, it seems preferable to treat them as two distinct species.

Locality and horizon : — This species is known only from
two small, probably immature specimens, ol)tained near the source
of the Poronai, Ishikari Prov. At this place, they were found
in a pebljle of a marly module, the exact horizon in which they
occur being unknown, but it is almost certain that the nodule
was derived either from the lower part of the upper Ammonite-
beds or from a laver 1)elow it.

42

ART. 2. H. YABE I CRETACEOUS

Bibliography.

F. M. Anderson:

D. Brauns:

P. Choffaï:
P. Choffat:

E. Favre :

A. Fritsch and U.

A. Fritsch:

L. C. Griesrach

A. DE Grossouvre

A. DE Grossouvre

A. DE Grossouvre

A. Hyatt

K. JiMBO

F. KOSSMAT.

Cretaceous Deposits of the Pacific Coast. Proceed.
Calif. Acad. Science,«. Ill series. Vol. II. No. 1. 1902.
Vorläufige Notiz, ü. Vorkommnisse der Juraforma-
tion in Japan. Mitth. d. deutsch. Gesell, f. Natur-
u. Völkerkunde Ostasiens. 1880.
Faune Crétaci(|ue du Portugal. Vol. I. Série IL 1898.
Contributions à la Connaissance Géologique des
Colonies Portugaises d'Afrique. I. Le Crétacique de
Conducia, 1903.

Description des Mollusques fossiles de la Craie des
environs de Leniberg. (By review).

ScHLOENBACH : Ccphalopoden der böhmischen Kreide.
1872.

Studien im Gebiete der böhmischen Kreideforma-
tion. Archiv der naturwiss. Landesdurchforschung
von Böhmen. IL Die Weissenberger u. Malnitzer
Schichten. 1877. III. Die Iserschichten, 1883. IV.
Die Teplitzer Schichten 1889. V. Priesener Schich-
ten. 1893.

Geology of Natal. Quart. Jour. Geol. Soc. London.
Vol. XXVII. 1871.

: Recherches sur la Craie supérieure. Mémoires pour
servir à l'explication de la Carte géologique détail-
lée de la France. 1893.

Sur r Ammonites perampluH. Bull. Soc. Géol. France.
Sér. III. Tome 27. No. 3. 1899.

: Sur quelques Fossiles Crétacés de Madagascar. Bull.
Soc. Géol. France. Sér. IIL Tome 27 1899.
Zittel's Text Book of Palaeontology, translated by
Eastjian. Cephalopoda. 1899.

Beiträge zur Kenntniss der Fauna der Kreideforma-
tion von Hokkaido. Pal. Abh. Bd. VI. (Neue Folge,
Bd. IL). Heft 3. 1894.
Untersuchungen fiber die Südindische Kreideforma-

CEPHALOPODA FEOM 'IHE HOKKAIDO. 46

tiou. Beiträge z. Pal. n. Geol. Oesterreich-Un-

garns ii. d. Orients. Bd. IX. 1895. Bel. XI. 1897.
¥. B. LIeek : Invertebrate Cretaceous and Tertiary Fossils of the

Upper Missiouri County. Rep. U. S. Geol. Surv. of

the Territories. Vol. IX. 1876.
tT. C. MerPvIAM : A ContriLution to the Geology of the John Day

Basin. Bull. Dep. Geol. Univ. Calif. Vol. II. No.

9. 1901.

E. Naumann : Ueber das Vorkommen der Kreideforraation auf der

Insel Jezo. Mitth. d. deutscli. Gesell, f. Natur-u.

V(ilkerkunde Ostasiens, 1880.
]\I. Neuäiayr : Die Ammoniten der Kreide u. Systematik d. Am-

monitiden. Zeitsch. d. deutsch, geol. Gesell, Bd.

XXVII. 1875.
M. Neumayr: Neues Jahrb. f. Min. Geol. u. Paléontologie. 1881.

Bd. II.
A. d'Orbigny: Paléontologie Française. Terrain Crétacés. Tome I.

1840-1841.
C. F. Parona e G. Bonarklli: P\)ssili alliiani d'Escragnolles, del Niz-

zardo e della Liguria occidentale. Pal. Italica. Vol.

II. 189(].
A. Peron : Les Ammonites du Crétacé supérieur de l'Algérie.

Mém. Soc. Géol. France. Tome VI. 1896.
L. Pervinquière: Sur un faciès particulier du Sénonien du Tunisie.

1898.
J. F. PiCTET et W. Roux : Mollusques fussiles des Grès Verts des environs

de Genève, 1846-49.
J. F. PiCTET et G. Campiche : Description des fossiles du terrain Crétacé

des environs de St. Croix. 1858-72.

F, A. Quenstedt: Petreftiktenkunde Deutschlands. Abth. I, Bd. I.

Cephalopoden, 1846-49.

A. Redtenbacher : Die Cephalopodenfauna der Gosauschichten in den
norddeutschen Alpen. Abb. d. k. k. Geol. Reichsanst.
Bd. V. 1875.

Ol. Schlüter : Cephalopoden der oberen deutschen Kreide. Pale-

ontographica. Bd. XXL, 1871. Bd. XXIV., 1874.

C L.Schlüter: Fodocrates im Senon von Brannschweig u. Ver-

breitung u. Benennung der Gattung. Zeitsch. d.
deutsch, geoloç;. Gesell. LI. 1899.

44

ART. 2. — H. Y A BE : CRETACEOUS

J). Sharpe :
F. Stoliczka
C. A. White

J. F. Whiteaves:
J. F. Whiteaves :

R. P. Whitfield:

H. Wooü :
M. Yokoyajia :
K. A. V. Z ITT EL ;

K. A. V. ZiTTEL

Mollusca of the Chalk. Palaeontograph. t^oc. L(Mi-
don. Vol. VII-IX. 18.53-185.").
The Fossil Cephalopoda of the Cretaceous Rocks of
Southern India. Palaeontologia Indica. 18G5.
Contributions to the Palaeontology of Brazil, com-
prising Descriptions of Cretaceous Invertebrate Fos-
sils, mainly from the Provinces of Sergipe, Per-
nambuco, Para and Bahia. Archives do Museu
Nacional do Eio de Janeiro. Vol. VII. 1888.
Mesozuic Fossils. Ah)1. I. Pt. 1-5. 1876-1902.
On some Fossils from the Nanaimo Group of the
VanC(Hiver Cretaceous. Trans. Roy. Soc. Canada.
2nd Series. Vol. I Sect. IV 189.5.
Observations on and emended Description of Hetero-
ceras simplicostafum Whitfield. Bull. Amer. Mus.
Nat. Hist. XVI. 1902. (By review).
The Molhisca of the Chalk Rock. Quart. Jour Geol.
Soc. LII. 1896.

Versteinerungen aus der Japanischen Kreide. Pale-
ontographica. Bd. XXXVI. 1890.
Handbuch der Paléontologie.
Grundzüge der Paleontolode. 1894.

CEPHALOrODA FK03I THE HOKKAIDO.

45

Contents.

Xarrilites

Turrililes cfr. Bergeri, BrongiART

„ Koiiiotdi ni

He/icocera-'i scnlare m

„ (?) venustuiii ni

]-rderocer(i-<i (?) Oshimai m

„ (?) Otsukai m

„ (?) „ var. inultimsttda

,, (?) japoidcuiii m

„ orientale m....

Nipponltes mirahi/is m

Jncertœ Sedis

Amuwnäes Kotoi m

■Olcostephanu'<

Olcostephanus iinirus m

ITauericeras

Hauericems Qurdeni Kaiia'

„ atKjuHlum m

Desiiioccra-^

Urftmooras Daiusonl Whiteaves var. japoniea

„ poronaicuin m.

Biljliography

Page.

1

4

7

9

]1

12

14

16

17

If)

20

25

26

27

28

29

;J2

33

35

35

39

42

-TS^-i

H. YABK.

CRETACEOUS CEPHALOPODA FROM THE HOKKAIDO.

PLATE I.

Turrilites.

Plate I.

Turrîlites Komotai m. Pp. 7-9.

Fig. 1. ISide view of the type, preserved in the Sei. Coll. Mus. From lower
Jcanthoc era fi-zone of the Ikuslumihets, Ishikari Prciv. Nat. size.

Jour. Sei. Coll. Vol. XX.. Art. 2. PI. I.

.1. IMznkl .1.1.

H. Y,.l..: Crcl. Ccph. U.M'iX,. 11. II.

II. Y ABIC.

CRETACEOUS CEPHALOPODA FROM THE HOKKAIDO.

PLATE IL

Turrilites.

Plate II.

TurrUites Komotai m. Pji. 7-9.

Fig. ], Upi)er view of the specimen drawn in PI. I.

Jour. Sri. („II. Vol. XX.. Ail. 2. PL 11.

.;. Mh„l.-: ,i..i

u. y.iin: tw.1. r,.pi,. ir.Jchwiû. i>t. u.

ir. YABE.

CRETACEOUS CEPHALOPODA FROM THE HOKKAIDO-

PLATE III.

Turrilites, Helicoceras and Heteroceras.

Plate III.

(All the specimens here figured are in Sei. Coll. Mus.)
Turrilites aff. Bergerl Bhgt. Pp. 4-7.

Fig. 1. Side view of the type. From lower Annuonite-l)eds of the Popets,
Il)uri Prov. Nat. size.

Helicoceras scdhire m. Pp. 9-10.

Fig. 2. Upper view of the type. From ScapJiites-hedn of the (Jpiraushibets,

Teshio Prov. Nat. size.
Figs. .3 a, b, c. Upper and lower views and cross-section of another specimen.

From Scaphi'tes-heds of the Opiraushibets. Nat. size.

Helicoceras venustum m. Pp. 11-12.

Fig. 4. Side view (slightly oblique) of a large specimen. From Scapliites-
beds of the Opiraushibets. Nat. size.

Heteroceros Oshimai m. Pp. 12-14.

Fig. .5. Side view of the type. From upper Amuionite-beds of the Iku-

shumbets, Ishikari Prov. Nat. size.
Fig. 6. Side view of a smaller si)ecimen. From upper Ammonite-beds of

the Shikuruki, Ishikari Prov. Nat. size.

Heteroceros orientale m. Pp. 19-20.

Fig. 7. Side view of the type. From the upper Ammonite-beds of Ura-
kawa, Hidaka Prov. Nat. size. In this drawing, the larger end
of the shell is shown below and the smaller end above.

Hetcroceras japonicurii m. Pp. 17-18.

Fig. 8. Side view of the type. From up})er Ammonite-beds of the Yii-
barigawa, Ishikari Prov. Nat. size.

Heteroceras Otsiikai var. multicostata m. Pp. 16-17.

Fig. 9. Side view of the type. From ScapJiites-heds of the Opiraushibets.
Nat. size. On this figure, the larger end of the shell is shown
below and the smaller end above.

Jour. Sei. Coll. Vol. XX., Art. 2. PI. JH.

•--.^T^'

//

/XK-- s.

J-l^- 7-

fic I.

Fi.^- 4.

Fig. Ç.

t'ii^.

Ù- j'

Fig. 6.

IL Tube: Orot. Ceph. Ilokkaidö. Ft. IL

PHOTO. BY K. OÜAWA.

H. Y A BE.

CRETACEOUS CEPHALOPODA FROM THE HOKKAIDO.

PLATE IV.

Heteroceras and Nippon ites.

Plate IV.

(All the specimens, here figured, are in Sei. Coll. Mus.)
Heteroccraf< OtmUMÎ m. I'p. 14-1(3.

Fig. 1. Side view of the type. From FicchyJiscus-heds of the Kiknmezawa,
Ishikari Prov. Nat. size. Slightly restored. In this figure, the
larger end of the shell is shown below and the smaller end ahove..

Fig. 2. Side view of a portion of whorls of another specimen. From upi>er
Ammonite-beds of the Yiibarigawa. The siphon comes nearer the
lower margin toward the anterior end. Nat. size.

Heterocera.'^ Otsiil-ni m. var. mulficodafa m. Pp. lG-17.

Fig. 3. Side view of a specimen with numerous ri])S. From upper Am-
monite-beds of the Yübarigaw^i, Nat. size.

Kij>i)<)iiiti'S mirahUis m. Pit. 20-25.

Fii;s. 4 and i^, A'iews from two different directions of the type. From

upper Annnonite-beds of the Opiraushibets.
Fig. I). Cross-section of the same. Nut. size. s. siphon.
Fig. 7. ^Yire-nlodel to show the mode of the coiling the whorls.

Jour. Sei. Coll Vol. XX., Art. 2. PI. IV.

%

^(c- .?•

/'7;

-^■.

/^v. <5.

V .-.f ■ -^

- n^

Fig. 7. Fig. I.

H. Yabc: t'rct. Ceph. IMLiUlj. Pi. If.

PHOTO. By K. ogawa.

II. YAIîE.

CRETACEOUS CEPHALOPODA FROM THE HOKKAIDO.

PLATE V.

Helicoceras, Desmoceras and Hauericeras.

Plate V.

(All the specimens, here figured, are in Sei. ColL Mus.)
Helicoceras vemtstiim m. Pp. 11-12.

Fig. 1. Upper view of the type. From Pachydiscu.s-heds of the Sanushibe^

Iburi Prov. Nat. size.
Fig. 2. Diagram of a suture line, drawn from the same specimen. s.

si phonal line. x 4.

Desmoveras Daicsoni Whiteaves var. japonica. Pp. 35-39.

Figs. 3 a, b. Side and front views of the type from the Thetis-zone of

tlie Ikushumbets. Nat. size.
Fig. 4. Diagram of a suture line, drawn from another specimen. Nat. size..

Hauericeras anriiistum m. Pp. 33-3.3.

Figs. 5 a, b. Side and front views of the type. From the Upper Am-
monite-beds of Ikandai, near Urakawa. Nat. size.

Fig. 6. Diagram of a suture line, drawn from the same specimen, g.
siphonal line; u. m. umbilical margin, x 4.

Jour. Sei. Coll. Vol. XX., Art. 2. PI. y.

& ]

1

a. b.

Fig. 6.

H. Yabe: Cret. CVpl. Hokkaido. Ft. U.

PHOTO. BY K. CGAWA.

H. YABE.

CRETACEOUS CEPHALOPODA FROM THE HOKKAIDO.

PLATE VI.

Desmoceras, Olcostephanus, Nipponites, Heteroceras
and an undetermined Genus.

Plate VI.

(AU the specimens, here figured, are in Sei. Coll. Mus.)
Desmoceras poronaicum. m. Pp. 39-41.

Figs. 1 a, b. Side and front views of the type from the Upper Ammonite-
beds of the Poronai, Ishikari Prov. Nat. size.

Fig. 2. Diagram of a suture line, drawn from the same specimen, s.
siphonal line ; u.m. umbilical margin. x 3.

Ammonites Kofoi m. F[). 26-27

Fig. 3. Side view of the type. From the Yûbarigawa. Slightly restored.

Nat. size.
Fig. 3 a. Cross-section of the whorl. Nat. size.
Fig. 4. Diagram of the suture line, drawn from the same specimen, s.

siphonal line ; u.m, umbilical margin. x 2.

Olcostephanus unicus m. Pp. 28-29.

Figs. .5 a, b. Side and front views of the type. From the Takambets,
Ishikari, Prov. Nat, size.

Nipponites; mirahilis m. Pp. 20-2.').

Fig. G. Diagram of a suture line, drawn from the specimen figured in PI.
IV. s. siphonal line. Nat. size.

Heteroceras Otsukai m. Pp. 14-16,

Fig. 7. Diagram of a suture line, drawn from the specimens figured in PI.
IV., fig. 2. s. siphonal line. Nat. size.

Hderocerm Otfiiûxii var, mnJUcoi^tato.. Pp. 16-17.

Fig. 8. Diagram of a suture Hue, drawn from the specimen figured in PI.
IV., fig. 3. s. siphonal line; a.s. antisiphonal line, x 2.

Jour. Sei. Coll. Vol. XX.. Art. 2. ri. VF.

H. Yale: Cret. Ceph. Hokkaido. Ft. II.

PHOTO. BY K. OGAWA.

JOUKNAL OF THE COLLEGE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 3.

Kevisio Euphorbiacearum et Buxacearum
Japonicarum.

aiictore
B. Hayata.

cum tahulis 6.

Collectiones plantarum quae hujns operis fundamentum ef-
ficiunt, in Universitatis imperialis Tokyensis herbario servantur.
Toto corde gratias publice ago viris doctissimis, qui peritissimis
consiliis magnum mihi auxilium prœbuerunt. Imprimis vero,
me illustrissimo Prof, Dri. J. Matsumur^, qui curam adhibuit
maximum ad hoc meum opus, gratiam maximam debere profiteor.

Z ART. 3. — B. HAYATA : REVISIO

EUPHORBIACE^.

Conspectus Geneimm.

Ovula in localis gemina.

Embryo magnas, parce minor qnam albumen,
Sepala "t iiiibricata.
Folia inrlivisa.

Fl. ^ ad nodos v. axillas glomerati.

Styli erecti v. recurvi, tenues v. apice solo dilatati, cocci 2-
valvati v. bacca 3-loculans.

Rudimentani ovarii evolatum.

Seminum testa tenuis ventie non cava. Embryo rectas,

1. Securinega.

Seminum testa Crustacea ventre duplicata inter

laminas cava. Embryo incurvas

2. Fluvaggea.

Rudimentum ovarii 0.

Discus e vol Utas 3. Plnjllanihns.

Discus 0.

Capsula. Calyx ^ dilatatiis. ...4. Glochidioii.
Bacca. Calyx 'V tarbinatus lobis v. dentibas

6 inflexis fere clausas 5. Breynia.

Styli valde dilataLi, patentes. Fratus di'upaceus, indehiscens,

abortu 1-spermus 6. Putravjiva.

Fl. 'V racemi 7. Antidesma.

Folia digitatim foliolata 8. Bischoffia.

Sepala '^ valvata 9. Bridella.

Embryo parvas, albumine quater ad sexies brevior. 10 Daphniphyllum.

Ovula in loculis solitaiia.
Involucrum non cyathiam.

Filamenta in alabastro apice inflexa 11. Crofon.

Filamenta in alabastro ei'ecta.

Fl. 't' sei:)alis valvatis, apetali. Spiere v. racemi.
Filamenta non ramosa.

Antlierœ subglobosœ v. oblongre.

Stamina 16-20. Antherarum lociili a basi distiucti
primuiu pendali demum ascendcntes, superne long-

EUPHOßBIACEAHÜM ET BUXACEARUM JAPONICARUM. 3

itudinaliter déhiscentes. Filaincnta libera

12. Mercurialis.

Stamina numerosa. Anuierœ prope bnsiu dorsifixœ,
localis ilistinctis parallelis connectivo lato separatis
V. connectivo parvo longitudinaliter dehiscentibus.

Filamcnia libera 13. Mallolus.

Stamina 8. Antherœ dorsifixa3, loculis distinctis
parallelis longitudinaliter dehiscentibus. Filamenta
basi brevissme in annulum connata. ...14. Alchornea.
Stamina numerosa. Anthera; dorsifixœ, connectivo
latiusculo breviter ulti-a loculos producto, loculis
lateralibus transverse didymis. Filninenta libera. ...

Ib. Cleidion.

Stamina 6-10. Anthera3 terminales, breves, 4-locuIares.

Filamenta liijera 16. 3Jacarcmga.

Antherie vermiibrmes 17. Acalypha.

Filami-nta ramosissima.

Herbœ 18. Eicinus.

Frutices 19. Homonoia.

Fl. "t sepalis imbricatis, petaliferi. Panicuhe corymbosaï

20. Aleurites.

Fl. "^ sepalis imbricatis, apetali. Spicie v. racemi.

Fl. glomerato-f asciculati 21. Gelonium.

Fl. spicati V. racemosi.

Calyx t a basi 8-partitus. 22. Excoecaria.

Calyx 'S" brevissinie lobatiis 2o. ^apinm.

Involucrum cyathium 24. Euphorbia.

1. Securinega Juss.

Bâillon, Hist. d. PI. V. 241 ; Benth. et Hook. Gen.
PL III. 27Ö ; Pax, in Engl, et Prantl. Nat. Pfl.
Farn. III. 5 18.
Flores dioici, apetali. Fl. J : sepala 5, imbricata. Disci

4 AKT. 3. B. HAYATA : EEVISIO

glandulœ 5, staminibus alterni. Stamina 5, sepalis alterna, filamentis
liberis ; antherse erectse, localis distinctis parallelis longitadinaliter
dehiscentibus. Ovarii rudimentum evolutum, apice 2-3-fidiim.
Fl. $ : calyx maris. Discus integer. Ovarium o-loculare, stylis
distinctis, recur vis, 2-fidis, ovulis in localis geminis. Cajpsula
sicca, in coccos 2-valves dehiscens. Seminum testa membranacea
minute reticulata, albumen carnosum ; embryo centralis, cotyledon-
ibus planis latis. Frutices ramosi. Folia alterna, intégra. Flores
ad axillas glomerati, J numerosi, ? pauciores, pedicellati.

1. S. fluggeoides Muell. Aeg. (Tab. I. A.). MuELL.

Aeg. in DC. Prodr. XV. 2, 450 ; Forbes et Hemsl. Ind. Fl.
Sin. II. 426 ; Henry, List. PL Form, in Trans. Asia. Soc.
Jap. XXIV. suppl. 82; L. Diels, Fl. centr. China, in Engl. Bot.
Jahrb. XXIX. 426.

S. japonica Miq. in Ann. Mus. Bot. Lugd. — Bat. III. 28 ;
Franch. et Savat. Enum. PI. Jap. I. 425.

Phyllanthus japonicus Muell. Arg. in Linnsea XXXII. 16.

Nom. Jap. Hitotsuba-hagi.

Ramuli foliigeri 1-5 pedes longi, graciles subflexuosi. Stipulse
ovatse, acuminatœ, fusca3. Petioli 4-5 mm. longi. Folia 3-5 cm.
longa, 1-2 cm. lata, oblongato-elliptica, obtusa, basi acuta, rigide
submembranacea, supra lucida, subtus glaucescentia. Flores ^ in
fasciculis 2-8, $ numerosi. Pedicelli $ 5-10 mm. longi apice in-
crassati, % 3-5 mm. longi, tenues. Calyx ^ 2 mm.; J li mm.
longus ; sepala oblongo-ovata, obtusa, margine latiuscule hyalina.
Ovarii rudimentum cylindricum patule et breviter trifidum glabrum,
1-2 mm. longum. Ovarium globosum 1-2 mm. longum, li mm.
diametro aequans, glabrum ; styli basi breviter connati, intus
canaliculati, apice bilobi. Fructus primum plus minusve carnosus,

EUPHORBIACEAEUM ET BUXACEARUM JAPONICAEUM. Ö

demum capsulari-aperiens, excarpio carnoso, enclocarpio osseo.
Capsulae 4^ mm. latje, oi mm. longse subteretes. Seraina obtuse
trigona, levia, reticulata.

DisTRiB. In Japonia mediana el australi vulgaris.

Hab. Liu-kiu : Kuniesima, leg. H. KuROiwA, anno 1898.

Kiushu : Kumamoto, anno 1901.

Hondo : prope Tokyo.

2. Fluggea Willd.

WiLLD. Sp. PI. IV. 757 ; Endl. Gen. PI. 1122 ; Benth.

et Hook. Gen. PI. III. 276 ; Pax. in Nat. Pfl. Fam.

III. 5, 18.

Flores dioici apetali. Fl. J: Fl. $: sepala 5, subpeta-

loidea, imbricata. Discus planus, annularis, dentatus. Ovarium
o-loculare ; styli distincti, recurvi, 2-fidi, ovula in loculis gemina.
Fructus subsuccosus, pericarpio tenui indéhiscente. Semina sub-
triquetra, dorso rotundata, testa Crustacea duplicata inter laminas
cava ; albumen baud copiosum, circa cavum incurvum ; embryo
incurvus, cotyledonibus latiusculis planis. — Frutices. Folia alterna,
disticha, petiolata, integerrima. Flores minimi, axillares, pedi-
cellati.

1. F. microcarpa Blume ; MiQ. Fl Ind. Bat. 1-2, p. 356;
Hook. f. Fl. Brit. Ind. V. 328 ; Forbes et Hemöl. Ind. FL Sin.
II. 427 ; Henry, List PL Form. 83 ; L. Diels in Engl. Jahrb.
XXIX. 427.

Securinega ohovata Muell. Arg. in DO. Prodr. XV-2, 449;
Benth. Fl. Austral. VI. 115.

6 ART. 3. — B. HAYATA : EEVISIO

Flueggea Sinensis " Batll. Etude Gene. Eupliorb. 592."

Phyllanthus Wightianus Müell Arg. in Liniuxea XXXII. 6.

P. virosus WiLLD. Sp. PL IV. 578; Roxb. Fl. Ind. III. 659.

P. retusus Roxb. Fl. Ind. III, 657.

Chorizandra pinnata Wight Ic. Ind. or. t. 1994.

Nom. Jap. Sliima-hitotsubahagi.

Frutices. Rami cinerascentes v. griseo-pallidi, graciles, glab-
res, plus minusve acute tetragoni. Folia petiolata, stipitata,
obovoidea, oblongo-elliptica, obtusa v. rotundata, basi acuta, costis
secundriis 10-12, subtus glaueescentia, 2i-4 cm. longa, 1-2 cm.
lata, petiolis 3-4 mm. longis. Fl. ^ : sepala I2 mm. longa.
Ovarium globosum, styli distincti, patuli, bifidi. Fructus globosus
3mm. diametro sequans. Pedicelli fructigeri 22 mm. longi. Semina
2 mm. longa, 1 2 mm. lata.

D18TRIB. Tropica et subtropica Asia, troj)ica Africa Australiaque.
Hab. Formosa: Tam-tsui, leg. T. Making, anno 1896; Kelung, leg. T.
Making, anno 1896 ; Kut-chbioli, leg. K. Miyake, anno 1899.

3. Phyllanthus L.

LiNN. Sp. PI. ed-2. 67 ; Endl. Gen. PL 1120 ; Muell.

Arg. in DC. Prodr. XV-2, 275 ; Baill. Hist. d.

PL V. 252 ; Benth. et Hook. Gen. PL III. 272 ;

Pax in Nat. Pfl. Fam. III-5, 18.

Flores mouoici, apetali. Discus varius. Fl. ^ : sepala 5-^),

V. 4 distincta, v. rarius basi breviter connata, imbricata, plus

minusve distincte 2-seriata, petaloidea. Stamina in centro flores

3, 2, V. 5, filamentis liberis v. in columnam centralem connatis ;

antlierarum loculi paralleli v. divergentes, extrorsum longitudinaliter

EUPHORBIACEAEUM ET BUXACEARUM JAPONICARUM. 7

V. liorizontaliter déhiscentes. Kiidimentiim ovarii 0. Fl ? : calyx
maris v. rigidior ; ovarium o-v. 6-loculare ; styli distincti v. plus
minus connati, integri sœpius 2-fidi, erecti v. recurvo-patentes,
subulati, V. ^^»lanodilatati, superne introrsum stigmatosi ; ovula in
loculis gemina. Capsula sicca v. carnosa, demum sœpissime in 2
coccos 2-valves dissdiens, rarius vix dehiscens. Semina trans-
versini o-augularia, dorso convexa, angulo interiore recta, erarun-
culata, testa membrancea v. Crustacea ; albumen carnosum, embryo
rectus V. leviter incurvus, cotyledonibus planis rectis. — Herbse,
fructices. Folia integerrima, alterna, disticha. Flores parvi, axillares
V. ad nodos vamulorum glomerati, J subsessilis v. distincte pedi-
cellati, pauci, Ç in eadem axilla v. in ramulo distincto pauci v.
solitarii, longius pedicellati.

Conspectus Phyllanthoriim Specter urn.

Herbœ v. siiffrutiees.
Stamina 3.

Filamenta tota longitudine monadelj^ha v. superne brevissime libera.

Sepala 5 1- P Nirvri.

Sepala 6 2. P. urinaria.

Filamenta libera, basi brevitcr connata 3. P. simplex.

Stamina 2.

Disci glandiilse liberœ 4. P. Matsivmurœ.

Disci glandulse connatœ 5. P. liukiuensis.

Frutices.

Stamina 2 6. P. flexuosus.

Stamina 5 7. P. reticulatus.

Arbores 8. P. Niinamii.

1. P. Niruri Linn. (Tab. I. B.). Linn.S p. PL ed.

-2, 1392; Thunb. Fl. Jap. 6Q', Aiton, Hort, kewensis, V. 335;
RoxB. Fl. Ind. 559 ; Hook, et Arn. Bot. Beech. Voy. 210 ; Sieb.

8 ART. 3. — B. HAYATA : REVISTO

et Zücc. Fl. Jap. Farn. Nat. Sec. Prim. 144 ; Wight, le. PI. Ind.
or. t. 1894; Miq. Fl. Ind. Bat. 1-2,369; Benth. Fl. Hongk.
311 ; MuELL. Arg. in Linnaea XXXII. 43; Grisebach, Fl. Brit.
West. Ind. 34; Miq. Prol. 24; Hemsl. Bot. cent. Ameri. III.
104 ; MuELL. Arg. in DG. Prodr. XV-2, 406 ; Hemsl. in Voy.
Ghallenger, Bot. 1-2, 63, Part 3, 187 ; Hillebrand, Fl. Hawai,
402 ; Hook. f. FI. Brit. Ind. V. 298 ; Forbes et Hemsl. Ind.
Fl. Sin. II. 422; Henry, List PI. Form. 82 ; L. Diels, in Engl.
Jahrb. XXIX. 29.

Nom. Jap. Kidachi-homihaiisô.

Nymi^hanthus Niruri, LouR. Fl. Gochin eh. ed-2, 545.

Herbœ v. snffrutices. Bamnli teretes levés. Folia oblonga
elliptica, obtusa, v. mucronata, 3-5 mm. longa, 2-2 è mm. lata,
snbtus pallidiora, membranacea. Stipulœ ovatœ v. lineari-lanceolatœ,
snbulato acnminatœ. Flores gemini, omnes breviter pedicellati. FI.
J : sepala 5 ; stamina 3, filamentis tota longitudine in columnam
connatis. Antherarum locnli horizontaliter déhiscentes. Glandulae
distincte, stipulœ peltata3. FI. Ç : calyx maris. Ovarium
globosum, trisulcatum : styli brevissimi apice revoluto-bilobi. Glan-
dulae connatse, 5-lobœ. Calyx fructiger 1,5 mm. longus, 1 mm.
latus. Capsulœ depresso-globosse, 2 mm. latœ, ambitu teretes,
lœves. Semina longitrorsum costulata, transversim minutissime
striolata.

DiSTRiB. in Guinea, Madagascaria, India orientali, Java, Borneo, Phillip-

pinis, Japonia vulgaris.
Hab. in Japonia australi.
Formosa : Kachirai-sha, leg. K. Miyaké, anno 1898 ; Pang-kang, Sha-gio,

leg. K. MiYAKE, anno 1898.
Hondo : Prope Tokyo, anno 1880.

2. P. urinaria Linn. (Tab. I. G.)— Linn. Sp. PI. ed-

EUPHORBIACEARUM ET BÜXACEAEUN JAPONICARTJM. 9

2, 1393 ; WiLLD. Sp. PL IV. 583 ; Lour. FL Cocliinch. 677 ;
MiQ. FL Ind. Bat. I.-2, 3G9 ; Muell. Arg. in Linnsea XXXII.
19 ; Benth. FL Hongk. 310 ; Müell. Arg. in DC. Prodr. XV.-
2, 364 ; Benth. FL AustraL 102 ; Miq. ProL 291 ; Franc h. et
Savat. Enum. PL Jap. I. 426 ; Hook. f. FL Ind. V. 293 ; Maxim.
in Engl. Jahrb. VI. 59 ; Forbes et Hemsl. Ind. FL Sin. IL 423 ;
Henry, List PL Formos. 82 ; L. Diels, in Engl. Jahrb. XXIX. 29.

P. lepidocarpus, Sieb, et Zucc. Fl. Jap. Farn. Xat. Sect.
Prim. 143 ; Wight, le. PI. Ind. or. t. 1895, f. 4 ; Miq. in Ann.
Bot. Mus. Lugd.-Bat. III. 127.

Xom. Jap. Komikansö.

Suffrutices 1-2-pedales, ramulis alto-angulosis. Hamuli flori-
geri puberuli. Folia in ramulis disticha, anguste oblongo-elliptica,
brevius acuminata, basi subacuta, margine plus minusve scabra v.
ciliolato-serrulata, vulgo dense approximata, in ramulis oblique
inserta, 5 ram.-l.cm. longa, 3-4 mm. lata. Stipulœ triangulari-
acutœ. Fl. J : sepala 6 ; stamina 3, filamentis tota longitudine
in columnam connatis, apice brevissime liberis; antherge verticaliter
birimosse, liberse. Glandulae liberie, subsessiles, peltatœ. Flores
J depauperato-fasciculati, in superne ramulorum dispositi, exgui
vix 2 mm. longi. FL ? : calyx maris. Ovarium globosum
sub-squamoso-papillosum, stylis complanatis, recurvis bifidis, ramis
hamato-recurvis. Glandulse in urceolum connatœ lO-lobœ, lobis
triangulari-acutis. Flores ? in axillis foliorum solitarii, in inferi-
oribus partibus ramulorum subsessiles. Capsulse subsquamoso- ver-
rucosa?, fere sessiles, depresso-globosœ, 2è ram. latse, ambitu teretes.
Calyx fructiger apertus vix 1.5 mm. diametro œquans. Semina
transversim costulata.

DisTRiB : In insula Ceylonia, Java, India orient., China, Japonia.
Hab. in Jajwiiia australi medianave.

10 ART. 2. — B. HAYATA : REVISIO

Formosa: Tong-si-kak, leg. K, Miyake, nnno 1898; Sin-tiam-koe, Kutsu-
shaku, leg. K. Miyake, anno 1898 ; Kelung, Taipea, leg. T. Making,
et C. OwATARF, anno 1896 ; Paehina, leg. T. Niinami, anno 1896.

Liu-kiu : Naha.

Hondo : Prov. Tötömi, in tractu Iwata, leg. K. Hisamatsu ; prope Tokyo,
Hort. Bot.; Prov. Suwau, in tractu Yoshiki, leg. G. I^ikai, anno 1901.

3. P. Simplex Retz. (Tab. I. D.) Willd. Sp. PL

IV. 573 ; RoxB. Fl. Ind. III. 634 ; Muell. Arg. in DC. Prodr.
XV.-2, 391 ; Hook. f. Fl. Brit. Ind. V. 295 ; Benth. Fl. Au-
stral. VI. 111; Seeman, Fl. Vitiensis, 220; Forbes et Hemsl.
Ind. Fl. Sin. IL 423 ; Hexry, List. PI. Formos. 82 ; L. Diels,
in Engl. Jahrb. XXIX. 427.

P. simplex y chinensis Muell. Arg. in Linnœa XXXII. 33.

P. anceps Vahl ; Franch. PI. David. 266.

Nom. Jap. Shima-koniihansd.

Herbœ v. sufTrutices semipedales, caulibas simplicibus v.
parce ramosis e caudice vulgo namerosis erectis v. adscendentibus
tota V. fere tota longitudine folii-florigeris. Folia disticlia, glabra,
l)reviter petiolata, intégra, penninervia, bistipitata, lineari-lanceolata
utrinqne obtusa v. apice acuta, 8-10 mm. longa, 3-4 mm.
lata, petiolis è-lèmm. longis. Stipulœ triangulari-ovatœ, acutœ.
Fl. J : sepala 6 ; stamina 3, filamenta oblique inserta, apice
recurva, basi leviter connata. Pimœ antherarum extrorsie, in-
clinatione connectivorum horizontales. Glandulae 6, liberal, or-
biculares, sessiles. Pedicelli J gemini v. terni è-1 mm. longi.
Fl. :Ç : calyx maris. Ovarium papillosum depresso-globosum ;
styli 2-fidi ; stylorum rami graciles, divergentes, ovario adpressi.
Glandulœ in urceolum crenato-lobatum, undulatum, connatœ. Pe-
dicelli longiores, 1-2 mm. longi. Capsulœ levés profunde trisulcatse.
Semina secus lineas longitrorsas reguläre puncticulato-verruculosa.

EUPHORBIACEARUM ET RUXACEARUM JAPONICARUM. 11

DiSïRiB : In India orient., Java, Borneo, Celebes, Philippini^ä, insulis Ma-

rianis, Formosa.
Hab. Formosa: Takao, leg. K. Miyake, anno 1898.

4. P. Matsumurse sp. nov. (Tab. I. E.)

Nom. Jap. Hime-mikansd.

Herbœ, caulibus simplicibus v. pauci ramosis. Rami florigeri
ancepites, subulato-compressi. Folia clisticlia, 8-10 mm. longa,
4-5 mm. lata, glabra, intégra, bistipitata, ovato-lanceolata, v.
lineari-lanceolata, acuta, basi obtusa, subtus pallidiora, petiolis h~
1 z mm. longis. Stipulœ basi retorsœ sagittatee, integrœ, lanceo-
acuminatse. Fl. J : sepala 4 ; stamina 2, filamentis fere connatis,
superne breviter liberis. Kimre antherarum inelinatione connec-
tivoruni horizontales. Glandulae 4, liberie, obconicie, sessiles.
Pedicelli 1-2 mm. longi, tenues. Fl. ^ : sepala 6 ; ovarium
globosum, levé ; styli patentes apice 2-lobi, lobis divergentibus,
non ovario adpressis. Glandulie liberie, 6, subcylindricse. Pedicelli
$ 1-2 mm. longi, acute transversimque 3-angulosi, superne valide
incrassati. Capsula depresso-giobosœ, glabrae, 2 mm. latse, 1 mm.
longœ, leviter papillosse. Semina secus lineas longi trorsas sparce
irregulariterque reticulata.

A P. simplici differt in floribus J, staminibus 2, sepalis 4,
glandulis liberis, stylis non ovario adpressis.

Hab. In Japonia australi vulgaris.

Prope Tokyo, anno 1880; Horinouchi, anno 1881. Prov. Iwashiro, in
tractu Aidsu, Yumoto, anno 1819. Prov. Suwau, in tractu Yoshiki, leg.
G. NiKAi, anno 1984 ; Insula Tsushima: prope Izugahara, Sliinimi, leg.
Y. Yabe, anno 1901.

5. P. liukiuensis ÄIatsumura. (Tab. I. E.)

Nom. Jap. Hana-komikanboku.

Frutices divaricati, ramis apice caulium multis, glabris,

12 ART. 2. B. HATAYA : REVISIO

gracilibus. Ramuli filiformes, teretes v. superne leviter compressi.
Folia falcato-recurva, oblique ovata v. oblonga, o-12 mm.
longa, 3-5 mm. lata, majuscule mucronata v. obtusa, subtus
glauca, penninervia, teuuiter membranacea, stipulis triangulari-
ovatis, subulato-acuminatis. Pedicelli filiformes, in pulvinulis
axillares. Sepala utriusque sexus oblongo-ovata, omnia denticulata.
Fl. J : sepala 4, 5 mm. longa v. breviora. Stamina 2 ; filamentis
brevissimis, circa ovarii rudimentum inserta. Antheree horizon-
taliter aperientes. Ovarii rudimentum brevissimum. Disci
glandulie in urceolum connatœ, 4-lob8e. Fl. Ç : sepala 6. Discus
integer v. margine répandus. Ovarium globosum levé, stylis
divaricatis, bipartitis oriuntur gracilibus. Pedicelli fl. Ç 5-7 mm.
subsequantes, filiformes. Capsulce depresso-globosœ, 4 mm. diametro
œquantes. Semina lievia.

A F. leptocladote Benth. differt ovarii rudimento evoluto,
seminibus lœvibus.

Hab. Liiikiu : Onnab, leg. J. Matsumura, auno 1897.

6. P. flexuosus MuELL. Arg. (Tab. I. G.) Muell.

Arg. in DC. Prodr. XV.-2, 324 ; Forbes et Hemsl. Ind. Fl.
Sin. IL 421 ; L. Diels, in Engl. Jahrb. XXIX. 427.

Cicca ßexuosa Sieb, et Zucc. Fl. Jap. Fam. Xat. Sect.
Prim. 143.

Glochidion flexuosum Franch. et Savat. Enum. PL Jap. I.
426.

Nom. Jap. KohannoJcl.

Frutex fasciculatim ramosus, ramis flexuosis, ramulis sub-
fasciculatis compressis. Pamuli florigeri tenues, 10-15 cm. longi.
Pulvinuli, e quibus oriuntur ramuli florigeri, densissime inferne
coriaceo superne membranaceo-squamato-perulati. Folia ovata v.

EUPHORBIACEARUM ET EUX ACE ARUM JAPONICARUM. 13

oblongo-ovata, acutiuscula, basi brevissime insequaliterque acuta,
membranacea, olivaceo-fuscidula, densissime reticulato-vencsa, 1-4
cm. longa, 7-25 mm. lata, petiolis 1-2 mm. longis, stipiilis angnstis,
spathulato-ovatis, persistentibus, 2-2 i mm. longis, scariosis, fuscis,
superne fimbriatis. Flores 5—ß fasciculati, uni g mixti. Sejoala 4,
œqualia, imbricata, 1.5 mm. longa, coriacea. Fl. ^ : sepala orbi-
culari-ovata, margine inflexa. Stamina 2, libera, filamentis validis ;
antbene distinctse verticaliter birimosœ. Glandulae 4, liberae,
truncatœ, interdem leviter connate. Pedicelli fl. ^ capillacei, 3-5—
mm. longi, basi articulati. Fl. Ç : sepala oblongo-spathulata, intus
nitida. Ovarium oblongo-ovoideum, glabrum; styli 3, liberi indivisi,
superne revoluti. Sepala oblongo-elliptica, decidua, omnia rotundato-
obtusa. Glandulae in urceolum connatse, margine 4-lobat8e, lobis
cum sepalis atteignis. Pedicelli ^ fl. masculis breviores, 5-3 mm.
longi, ad basin calycis crassati. Bacca spuria sicca, depresso-giobosa,
4 mm. longa, 6 mm. lata. Semina trigona dorso leviter alveolato-
reticulata. Embryo complanatus ad dorsum seminum affixus, re-
curvus, cotyledonibus spathulato-complanatis.

DisTRiB.: in Japonia australi.

Hab. Kiushii : Prov. Buzen, in monte Iwatake ; prov, Hiiiga, in monte

Kirisliima, auno 1882. In insula Oshima, leg. S. Matsuda, anno 1896.
Shikoku : Prov. Tosa, in monte Tsuetate ; in traetu Sngawa, leg. T. Making,

anno 1887 ; prov. lyo, in monte Wariishi ; prov. Awa, in ]^îi--hu-mura,

anno 1888.
Hondo : Prov. Nagato, in traetu Otsu, leg. G. Nikai, anno 1902 ; Prov.

Kii, in monte Köya, anno 1883. Tokyo Hort. Bot. cult.

7. P. reticulatus Poir.; Muell. Arg. in Linndea XXXII.
12 et in DC. Prodr. XV.-2, 344 ; Benth. Fl. Austral. VI. 101 ;
Hemsl. in Vog. Challenger, Bot. I.-3, 187 ; Hook. f. Fl. Brit.
Ind. V. 288; Forbes et Hemel. Ind. Fl. Sin. II. 422; Henry,
Lis. PI. Formos. 82.

14 ART. 2. E. HAYAÏA : EEVISIO

Cicca microcarpa Benth. Fl. Hongk. 312.

Phyllanthiis microcarpus Muell. Arg. in Linn?ea, XXXII.
51 et in DC. Prodr. XV.-2, 343 ; Hange, in Journ. Linn. Soc.
XIII. 121.

Kirganelia sinensis, Baill. " Etude Gén. Eupliorb. 614."

Phyllanthus sinensis Muell. Arg. in Linnœa, XXXII. 12.

Nom. Jap. SJmna-lcohannoki.

Frutices. Rami graciles, ramulosi. Folia disticba, 3 cm.
longa, 2 cm. lata, ovata, utrinque obtusa v. emarginata. Petioli
2-3 mm. longi. Stipulée longue, lanceolatse, crassiusculœ. Flores
ad axillas glomerati, articulato-decidui, J 6-Q> glomerati ^ uni
mixti, longe pedicellati. Bracteœ membranaceie. FI. J : sepala 5,
imbricata, crassiuscula, ovata basi leviter angustata, intus nitida,
extus glabra vel leviter pubescentia. Stamina 5, duo interiora
connata, longiora, très exteriora libera, filamentis brevissimis, an-
tlieris oblongo-ovatis apiculatis. Glandula? cum staminibus alternée.
FI. Ç : sepala 2-5, incrassata. Ovarium globosum glabrum,
7-8-loculare ; ovula in loculis gemina, superposita, stylis stigma-
tis. Drupa. Semina triangulari-ovata, sicco-rugosa ; embryo
rectus, cotyledonibus complanatis.

DiSTRiB.: in Asia ttopica et Africa, vulgaii.s,

Hab. Formosa : Tai-tong-thian, Pi-lara et Tai-lani-thinn, leg. K. Miyaké,
anno 1899.

8. P. (Sect. Embrica.) NÜnamÜ sp. nov.

Nom. Jap. Shima-ammaroku.

Arbores ? ramosœ, ramulis gracilibus, tomentosis. Folia in
ramulis disticha, oblongo-elliptica, 12-18 mm. longa, 5-8 mm.
lata, margine integerrima recurva, apice rotundata, glanduloso-
apiculata, basi rotunda to-cordata, subtus glabra, petiolis brevissimis

EUPHORBIACEARUM ET BUXACEARUM JAPONICARUM. 15

i mm. longis, stipulis riifis, triaugularibus, crassiusculis, margine
laciniatis, minimis, 1 mm. longis. Flores numerosi, in axillis
foliorum inferiorum parvorum siti. Fl. J : sepalis 6, spathulatis,
membranaceis, 23 mm. longis, 1 mm. latis ; glandulis 0 ; staminibns
3, filamentis tota longitudine in columnam 1 mm. longam connatis,
antheris erectis, extrorsis ; pedicellis geminis v. ternis, gracilibus,
4 mm. longis. Fl. $ : sepalis 6, spathulatis, crassiusculis, ma-
joribiis, 3 mm. longis, limm. latis, glandulis in urceolum crenato-
dentatum connatis ; ovario gioboso-ovoideo, stylis basi in columnam
ovarium subœquantem connatis, 2-fidis, ramis gracilibus, 3 mm.
longis, recurvis ; pedicellis brevissimis, * mm. longis, solitariis.
Fructus ignotus.

A P. Embrica L, foliis oblougo-ellipticis, stylis 2-fidis,
gracilibus, longioribus, facile distincta. Ceteris partibus vero
utraque species persimilis.

Hab. Fornioa : Shokwa, leg. T. Niinami, anno 1902 (Aug. fl.)

4. Glochidion Foest.

Endl. Gen. PI. 1121 ; Pax, in Nat. Pfl. Fam. III.-5, 23.
Flores monoici, apetali. Discus 0. Fl. J : sepala 6, distincta,
imbricata plus minusve 2-seriata, omnia simila v. exteriora bre-
viora. Stamina in centro floris 3-6, filamentis connatis brevissima,
antheris erectis in columna centrali sessilibus, loculis distinctis
parallelis longitudinaliter dehiscentibus, connectivis prominenter
apiculatis. Ovarii rudimentum 0. Fl. Ç : calyx rigidior, brevis,
incrassatus. Ovarium subglobosum, stylis brevissimis ; ovarium 3-

16 ART. 2. — B. HAY ATA : RE VISIO

8-loculare ; ovula in loculis gemina. Capsula sicca, demum in
coccos 2-valves dissiliens. Semina ecarunculata, testa Crustacea ;.
albumen carnosum ; embryo leviter incurvus, cotyledonibus planis.
Frutices. Folia coriacea, integerrima, alterna, ssepius disticha.
Flores parvi, axillares, J pedicellati, ? in eadem axilla glomerati^
subsessiles.

Conspectus Glochidionis Specierum.

Stamina 6-4.

Folia lanceolata ; ovarium 5-4-loculare 1. G. lanceolatwn.

Folia elliptica ; ovarium 6-8-loculare.

Tota glabra 2. G. zeylanicmn.

Tota liirsutfi. 3. G. hirsutum.

Stamina 3.

Ovarium 3-loculaie 4. G. bicolor.

Ovarium G-7-loculare.

Folia glabra 5. G. obovatwn.

Folia pubesceutia 6. G. formosanum.

1. G. lanceolatum. sp. nov. (Tab. II. C.)

Nom. Jap. Kiirun-kanhonoki.

Kami teretes, omnino glaberrimi, foliosi. Folia oblongo-ovata,
v.-lanceolata, obtuse accuminata, basi acuta, incequilateralia,
utrinque 7-8-costata, 5-6 cm. longa, 2-3 cm. lata, petiolis 4 mm.
longis, glabra, supra pallidiora, subtus fusco-virida, stipulis cras-
siusculis triangulari-acutis. Flores axillares, superne Ç, inferne
J. Inflorescentia breviter pedunculati, pedunculis 8 mm. longis.
Fl. % : sepala ovata, 2.5 nun. longa. Stamina ssepissime 4, rarius
5 V. 6. Endimentum parvissimum ovarii in centro staminum
prseditum. Flores J fasciculati, breviter pedicellati, pedicellis 4-
5 mm. longis. Fl. ? : sepala rotund ato-o va ta v. triangulari- ovata,.
1.5 mm. longa. Ovarium globosum, 2 mm. longum. Stylorum

EUPIIORBIACEARUM ET BUXACEARUM JAPONICARUM. 17

columna hemisplierica, leviter 8-10-sulcata. Ovarium 4-5-loculare,
rarius 6-loculare. Capsula depresso-globosa, leviter 8-sulcata.

A G. cacumino differt pedunculis pedicellisque florum J
longioribus.

Hab. Formosa : Kelimg, leg. T. Making et C. Owatari, anno 1896.

2. G. zeylanicum A. Juss.; Muell. Arg. in Linnœa,
XXXII. 60 et in DC. Prodr. XV.-2, 281 ; Hook. f. Fl. Brit.
Ind. V. 311.

Nom. Jap. KaJcibano-hanhonohi.

Rami grabri, foliosi. Folia oblongo-ovata, acuta v. obtusa, basi
obtusa, glabra, v. nitida, supra pallidiora, subtus fusca. Stipulée
minutée, triangulari-ovatse, acuminatœ, rigidse, arete recurvse,
persistentes. Petioli 6 mm. longi. Inflorescentise pedunculatee ;
pedunculi 15 mm. longi, spatio circ. 6-7 mm. longo ab axilla
remoti. Flores utriusque sexus in quaqua inflorescentia mixti, J
longius, Ç brevius pedicellati. Fl. J : sepala oblongo-ovata, 3 J
mm. longa, levia. Stamina seepissime 5. Fl. $ : glomerato-
fasciculati. Sepala ovata, acuta, levia, 2 mm. longa, interiora
angustiora. Ovarium globosum, glabrum, 8-9-loculare ; columna
stylaris conica, truncata, apice 5-6-fida. Capsula depresso-globosa,
circ. 18-sulcata, ambitu teres, 7 mm. lata, 4 mm. longa.

DisTRiB. In insula Ceylonia, Banca juxta Sumatra, Borneo, peninsula Indiœ

orientalis, Formosa, Liu-kiu.
Hab. Formosa : Taipea, leg, T. Making et K. Honda, anno 1896. Kelung,

leg. C. Owatari, anno 1896. In Tai-tong-thian, Ka-leng-oan, leg. K.

Miyake, anno 1899.
Liu-kiu : Yontanzan, leg. J. Matsumura, anno 1897.

3. G. hirsutum Muell. Arg. (Tab. II. D.) Muell.

Arg. in Linnœa, XXXIL 61 ; Hook. f. Fl. Brit. Ind. V. 311.

18 ART. 2. — B. HAYATA : EEVISIO

Bradleia hirsuta Eoxb. Fl. Ind. III. 699. " Baill. Etude
Gen. Euphorb. 638."

Phylla7iihus hirsutus Muell. Arg. in DC. Prodr. XV.-2, 283.

Nom. Jap. Obano-kehankonoM.

Hamuli crassiusculi oligophylli, dense puberuli, apice fulvo-
tomentelli. Petioli 3 mm. longi. Stipulée lineari-subulatœ, arcuato-
reflexse, 3 mm. longse. Folia oblongo-ovata, 10-15 cm. longa, 6-
8 cm. lata, obtusa, basi oblique truncato-obtusa, latere interiore
obtusa V. leviter retusa, exteriore sfepius acuta, supra paucius
subtus densius pubescentia. Inflorescentiœ paulo supra-axillares,
cum ramulo basi connatae, et 5-6 mm. supra axillam folii quasi
insertse, pedunculatœ. Fl. J fasciculati, pedicellis 13 mm. longis.
Sepala 2 mm. longa, late elliptica, cum pedicello tenui-vel griseo-
pubescentia. Stamina 6. Fl. Ç : calyx 2.5 mm. longus, 6-lobatus,
lobis crassis, acutis ; ovarium globosum, cum pedicello omnino
pubescens. Pedicelli 3 mm. longi. Capsula depresso-globosa, 8ï
mm. diametro sequans, 7 mm. longa, pubescens, non sulcata,
12-sperma, v. abortu 6-sperma. Semina valde compressa, 4 mm.
longa, 2è mm. lata, 1 mm. crassitudine ?equantia.
DiSTRiB. In insula Penang, Indiae orient, et Formosa.
Hab. Formosa : Taipea, leg. K. Honda, anno 1896 ; Po-li-sia, Tsui-sia, leg.
K. MiYAKE, anno 1898.

4. G. bicolor (Muell. Arg.) (Tab. II. E.)

G. acuminatum Muell. Arg. in Linnœa, XXXII. 68.

Phyllanthus bicolor Muell. Arg. in DC. Prodr. XV.-2, 311.

Nom. Jap. Uraziro-kankonoki.

Kamuli apice pallide fulvo-sericeo-pubescentes. Petioli 2-3
mm. longi. Folia 4-8 cm. longa, 1.5-3 cm. lata, lanceolata,
utrinque acuta, leviter inœquilateralia, supra in costis pubescentia,
cseterum glabra, subtus plus minusve argeuteo-pallida vel albicantia,

EÜPHOEBIACEARUM ET BUXACEARUM JAPONICARUM. 19

brevissime pubescentia. Flores J fasciculati, pedicellis pubescentibus

4h mm. longis. Sepala 2 mm. longa, extus pubescentia. Stamina 3.

Flores ? primum subglomerati. Sepala oblongo-elliptica, acuta,

i-1 mm. longa, interiora angustiora, parviora ; ovarium 3-loculare,

globosum sericeum, cum stylis limm. longum ; columna stylaris

cylindrica, superne paulo latiora, breviter 3-fida, ovarium longitudine

ter V. quater œquans. Capsula parva, depressa, profunde sulcata,

basi et apice umbilicato-depressa, 4mm. longa, 7mm. lata. Semina

ovoidea, 3 mm. lata, 3i mm. longa. Fasciculi fl. J in axillas

inferiores, fl. Ç primum glomerati, in axillas superiores dispositi.

DiSTRiB.: In Nepalia, Sikkim, Bengalia orientali, Formosa, et Liu-kiu.
Hab. Formosa : Pak-kang-khoe, Tsuisia, Po-li-sia, leg. K. Miyake, anno

1898.
Liu-kiu : Leg. J. Matsumura, anno 1897 ; in Insula Orfhima, Nase, leg. T.

UcHiYAMA, anno 1900.

5. G. Obovatum Sieb, et Zucc. (Tab. II. F.) Sieb.

et Zücc. Fl. Jap. Fam. Nat. Sect. Prim. 143 ; Müell. Arg. in
Linnsea, XXXII. 67 ; Miq. Prol. 292 ; Forbes et Hemsl. Ind.
Fl. Sin. II. 425.

Phyllanthus obovatus Müell. Arg. in DC. Prodr. XV.-2, 307.

Nom. Jap. Kankonohi.

Hamuli pauce et tenuiter pubescentes. Petioli I2-2 mm.
longi. Stipulge minimse, ovatie, subacuta. Folia 3-6 cm. longa,
2-3 cm. lata, obovata, oblongo-ovata, vel obovato-lanceolata,
obtusa, rotundata v. breviter acuminata, basin versus sensim
cuneato-angustata, glabra. Flores J pedicellati, pedicellis evolutis,
10 mm. longis. Calyx 2 mm. longus ; sepala oblongo-ovata, interiora
angustiora ; stamina 3. Flores ? longe pedicellati ; pedicelli 8-9
mm. longi. Calyx 6-lobatus, lob is crassiusculis, acutis, interioribus
angustioribuö. Columna stylaris cylindrico-obovoidea, apice obtusa,

20 AET. 2. — 'B. H AY AT A : RE VISIO

6-fida, sursum latior, deorsum aDgustior, ovarium longitudine bis
sequans, crassa. Ovarium 6-loculare, glabrum v. leviter pubescens,
depresso-globosum. Calyx le mm. loiigus. Capsula depresso-
globosa, profunde sulcata, 82 mm. longa, 6 mm. diametro sequans.
Semina

DisTRiB. in China, Japonia australi.

Hab. Formosa : Tamtsui, Kelung, leg. T. Makino, anno 1896 ; Takao,

leg. C. OwATARi, anno 1896 ; Hok-ki-tsng, leg. K. Miyake, anno

1897; Kachilai et Ka-chin-lo, leg. K. Miyake, anno 1898.
Liu-kiu : in insula Oshima, leg. T. Uchiyama, anno 1900.
Kiushiu : prov. Higo, prov. Tikuzen, in tractu Kasuya, leg. K. Nagano ;

prov. Hiuga, in Miniitsu; prov. Hizen, in monte Onsen-gatake, anno 1892.
Hondo : prov. Suwau, in tractu Yoshiki, leg. G. Nikai, anno 1899 ; prov.

Kii in Miuamichi, anno 1883. Tokyo Hort. Bot., cult.

6. G. formosanum sp. nov. (Tab. II. G.)

Nom. Jap. Ke-hanhonoki.

Plante omnino pubescentes. Kami teretes, oligopbylli. Folia
oblougo-ovata v. oblongo-elliptica, apice obtusa v. acuta, basi oblique
truncato-obtusa, latere interiore obtusa, exteriore acuta, supra pauce,
subtus densiuscule pubescentia, 7-8 cm. longa, 3-4 cm. lata, petiolis
4 mm. longis. Flores o fasciculati ; longe pedicellati, pedicellis
8 mm. longis ; sepalis 6, interioribus parvioribus, ovatis acutis,
intus et extus brevissime pubescentibus, 0 mm. longis. Stamina
3, rarissime 4. Flores g breviter pedicellati, pedicellis 2.5 mm.
longis ; calyx 1 mm. longus, 6-lobatus, lobis brevissimis ; ovarium
ovoideum cum stylis 2 mm. longum ; columna stylaris depresso-
globosa, leviter 7-8-sulcata, basi constricta, ovario brevior ; ova-
rium 6-7-loculare. Capsula depresso-globosa, 10-12-sulcata,
ambitu teres.

Hab. Formosa ; Tai-tong-thian, Pi-lam, leg. K. Miyake, anno 1899.

EÜPHOEBIACEARUM Eï BUXACEAKÜM JAPONICARUM. 21

5. Breynia Forst.

Endl. Gen. PI. 1120; Muell. Arg. in DC. Prodr. XV.-
2, 438; Baill. Hist. d. PL V. 254; Benth. et Hook.
Gen. PL III. 276; Pax, in Nat. Pfl. Farn. III.-5, 24.
Flores monoici, apetali. Discus 0. FL J : calyx breviter turbina-
tus, supra planus, lobis v. dentibus inflexis fere clausus. Stamina
3, filarnentis in columnam centralem connatis ; antlierse ad apicem
columnse sessiles adnatœ, loculis parallelis longitudinaliter deliiscenti-
bus. Kudimentum ovarii 0. FL ^ : calyx campanulatus, fructiger
interdem auctus, explanatus, lobis brevibus nunc ad dentés minutos
reductis. Ovarium sessile v. breviter stipitatum, 3-loculare, superne
carnosum ; styli indivisi v. apice 2-fidi ; ovula in loculis gemina.
Bacca globosa, indehiscens. Semina triquetra, dorso rotundata,
testa Crustacea, ventre duplicata inter laminas cava ; albumen baud
copiosum, circa cavum incurvum ; embryo curvus, cotyledonibus
latiusculis planis. — Frutices. Folia alterna, petiolata, ssepius dis-
ticha, siccitate nigricantia. Flores axillares, parvi, pedicellati,
fasciculati, fasciculo rarius in racemum brevissimum excurrente.

Conspectus Breyniarum Specierum.

Calyx fructiger accrescens, dilatatus.

Ovarium ioclusum ; bacca apice producta 1. B. accrescens Jo nn. a.

Ovarium exsertum ; bacca ?i])ice depressa 2. B. accrescens form. ß.

Calyx fructiger vix accrescens, campanulatus 3. B. rhamnoides.

Calyx fructiger non accrescens.

Stipulée breves triangulares; calyx 't turbinatus, staniinibus exsertis

4. B. stipitata \ar. forviosana form. a.

Stipulée subulato-acuminatse ; calyx 't subglobosus, staniinibus inclusis

5. B. stipitata ytir. formosana form. ß.

22 ART. 2. — B. HAYATA : REVISIO

1. B. accrescens sj^. nov. form. «. (Tab. I. H.)
Nom. Jap. Hime-kobannohi.

Rami graciles, flexuosi. Folia disticlia, membranacea, breviter
petiolata, ovata, apice rotundato-obtusa, basi acuta, 2 cm. longa, 1.5
cm. lata, petiolis 2-3 mm. longis, stipulis minimis, subulato-lance-
olatis, subtus pallidiora. Flores J pedicellati, pedicellis 8mm. longis;
calyces turbinati 2 mm. longi, I2 mm. in diametro sequantes ; 2-
3-fasciculati, 2-3-bracteati. Flores ^ ad axillas solitarii, pedicellis
3 mm. longis, calycem œquantibus. Calyx turbinato-campanulatus,
margine obscure 6-fractus. Ovarium ovoideum, truncatum, vertice
foveolatum, inclusum. Calyces fructigeri dilatati, bemispherico-
deplanati, subintegri, leviter 2-lobati, 5 mm. diametro sequantes,
pedicellis 5 mm. longis. Baccse globosse 4 mm. diametro sequantes,
apice productse. Semina triquetra, 02 mm. longa, 1.5 mm. lata.

Similans B. coronatam, Hook, f.; sed floribus J fasciculatis,
foliis minimis, fascile distincta.

Hab. Formosa : Kotöshö, leg. K. Miyake, anuo, 1899.
Nein. Formos. Kakaridd-yariggudd

2. B. accrescens sp. nov. form. ß. (Tab. I. I.)

Fl. ? : calyces campanulati, 1 mm. longi, distincte 6-lobati,
lobis acutis ; ovarium exsertum, ovoideum, truncatum, vertice con-
cavura, in centro stigmatibus minimis. Calyces fructigeri dilatati
deplanati, margine breviter crenulati, 3è mm. diametro aequantes.
Baccae globosae, 5 mm. diametro aequantes, apice depressae.
Hab. Formosa : Tamsui, leg. T. Marino, anno 1896.

3. B. rhamnoides Muell, Arg.; (Tab. I. J.) MuELL.

Arg. in DC. Prodr. XV.-2, 440 ; Hook, f Fl. Brt. Ind. V. 330 ;
Forbes et Hemsl. Ind. Fl. Sin. IL 428 ; L. Diels, in Engl.
Jahrb. XXIX. 427.

EUPHORBIACEAHÜM ET BUXACEARUM JAPONICARUM. 23

Phyllanlhus rhamnoides Willd. Sp. PI. IV. 580 ; Hook, et
Arn. Bot. Beech. Vog. 270.

P. vitis-idaea Koxb. Fl. Ind. HI. Q?)5.

3Jelanthesa rhamnoides Wight, Ic. Fl. Ind. or. t. 1898 ; MiQ.
Fl. Ind. Bat. I.-2, 370.

Nom. Jap. Oshima-kobannoki.

Frutices glaberrimi. Folia brevissime petiolata, petiolis 3 mm.
longis, membranacea, ovata vel orbiculari-obovata, rotundato-obtusa,
3 cm. longa, 21 cm. lata. Flores J 3-4, fasciculati, pedicellis
7 mm. longis ; calyces turbinati v. subcampanulati, margine obscure
lobati, 2jmm. longi, IJmm. diametro aequantes. Flores Ç soli-
tarii ; calyces 2 mm. longi, hemisplierico-carapanulati, 6-dentati ;
ovarium subglobosum, apice truncatum, concavum, vix exsertum.
Pedicelli 4 mm. longi. Baccae globosae, vertice obsolete disciformi-
tumidae, 3? mm. longae, 6 mm. latae, breviter stipitatae. Calyx
fructiger vix accrescens. Semina triquetra, 3i mm. longa, 2? mm.
lata.

DiSTRiB. In tropica Asia, vulgaris.

Hab. Formosa: Ang-thau-su, leg. K. Miyake, anno 1899.

Liu-kiu : Naha ; Oshima, leg. T. Uchiyama, anno 1900.

4. B. stipitata Muell. Arg. var. formosana var. nov.
form. «. (Tab. II. A.)

Nom. Jap. Daiwan-kobannoki.

Rami graciles. Folia coriacea v. membranacea ; stipulae
triaugulari-ovatae, subulato-acuminatae. Petioli 3 mm. longi. Folia
1^-21 cm. longa, 1.5-1 cm. lata, oblongo-ovata, apice ob-
tusa, basi inaequilateralia, acuta, supra olivaceo-viridia, snbtus
pallidiora. Flores J 2-3, fasciculati, pedicellis 6 mm. longis.

24 ART. 2. — B. HAY ATA I REVISIO

Calyces turbinati, 6-lobati, lèmm. longi. Flores Ç ad axillas
solitarii, breviter pedicellati. Calyx 2 mm. longus, pedicellum
œquans, obconicus acute breviterque 6-lobatus, sub fructu non
accrescens. Ovarium turbinato-obconicum, apice latum, concavum
tumido-stigmatum. Baccae subdepresso-ovoideae, basi distincte
in stipitem calyci urceolari subaequilongum gracilem attenuatae,
4 mm. longae, 6 mm. latae. Stipes 2è-3mm. longus, apice abrupte
in fructum abiens. Semina.

Hab. Formosa: Shinteklui, leg. T. Marino, anno 1896.

5. B. stipitata Muell. A eg. var. formosana var. nov.

form. ß. (Tab. II. B.)

Stipulae lineari-lanceolatae, acuminatae. Folia ovata, utrinque
subacuta, supra nigra, subtus glauca. Flores J : calyx subglobosus,
margine 6-dentatus, staminibus inclusis. Fl. Ç : ovarium cylind-
ricum, truncatum, apice leviter trisulcatum.

Hab, Formosa: Taipen, anno 1896.

6. Pulranjiva Wall.

Endl. Gen. PI. 287 ; Muell. Arg. in DC. Prodr. XV.-

2, 443 ; Baill. Hist. d. PI. V. 249 ; Benth. et

Hook. Gen. PI. III. 277 ; Pax. in Nat. Pfl. Fam.

III.-5, 27.

Flores dioici apetali. Discus 0. Fl. J : calyx tenuis, 4-6-

partitus, segmentis inaequalibus, imbricatis. Stamina 2, filamentis

EÜPHORBIACEARUM ET BUXACEARUM JAPONICARUM. 25

compressis, in centro floris basi connatis ; antlierae globoso-
ellipsoideae, crassiusculae, ntrinque obtusae, erectae, extrorsae,
loculis distinctis, parallelis longitudinaliter dehiscentibus. Kudi-
mentum ovarii 0. Fl. Ç : calyx tenuis, 3-partitus, segmentis
angustis. Ovarium oblongo-ovoideum, 3-loculare ; ovula in loculis
gemina. Styli longiusculi, patentes, in ramos totes carnosos
papillosos expansi. Drupa oblongo-ovoidea, endocarpio duro, fere
osseo, abortu 1-locularis, 1-sperma. Semina oblongo-ovoidea ; testa
tenuis ; albumen carnosum ; embryo rectus, magnus, cotyledonibus
latis planis, palmatinerviis. — Arbores ? Folia alterna, bistipulata,
crenulato-serrulata, rigide membranacea v. coriacea, dense tenuiter-
que pennivenia et reticulato-venosa. Flores axillares, J in racemes
glomerati, brevissirae pedicellati, Ç longius pedicellati, solitarii.

1. P. Koxburghii Wall. (Tab. II. H.) Wight, Ic.

PI. Ind. or. t. 1876; Muell. Arg. in DC. Prodr. XV.-2, 443;
Beddome, Fl. Sylv. Sou. Ind. t. 275 ; Hook. f. Fl. Brit. Ind.
V. 336; Matsumura, in Tokyo Bot. Mag. XII. 61.

Nayeia Putranjiva, Roxb. Fl. Ind. III. 766.

Nom. Jap. Tsugemodohi.

Kami graciles, teretes, superne angulosi, glabrati. Hamuli
acute sulcato-angulosi, obscure pubescentes. Petioli 7-10 mm.
longi, firmi, supra canaliculati, pubescentes. Folia ellip-
tica, apice obtuse acuminata, basi acuta, 3-5 cm. longa, 1-
3 cm. lata, elliptica, juniores membranacea, evoluta chartacea,
margine adpresso-pubescentia, denum glabrata. Fl. J racemoso-
spicati. Bracteae multiflorae. Calyx 1-2 mm. longus ; laciniae
margine densius ciliato-pubescentes, dorso paucius puberulae. An-
therae crassae, exsertae, 1 mm. longae, hirtellae. Fl. Ç pedicellis
5-lOmm. longis, sub fructu circ. 12mm. attingentibus, validiusculis.

26 ART. P>. — B. HAYATA I REVISIO

Calycis Ç laciniœ 1 3-2 mm. longœ, sub fructu deciduse, oblongo-
ellipticœ, apice rotundatse. Ovarium oblongo-cylindricum, calycem
tri-quater œquans, sub-fulvo-sericeum. Drupa oblongo-ellipsoidea
utrinque angustata, 20 mm. longa, 13 mm. lata, pilis densis pallidis
brevissime tomentella.

DiSTRiB. In peninsula Ind. orient.; Formosa et Liu-kiu.

Hab. Formosa: Ang-thau-su, leg. K. Miyake, anno 1899.

Liu-kiu : Ins. Okinawa, leg. J. Matsumura, anno 1897 ; Shuri, leg. K.

Miyake, anno 1899; Ins. Miyakojima, leg. S. Tanaka, anno 1879;

Ins. Okinawa, leg. Y. Tasiiiro, anno 1897.

7. Antidesma Linn.

Linn. Sp. PI. ed.-2, 1455 ; Endl. Gen. PI. 287 ; Muell.

Arg. in DC. Prodr. XV.-2, 248; Benth. et Hook.

Gen. PI. III. 284 ; Pax, in Nat. Pfl. Fam. IIL-5,

30.
Flores dioici, apetali. Fl. J : calyx alte 3-4-lobatus, lobis
imbrieatis. Discus evolutus, glandulis staminibus alternis, concre-
tis. Stamina 4-5, lobis ealycis opposita, filamentis liberis, longis,
sub anthesin calycem vulgo 2-4-plo superantibus. Antlierarum
loculi in alabastro extrorsum aperientes, loculi post osccillationem
superne divergentes, introrsura aperti. Kudimentum ovarii 0.
Fl. ? : calyx 5-6-partitus. Disci annulares. Ovarium 1-loculare ;
styli 3, brevissimi ; ovula in loculo gemina. Drupa parva,
subcarnosa, obliqua, indehiscens. Putamen osseo-coriaceum,
foveolatum, carinatum. Semina abortu unica, ecarunculata ; albumen
carnosum ; embryo rectus, cotyledonibus planis latis. — Frutices.
Folia alterna, bistipulata, breviter petiolata, penninervia ; costae

EUPHORBIACEARUM ET BUXACEARUM JAPONICARUM. 27

ante marginem integrum arcuato-adscendentes, anastomosantes.
Flores parvi, racemosi, racemis tenuibus, ad axillas solitariis.
Bracteae parvae.

1. A. japonica Sieb, et Zucc; (Tab. II. I.) Benth.

Fl. HoNGK. 318 ; Muell. Arg. in DC. Prodr. XV.-2, 258 ; Maxim.
in Engl. Jalirb. VI. 59 ; Forbes et Hemsl. Ind. Fl. Sin, II.
432 ; Henry, List. PI. Formos. 83.

Nom. Jap. Uguyoshi vel yama-hihatsii.

Eamuli tenues glabri v. pilis brevibus scabridi. Folia 4-
9 cm. longa, 1-3 cm. lata, lanceolato-elliptica v. latins angustiusve
lanceolata, acuminata, basi attenuata, glabra, subtus ad nervös
parce pubescentia, fusco-viridia v. dein fuscidula, stipulis lineari-
lanceolatis membranaceis deciduis. Petioli pubescentes, 4-8 mm.
longi. Bacemi omnino pubescentes. Bracteae lanceolatœ, ovatse,
acutae. Stamina valde exserta, filamentis validis. Calyx dilata-
tus, 3-4-lobatus, lobis brevissimis. Flores ? pedicellis gracilibus,
liliformibus, liorizontaliter patentibus. Calyx urceolatus, lobis
brevibus. Drupa elliptica subincurva, pedicello vix duplo longior,
6 mm. longa, 5 mm. lata, sicco grosse subfoveolato-rugosa.

DiSTRiB. In Hongkong, Japouia australi.

Hab. Formosa : In monte ïaiton ad 300 m. alt., Rev. U. Faurie, anno
1903.

Liu-kiu : Okinawa, Yaeyama, Miyakojima, leg. Y. Tashiro, anno 1887;
Yseyama, in monte Urabu, Yonakuni, Tamatsuku, leg. S. Tanaka,
anno 1891 ; in monte Ubashiinata, Kiinchan, leg. J. Matsumura,
anno 1897; in insula Kumesliima, leg. K. KuROiwA, anno 1898; in
insula Osliima, in inonte Takamine, leg. T. Uchiyama, anno 1900.

28 ART. o. — B. HAYAïA : REVISIO

8. BischofTia Blume.

Endl. Gen. PL 1149; Muell. Arg. in DG. Prodr.

XV.-2, 478 ; Baill. Hist. d. PI. V. 227 ; Benth.

et Hook. Gen. PI. III. 281 ; Pax, in Nat. Pfl.

Farn. III.-5, 33.
Flores dioici, apetali. Discus 0. Fl. J : sepala 5, imbricata,
margine induplicato-cucullato-concava. Stamina 5, filamentis bre-
vibus, circa rudimentum ovarii horizontaliter inserta, sepalis op-
posita ; antherae magnae, localis parallelis distinctis versus medium
affixis longitudinaliter deliiscentibus. Ovarii rudimentum brève,

latum. Fl. ^ : Fructus subbaccatus, globosus, indebiscens,

3-locularis, mesocarpio carnoso, endocarpio pergameno. Semina
ecarunculata, testis crustaceis ; albumen carnosum ; embryo rectus,
cotyledonibus planis latis, corda to-o va tis complanatis, le vi ter
recurvis. — Arbores. Folia alterna, 3-foliolata, foliolis longe petio-
lulatis, majusculis, crenatis. E-acemi laterales ; J paniculato-ramosi,
brevius pedicellati.

1. B. javanica Blume ; Miq. Fl. Ind. Bat. I.-2, 363 ;
Benth. Fl. Hongk. 316; Muell. ArCx. in DG. Prodr. XV.-2, 478 ;
Seeman, Fl. Vitiensis, 221 ; Beddome, Fl. Sylv. Sou. Ind. t.
259 ; Maxim, in Engl. Jahrb. VI. 59 ; Hook. f. Fl. Brit. Ind.
V. 345 ; Forbes et Hemsl. Fl. Ind. Sin. II. 428 ; Henry, List
PI. Fornios. 83 ; L. Diels, in Engl. Jahrb. XXIX. 428.

Ändrachne trifoliata Roxb. Fl. Ind. III. 728.

Bischoffia trifoliata FIook. Ic. PL. t. 844.

B. Eoeperlatia Decne; "Baill. Etude Geu. Euphorb. t. 26."

EUPHOEBIACEAKUM ET BÜXACEARUM JAPONICAEUM. 29

Nom. Jap. Akagi.

Kami teretes grabri. Folia trifoliata, foliolis petiolulatis ;
petiolulus terminalis reliquis longior ; foliola ovata, penninervia,
margine crenato-dentata, rigide membranacea, basi acuta, apice
cuspidato-acuminata, 5-10 cm. longa. Petioluli laterales 5-2.5
cm. longi, terminales 5-7cm. longi. Filamenta libera ; antherse late
ovoideae, utrinque emarginatse, extrorsum aperientes. Ovarii
rudimentum cylindricum, brève, vertice in discum crenulatimi et
planum dilatatum. Pedicelli masc. calycem semul v. bis aequantes,
ad medium articulati. Pedicelli fructigeri incrassati, fructum long-
iuscule superantes. Fructus ovoideus, rugulosus, basi brevissime
contracto-angustatus. Semina semiglobosa.

DisTRiB. India, Malaya, Polynesia, Formosa et Liu-kiu.

PIab. Formosa : Aug-thau-su, ïai-toug-thiau, Pi-lam, leg. K. ÄIiyake, anno

1899; Cliip-chip-koe, Lim-ki-po, leg. C. Owatari, anno 1898; Pachina,

Tarasui, leg. T. Making, anno 1896.
Liu-kiu : Insula Okinawa, Naha.

9. Bridelia Willd.

WiLLD. Sp. PI. IV. 978 ; Endl. Gen. PI. 1119 ; Muell.

Aeg. in DC. Prodr. XV.-2, 493 ; Benth. et Hook.

Gen. PL III. 267 ; Pax, in Nat. Pfl. Fam. III.-

5, 35.

Flores monoici, petaliferi. Fl J : sepala 5, valvata ; petala 5,

brevia, squamiformia, spatliulata. Discus margine subinteger.

Stamina 5, filamentis in medio disco basi in columnam connatis,

superne liberis patentibus ; antherae obovoideae, loculis parallelis,

longitudinaliter dehiscentibus. Pudimeutum ovarii ad apicem

30 ART. 3. — B. HAYATA I EEVISIO

columnse styliforme. Fl. Ç : calycis segmenta quam J angnstiora.
Petala maris. Discus duplex, interiore cupulato 6-lobato, ovarii basio
arcte cingente. Ovarium 2-loculare. Baccse indéhiscentes, meso-
carpiis carnosis, endocarpiis osseis. Semina in pyrenis abortu soli-
taria ; albumen carnosum facie interiore late profundeque excavatum ;
embryo cavo albuminis parallelus, cotyledonibus latis tenuibus ;
radicula brevissima. — Frutices. Folia alterna, breviter petiolata,
integerrima, venis primariis pinnatis, secundariis transversis, utrin-
que prominulis. Flores ad axillas glomerati, bracteati, J numerosi,
Ç solitarii, breviter pedicellati. Bracteae stipulares.

1. B. tomentosa Blume; (Tab. III. A.) Miq. Fl.

Ind. Bat. I.-2, 364 ; Benth. Fl. Hougk. 309 ; Muell. Arg.
in DG. Prodr. XV.-2, 501; Benth. Fl. Austral. VI. 120; Hook.
f. Fl. Brit. Ind. V. 271 ; Forbes et Hemsl. Ind. Fl. Sin. II.
420; Henry, List. PL Formos. 82.

Bridelia Loureiri Hook, et Arn. Bot. Beecli. Voy. 211.

Nom. Jap. Kankomodoki.

Frutices ramosi, ramulis pilosis, rufescentibus, tenuibus
llexilibus. Folia submembranacea, ovoidea, lanceolata, utrinque
subacuta, breviter petiolata, 7-3 cm. longa, 4— le cm. lata, margine
subcrenata, costis secundariis utrinque latere 7-8, supra glabra,
subtus molliter ferrugineo-puberula, petiolis 5-3 mm. longis.
Flores J aperientes 7 mm. diametro aequantes, sepalis triangulari-
ovatis, demum patulis, intus parce hirsutis, 1 h mm. longis, 1 mm.
latis; petalis angustioribus, plicato-bigibbosis, Ih mm. longis, stipi-
tatis. Kudimentum ovarii 2 mm. longum, apice stigmatum. Co-
lumnae staminum et antherae parce hirsutae. Discus complanatus,
margine leviter sinuatus. Pedicelli z mm. longi. Flores Ç :
sepala et petala more J, sed parviora. Calyx ext us parce pubes-

EUPHORBIACEAEUM ET BUXACEAEUM JAPONICARUM. ol

cens. Discus duplex, exteriore annulari, interiore cupuliformi 6-
lobato. Ovarium oblongo-globosum, stylis bifidis. Baccae 7 mm.
longae, 6 mm. latae.

Norn. Formos. Tu-mi-shu. (fide Henry).

DisTRiiî. In Javn, ludiae orient., Hongkong, Australia et Formosa.
Hab. Formosa: Hong-san, Tong-si-kak, leg. K. Miyake, anno 1898.

10. Daphniphyllum Blume.

Endl. Gen. PI. 1104; Muell. Arg. XVI.-l, 1 ; Baill.

Hist. d. PL V. 292 ; Benth. et Hook. Gen. PL III.

282; Pax, in Nat. Pfl. Fara. in.-5, 36.
Flores dioici, apetali. Fl. J : calyx obsoletus v. evolutus,
segmentis 5-8, parvis. Stamina 5-10, filamentis brevibus, liberis, e
basi radiato-divergente adscendentibus ; antherae magnae, erectse,
tetragome, a dorso compressae, localis parallelis, adnatis, lateraliter
2-valvatim dehiscentibus. Budimentum ovarii 0. Pedicelli mox
articulato-decidui. Disci glandulae 5 v. 0. FL ? : calyx maris.
Ovarium vix perfecte 2-loculare ; styli distincti, crassiusculi,
recurvo-patentes, indivisi ; ovula in loculis gemina. Drupa
oliviformis, indehiscens, abortu 1-sperma, mesocarpio crassiusculi,
endocarpio rigide cbartaceo, intus albo, sericeo-fibroso. Semina
ecarunculata, testis membranaceis ; albumen crasso-carnosum,
oleagineum, olivaceo-nigrescens ; embryo apicalis, albumine cire,
sexies brevior, cotyledonibus semicylindricis, latitudine radiculam
aequantibus. Frutices v. arbores glabrae. Bami teretes densius-
cule foliosi, glabri. Folia alterna, petiolata, integerrima, coriacea,
saepe angusta, penninervia, reticulato-venosa, subtus glaucesceutia.

32 ART. P.. — B. HAYATA I REVISIO

Racemi utriusque sexus axillares, breves, laxiusculi, bracteis minu-
tis. Flores pedicellati, secus rhachin sparsi.

Conspectus Daphniphylli Specierum.

Calyx obsoletus 1. D. macropodum.

Calyx evolutus.
Stamina 8.

Folia elliptica 2. D. glaucescens.

Folia oblongo-lanceolata, acuminata 3. D. glaucescens var. Oldhami.

Stamina 5 4. D. himalayense.

1. D. macropodum Miq.; (Tab. II. J.) Miq. in Ann.

Mus. Bot. Lugd.-Bat. III. 129 et in Prol. 393 ; Muell. Aeg. in
DC. Prodr. XVI.-l, 5; Forbes et Hemsl. Ind. Fl. Sin. 429;
L. DiELS, in Engl. Jalirb. XXIX. 428 ; Shirasawa, Nippon shin-
rinjumokuzusetz, vol. 1, t. 54.
Nom. Jap. Yuzuriha.

Hamuli validi, superne angulati et foliosi. Folia coriacea 9-
18 cm. longa, 3-7 cm. lata, e basi acute elliptico-oblonga, v. sub-
lanceolata, apice abrupte brevissime acuteque duro-apiculata, glabra,
subtus subglauca, margine extremo leviter recurva, costulae utrin-
que 12-15, erecto-patulae, tenerae. Petioli vulgo 3-5 cm.
longi, canaliculati. Fl. J : calyx obsoletus. Stamina 8-10 ;
antherae rhomboideae, 3 mm. longae. Fl. J aperientes 7 mm.
diametro aequantes. Pedicelli 10 mm. longi. Fl. Ç : calyx obsoletus ;
staminodia 10, crasso-carnosa, digitaliformia, 1 mm. longa. Ovarium
globosum, glabrum, stylis 2, divaricato-adpressis, lanceolatis,
intus canaliculatis, leviter papillosis. Fructus oblongo-globosus,
sicco nigricans ; endocarpium firmum, foveolatum.

DisTRiB. in China australi, Japon la.

Hab. Hondo : jn-ope Tokyo, Insla Hacliijojima, anno 1881 ; prov. Kii, anno

1883. Prov. Kagn, anno 1881. Insula Sado, anno 188G. Insula Közu-
shima, anno 1887. Piov. Shinano, anno 1884. Prov. Iwasliiro, anno
1879. Prov. Izu, insula Oshima, in monte Mihara, anno 1887, Prov.
Suwo, in traetu Yoshiki, leg. G. Nikai, anno 1893.

2. D. glaucescens Blume ; (Tab. II. K.) Miq. Fl.

Ind. Bat. I.-2. 431 ; Beddo.mf, Fl. Sylv. Sou. Ind. t. 288 ;
MüELL. Arg. in DC. Prodr. XVI.-l, 3; Feaxch. etSAVAT. Enum.
PI. Jap. I. 427 et IL 488 ; Hange, in Journ. Linn. Soc. XIII.
120 ; IMaxim. in Engl. Jahrb. VI. 59 ; Hook. f. Fl. Brit. Ind. V.
3Ö3 ; Forbes et Hemsl. Ind. Fl. Sin. IL 429 ; Henry, List PI.
Formos. 83; L. Diel?, in Engl. Jahrb. XXIX. 428.

GompJiia neilgherreyisls Wight, Ic. PL Ind. or. t. 1878-
1879.

DapJiniphyllum Roxhurghu Batll. " Etud. Gen. Enphorb.
bQù'') Benth. Fl. Hongk. 316; Miq. in Ann. Mus. Bot. Lugd-
Bat. III. 129.

Nom. Jap. Hime-yuziiriha.

Frutices. Folia oblongo-ol)Ovata, v. obovato-lanceolata, apice
nunc rotundato-obtusa nunc acuta, basi subacuta, 8-10 cm. longa,
3-4 cm. lata, margine integro-recurva, ssepius apice sinuato-serrata,
subtus glaucescentia. Flores J : calyx aperiens planus, quadratus,
laciniis 4 acutis. Stamina 8 ; antherae introisœ, oblongo-globos?e,
a dorso compressée, 2 mm. longue, filamentis brevioribus, plus
minusve apiculat?e. Flores Ç : calyx tenuis, laciniis 6-8, acutis,
membranaceis ; ovarium oblongo-ovoideum, stylis basi connatis, re-
curvis. Fructus obovoideo-ellipsoideus, 9 mm. longus.

DiSTRiB.: In India orient., Cey Ionia, Japonia.

FIab. Formosa: Su-hun, leg. C. Owatari, anno 1897. Tsui-sia, leg. K.

ÎMiYAKE, anno 1898.
Liu-kiu : Okinawa, leg. Y. TashiBO, anno 1887. Kumeshima, leg. H.

KuROiwA, anno 1898,

34 ART. 2. — B. IIAYAÏA : REVIS lO

Tlotidö : Niijima, anno 1887; Hachijojima, anno 1887. Prov. Idzu, anno 1883.
Prov. Setsu, anno 1880. Prov. Kii, anno 1873. Prov. Nagato, leg. G.
NiKAi, anno 1900.

3. D. glaucescens Bl. vnr. Oldhami Hemsl. in Forbes
et Hemsl. Ind. Fl. 8in. IL 429.

Nom. Jap. Nagabano-himeyuzuriha.

Folia 8-16 cm. longa, 1.5-3 cm. lata, acutissime acuminata,
subtus pallidiora.

Hab. Formosa: Lok-ko-tsng, Tsui-sia, log. K. RTiyakk, anno 1898; Po-li-
sia, leg. K. JMiyake, anno 1898.

4. D. himalayense Muell. Arg. (Tab. II. L.)

MuELL. Arg. in DC. XVI.-l, 4; Hook. f. Fl. Biit. Ind. V. 354.
D. Benthami "Baill. Etude Gen. Euphorb. 565."
Folia supra impresso-reticulato-venosa, 8-11 cm. longa, 2?-
3 cm. lata, lanceolata, acuminata, basi acuta, intégra, subtus fusco-
glaucescentia. Petioli 22-3 cm. longi. Flores J : calyx 5-lobatus,
lobis triangularibus. Stamina 5, rarius 6-7, glandulis clylindraceis
alterna; antlierœ introrsie, filamentis antlieras œquantibus, connectivis
apiculatis. Flores Ç

DiSTRiB.: In Indiœ orientalis regione teni[)erata, et Formosa.
Hab. Formosa : Shihun, leg. K. IMiyake, anno 1898.

5. D. humile Maxim, ex Franch. et Savat. Enum. PI.
Jap. II. 488.

Planta non satis nota.

11 Croton L

INN.

Linn. 8p. PI. ed.-2, 1424 ; Endl. Gen. PI. 1117 ; Muell.
Arg. in DC. Prodr. XV.-2, 513; Baill. Hist.

EUPHOEBIACEARUM ET BUXACEARUN JAPONICARUM. oO

a. PL V. 225 ; Benth. et Hook. Gen. PL III. 293 ;

Pax, ill Nat. Pfl. Fam. I1I.-5, 37.
Flores monoici, petaliferi. FL J : calyx 5-partitiis, segmentis
angiiste iiiibricatis. Petala segmentis ealycis isomera, breviora.
Disci glandulaB ealycis basi affix» v. adnatœ, staminibus
exterioribiis alternie. Stamina 15-20 receptaculo afRxa, exteriora
ealycis segmentis isomera iis alterna, filamentis liberis apice
in alabastro inflexis, per authesin erectis ; antherœ aclnatœ, in
alabastroinfracto-incurva?, apice inferio basi superio, dein oscillando
erectie, loculis parallelis introrsum contignis longitudinaliter
dehiscentibus. Rndimentum ovarii (). Fl. ? : ealycis segmenta
quajn in mare sa?pius angustiora, calyx tamen post anthesin auctus
dilatatusque. Petala vulgo subuliformia quam in mare minora.
Disci glanduhe distinct^e, v. in annulnm connatœ. Ovarium 3-
loculare ; styli dichotome multifidi, v. bifidi. Ovula in loculis
solitaria. Capsula in coccos 2 valves dissiliens. Semina laevia ;
testa Crustacea v. indurata ; albumen copiosum, carnosum ; embryo
rectus, cotyledonibus latis planis. Indumentum e pilis stellatis
sœpe ferrugineis, canescentibus v. squamellis metallice uitentibus.
Folia alterna v. sub inflorescentia subverticillata, integerrima,
pennivenia v. 5-nervia, basi biglandulosa. Flores ad apices
ramorum v. in axillis summis spicati, fl. J breviter pedicellati, fl. $
sessiles v. pedicellati. Bractese parvie, v. sub fl. $ auctse.

Conspectus Grotonum Specierum.

Folia penniiurviii, siibtus iiidumento albido-fulvo vestn 1. C Cuiningii.

Folia palmaLinervia, .^ubtus pilis rigidulis parce vesta 2. C. Tiglium.

1. C Cumingii Ml ell. Arg. (Tab. III. B.) Muell.

36 ART. 2. B. HATAYA : REVISIO

Arg. in Linnœa, XXXIV. 101 et iii DC. Prodr. XV.-2, 566 ;
Forbes et Hemsl. Tnd. Fl. Sin. IT. 434; Henry, List PI. Formos.
84.

G. 2^olystachyum Willd.; Hook, et Arn. Bot. Beech. Voy. 270.

Nom. Jap. Chankanii.

Folia alterna, breviuscule petiolata, penninervia, subtus dense
albido-lepidota, in apice ramorum conferta, 5-11 cm. longa, 2è-5
era. lata, obovata v. elliptica, lanceolata, acuminata, basi anguste
obtusa, fere intégra. Fl. J : calycis lacinise ovatœ, margiiie
membranaceœ, fuscœ, squamoso-stellato-pilosœ. Petala subulata,
intus basi margineque pilosa, glandulis sparsa. Stamina 15-20,
receptaculo lanato-pubescenti inserta, filamentis rubris inferne
pilosis, connectivis nigris. Pedicelli breves, calycis lacinias sub-
sequantes. Fl. $ : calyx ut in $ ; petala glanduligera, truncata,
calycem subaequantia, spathulata, margiiie et intus basi longe pilosa.
Ovarium dense lepidoto- squamosum, stylis multifidis gracilibus.
Capsula3 7 mm. longse, 6 mm. latie, tricocc^e, coccis apice
gibboso-productis. Semina oblongo-globosa, apice acuta, elegante
reticulata.

DisTRiB.: In insulis Philippiiiis, Formosa, l.iu-kiii.

Har. Füiraosa : Sübougsha, Thoug-po-tsiig, leg. K. Miyake, anuo 1898;

Bo-(an-oan, Pak-kang-khœ, C. Owatari, aniiü 1896 ; Heng-chhun, Y.

Tashiro.
Liu-kiu : Miyakujima, Yaeyama, Okinawa, leg. Y. Tashiro, anno 1887.

2. C. Tiglium Linn. (Tab. III. C.) Linn. Sp. PI.

ed.-2, 1426; Lour. Fl. Cocliinch. 714; Willd. Sp. PI. IV. 453;
AiTON, Hort. Kew. V. 327; Miq. Fl. Ind. Bat. I.-2, 379; Roxb. Fl.
Ind. III. 682 ; Muell. Arg. in DC Prodr. XV.-2, 600 ; Hook.
f. Fl. Brit. Ind. V. 393 ; Forbes et Hemsl. Ind. Fl. Sin. IL

EUPHORBIACEARUM ET BUXACEARUM JAPONICARUM. 37

435 ; Henry, List PL Formos. 84 ; L. Dielp, in Engl. Jahrb.
XXIX. 428.

Nom. Jap. Ha zu.

Ramiili et petioli glabri. Folia 5-nervia, firma, basi

biglandulosa, 5-14 cm. longa, 2-7 cm. lata, ovata, acuta, basi

obtusa, margine plus minusve crenulato-serrulata, v. fere intégra,

petiolo vulgo duplo longiora, pilis, adpressis, vulgo brevibus,

rigidulis, sparsa. Petioli 3-4 cm. longi. StipuUe subulatœ

deciduissimœ. Racemi multiflori, bracteis lanceolato-subulatis. Fl.

J : sepala ovata, apice incras^ata, intus glabra, lucida, extus pilosa,

membranaceo-albida. Petala subulata, apice capitato-incrassata.

Stamina 15-18, filamentis glabris, receptaculo piloso conferta :

Pedicelli calyce longiores v. eum œquantes, tenues. Fl. $ : sepala

subulata, crassata. Petala angusta, sejmlis alterna. Glandulaß

graciles, virides. Ovarium stellato-liirtellum, stylis elongatis

apice bifidis. Capsula magnte, leptodermse, pallidiores, matune

2 cm. longœ, 1.5 cm. latœ, ellipsoideœ, obtus^e, trigonse, leviter

longitrorsum trisulcatee. Semina oblongo-ovoidea, apice apiculata,

12 mm. longa, 6 mm. in diametro œquantia, leviter reticulata.

DisïRJB. In India orient., Borneo, insulis Pliiüppinis, Malaya, Formosa.
Hab. Formosa: Lok-ko-tsung, Po-li-sia, Chip-chip-koe, leg. K. Miyake, anno
1898.

12. Mercurialis Linn.

Linn. Sp. PI. ed.-2, 1465 ; Endl. Gen. PI. 1111 ; Muell.
Arg. in DC. Prodr. XV.-2, 794 ; Baill. Hist. d.
PL V. 210; Benth. et Hook. Gen. PL IIL 309;
Pax, in Nat. Pfl. Fam. III.-5, 49.

38 ART. 2. — B. HAYATA I EEVISIO

Flores monoeci, apetali. Fl. J: calyx tenuiter membraDaceu?,
in alabastro globosus, clausus, per anthesin valvatim 3-partitus.
Discus 0. Stamina 16-20, in receptaculo elevato ceiitralia, fila-
meutis teiiuibus, erectis, liberis ; anlherarum loculi a basi distincti,
pyriformes, ex apice primum pencluli demum ascendentes, superne
longitudiiialiter déhiscentes. Kimae antherarum extrorsse. Ovarii
rudimeiitum 0. Fl. $ : sepala 3. Disci lobi 2, lineari-subulati,
carjiellis alternantes. Ovarium 2-loculare ; styli vix basi connati,
longiueculi, divergentes, intus prominenter papilloso-stigmatosi,
indivisi ; ovula in loculis solitaiia. Capsula didyma, in coccos 2-
valves dissiliens, endocarpio crustaceo, Semina ovoidea. Plerbie
perennes, glabrae, nitidœ. Folia oj)posita, stipulata, petiolata, vulgo
crenato-serrulata, pennivenia. Eacemi axillares. Acribus supra
medium in glomerulos paucos distantes collectis.

1. Mercurialis leiocarpa Sieb, et Zucc. (Tab. III. D.)

— Sieb, et Zucc. Fl. Jap. Fam. Nat. Sect. Prim. 145 ;

MuELL. Arg. in DO. Prodr. XV.-2, 795; Miq. Prol. 291; Franch.
et Savat. Enum. PI. Jap. I. 425 ; Forbes et Hemsl. Ind. Fl.
Sin. IL 436 ; L. Diels, in Eng. Jahrb. XXIX. 428.

Nom. Jap. Yama-ai, Kwa-wi, herb., vol. 4, fol. 24 ; Somoku-
dusetsu, vol. 20, fol. 63.

Folia longe petiolata, membranacea. Petioli limbum sub-
aequantes v. eo stepius circiter duplo breviores, 2-3 cm. longi.
Folia 3-10 cm. longa, 2-4 cm. lata, oblongo-ovata, v. ovato-lanceo-
lata, longius breviusve acuminata, basi obtusa, acutiuscula, rarius
leviter cordata, crenato-serrata, supra nitida, purpurascentia,
parce pubescentia, crenis apice glanduligeris, basi biglandulosa.
Flores J : glomerato-spicati ; alabastri trigoni, acuti, 2-3 glomerati.

p:UPHOPtRTACEARUM ET RUXACEAKUM JAPON FC AHUM. oO

Flores Ç : sepalis ovntis, a cutis ; ovarium dorso geminatim 2-4-

muricatuni, stylis ovario adpressis ; disci glandulse auguste, lougce,

ovarium superantes.

DisTHiB.: In Japonin, Corea, China cent. .

Hab. Lin-kiu : insula Okinawa, log. Y. Tashiro, nnno 1887.

Kiushu : piov. Hyûga, prov. Buzen, in monte Inugatake, anno 1882 ; prov.

Chikuzen, in monte Taehibana, leg. K. Nagano.
Hondo: Prov. Suwö, in tractu Yoshiki, leg. G. NiKAi, anno 1803. Prov.

Kii, anno 1883; Prope Tokyo. Hikawa, ipse, anno 1904.

13. Mallotus Lour.

Lour. FI. Cochinch. 780 ; Endl. Gen. PL 1116 ;

MuELL. Arg. in DC. Prodr. XV.-2, 256; Benïh.

et Hook. Gen. PI. III. 319; Pax, in Nat. Pfl.

Fam. III.-5, 53.
Flores dioici, apetali. Discus 0, rarius evolutus. Fl. J : calyx
in alabastro globosus, clausus, per antliesin valvatim 5-partitus.
Stamina oo^ recej^taculo subdilatato parum elevato imposita, filamentis
liberis ; anther«) birimosœ longitudinaliter introrsum déhiscentes.
Ovarii rudimentuni 0. Fl. Ç : calyx alte v. breviter 3-5-lobatus,
lobis valvatis v. vix imbricatis, v. 5-partitus. Ovarium 2-3-
loculare ; styli distincti v. basi connati, recurvo-patentes, indivisi,
intus fimbriati v. prominenter papillosi ; ovula in loculis solitaria.
Capsula globosa v. di-tri-dyma, in coccos 2 valves dissiliens. Semina
globosa V. ovoidea, testa Crustacea ; albumen carnosum ; embryo
rectus, cotyledonibus planis, latis. — Arbores v. frutices. Folia
alterna, opposita, bistipulata, petiolata, lata, sœpe ampla, integerrima,
dentata v. late 3-loba, basi 3-7-nervia, glandulis disciformibus,

40 ART. 2. — B. HAYATA : REVISIO

nitentibus, suhaureis demum fuscis v. rufescentibus, adspersa, prope
basin supra 2-v. miilti-glandulosa. Eacemi v. S[)icœ simplices v.
paDiculato-ramosi, axillares, v. ad apices ramorum paniculati. Fl.
J sœpius parvi, breviter pedicellati, secns rhachin glomeriilati, fl. ?
pauciores, sub quaque bractea solitarii, brevius v. loiigius pedicellati.

Conspectus Mallolorum Specierum.

Folia basi pluriglandulosn ; discus fl. Ç evolutus 1. M. moluceanus.

Folia basi biglandulosa ; discus obsoletus.

Folia opposita 2. M. Playfahii.

Folia alterna.

Ovarium inerme.

Ovarium 2-locularo 3. 31. rcpandiis.

Ovarium 3-loculare 4. M. jihilippinensis.

Ovarium molliter eehinntum.

Folia subtus minute flaveo-glanduligera, demum utrnque pagine

glabrata 5. M. japonicus.

Folia fenuginen, nigra, subtus densissime breviterque stellato-
tomentosa 6. M. cochinchinensis.

1. M. moluceanus Muell. Arg. (Tab. III. E.)—

MuELL. Arg. in Linnœa, XXXIV. 185 ; Muell. Arg. in DC.
Prodr. XV.-2, 958; Hemsl. in Voy. Challenger, Bot. I.-3,
191 ; Forbes et Hemsl. Ind. Fl. Sin. II. 440; List PI. Formos.
84.

Jlelanolepis multiglandulosa Reichenb.; Zoll, in Linnœa,
XXVIII. 824.

Nom. Jap. Yanbaru-akaynegashiwa.

Ramuli, cum juvenilibus et inflorescentibus, tomento sub-
floccoso, fulvo, dense tecti, dein glabrati. Folia ampla, subtus
stellato-tomentosa, ambitu suborbicularia, 5-nervia, longe petiolata,

EUPHORBIACEARU.M ET BUXACEARUM JAPONICARÜ.Ar. 41

petiolis linil)iini foliornin subœquantibns, sinnato dentata, dentibus
miicronatis, basi pluriglaiidalosa, elobata v. subtrilobata, 4-1 ö cm.
longa, late rhorabeo-ovata, acuminata, basi breviter contracta v.
subcordata, utraque pagine sœpe tomento flocco tecta, demnm
glabra. Racenii panicnlato-ramosi ; pedicelli longi, calycem sn-
perantes. Flores dioici casii monoici. Fl. ^ : calyx 5-partitus ;
ovarium oblongo-globosum, 2-3-loculare, stylis minutis, simplicibiis;
discus evolutus, urceolaris, crassiusculus, crenatus. Fl. J : calyx
5-partitus, segmentas valvatis, crassiusculis. Stamina oo, rubella,
fijamentis liberis, filiformibus, connectivis coccineis vulgo obconicis
productis, antheris versatilibus ; receptaculum stellato-pubescens.
Capsulse dense subpulveraceo-stellato-tomentelUr, circ. 1 cm. latie,
8 mm. longse, stylis persistentibus. Semina foveolata, oblongo-
globosa, apice acuta, diametro 4 mm. œquantia, arillis spuriis
violaceo-purpureis.

DisTRiB. Ill iii.<ul;i Java, Borneo, Timor, Philippinif-, India orieiUali, For-
mosa, Lia-kiLi.

Hab. Formosa: Takno, Pati-te-chhu, leg. K. Miyake, anno 1898; Kciig-
eliio-kha-tsng, Hosonng, K. Miyake, C. Owataki, anno 1896.

2. M. Playfairii, Hemsl. (Tab. III. F.) Forbes

et Hemsl. Ind. Fl. Sin. II. 411 ; Henry, List PL Formos. 84.

Nom. Jap. Shima-azvsa.

Frutices parvi, ramulis crassis compressis dense et brevissime
stellato-tomentosis pallidis. Folia opposita, subcoriacea, rotunda-
to V. cordato-ovata, 5-14 cm. longa, intégra v. ol)scure calloso-
dentata, abrupte acuminata, suj^ra cristis pilorum brevissimorum
dense congestorum parce conspersa, sub tus simul brevissime albido-
tomentosa, venis primariis lateralibus utrinque circiter 6, secundariis
subrectis conjunctis. Petioli tcretes, limbo breviores. Fl. J
aperientes 5 mm. diametro œquantes ; calyx 4-partitus, segmeutis

42 ART. '?,. — B. HAYATA : REVISIO

late ovatis vix acutis. Stamina oo, extrorsa, loculis antheranim
connectivum sajîerantibus. Flores brevissime pedicellati, peclicellis
sepala gequantibus, fasciculatim racemosi, racemis foliis brevio-
ribus.

Hab. Formosa : Takao, leg. C. Owataei, amio 1897.

3. M. répandus JMuell. Arg. (Tab. III. G.)

MuELL. Arg. in Linniea, XXXIV. 197, et in DC. Prodr. XV.-
2, 981 ; Benth. Fl. Austral. VI. 142 ; Hemsl. in Voy. Cliall.
Bot. I.-3, 191 ; Hook. f. Fl. Brit. Ind. V. 442 ; Forbes
et Hemsl. Ind. Fl. Sin. II. 441 ; Henry, List PI. Formos. 84 ;
L. DiELS, in Engl. Jahb. XX [X. 429.

Croton rhombifoUus Willd. Sp. PI. IV. 555.

Rottlera trinervis Zipp. in Linnsea, XV. 348.

E. iricocca Roxb. Fl. Ind. III. 829.

R. cordifolia Benth. Fl. Hongk. 307.

R. scdhrifolia A. Juss.; Keiciib. et Zoll, in Linuîea,
XXVIII. 319.

Trewia nvcUffora Hange, in Journ. Bot. 1878, 14.

Mallotus contubernaUs Hange, in Journ. Bot. 1882, 293.

Xom. Jap. Tsiiru-akamegashiiva.

Ramuli juniores, cum petiolis, foliis, inflorescentiis, brevius
stellato-fulvo-tomentelli. Folia longe trinervia, basi subo])solete
peltata, 4-7 em. longa, 4-5 cm. lata, ambitu varia vulgo triangulari-
ovata, abrupte acuüi, basi obtusa, subcordata, subintegra v. repando-
denticulata, lutescenti-viridia, supra subglabra, subtus stellato-
tomentella demum glabrescentia, glandulis ceraceis aureo-nitentibus
ad pressa, basi supra biglandulosa. Spicse in ramulis terminales,
foliis siepe longiores, fulvse. Bracteœ J S-fîorse, Ç l-florse,
tiiangulari-ovatœ, acurainatse. Fl. J: calyx 4-partitus, segmentis

EUPHORBIACEARUM ET EUX ACE ARUM JAPONICARUM. 43

ovatis, acutis, glaiidulis pallidis majusciilis adspersis. Stamina circ.
50, rcceptaculo parvo congesta. Fl. 2 : calyx 4 lobatus, lobis
crassiusculis, ovatis dein lanceolatis ; ovarium globosum, 2-loculare,
glandulis adpressum, pubescens, stylis basi connatis intus ramuloso-
papillosis. •

DisTHTB. In India orientali, Ceylonia, Java, Philippinis, Formosa.
Hab. Formosa : Takao, leg. C. Owatari, anno 1897.

4. M. philippin ensis Muell. Arg. (Tab. III. H.)

MuELL. Arg. in Liunœa, XXXIV. 196 et in DC. Prodr. XV.-
2, 980 ; Benth. Fl. Austral. VI. 141 ; Beddome, Fl. Sylv. Sau.
Ind. t. 289 ; Hange, in Journ. Linn. Soc. XIII, 122 ; Hemsl.
Voy. Cball. Bot. I.-3, 190 ; Hook. f. Fl. Frit. Ind. V. 442 ;
Forbes et Hemsl. Ind. Fl. Sin. IL 440; Henry, List PI. Formos.
84; L. DiELS, in Engl. Jabrb. XXIX. 428.

Croton moîilanus Willd. Sp. PI. IV. '545.

RoUlera tlncioria Roxb. Fl. Ind. III. 827 ; Benth. Fl.
Hongk. 307.

R. aurantiaca Hook, et Arn. Bot. Beecli, Voy. 270.

Xom. Jap. Kusunoliagashhoa.

Bamuli juniores, petioli et infiorescentite pilis stellatis brevibus
ferrngineo-tomentelli. Folia alterna longe triplinervia, 5-20 cm.
longa, 1-4 rarius 7 cm. lata, ovato-lanceolata, v. rhombeo-lanceolata,
acuminata, basi acuta, supra biglandulosa, intégra v. denticulata,
subtus inter indumentum brevissime tomentellum coccineo-
glanduloso-pulveracea, supra glabra, demum sensim expallentia,
petiolis limbo foliorum 2-4-plo brevioribus. Spicse axillares v.
terminales, ferruginese, subpulveraceo-tomentelliç. Discus utriusqiie
sexus 0. Bracteîc 1 mm. longue, basi incrassatae, l-flone, glandulis
cocciiieis adpressie. Fl. J : sepala lanceolata, crassiuscula ; stamina

44 RAT. :;. B. HAYAÏA : REVISIO

cire. 30. Fl. ^ : calyx '2-21 nnii. loiigus, irregulariter 3-lobatus,
lobis ovatis ; ovarium dense coccineo-glandulosum, 3-loculare,
stellato-loinentellum ; styli basi connati ovarium sub^equantes, intus
fimbriato-papillosi. Capsula' inermes depresso-globosa^, profunde
trisulcatir, pedi cello fructigero 2-3-plo lougiores, 8-9 mm. lata',
G-7 mm. longœ, dense coccineo-glanduloste. Semina globosa,
4 mm. longa.

DiSTKiij. In Indiii orientali latissimc distiibuta ; Ceylonia, Java, insulis

Philippinis, Hongkong, Forjnosa, Liu-kiu.
Hab. Formosa : Kacliiraisha, Thong-po-tsng, Pang-liau, K. Miyaké, anno

1898. Taipea, Pachina, leg. T. Making, anno 1896, et C. Owataki,

anno 1897.
Liu-kiu : Okinawa, leg. Y. Tasiiieo, anno 1887. Yaeyania, leg. »S. Taxaka,

anno 1891. Kobito, leg. M. Miyajima, anno 1900.

Ö. M. japonicus Muell. Arg. (Tab. III. I.)

MuELL. Arg. in Linnaa, XXXIV. 189 et in DC. Prodi-. XV.-
2, 966 ; Maxim, in Engl. Jalirb. VI. 59 ; Forbes et IIemsl.
Ind. Fl. Sin. 11. 440 ; Henry, List PL Formos. 84 ; L. Diels,
in Engl. Jahrb. XXIX. 428 ; Shirasawa, Xipponshinrinju-
mokudusetsu, t. 55.

Ci'oton japonieuDh Thunb. Fl. Jap. 270.

Rottlera japouica 8ieb. et Zucc. Fl. Jap. 147, t. 79.

M. japouica Spreng.; Hook, et Arn. Bot. Beecli. Voy. 270 ;
Sieb, et Zucc. Fl. Jap. Fam. Xat. Sect. Prim. 144; Miq. Prol.
291.

Nom. Jap. Äkamegashiwa, Kwawi, arb., vol. 1, fol. 18, sub.

Arbores. Eami lignosi, teretes, glabri, ramuli cum partibus
omnibus juvenilibus pilis stellatis tecti. Folia 7-14 cm. longa,
5-10 cm. lata, membranacea, suborbiculari-ovata, rhombeo-ovata,
V. rhombeo-orbicularia, cuspidato-acuminata, supra repando-
tricuspidata, lobata, v. interdum elol)ata, basi liaud peltata, cuneato-

EÜPHORBIACEAKUM ET BUXACEARIM JAl'ONICARlJM. 40

subacuta, obtusa v. subcordata, triplinervia, margine intégra, sul)tus
minute flavido-glanduligei-a, deraum utraque pagine glabrata,
petiolis limbum subœquantibus, stellato-puberulis, glabrescentibus.
Flores ramoso-spicato-racemosi, subferrugineo-tomentosi, terminales,
bracteis triangularibus acutis interdum fidis parvis. Flores J :
stamina 50-60, filamentis 2 mm. longis, localis antlierarum superne
subpatulis. Flores o ; calyx 3-Ö- irregulariter fidus ; ovarium
dense coccineo-glandulosum, 3-loculare, stellato-tomentosum,
molliter ecliinatum ; styli basi connati ovarium bis superantes
intus dense longeque papilloso-plumosi. CapsuUe molliter echinatai,
glandulis adpressis, aculeis mollibus tenuibus stellato-puberulis,
absque aculeis 8 mm. latœ, 5 mm. longas, tridymœ, vinosorubellœ.
Semina compresso-globosa.

DisTRiB.: in China boreali, J;)poni;i.

Hab. Foiniosa : Tai-tong-thiun, Pliok-cliioli-koh, leg. K. Mivakk, anno 189;>.
Tiit-chliioh, leg. K. MiYAKE, anno 1899; Sia-lian-to, kelung, Pilaiu,
Tnmt!<ui, leg. T. Making, et C. Owatahi, anno 1896.

Liii-kiu : kumesiina, leg. H. Ktjruiwa, aimo 1898.

6. M. cochinchinensis Lour. (Tab. II r. j.)

Loue. Fl. cochinch. 635; Muell. Arg. in Linmea, XXXIV.
180; Hook. f. Fl. Brit. Ind. V. 430; Forbes et Hemsl. Ind.
Fl. Sin. II. 439 ; Henry, List PL Formos. 84.

Trewia trlcuspidata Willd. Sp. PL IV. 831.

Rottlera paniciilata A. J.; Benth. Fl. Hongk. 307 ; Hook.
et Arxx. Bot. Beecli. Voy. 212; M]q. Fl. Ind. Bat. L-2, 395.

Jlallotus paniculatus Muell. Arg. in Linnœa, XXXVI.
189 et in DC. Prodr. XV.-2, 965 ; Benth. Fl. Austral. VI. 140.

Xom. Jap. Annan- akamegashiiva.

Folia longe petiolata, rhombeo-ovata, acute cuspidata, interdum
lobato-tricuspidata, basi abrupte acuta, epeltata, basi biglandu-

46 ART. o. JJ. HAYAÏA : IIAVI.SIO

losa, luargiiie subiiitegra, subtiis densissime breviterque stellato-
tomentosa et ferruginea. Spictie paniculato-raraosae, terminales.
Bractese J o-florœ, $ l-florse, triangulares, acutie, bi'cves. Calyces
pilis stellatis ferrugineo-tomentelli. Fi. J staminis niimerosis,
antheris parvis. Capsulse multo validius appendiculato-subechinatae,
absque aculeis 7 mm. latœ, o mm. longœ ; aculei 4-5 mm. longi.
Seraina 3 mm. longa, ovoidea, vivo glabra, sicco rugulosa.

DisTRiB.: In Java, Borneo, Älalaya, Hongkong, vnlgaris.
Hab. Formosa : Su-hun, leg. C. Owatari, anno 1897. Png-lcang, Sûhonslia,
Ang-thau-bU, leg. K. Miyake, anno 1898.

14. Alchornea s. w.

Endl. Gen. PI. 111:3; Muell. Arg. in DC. Prodr.

XV.-2, 899; Baill. Hist. d. PI. Y. 212 ; Benïh.

et Hook. Gen. PI. III. 314; Pax, in Kat. Pfl.

Fam. III.-ö, ö'~).
Flores dioici casu monoici, apetali. Discus 0. Fl. J : calyx
in alabastro globosus, clausus, per antbesin valvatim 4-partitus.
Stamina 8; filamcntis liberis v. ssepius basi brevissime in annulum
connatis; antbene introrsœ, loculis distinctis parallelis longitudi-
naliter debiscentibus. Ovarii rudimentum 0. Fl. $ : sepala ö,
imbricata. Ovarium 3-loculare ; styli distincti, lineares, indivisi ;
ovula in loculis solitaria. Folia alterna, serrulata, 3-nervia,
basi supra 2-glandulosa. Spic?e terminales, simplices, floribus J
parvis secus rbacbin glomeratis. Bracteie utriusque sexus minuta',
Horil)us 2. sul^ bractea solitariis. — Frutices. Folia alterna, petiolata
btipulata.

EUPHORBIACEAEUM ET BUXACEARUM JAPONICARFM. 47

1. A. trewioides Muell. Arg. (Tab. IV. A.)-

MuELL. Arg. in Linnœa, XXXIV. IßS et in DC. Prodr. XV.-
2, 901.

Nom. Jap. Amigasagiri.

Folia longe petiolata, cordata, ovata, ciispidato-acuminata,
7-15 cm. longa, 6-14 cm. lata, basi biglandulosa, margine serrnlata,
stipulis lanceolatis, linearibiis. Flores ,$ aperientes 5 mm. diametro
œquantes, sepalis 1 è mm. longis, triangularibus acutis. Stamina 8
biseriatim disposita, filamentis crassis, Fl. o : sepala lanceolata
acuminata, 21 cm. longa, ovarium superantia ; ovarium globosum,
stylis basi connatis patulis, intus papillosis, 5 mm. longis. Flores
utriusque sexus subsessiles.

DiSTRiB.: In Hongkong, Liu-kiu.

Hab. Liu-kiu : leg. J. Matsumura, anno 1897; kunie.-^inin, leg. H. Kuruiwa,
anno 1898.

15. Cleidion Blume.

Endl. Gen. PI. 1113; Muell. Arg. in DC. Prodr.

XV.-2, 984 ; Baill. Hist. d. PL V. 203 ; Benth.

et Hook. Gen. PL HI. 320 ; Pax, in Nat. Pfl.

Fam. III.-5, r>8.
Flores dioici, apetali. FL J : calyx in alabastro globosus,
clausus, per anthesin valvatim 3-partitus. Stamina receptaculo
convexo v. conico dense conferta, filamentis liberis ; anther«)
dorsifixœ, connectivo sœpius breviter ultra loculos producto, loculis
lateralibus transverse didymis. Ovarii rudimentum 0. FL ^ ...
Folia alterna, petiolata, ampla, dentata, trinervia. Racemi J
terminales, paniculati, floribus glomeratis.

48 AET. ?>. — B. HAYATA : RAYTSTO

1. C. Ulmifolium IMuell. Arg. (Tab. TV. B.)

MuELL. Akg. in DC. Prodr. XV.-2. 984; Forbes et Hemsl.
Iiid. Fl. Sin. II. 442.

Nom. Jap. Enohifuz}.

Ramnli teretes, densius foliosi, juniores superne leviter flexuosi,
rnbelli, stipnlis siibnlatis, 2? nnn. longis, cadocis. Folia 5-10 em.
longa, 2.;j-5 cm. lata, oblongo-ovata, acuminata, basi rotundata,
leviter cordata, v. obtusa, grosse crenato-serrata, membranacea,
obscure viridia, graciliter reticulato-venosa, petiolis 1-3 cm. longis.
Racemi J l^racteis ovatis, acuminatis, 3-5-floribus; sepala aperientia
2 mm. longa, ovata-acuta.

Hab. Liu-kill : Knme.-ima, leg. H. Kuruiwa, anno 1898.

16. Macaranga Thouars.

Endl. Gen. PI. 1112; Muell. Arc;, in DO. Prodr.

XV.-2, 988 ; Batll. Hist. d. PI. V. 208 ; Benth.

et Hook. Gen. PI. III. 320; Pax, in Nat. Pfl.

Fam. III.-5, 59.
Flores dioici, apetali. Discus 0. Fl. J : calyx in alabastro
globosus, clausus, apice punctatus, per antliesin valvatini 4-partitus.
Stamina 7-9, in receptaculo piano conferta, filamentis brevibus
liberis; antlierœ terminales, breves, 2-loculares. Ovarii rudimentum
0. Fl. -f : calyx per anthesin oblique spatbaceus. Ovarium 2-
loculare. Styli breves crassiusculi patentes, indivisi, basi breviti']'
c )nnati ; ovala in loculis solitaria. Capsula sparse moUiter ecbinati.
Folia alterna, petiolat.i ampla integerrima, basi peltata. Flores ?
spicati ; J racemosi, axillares, ramosi, intra bracteas numéros:,

EUPHORBIACEARUM ET BIJXACEARUM JAPONICARUäT. 49

parvi, secus rliacliin sjlomerati, fl. ^ intra bracteas solitarii.
Bracteœ flores amplectantes, fimbriatœ.

1. M. Tanarius AIuell. Arg. (Tab. IV. C.)

]\ruELL. Arg. in DO. Prodr. XV.-2, 997 ; Benth. FI. Austral.
VL 14G; Hemsl. in Voy. Cballeng. Bot. I.-3, 191; PIook. f.
FL Brit. Ind. V. 447 ; Forbes et Hemsl. Ind. FL Sin. II. 443 ;
Henry, List PL For mos. 84.

Ricinus Tanarius Linn. Sp. PL ed.-2, 1430.

R. Mappa'JRoxB. FL Ind. III. 690.

Mappa moluccana Wkjht, Ic. Ind. or. t. 816.

31. Tanaria Spreng.; Benth. FL Flongk. 304 ; Zolling, in
Linnœa, XXVIII. 307 ; MiQ. FL Ind. Bat. I.-2, 401.

Nom. Jap. Ohahi.

Hamuli cura petiolis glabri, glauco-pruinosi. Folia 10-30
cm. longa, evoluta vix longiora quam lata, triangulari-orbicularia,
late peltata, acuminata, basi rotundata v. obtusa, repando-
denticulata v. intégra, subtus granulis nitidis exguis adspersi,
moUiter tomentosa, stipulis 12-15 mm. longis, Lmceolato-oyatis
acuminatis, scariosis, fuscis, petiolis longis limbum suboequantibus.
Flores J paniculati, paniculis longe pedunculatis, bracteis fimbriato-
dentatis, ovatis, flabellato-nervosis, cucuUato-concavis, flores
amplectantibus. Cal3'x 2-3-4-partitus, segmentis ovatis acuminatis.
FL ? : ovarium depresso-globosum, stigmatis exsertis, granulis
pulchre pallide cereis et ceraceis dense adspersum.

DiSTEiB.: In Malnva, Foimosa, Liu-kiu.

Hab, Formjsa : Sia-liau-to, Paclnna, kelung, leg. T. Making, annolSQG;
Leng-nga-liau, Hoetong, leg. C. Owatari, anno 1896; Subonslia, Tai-
tong-thian, To-loan-sia, leg. K. Miyake, anno 1898.

Liu-kiu : Okinawa, leg. Y. Tashiho, anno 1887 ; Yaeyama, leg. S. Tanaka,
anno 1893; Kobitö, leg. M. Miyajima, anuo 1900.

50 ART. 3. — B. HAYATA : KEVISIO

17. Acalypha Linn.

Ltxn. Sp. PL ed.-2, 1423; Endl. Gen. PI. 1111;

MuELL. Arg. in DC. Prod. XV.-2, 800 ; Batll.

Hist. d. PI. V. 212 ; Benth. et Hook. Gen. PI.

III. 311 ; Pax, in Nat. Pfl. Fam. III.-o, 60.
Flores monoeci, apetali. Discus 0. Fl. J : calyx minimus,
tenuiter membranaceus, in alabastro clausus, per anthesin valvatim
4-partitus. Stamina 8, 2-scriatim disposita, exteriora sepalis
alterna, receptaculo parum elevato affixa, filamentis liberis ; antherœ
biriraosœ ; loculi subvermiformes, divaricati, oblongi, in flore
aperto saepissime flexuosi. Ovarii rudimentum 0. Fl. ? : sepala
3, imbricata minima ; ovarium 3-loculare ; styli liberi, filiformes,
in lacinias filiformes divisi ; ovula in loculis solitaria. Capsula
tridyma, parva, in coccos 2-valves dissiliens. Semina subglobosa,
testa Crustacea ; embryo rectus ; cotyledones latœ planœ. — Herbre.
Folia alterna, ovata, plus minusve dentata, petiolis longis. Flores ^
parvi, sub bracteas parvas glomerati, in spicas amentiformes,
demum articulato-decidui. Flores o intra bracteam demum auctam
f)Iiacearaque 2-4--glomerati.

1. A. australis Linx. (Tab. IV. D.) Linn. Sp. PI.

cd.-2, 1424 ; Muell. Arg. in Linngea, XXXIV. 41 ; Forbes
et Hemsl. Ind. Fl. Sin. II. 437.

A. virgata Thunb. Fl. Jap. 268.

A. chinensls Roxb. Fl. Ind. III. 677 ; Hook, et Arx. Bot.
Beech. Voy. 213.

EUPHORBlACEARüM Eï BUXACEARUM JAPONIC ARUM. Ol

.1. paucijiora Horxem.; Maxim. Prim. Fl. Amur. 240 ; IMiQ.
Prol. 201 ; Franch. et Savat. Eiium. Fl. Jap. I. 421 ; Francii.
PI. David. 2G4.

Ä. gemina Muell. Arg. in DC. Prcdr. XV.-2, 880.

Nom. Jap. Enokigiisa, Sômokuzusetsu, XX. fol. 12.

Folia 3-7 cm. longa, 1-4 cm. lata, snbrliombeo-ovata, obtusa
V. acuta, apice acuminata, tenuia nunc firma, serrulato-dentata.
Flores J : calyx 4-partitus, sepalis extus verrucosis glanduligeris ;
filamenta basi connata, spathulata, antheras vermiformes in longi-
tudido œquantia. Pedicelli articulati, caducissimi. Flores Ç : calyx
3-partitus, sepalis ovatis acutis margine ciliolatis glanduligeris
pubescentibus ; ovarium globosura tuberculato-scabrum pubescens ;
styli fimbriati. Bractei« triangulari-ovatie. Bracteie fructigene
10-15 mm. longae, virides, parce pubescentes. Capsuhe globosie
profunde trisulcatse pubescentes, 2 mm. longte. Semina ovoidea
levia 2 mm. subœquantia.

DiSTRiB.: Muiulsluiri;i et Japonia.
Hab. Formosa : Keluiig, leg. T. Making, anno 1896.
Liu-kiu : Okinawajima, leg. Y. Tashiro, anno 1887.

Hondo : öuwö, in tractu Yoshiki, leg. G. NiKAi, iinno 189-1 ; prov. Tushima,
leg. Y. Yabe, anno 1901 ; propo Tokyo.

2. A. australis L. var. lanceolata var. nov.

Nom. Jap. Hosobano-enokigusa.

Folia angustiora, 2.5-5 cm. longa, 5-7 mm. lata, lanceolata,
oblongo-lanceolata, apice obtusa v. mucronata, repando-dentata v.
crenata. — Rami inferne fasciculati.

Hab. Formosa : Kelung, leg. Ï. Making, nuno 1896.

52 ART. ;J. — ü. HAYATA : IIEYISIO

18. Ricinus Linn.

LiNN. Sp. PL ecl.-2, 1480; E.xdl. Gen. PI. 1115;

MüELL. Aeg. in DC. Prodr. 1017 ; Baill. Hist.

(1. PI. V. 178 ; Benth. et Hook. Gen. PI. III.

321 ; Pax, in Nat. Pfl. Farn. IIL-5, 70.
Flores monoici, apetali. Discus 0. Fl. J : calyx membranaceus,
in alabastro globosa^i, per anthesin valvatim 5-partitus. Stamina
in receptaculo plano-convcxo conferta, numerosa, filamentis repetito-
ramosissimis ; antheraruni loculi distincti didymo-subglobosi
divaricati, longiiudinaliter déhiscentes. Ovarii rudimentura 0. Fl.
? : calyx 3-spatliaceo-fissus, caducissimus. Ovarium 3-loculare ;
styli basi connati longiusculi, patentes, ultra medium bipartiti,
intus valde papillosi ; ovula in locnlis solitaria. Capsulse tridymte,
in coccos 2-valves déhiscentes, magme, o-sperma3. Semina ovoidea,
testa Crustacea, carunculata ; albumen oleosum ; cotyledones planie,
latic. — Herbie elatte, annucC, glabra?, glaucœ. Folia alterna, ampla,
longe peltata, palmatim 7-8 lobata, lobis serratis. Kacemi ad
apices ramorura subpaniculati. Acribus majusculis, superioribus fi. $
confertis, inferioribus fl. J breviter pedicellatis.

1. R. communis Linn. Sp. PI. ed.-2, 1430; Willü.
Sp. PI. IV. 564 ; LouK. Fl. Cochinch. ed.-2, 716 ; Aiïon, Hort.
Kewensis, V. 331 ; Hook, et Arn. Bot. Beech. Voy. 212 ; SiEn.
et Zucc. Fl. Jap. Farn. N-^t. Sect. Prim. 145; Hook. Nig. Fl.
178; MiQ. Fl. Ind. Bat. I.-2, p. 31)1; Bentii. FL Hongk. 307;
Bot. Miig. t. 2201); Grisehach, FL Brit. West lud. 37 ; Muell.

EUPHORBIACEAllUM ET BUXACEARUM JAPON ICARÜM. Oo

Arg. in DC. Prodi-. XV.-2, 1017; Miq. Prol. 291 et in Ann.
Mus. Bot. Lugd.-Bat. IIL 127 ; Seem. Fl. Vitiensis, 229 ; Hemhl.
in Voy. Chall. Bot. T.-2, G4 et I.-3, 191 ; Hemsl. Biol. Cent.
America, III. 131 ; Hillebrand, Fl. Hawai. 399 ; Hook. f. Fl.
Brit. Ind. V. 457 ; Forbes et Hemsl. Ind. Fl. Sin. II. 443 ;
Henry, List PI. Formos. 84.

Nom. Jap. Tögoma, Sömokuznsetsu, vol. 20, fol. 33.
ffima, Honzôzufu, vol. 21, fol. 23.

Folia lo-2öcm. longa, ultra medium palmato-lobata, lobis
ovato-lanceolatis acuminatis intequaliter serrato-dentatis. Flores t :
sepala triangulari-ovata. Alabastra subglobosa leviter acuta.
Pedicelli inferne articulati. Flores $: ovarium ecbinatum, stylis
basi connatis. Pedicelli superne articulati. CapsuUe longitrorsum
trisulcatie. Semina kevia, eleganter maculata, valide carunculata.
DiSTEiB.: Ubique loci cult.
Hab. Formosa: Taipea, leg. T. Marino, anao 1896 ; ïakar) leg. C. Owatai::,

auijo 1890; Heug-chlmu., Tümoii, leg. K. Miyake, anno 1898.
Liu-kiu ; Xuha, auiio 1882.
Hondo : Prope Tokyo.

19. Homonoya Lo

UR.

LouR. Fl. Cochinch. 782; Ende. Gen. PI. 1125; Muell.

Arg. in HC. Prodr. XV.-2, 1022 ; Baill. Hist. d.

PI. V. 179 ; Benth. et Hook. Gen. PI. III. 322 ;

Pax, in Nat. Pfl. Farn. III.-5, 71.

Flores dioici, apetali. Discus 0. Fl. J : calyx in alabastro

globusus, clausus, per antliesin valvatim 3-partitus. Stamina

centralia numerosa ; antheric in tilamentis repetiter ramosis dense

54 ART. 2. — B. HAYATA : REVISIO

coiiferta", loculi« subglobosis divaricatis in filaniento sessilibus,
unilocularibus. Ovarii rudimentum 0. Fl. ^: sepala 5, aiigusta,
imbricata, inoequalia. Ovarium 3-loculare ; styli iiidivisi, patentes,
papilloso-plumosi, vix basi connati ; ovarii loculi 1-ovulati, duo
anteriores, bracteam spectantes, un us posterior. Frutices puberuli.
Folia alterna, bistipulata, petiolata, penniuervia, venosa, angusta,
dentata, subcoriacea, subtus gland uloso-lepidota. Stipulte cadueis-
sim*. Spicœ axillares, utriusque sexus densiflone. Bracteœ l-florie.

H. riparia Lour. Fl. Cochinch. ed. -2, 637; Muell. Arg. in
Linnœa, XXXIV. 200 et in DC. Prodr. XV.-2, 1023 ; Beddome,
Fl. Sylv. t. 212; Hook. f. Fl. Brit. Ind. V. 455; Forbes et
Hemsl. Ind. Fl. Sin. II. 443 ; Henry, List PI. Formos. 84.

Adelia nerlifolla Roth.; Roxb. Fl. Ind. III. 849 ; Wight,
le. PI. Ind. or. t. 1868.

Hemotospermum saUclnani " Baill. Etude Euphorb. 293."

Nom. Jap. Nanban-yaiiagl.

Eami teretes pubeseentes. Folia 7-18 cm. longa, 1-2 cm.
lata, angusta, lineari-lanceolata, acuminata, intégra v. denticulata,
saliciana, supra fusco-nigricantia, glabra, subtus parce pubescentia
et ferrugineo-lepidota. Petioli breves, 3-5 mm. longi. Spicœ
erectœ, graciles. BractciC 3-partitie, imbricatie. Fl. J : alabastra
2è-3 mm. diametro a?quantia, globosa, j^uberula ; sej^ala 3 mm.
longa, ovata, acuta. Flores ^ : sepala angusta, subulato-lanceolata ;
ovarium globosum leviter trisulcatum muricatum puberulum, stylis
iutus longe papillosis, ovarium 5-6-plo superantibus.

DiSTRiB.: In India, Java, Ceylonia, Formosa et Liu-kiii.

Hah. Formo.^a : Bi-long-tsng, Ang-thau-su, leg. Y. Tasiiiho, anno 1896.

liiu-kiu : Tamitsuki, leg. K, Mivake, anno 1698.

EÜPH0RBIACEARÜ3[ ET BUXACEARLIM JAPONICARUM. 55

20. AleuriteS Forst.

Endl. Gen. PI. 1114; Muell. Arg. in DC. Prodr.

XV.-2, 723 ; Batll, Hist. d. PI. V. 183 ; Bfnth.

el Hook. Gen. PI. III. 292 ; Pax, in ^^at. Pfl.

Farn. III.-5, 72.
Flores monoici, petaliferi. Fl. J : calyx junior globosus,
clausus, per anthesin in lobos 2-valvatim ruptus. Petala 5 ;
œstivatione contorta, calyce longiora. Disci glandulse petalis altern?e.
Stamina 10 biseriatini disposita, exteriora petalis opposita, recep-
taculo conico aflixa, glandulis parvis alterna, filamentis elongatis ;
antlier<T erectic, adnatœ, birimos?e, localis parallelis longitudinaliter
dehiscentibiis. Ovarii rudimentum 0. Fl. 2 : calyx et discus
maris. Ovarium 4-loculare, styli bidivisi ; ovula in loculis solitaria.
Fructus magnus indeliiscens, crassus. Seminum testa lignosa ;
allmmen crassum ; embryo rectus, cotyledonibus latis planis. —
Arbores indumento stellari. Folia alterna, longe petiolata, ampla,
basi 7-nervia, petiolo apice 2-glanduloso. Flores laxe cymosi,
cymis ad apices ramorum paniculatis.

1. A. COrdata Steud.; ]\[uell. Arc. in DC. Prodr. XV.-2,
724 ; Forbes et Hemsl. Ind. Fl. Sin. IL 433 ; Henry, List PI.
Formos. 83 ; L. Diels, in Engl. Jalirl). XXIX. 430 ; Spiirasawa,
Nippon-Shinrinjumokudufu, I. t. 57.

Dryahdra cor data Thunb. Fl. Jap. 267, t. 27.

Vernicia montana LouR. Fl. Cocliinch. ed. -2, 587.

Elœococca cordata Blume ; Franch. et Savat. Enum. PI.
Jap. I. 425.

5ß ART. 2. — B. HAYATA : REVISIO

Nom. Jap. Abura-giri , Honzo-zufu, vol. 83, foL 7.

Kami ramosi glabri, densiuscule verrucoso-lenticellati. Folia
9-18 cm. longa, 7-10 cm. lata, late ovata, acuta vel acuminata,
basi truncata v. levius v. profunde cordata, ambitu intégra v.
superne tricuspidata, firme membranacea, petiolum sequantia v.
superantia v. eo breviora, basi biglandulosa, glandulis gyalectifor-
mibus, concavis longe stipitatis, utrinque glaberrima. Pedicelli
calycem œquantes, inferne sericeo-pubescentes, superne glabri.
Fl. J : alaliastrum oblongo-ovoideum, apice obtuse cuspidatum.
Calyx 2-partitus, segmentis ol)longo-ovatis, acutis, apice cras-
siusculis, 10-12 mm. longis. Petala spathulato-lanccolata, striata,
1 0-17 mm. longa, 4 mm. lata; stamina filamentis subconnatis,
intus pilosis. Antherae introrsum birimospe, in alabastro erectœ,
dein post emisionem pollinis oscillando extrorsum reflex», tum
subpendulcT, extus l)irimostTe, versatiles, connectivis subdilata tis.
Glandula? conico-ovoidete acut«'. Fl. 2 : ovarium oblonsfo-
ovoideura -5 mm. longum, 3-32 mm. latum, subsericeum, stylis
liberis. Fructus magni, verrucosa Semina compresso-globosa,
12mm. longa, 13mm. lata, oleosa

DiSTRiB.: In Japonia, China.
Hab. Formosa : Loco non indicate.

Hondo: Insnla Tsu>himn, le^. K. Hirata, anno 1902. Prov. Sni'uga,
prope Oiniyn, anno 1881 ; ])rope Tokyo, Bot. HdiI.

21. Gel on Î urn Roxe.

Eoxjj. in WiLLD. 8p. PI. IV. 831 ; Endl. Gen.
PI. 1116; MuEi^L. Arg. in DC Prodr. XV.-2,

EÜPHORBIACEARUM ET BüXACEARUM JAPON ICARUM. 57

1126; Batll. Hist. d. PL V. 200; Benih. et

Hook. f. Gen. PI. III. 324 ; Pax, in Nat. Pfl.

Farn. III.-5, p. 88.

Flores dioici, apetali. Fl. J : sepala 5, lata, valde imbricata.

Stamina oo, libera, supra discum carnosum dense inserta, filamentis,

linearibus, antheris oblongis, basifixis, 2-locularibus, longitudinaliter

dehiscentibus. Ovarii rudimentum nullum. Fl. ^ : sepala maris.

Discus membranaceus margine crenato-dentatus. Ovarium 3 locu-

lare ; styli breves, patentes, 2-fidi, ramis recurvis ; ovula in

loculis solitaria. Fructus globosus, endocarpio duro 3 loculari,

indehiscens. — Arbores glabrœ. Folia alterna brevissime petiolata,

coriacea, integerriraa, bistipulata, stipulis connatis, caducissimis,

annulum prominulum ad nodum relinquentibus. Flores parvi.

Pedicelli fasciculati, brevissimi, oppositifolia.

1. G. aequoreum Hange, in Journ. Bot. (1866), 173 ;
FoRB. et Hemsl. Ind. Fl. Sin. II. 444 ; Henry, List PI.
Formos. 84.

Owalaria formosana Matsüm. in Tokyo Bot. Mag. XIV. 1.

Nom. Jap. Oba-tsuge.

Arbor ultra 10 ped. alta ; truncus 1 poll, in diametro sequans,
ramis griseo-suberosis, ramulis viridibus subangulatis, nodis suberosis
annulatis. Folia brevissime petiolata, elliptica vel obovato-
oblonga, apice rotundata, basi attenuata, 3è-9 cm. longa, 2-33 cm.
lata, integerrima, margine subrevoluta, penninervia, nervis utrinque
prominulis, nervulis reticulatis, supra elevato-granulata, glaberrima.
Flores 6 mm. in diametro œquantes. Sepala carneo-alba, chartacea,
late orbicularia, 3 mm. in diametro œquantia, concava, margine
ciliolata, glabra. Stamina oo, filamentis 3 mm. longis calycem
excellentibus. Ovarium ovoideo-globosum, 4èmm. longum, glabrum.

58 ART. 3. B. HAYATA : REVISIO

Pedicelli subcrassi, 2 mm. loiigi, basi resinam exsudantes. Fructiis

globosus, glaber, piso major. Semina

Hab.: Kosliün, Fukö, leg. Y. Tashieo, anno 189G ; Kaisa, Shajö, Slio-
liukiütö, leg. C. Om^atari, anno 1898.

22. Excoecaria Linn.

LiNN. Sp. PI. ed.-2, 1451 ; Endl. Gen. PI. 1109; Muell.

Arg. in DG. Prodr. XV.-2, 1213 ; Baill. Hist.

d. PI. V. 227 ; Benth. et Hook. Gen. PI. III.

337 ; Pax, in Nat. Pfl. Fam. III.-5, 95.
Flores dioici v. monoici, apetali. Discus 0. Fl. J: sepala 3, parva
distincta, imbricata ; stamina 2-3, filamentis liberis ; antherarum
loculi distincti, ovoidei, longitudinaliter déhiscentes. Pudimentum
ovarii 0. Fl. Ç : calyx 3-ßdus, v. o-partitus. Ovarium 3-loculare;
styli basi breviter connati, recurvo-patentes, indivisi ; ovula in
localis solitaria. Capsula tridyma, in coccos 2-valves a columella
minuta dissiliens, endocarp-o crustaceo. Semina globosa, testa
Crustacea, albumen carnosum ; cotyledones planse, latœ. Frutices.
Folia alterna longiuscule petiolata, intégra, coriacea, penninervia,
stipulata. Pacemi terminales. Flores J sub bractea 2-3 glomerati
sœpius breviter pedicellati, pedicello basi 2-bracteolato.

Conspectus Excoecariarum Specierum

Folia coriacea 1. E. Agalloclia.

Fol ia niembran acea

Folia niargine intégra 2. F. jupontea.

Folia margine crenato-serulata 3. E. crenulata.

1. Ex. Agallocha Linn. Sp. PI. ed. -2, 1451 ; Wight,

EüPHORBIACEAEUM ET BUXACEARUM JAPONICARUM. 59

Ic. PI. lud. or. t. 1865 ; Muell. Arg. in Liiiuœa, XXXII.
124 et in DC. Prodr. XV.-2, 1220; Seemann, Fl. Vitiensis,
232 ; Benth, Fl. Austral. 152 ; Hange, in Jour. Linn. Soc.
XIII. 123; Hemsl. in Voy. Chall. Bot. I. 192; Hook. f.
Fl. Brit. lud. V. 472 ; Forbes et Hemsl. Ind. Fl. Sin. II.
44G ; Henry, List PI. Formos. 85 ; L. Diels, in Engl. Jahrb.
XXIX. 430 ; Matsumura, in Tokyo Bot. Mag. XII. 62 et 61.

Ex. Camettia Willd. Sp. PL IV. 864.

Stillingia Agallocha " Baill. Etude Gen. Eupliorb. 518."

Nom. Jap. Shimashiraki.

Rami firmi, teretes, deusiuscule et patenter foliosi. Folia ovata
V. elliptica 3-9 cm. longa, 2-5 cm. lata, basi obtusa, apice ssepius bre-
viter acuminata, margiue intégra v. crenata, petiolis 1-2 cm. longis.
Eacemi axillares, solitarii. Bractese l-florse, truncatse, dentatie.
Fl. $ : sepala ovata, acuta ; ovarium oblongo-globosum, trisulcatum,
glabrum, nigricans ; styli ovario breviores, basi breviter connati.
Capsulse circ. 6 mm. latœ, 4 mm. longse profunde trisulcatœ.

DiSTRiB.: In Asia meridionali late difilisa ; Australia, Formosa et Liu-kiu.
Hab. Formosa : Takno.

Liu-kiu: Insula Okinawa, leg. J. Matsumura, anno 1897; Yaeyama, leg.
S. Tanaka, anno 1891.

2. Ex. japonica Muell. Arg. (Tab. IV. E.) Muell.

Arg. in Linnœa, XXXII. 123 et in DC. Prodr. XV.-2, 1217 ;
MiQ. in Ann. Mus. Bot. Lugd.-Bat. III. 126 et Prol. 290;
Hange, in Jour. Linn. Soc. XIII. 123 ; Forbes et Hemsl. Ind.
Fl. Sin. IL 446; L. Diels, in Engl. Jahrb. XXIX. 430;
Matsumura, in Tokyo Bot. Mag. XII. 61 et 62.

Stillingia japonica Sieb, et Zucc. Fl. Jap. Farn. Xat. Sect.
Prim. 145; Benth. Fl. Hongk. 303.

Nom. Jap. Shiraki, — Kwawi, arb., vol. 3, fol. 15, sub.

60 AET. 3. — B. HAYATA : EEVISIO

Rami graciles. Folia firma, membranacea, 7-17 cm. longa,

4-10 lata, ovata, basi subobtusa v. cordata, apice acuminata," costis

secundariis distantibus, subtus pallidiora, stipulis longis lineari-

lanceolatis, petiolos semisequantibus, basi 22 mm. latis, fere scariosis,

petiolis lè-3cm. longis. Folia subtus a margine 10-12 glandulosa,

glandulis minimis, basi bi-glandulosa, glandulis majoribus. Racemi

terminales. Bractese oblongo-ovatse dentatœ, glandulis crassis

disciformibus subobliquis semilunaribus, fl. J 3-4-florœ, fl. Ç 1-

florse. Fl. J: calyx cupuliformis demum sub-deplanatus, 1mm. longus.

Stamina 2-3, exserta. Pedicelli 2-21 mm. longi. Fl. ^ : sepala

3 parva, triangularia, acuta ; ovarium ovatum, trisulcalum ; styli

basi connati ovarium œquantes. Pedicelli crassiusculi 10 mm. longi,

ovarium 3-plo superantes. Capsulae magnse, profonde trisulcatse,

10 mm. long*, 7 mm. diametro sequantes, stylis persistentibus.

Pedicelli fructigeri 20 mm. longi. Semina globosa, 9 mm. diametro

sequantia, nigro-maculata.

DiSTRiB.: In China, Jaj^ouia.

Hab. Li u- kill : Insula Okinawa, J. Matsumura, anno 1897.

Kiiisiu ; Prob. Buzen, in monte Iwatake, anno 1882.

Siiikoku : Prov. Tosa, in monte Tsuitate, anno 1888. Piov. Tosa, in tractu

Wagawa, leg. T. Making, anno 1889. Prov. lyo, in monte Ishizuchi.

Prov. Tosa, Yahazu, anno 1888.
Hondo : Prov. Kii, in monte Gujotoge, anno 1883. Insula Tsusima, in tractu

Shishimi, leg. Y. Yabe, anno 1901. Prov. Nagato, in tractu Abu, leg.

G. NiKAi, anno 1893. Tokyo Bot. Hort.

3. Ex. crenulata Wight, Ic. PL Ind. or. t. 1865.

Ex. cochinchinensis Lour. Fl. Cochinch. 750 ; Muell. Arg.
in DC. Prodr. XV.-2, 1215.

Nom. Jap. Shima-seishiboku.

Kami validiusculi. Folia opposita, oblongo-lanceolata,
lanceolato-acuminata, basi acuta, margine crenato-serrulata, 8-12 cm.

EüPHORBIACEARUM ET BÜXACEARUM JAPONICARUM. 61

longa, 2.5-4 cm. lata, coriacea, petiolis brevibus, 5-6 mm. longis,
stipulis subulatis, parvis. Bractese omnes l-florse, breves. Fl. J :
calyx 3-partitus, sepalis lanceolato-ovatis ; stamina 3, filamentis

parum exsertis. Fl. $ Capsulée depressse, 5 mm. longse,

8 mm. latœ. Semina subglobosa, 4 mm. lata, 5 mm. longa, levia,

elegante reticulata.

DiSTRiB.: In Cochiuchina, India orientali, Formosa.
Hab. Formosa: Kachinrö, leg. C. Owatari, anno 1897.

23. Sapium p. Br.

MuELL. Arg. in DC. Prodr. XV.-2, 1202 ; Benth. et
Hook. Gen. PI. III. 334 ; Pax, in Nat. Pfl. Farn.
III.-5, 97.
Flores monoici, apetali. Discus 0. Fl. J : calyx parvus,
membranaceus, dentatus. Stamina 2, filamentis liberis ; antherarum
loculi ovoidei, distincti, contigui, paralleli, longitudinaliter déhis-
centes. Ovarii rudimentum 0. Fl. ? : calyx 3-fidus, ovarium
3-loculare ; styli mediocaliter connati, recurvo-patentes, indivisi ;
ovala in loculis solitaria. Capsula columella centrali bene evoluta
prsedita, in coccos 2-valves dissiliens. Semina globosa, testa
Crustacea ; embryo rectus ; cotyledones planœ latœ. — Arbores glabrae.
Folia alterna, petiolata, intégra, pennivenia, petiolo apice 2-glandu-
loso. Racemi terminales, bisexuales. Flores J sub bractea 3 ;
flores ? ad basin spicse plures sub bractea solitarii. Bractese
parvse, ssepe basi glanduliferse.

1. Sapium sebiferum Boxb. Fl. Ind. III. p. 693 ;
Hook, et Arn. Bot. Beech. Voy. 213 ; Muell. Arg. in Linnsea,

62 ART. o. B. HAYATA : REVISIO

XXXIL 121 ; MiQ. in Ann. Mus. Bot. Lugd.-Bat. III. 126
et in Prol. 290 ; Franch. PL David. 266 ; Hook. f. Fl. Brit.
Ind. III. 470; Forbes et Hebisl. Ind. Fl. Sin. IL 445; Henry,
List PI. Formos. 85; L. Djels, in Engl. Jalirb. XXXIX. 430;
Matsumura, in Bot. Mag. XVI. 163.

Croion sebifcrum Linn. Sp. PI. ed.-2, 1425.

Triadeca sinensis, Lour. Fl. Cochincli. ed.-2, 748.

Still'ingia sehifera Michx.; Hook, et Arn. Bot. Beech. Voy.
213 ; Benth. Fl. Hongk. 302 ; Sieb, et Zucc. Fl. Jap. Fam.
nat. Sect. Prim. 145!

Excoecaria sehifera Muell. Arg. in DC. Prodr. XV.-2, 1210.

Siillingia sinensis, " Baill. Etude Gen. Euphorb. 512, t.
7, f. 26-30."

Xom. Jap. Nanhinhaze.

Ramuli graciles, teretes, glabri. Folia louge petiolata dilatato-
rhombeo-ovata, intégra, ssepissime paulo latiora quam longa, basi
acuta, apice acute acuminata, submembranacea, olivaceo-fuscescentia,
subtus pallidiora. Petioli bene evoluti, 2-5 cm. longi. Pacemi
spiciformes, terminales, folium demum longe superantes, basi
floribus Ç vulgo 3-4, cœterum floribus J dispositi. Bractese
ovatœ, acuminatœ, glandulis crassis globoso-subreniformibus.
Flores ^ : pedicellis medio articulatis, circ. 5 mm. longis.
Calyx 1 mm. longus, cupulitbrmis, dentatus. Stamina exserta.
Flores ^ : pedicellis 3 mm. longis. Calycis lobi triangulares
acuminati, membranacei ; ovarium 3 mm. longum ovoideum, stylis
longis ovarium 4-23lo superantibus. Capsuke 10 mm. longte, 13 mm.
latœ, nigrae, globoso-ellipsoideie, subacutœ. Semina incrasso-
subarillata, albo sebo prœdita, testis lœvibus.

DiSTRiB.: lu Japonia, China spontanea ; in In(li:im oiientalem et partes
culidas utriusque hemisphserii iutrodueta.

EUPHOKBIACEARUM ET BUXACEARUM JAPONICARUM. 63

IIab. Formosa: Taipea, Yensan, leg. K. INIiyake, anno 1898; Hong-soang,
leg. C. OwATARi, anno 1896 ; kcliuig, Tamsui, leg. T. Making, aniv)
1806 ; inter Kusshnku Sliinshogai, leg. K. Miyake, anno 1899,

Hondo : Prope Tukyo, Bot. Hort. (Cult).

24. Euphorbia Linn.

Linn. Sp. PI. ed.-2, 246 ; Endl. Gen. PI. 1108 ;

BoissiEE, in DC. Prodr. XV.-2, 8 ; Baill. Hist.

d. PI. V. 177 ; Benth. et Hook. Gen. PI. HI.

258; Pax, in Nat. Pfl. Fam. 111.-5, 103.
Involucrum subregulare, campanulatum, turbinatum vel liemi-
spliericum, lobis 5 v. 4, membranaceis, glandulis alternantibus,
glandiilis semper 4 rarius 5, substipitatis, crassiusculis, carnosis,
sœpius horizontalibus v. extrorsum inclinatis, vulgo latioribus quam
longis, planis v. concavis, margine exteriori rotimdatis v. truncatis,
semilunatis, v. bieornutis. Flores J : in involucro numerosi,
pedicellati. Perianthium 0. Stamen 1, filamento brevi cum
pedicello articulate ; anthera erecta, loculis distinctis. Flores $ :
in centro involucri 1, pedicello demum elongate ex involucro
exserto. Perianthium 0. Ovarium 3-loculare, styli 3, distincti
V. plus minus coaliti, apice patentes, indivisi v. 2-fidi, ramis apice
V. introrsum stigmatosis ; ovula in loculis solitaria. Capsula in
coccos 2 valves a columella persistente dissiliens, endocarpio duro
V. cartilagineo. Semina albuminosa, cotyledonibus latis planis, v.
linearibus.

Conspectus Euphorhiarwn Specierum.

Folia estipuUita, alterna ; sernina carunculata.

Herbse; folia caulina decussata; involucri glandulœ bieornutœ...l. E. Lathyris.

64 AET. 3. — B. HAYATA : EEVISIO

Frutices, ramis crassiusculis iiiferne denudatis siiperne foliosis ; involucri glandulj«

rotundatse 2. E. dendroides.

Herbse ; folia sparsa, involucri glandulse margine rotundatœ.
Bracteolœ inter flores raasculos prœditse.
Folia umbellaria 5-10.

Folia serrulata; ovarium levé 3. E. Helioscopia.

Folia Integra ; ovarium verrucosum.

Caules ramosi 4. E. adenochlora.

Caules simplices 5. E. Jolkini.

Folia serrulata, ovarium verrucosum 6. E. pelcinensis.

Folia intégra, ovarium sub-lseve, hirsutum. ...7. E. togakusMensis.

Folia umbellaria 4 8. E. orientalis.

Bracteolse inter flores masculos non prseditse 9. E. ehradeolata.

Herbœ ; folia caulina sparsa, involucri glandulse margine relusse v. cornutae.

Glandulse semilunatse longe bicornutse 10. E. Sieboldiana.

Glandulae retusse 11. E. Esula.

Folia stipulata, opposita, disticha, semina ecarunculata.

Suffrutices 12. E. Sparmanni.

Herbse.

Herbse elatiores erectse.
Caules pubescentes.

Semina transverse rugosa 13. E. piluUfera.

Semina Isevia 14. E. hypericifolia.

Caules glabri 15. E. serrulata.

Herbse humiliores, prostratse.

Glandula appendice angustiori ; capsula hirsuta.

Folia superne facie non mnculata 16. E. thymifvlia.

Folia superne facie maculata 17. E. maculata.

Glandula appendice latiori ; capsula glabra.

Glandula appendice elobata ; folia magna; caules hirsuti

18. E. humifusa.

Glandula appendice trilobata ; folia minora; caules glabri

19. E. microphylla.

1. E. Lathyris Linn. (Tab. IV. F.) Linn. Sp. PI.

ed.-2, Q65', Thunb. Fl. Jap. 196; Willd. Sp. PI. IL 906;
AiTON, Hort. Kewinsis, ed.-2, IIL 164 ; Hook, et Akn. Bot. Beech.
Voy. 44 ; Ledeb. Fl. Ross. III. 572 ; A. Kanitz, in Linnsea,
XXXII. 555 ; Sowerby, English. Bot. VIII. 113 ; Boissiee, in

EUPHORBIACEARÜM ET BÜXACEARUM JAPONTCAPvUM". 65

DC. Piodr. XV.-2, 99; ]\[iq. in Ann. Mus. Bot. Lngd.-Bat.
III. 125 et Prol. 289; Feanch. et Savat. Enuni. PL Jap. I.
420; Maxim, in Mel. Biol. XI. 833; Hemsl. Biol. cent, Americ.
III. m.

Nom. Jap. Hanshiren, Kwawi, lierl)., vol. 3, fol. 25.
,, Zokuzuishi, Honzözufu, vol. 21, fol. 20.

,, Hovutosö, Sömokuzusetsu, vol. 9, fol. 23.

Gaules glnbri erecti crassiusculi stricti, umbellöB 4-fidiç, radiis
semel bisve bifidis. Folia patala sessilia oblongo-lanceolata, intégra,
obtusa vel acuta, inferiora linearia, conferta, superiora remotiuscula,
umbellaria conforniia majora, floralia subcordata ovato-oblonga
acuminata, caulina deccusata. Involucra 2h mm. diametro sequantia,
campanulata, intus et extus glabra, lobis ovatis truncatis fimbriatis,
glandulis semilunatis, bicornubus subspathulatis apice rotundatis
subcrassatis maculatis, bracteis inter flores masculos obsoletis.
Flores J : pedicel lis hirtillis. Antherse ssepe coufluentes. Flores
^ : ovarium globosum, stylis longis basi coalitis bifidis apice
subdilatatis planis recurvis. Capsulse maximae ovato-depressœ
trisulcat?e, 13 mm. latse. 10 mm. longse, coccis rotundatis. Semina
ovoidea minute reticulato-rugosa, 6mm. longa, 4mra. lata; carnncul^e
orbiculares conic?e, margine lobulatie, stipitatœ ; albumen copiosum
oleaginum. Embryo centralis ; cotyledones crassatfe lineari-
lanceolatie radicula non latiores.

DiSTRiB. In Europa aiistrnli S[)outanea.
Hab. Tokyo (culta).

2. E. dendroides Linn. 8p. PI. ed.-2, 662 ; WiLLD. Sp.
PI. IT. 924 ; BoissiER, in DC. Prodr. XV.-2, 109.
K lœta Ait. Hort. Kew. ed.-2, III. 164.
Nom. Jap. Kidachi-taigeki.

6(3 ART. ?.. B. HAYATA : REVISlO

Suffrutices glabri, ramis rubellis nitidis superne dense foliosis,
umbellie 4-7-fidîe, radiis brevibus, 1 cm. longis, bifidis. Folia
dense congesta, sessilia, attenuata, lineari-lanceolata, 22-5 cm. longa,
5-ß mm. lata, subtus pallidiora, umbellaria breviora, 5-6-
verticillata, floralia 2-4-verticillata, rhombeo-semiorbicalata mucro-
nata lutescentia involucrnm o-4-plo superantia, 8 mm. longa, 6
mm. lata. Involucra campanulata, 2 mm. diametro œqiiantia,
lobis oblongis, fractis, glandulis subpeltatis, transverse ovatis
truncatis, concavis. Styli elongati, tenues, connati, bifidi, apice
leviter incrassati. Ovarium verrncosum.

DiSTKiB. Rupestris regionis mediterraneae calidioris.

Hab. Formosa : Tai-toug-thian, Taug-si-kak. Suiteiriyo, leg. K. jMiyake,
anno 1898.

3. E. Helioscopia Lixn. (Tab. IV. G.) LiNN. Sp.

PI. ed.-2, 658 ; Thünb. Fl. Jap. 197 ; Willd. Sp. PI. IL 914 ;
Atton, Hort, kewensis, ed.-2, III. 167; Ledeb. Fl. Ross. III.
562; Boiss. in Linnä^a, XXXII. 559; Benth. Fl. Hongk.
361 ; Boiss. in DC. Prodr. XV.-2, 136 ; Sowerby, English Bot.
VIII. 99; MiQ. in Ann. Mus. Bot. Lugd.-Bat. III. 126,
et in Prol. 290 ; Franch. et Savat. Enum. PL Jap. I. 422 ;
Franch. PI. David. 263 ; Maxim, in Mel. Biol. XL 338 ; Hook.
f. Fl. Brit. Ind. V. 262 ; Forbes et Hemsl. Ind. Fl. Sin. IL
413 ; L. DiELS, in Engl. Jahrb. XXIX. 430.

Nom. Jap. Tüdaigusa, Sömokuzusetsu, vol. 9, fol. 16.

Tahusitsu, Honzözufu, vol. 21, fol. 17.

Annuales, erectte, superne dichotome ramosœ. Folia glabri-
uscula, obovato-cuneata vel spathulata, obtusi vel mucronata,
sessilia, serrulata, 1 -i cm. longa, 1 cm. lata, umbellaria conformia
V. obovata, majora, 2cm. longa, la cm. lata, floralia elliptica v.

EUPHORBIACEARÜM ET BUXACEARUM JAPONICARUM. 67

orbiculata. liivolucra stipitata, turbinata, extus parce liirsuta,
intus sub glandulas dense barbata, 2 mm. diametro {equantin,
bracteis interfloribus masculinis gracilibus barbatis. Styli basi
coiniati divergentes apice breviter incrassati. Capsulée globoste,
trisulcatie, heves, o.ö mm. in diametro ?equantes. Semina ovoidea
2 mm. longa apice compressa, favoso-reticulata ; carunculne oblique
obsittie, concavie, ovatie, cordatœ, orbiculares.

DisTRiB. Tola Eiii-opa et Asia.

Hab. Liukiii : Insula Okinawa, leg. Y. Tashiro, anno 1887.

Hondo : Insula T.'^usliinin, leg. K. Hirata, anno 1902. Pj-ov. Nagato, in

tractu Tüvoura, leg. G. Nikai, anno 1898. Prov. Kü. Prov. Musashi,

in tractu Tichibu, anno 1888, Tokyo.

4. E. adenochlora Morr. et Decne. (Tab. IV. H.)

MiQ. in Ann. Mus. Bot. Lugd.-Bat. III. 26; et Prol.

290 ; Boiss. in DC. Prodr. XV.-2, 176 ; Franch. et Savat.
Euum. PL Jap. I. 422.

Euphorbia japonica Boiss. in DC. Prodr. XV.-2, 1266.

Nom. Jap. JVourushi, Sömokuzusetsu, vol. 9, fol. 10 et 11.
,, Riyo-giyo, Sömokuzusetsu, vol. 9, fol. 12.

Caules elati, erecti, crassi, foliati, glabri, apice ramosi. Folia
subtus parcissime longipilosa, inferiora 3-7 cm. longa, 8-12 mm.
lata, squamiformia, reliqua 7-12 cm. longa, 2-3-2 cm. lata,
oblongo-elliptica, obtusa, intégra, umbellaria ovato-oblonga sub
anthesin radios superantia, floralia rotundato-ovata, acutiuscula ;
umbelliß radii 6-Q, trifidi, glabri. Involucra 2i mm. diametro
œquantia, turbinata, glabra, intus barbata, lobis ovatis, undulato-
repandis glabris, glandulis transverse ovatis, subreniformibus,
rotundatis, substipitatis, bracteis interfloribus masculinis lanceolatis
biüdis hirsutis. Styli erecti crassi, apice bifidi. Capsuhc 5 nnn.
diametro sequantes, depresso-globosie, profunde trisulcatie dorso parce

68 ART. 3. B. HAYATA : REVIÖIO

acuteque muricatie. Semina subgiobosa apice subcompressa ; Irevia,
3 mm. cliametro gequautia ; carunculne miuntœ fungiformes, caducée.
Hab. Hondo : Prov. Musn.slii, Wada, prope Tokyo.

5. E. Jolkini Bois«. in DC. Prodi-. XV.-2, 121 ; Miq.
in Ann. Mus. Bot. Lugd.-Bat. III. 126 et Prol. 290; Fkanch.
et Savat. Enum. PI. Jap. I. 421 ; Maxim, in Mel. Biol. XL
836 ; Boiss. cent. Euphorb. 32 et Ic. Euphorb, t. 71 ; Kuez,
in Journ. Bot. (1873), 193.

Nom. Jap. Iwa-taigeki, Sömokuzusetsu, vol. 9, fol. 22.

Gaules glabri, crassiusculi, erecti, inferne denudati cicatoricosi,

superne dense foliosi, apice ramulosi, umbellse radiis 5-7 breviter

bis biüdis dense corymbosis. Folia intégra e basi sessilia attenuata

lineari-lanceolata, obtusissima, umbellaria oblongo-elliptica radios

subœquantia ; floralia e basi rotundata ovato acutiuscula. Involucra

2 mm. diametro ^equantia, campanulata, stipulata, extus glabra,

intus sub glandulas hirta, lobis ovatis répandis integris, glandulis

transversis orbicularibus cyathiformibus, substipitatis. St3di

CapsuUe ö mm. diametro œquantes, depresso-globosœ, profunde

trisulcatœ, verrucis obtusis, dense obsitse, secus coccorum dorsum

Iseviter sulculosae. Semina subgiobosa kevia 3 mm. in diametro

lequantia ; caruncula? orbiculares, conico-depressœ, fungiformes.
DisïRiB, In Japonia.
Hab. Liu-kiu : leg. J. Matsu:mura, anno 1897; Kunigiuni, leg. IS. Tanaka,

anno 1889.
Kiushiu : prope Naga«iki.
Hondo : Insula Hachijö.

6. E. pekinensis Kupr. (Tab. IV. J.) Rupr. in

Maxim. Prim. Fl. Amur. 239; Boiss. in DC. Prodr. XY.-2,
122; Maxim, in Mel. Biol. XI. 834; Forbes et Hemsl. Ind.
Fl. Sin. IL 416 ; L. Dielö. in Engl. Jahrb. XXIX. 430.

EUPHORBIACEAKTM ET BUXACEARUM JAPONICARUM. 69

E. lasiocaula Boiss. in DC. Prodr. XV.-2, 1266; Franch.
et Sayat. Eiium. PI. Jap. I. 421 et II. 485 ; Franch. PI.
David. 263.

E. Sampsini Hance ; Maxim. Mel. Biol. XI. 835.

E. corcdlokles Thuxb, Fl. Jap. 197, ex jNIaxim.

Nom. Jap. Takatödai, Sömokuzusetsu, vol. 9, fol. 17.

Gaules herbacei nibelli molliter hispiduli ramosi, umbellse
radiis quinis 2-3-fidis dein bifidis. Folia firmula saliciformia
sessilia e basi valde atteiiuata lanceolata acutiuscula v. acuta
subintegra v. seirulata, 3-6 cm. longa, 6-16 mm. lata, superne
glabra vel sparce hirsuta, subtus dense liirsuta, glaucesceutia,
umbellaria breviora oblongo-lanceolata, floralia ovato-rhomboidea.
Involucra II mm. diametro lequantia, campanulata extus glabra
intus liirta, lobis ovatis retusis, glandulis transverse subreniformibus,
bracteolis interfloribus masculinis barbatis. Ovarium depresso-
globosum. Styli basi connati, apice bi-fidi dilatati. Capsulœ 3 mm.
in diametro sequantes, globoso-depressse, verrucis conicis obtusis
sparce obsit^e. Semina 1 s mm. longa, globoso-ellipsoidea ; carunculte
oblique conicae subsessiles.

DiSTRiB. In China, Japonia.

Hab. Kiusliiu : Pro v. Higo, in monte Aso, leg. Y. Yabe, anno 19U1.

Shikoku : Tosn, leg. T. Makin(X

Hondo : Pi-ov. Öiui, in monte Ihuki, anno 1881. Prov. Musashi, piope Tokyo,

Prov. Hitachi. Prov. Shinano, in montibiis Omine, ïogakushi, Ai-anui.

Prov. Sagami, Hakone,

7. E. togakusensis sp. nov. (Tab. IV. A.)

Nom. Jap. Jli i/ama-nournshi.

Caules glabii crassiusculi erecti herbacei rubelli apice ramosi,
umbell« itdiis quinis bifidis, glabris. Folia tenuia sessilia basi
rotundata oblongo-elliptica apice obtusissima v. retusa 5-7 cm.

70 ART. 3. B. HAYATA I REVISIO

longa, 2 cm. lata glabra, iimbellaria oblongo-ovata, floralia ovata v.

rhombeo-ovata. Involucra 22 mm. diametro seqiiantia campanulata

extiis glabra intus sub glandulas hirta, lobis ovatis retusis,

gland ulis transverse orbicularibus reniformibus substipitatis. Styli

basi coaliti apice bifidi subincrassati. Ovarium depresso-globosum

hirsutum, verrucis conicis obtusis sparce obsitum, trisulcatum,

secus sulcum costata. Capsulye 5 mm. in diametro arquantes

globoso-depressae verrucis conicis obtusissimis obsitpe v. sublseves.

Semina 2imm. longa globoso-compressa; carunculœ oblique conico-

depressse substipitatse.

Ab. JlJ. pekinensi Rupr. differt pro caule glabro, crassiusculo,

foliis oblongo-ellipticis integris glabris, ovario liirto costato, facile

distincta.

Hab. Hondo : Prov. Kaga, in monte Hakusau, anno 1881. Prov. Eccliiu
piov. Sliinano, in monte Togakushi, (ipse) anno 1900.

8. E. Orientalis Linn. Sp. PL ed.-2, 660; WiLLD. Sp.
PI. II. 917 ; AiTON, Hort, kewensis, ed.-2, III. 168 ; Ledeb. Fl.
Ross. III. 566 ; Boiss. in DC. Prodr. XV.-2, 121, et Ic. Euphorb.
19, t. 72.

Nom. Jap. Shima-taigeki.

Gaules glabri crassi erecti, umbelke radiis 7-8 simplicibus.
Folia sessilia membranacea uninervia lanceolata, acuminata, intégra,
7-8 cm. longa, 8-lOmm. lata, umbellaria breviora oblongo-lanceolata,
tioralia ovato-rhombea mucronata. Involucra 2i mm. in diametro
lequantia, turbinato-campanulata extus glabra sub glandulas liirsuta,
lobis oblongis acutis breviter ciliatis, bracteolis interfloribus
masculinis. Styli a medio connati apice bifidi. Ovarium globosum
trisulcatum, secus dorsa loculorum ad latus sulculorum verrucis
compressis brevibus conicis obtusissimis obsitum.

Hab. Formosa : Taichù-perfect., Daiboliosho, leg. K. Satake, auno 1898

EUPHORBIACEARUM ET BUXACEARUM JAPONICARUM. 71

0. E. ebracteolata sp. nov. (Tab. IV. I.)

Nom. Jap. 3farumi-7ifnimshi.

Caules erecti crassati glabri v. sparce pubescentes superne
ramosi, umbellae radiis quinis bifidis, 9-10 cm. longis. Folia magna,
7-12 cm. longa, 2-3 i cm. lata, membranacea, oblongo-elliptica,
oblongo-ovata, ovato-spatbulata, basi leviter attenuata, apice obtusa
V. retusa, glabra, subtus j)arce pubescentia, umbellaria conformia
breviora angustiora, floralia basi truncata v. subcordata triangularia,
ovato-triaiigularia, obtusa, 3-3icm. longa, 2-2icni. lata. Involucra
campanulata, majora breviora, sessilia, 4 mm. diametro œquantia,
extiis glabra, intus glabra v. faiice leviter hirtilla, lobis ovatis
fimbriatis, glandulis transversis semiorbicularibus integris, bracteolis
interfloribus masculinis obsoletis v. minutissimis. Styli e basi
connati apice breviter bifidi subincrassati. Ovarium pedicellis
longis apice dilatatis rudimentum calycis formantibus, globosum,
Ueve, non sulcatum. Capsulte leviter trisulcatœ, l?eves.

Pro bracteolis interfloribus masculinis obsoletis, ovario levi
facile distincta.

Hab. Hondo : Prov. Musashi, Tichibu, in monte Bukosan, leg. T. Making,

anno 1888,
Yezo : Sapporo, leg. K. INIiyake, anno 1891 ; Prov. Hidaka, Kirimafu,

anno 1884.

10. E. Sieboldiana Morr. et Decne. (Tab. VI. B.)

MiQ. in Ann. Mus. Bot. Lugd.-Bat. III. 126 et in Prol. 290; Boiss.
in DC. Prodr. XV.-2, 158 ; Franch. et Savat. Enum. PI. Jap. I.
422; Maxim, in Mel. Biol. XL 838; Boiss. Ic. Euphob. t. 104.

K coralloides L. Sp. PI. ed.-2, 569 ; Willd. Sp. PI. II.
916 ; Thunb. Fl. Jap. 197 ; Aiton, Hort. Kevv. ed.-2. III. 168.

E. Guilielmi A. Gray. Bot. Jap. 406.

Nom. Jap. Nalsutddai — Sömokuzusetsu, vol. 9, fol. 13.

72 ART. 3. B. HAYATA ! REVISIO

Cailles glabri tenui herbacei simplices erecti, umbellse radiis
tenuibus 5-6, divergentibus, 3-5 cm. longis, bifidis, trifidis dein
bifidis. Folia intégra membranacea superne subsessilia inferne
petiolata brevissima subtiis glaucescentia e basi attenuata oblongo-
spathulata v. oblonga obtnsa v. retiisa, 3-6 cm. longa, 1-2 cm.
lata, umbellaria conformia majora oblongo-spatbulata v. oblongo-
rhombea, 4-7 cm. longa, 2-3 cm. lata, floralia e basi trimcata v.
subcordata, triangularia v. oblongo-triangularia acuta vel acutiuscula.
Involucra 3 mm. longa, 2i mm. diametro lequantia, turbinato-
campanulata extus et intus glabra, lobis ovatis ciliatis, glandulis
transverse elongatis bicornubus subparallelibus 2 mm. longis. Styli
longi ad medium coaliti longe bifidi, Ismm. longi. CapsuUe
globosse, profunde trisulcatse, Ineves, 4 mm. in diametro requantes.
Semina ovoidea kevia, fusca, glabra, 2 mm. longa, I2 mm. diametro
a?quantia ; caruncube suborbiculares convexiuscube.

DiSTRiB.: In Japonia vulgaris.

Hab. Hondo : Prov. Musashi, Kokubungi, leg. T, Making, anno 1894.
Prope Tokyo, Kamiitabashi, anno 1878 ; Prov. Sagami, in monte Öyama,
leg. S. Matsuda, anno 1900. Prov. Shinano, in monte Togakushi, anno
1884. Prov. Uzen, in monte Yodonosan, anno 1887. Prov. Izu, Insula
Kozushima, insula Osliima.

Hokkaido: Sapporo, leg. K. Miyabe, anno 1891.

11. E. Esula LiNN. Sp. PL ed.-2, 6G0; Willd. Sp. PI.
IL 919; AiTON, Hort. Kew. ed.-2, III. 169; Ledeb. Fl. Ross.
III. 576 ; Maxim. Prim. Fl. Amur. 238 ; Baker et Moore, in
Journ. Linn. Soc. XVII. 386; Schmidt, in Memoir Acad. imp.
Seien. st.-Peter. VIL ser XII.-2. 60 ; Boiss. in DC. Prodr. XV.-
2, 160 ; Maxim. Mel. Biol. XL 839 ; Sowerby, English Bot. VIII.
105; Hange, in Journ. Bot. (1875) 184, et (1878) 14; Debeux,
Fl. Shangh, 53 ; Forbes et Hemsl. Ind. Fl. Sin. II. 412 ; L.
Diels, in Engl. Bot. Jahrb. XXIX. 430.

EUPHORBIACEARUM ET BUXACEARUM JAPONICARUM. 73

Nom. Jap. Hagikim, — Sömokuzusetsn, vol. 9, fol. 15.

Cailles erecti, glauceseentes ; umbellœ racliis 6-7, bis bificlis.
Folia oblongo-lanceolata v. lineari-lanceolata v. spatliulata
obtusissima v. nuicronata basi attenuata inferiora sœpe subpetiolata,
2 cm. longa, 3 mm. lata, iimbellaria breviter obovata vel obovato-
lanceolata, floralia rhombeo-ovata v. reniformia retiisa. Involucra
2 mm. diametro œquantia campa nulata, extus et in tus glabra,
lobis triangularibus bificlis fimbriatis, glandulis brevissime bicorniitis,
bracteolis interfloribus masculinis gracilibus, pedicellis fl. J parce
pubescentibiis. Capsula

DiSTRiB.: In Euiopre pratis et humidis.

Hap, : Ilouilo, pnA'. jNlikawa. in tiaetu Akumi, leg. Nakura, anno 1S92.

12. E. Sparmanni Boiss. (Tab. Y. A.) Boiss.

Cent. Euphorb. ô ; Bentft. Fl. Austral. VI. 46 ; Maxim. Mél.
Biol. XL 831 ; Forbes et Hemsl. Ind. Fl. Sin. IL 417.

E. ramosmima Boiss. in DC. Prodr. XV. -2, 14; Hook, et
Arn. Bot. Beecli. Voy. 69.

Nom. Jap. Oagari-nishihi so .

Caules glabri, rubelli, denudati, ad nodos incrassati, ramulis
tenuibus foliosis. Folia opposita intégra, basi cordata, disticha
ovata, mucronata subtus glauca palmatim 3-5-nervia, 1-2 cm.
longa, 6-10 mm. lata, stipulée oblongœ fimbriato-ciliat?e magn.'ç
rufescentes. Cynife terminales densse v. axillares. Involucra longe
pedicellata turbinata campanulata extus glabra intus sub glandulas
hirtilla, lobis triangularibus seepe tridentatis, glandulis substipitatis
rotundatis, appendice alba eis angustiori, bracteolis interfloribus
masculinis spatbulatis ciliatis. Styli breves bifidi. Capsube 2h
mm. diametro ?equantes, lœves globoso-depressœ, coccis subcarinatis.

74 ART. ;".. — B. H AY ATA : RE VISIO

Semina lœvia globoso-tetragona, ad medium faciei exterioris

elevatione longitudinali.

DiSTRiB. In India orientali, insulis Pitcain, Liu-kiii, Formosa.

Hab. Formosa: Ang-thau-sn, K. Miyaké, auno 1897.

Liu-kiu : leg. J. Matsumura, anno 1897, et leg. S. Tanaka, anno 1891;

insula Okinawa, leg. Y. Tashiro, et insula üagari, leg. Y. Tashiro,

anno 1887.

13. E. pilulifera Linn. (Tab. V. B.) Linn. Sp. PL

ed.-2, 651 ; Willd. Sp. PL IL 897 ; Hook, et Arn. Bot. Beech.
VoY. 213; MiQ. FL Ind. Bat. I.-2, p. 420; Benth. Fl. Hongk.
302 ; Grtsebach, FL Brit. West. Lid. 54 ; Boiss. in DC. Prodr.
XV.-2, 21 ; MiQ. in Ann. Mus. Bot. Lugd.-Bat. HL 125 ; et
ProL 289 ; Seemann, FL Vitiensis, 216 ; Benth. FL AustraL VI.
51 ; Franch. et Savat. Enum. PL Jap. I. 420 ; Maxim, in MéL
BioL XL 831 ; Hemsl. in Voy. Challeng. Bot. L-2, 22 et I.-3,
187; Hemsl. Bot. Cent. America, III. 98 et IV. 108; Hillebrand,
Fl. Hawai. 397 ; Hook. f. FL Brit. Ind. V. 250 ; Forbfs et
Hrmsl. Ind. Fl. Sin. IL 416; Henry, List PL Formos. 81.

E. hirta Thunb. Fl. Jap. 196 ; Roxb. FL Ind. IL 472.

Nom. Jap. Shiina-nishihisö.

Caules erecti v. ascendentes simplices vel parce ramosi, toti
crispulo-pubescentes, superne dense flavido-setulosi. Folia basi
breviter petiolata valde insequalia cuneata v. oblongo-lanceolata v.
ovato-rbombea acuta serrulata dentatave, 1.5-4 cm. longa, 5-15
mm. lata, stipulée minimœ lineares acuminata?. Involucra l mm.
in diametro œquantia, minima, turbinata, hirsuta, intus glabra, lobis
ovatis fimbriatis, glandulis orbicularibus^concavis, appendicibus
obsoletis v. angustissimis, bracteolis interfloribus masculinis spathu-
latis fimbriatis. C3^ma3 axillares sessiles vel breviter pedunculat?e
globoso-capituliformes, polyceplialœ. Styli breviores bilobi apice

EUPHORBIACEARUM ET BUXACEARUN JAPONICARUM. 70

incrassati. Caj'tsuLe H mm. in diametro œquantes clepresso-globosa^
profunde trisulcaUu iiavido-hirsuüp, coccis compresso-carinatis.
Semina 2 mm. longa i mm., lata, rubella, acute oblongo-tetragona,
transverse irregulariter rugulosa.

DiSTRiB. In tota America tropica a Florida et Novo Mexico ail Brasiliani
raeridionalem ; in Africa tropica occirlentali et oiientali, India orientali
omni insulis Sundaicis, Philippinis, China, Formosa et Hougk.

Hab. Formosa : Taipea. Kluug, leg. T. Making, anno 1896. Tai-tong-
thian, Tai-kah-kbœ, Gi-lan-tbian, Toa-o-tsng, leg. K. Miyake, anno
1899 ; Tsui-tng-lau, Hokukokei, leg. C. Owatari, anno 1898. Heiig-
chhun. Sei-mon, leg. Y. Tashiro, anno 1898.

Liu-kiu : insula Miynkosima, Okinawa, Yaeyama, leg. Y. Tashiro, anno
1887.

14. E. liypericifolia Linn. (Tab. V. C.) Linn. Sp.

PI. ed.-2, 650 ; Willd. Sp. PI. II. 895 ; Aiton. Hoit, kew. III.
1(31 ; Benth. Fl. Hongk. 301 ; Hook. Niger Fl. 176 ; Grisebach,
Fl. Brit. West. Ind. 54 ; Boiss. in DC. Prodr. XV.-2, 23 ; Maxim.
in Mel. Biol. XL 832; Hemsl. in Voy. Challeng. Bot. I.-l. 63,
et I.-2, 22 ; Hook. f. Fl. Brit. Ind. V. 249 ; Hemsl. Biol. Cent.
America, HI. 96 ; Forbes et Hemsl. Ind. Fl. Sin. II. 414.

E. parviflora Linn.; Boxb. Fl. Ind. IL 472 ; Boiss, in DC.
Prodr. XV.-2, 22.

Nom. Jap. Unrin-nishikiso.

Caules pubescentes erecti simplices vel parce dicbotome ramosi,
rarais tenuibus. Folia brevissime petiolata oblonga vel oblongo-
elliptica obtusa vel acutiuscula serrulata, 8-10 mm. longa, 4-5 mm.
lata, basi rotundata vel subcordata, stipulée triangulares bifida'.
Cymœ axillares pedunculatœ oligocepliabe, tenues folio breviores.
Involucra minima, è mm. in diametro œquantia, turbinata extus
glabra intus fauce liirtilla, lobis lanceolatis acuminatis integris,
glandulis orbicularibus concavis, appendice alba rotundata glandula

76 ART. 3. B. HAY ATA ! RE VISIO

niLilto latiori, sa^pe biloba, bracteolis iiiterfloribus masculinis
gracilibus trifidis. Styli breves profunde bifidi. Capsulae le mm. in
diametro nequantes liirtilbie profunde trisulcatœ, cocci s rotundatis.
Semina 1 mm. longa, rubella ovato-quadrangula lîevia v. minute
tiiberculata ad microscop)ium spectantia.

DisTRiB, In utriusque hemispherici tropicis abuude ditiusii.
Hab. Fonuüsa : Tha-li-bii-koe, leg. Y. Tashiro, anno 1895.

15. E. serrulata Reinw. (Tab. V. D.) Willd. Sp.

PL IL 892 ; MiQ. Fl. Ind. Bat. I.-2, 421 ; Eoiss. in DC. Prodr.
XV.-2, 25 ; Benth. Fl. Austral. VI. 51 ; ]\Iaxiäi. in Mel. BioL
XL 832 ; Foebes et Hemsl. Ind. Fl. Sin. II. 417 ; Henry,
List PL Formos. 81.

E. Vachellii, Hook, et Arn. Bot. Beech. Voy. 213.

Nom. Jap. Miyakojima-nishikisö.

Gaules erecti simpliciusculi gl abri. Folia breviter petiolata
basi subcordata vel rotundata lineari-lanceolata acutiuscula remote
serrulata uninervia, 2-2.5 cm. longa, 4-5 mm. lata, stipuL«
interpetioLe dentatœ. Cymœ axillares et terminales oligocepbalge
capitatse foliolis linearibus intermixtse. Tnvolucra § mm. in
diametro sequantia, campannlata extus glabra intus hirtilla, lobis
lanceolatis denticulatis, glandulis minimis concaviusculis subrotun-
datis, appendice alba petaloidea orbiculari emargiuata intégra
multo latiori, bracteolis interfloribus masculinis lineari-spathulatis
bi-trifidis hirtillis. Styli breviores apice bifidi. Capsulse 1 mm.
in diametro ?equantes, globoso-depress?e, in peripliero triangulares,
acute carinatie. Semina lï mm. in longitudine requantia, nigra
quadrangulo-ovoidea transverse rugulosa.

DiSTRJB. In insulis Philippinis, 'J'imor, Celebes, Fornio.-;;!, Liu-kiu.
Hab, Formosa : Sin-tek, leg. T. Making, anno 1896.

EUPHORBIACEARUM ET RÜXACEARU3I JAPONICARÜM. 77

Liu-kill : Prope Sliuri, leg. K. Miyake, anno 1898; Insula Okinnwa, leg.
Y. Tashiro, anno 1887.

E. thymifolia Ltnn. (Tab. V. E.) Linn. Sp. PI.

ed.-2, 651 ; Thunb. Fl. Jap. 196 ; VVilld. Sp. PL II. 898 ;
AiTON, Hort. Kew. III. 162 ; Roxb. Fl. Ind. II. 473 ; Benth.
Fl. HoHgk. 302 ; Boiss. in DC. Prodr. XV.-2, 47 ; Hemsl. in
Voy. Challeng. Bot. L-2, 22 ; Hemsl. Biol. Cent. America, III.
101 ; Maxim, in Mel. Biol. XI. 833 ; Hook, f, Fl. Brit. Ind.
V. 252 ; Forbes et Hemsl. Ind. Fl. Sin. IL 417 ; Heinry, List
PL For mos. 81 ; PIook. et Arn. Bot. Beech. Voy. 213 ; Miq.
FL Ind. Bat. I. 420.

Nom. Jap. Iriomole-7iishikisö.

Caules filiformes prostrati ramosissimi liirsuti. Folia petiolata
basi subœqnalia oblonga v. oblongo-elliptica obtusa v. acutiuscula
remote serrulato-crenulata, 6-9 mm. longa, 3-4 mm. lata, pagina
supra glabra subtus s^epe parce hirtilla, stipulée lanceolata^
fimbriato-ciliatse. luvolucra i mm. diametro œquantia, axillaria,
solitaria, a internodiorum brevitate racemulos axillares brevissimos
formantia, turbinata extus adpresse liirta intus hirtilla, lobis
triangularibus acutis, ciliatulis, glandulis rotundatis concaviusculis
stipitatis, aj^pendice angusta v. obsoleta, bracteolis interfloribus
musculinis gracilibus ciliatis. Styli longiores bifidi. Capsulte
subglobosœ, 1 mm, diametro œquantes, adpresse hirtse, obtuse
carinatse, erectœ, brevissime stipitatie. Semina rubella f mm.
longa, oblongo-tetragona transverse 5-6-sulcata.

DiSTKiB. In tropicis totius fere orbis.

Hab. Formosa : Kelung, Tai pea, Tamsui, leg. T. Marino, anno 1896 ;
Heng-chhun, leg. Y. Tashiro, anno 1868. Tsui-tng-lau, Pak-kang-khoc,
leg. C. OwATARi, anno 1898 ; Takao, Gi ran -perfect.: Tai-ko-shö, leg.
K. Miyake, anno 1899.

Liu-kiu : Insula Yaeyama, leg. Y. Tashiko, anno 1887.

78 ART. '.]. — B. HAYATA : REVISIO

17. E. maculata Linx. (Tab. V. F.) LinxX. Sp. PI.

ed.-2, 652 ; Willd. Sp. PL II. 896 ; Atton, Hort. Kew. III. ed.-2,
162; Grisebach, Fl. Brit. West. Ind. 53; Boiss. in DC. Prodr.
XV.-2, 46 ; Hemsl. in Voy. Challeng. Bot. I.-2, 63, et Biol.
Cent. America, III. 97.

Nom. Ja]). Konishihisö.

Gaules filiformes, prostrati, crispuli hirtuli ramosissimi. Folia
breviter petiolata basi valde inœqualia oblongo-elliptica v. oblongo-
lanceolata acutiuscula versus apicem serrulata, 6-7 mm. longa, 2-3
mm. lata, supra brunneo-maculata, stipulée lanceolaÜTe denticulatte
fissœ. Involucra axillaria internodiorum brevitate racemulos breves
formautia, pedicellata turbinata extus et intus liirtilla lobis lanceo-
latis, fimbriatis, glandulis transverse rotundatis concavis, appendice
angustiori v. obsoleta, bracteolis interfloribus masculinis gracilibus
liirsutis. Styli breviores, bifidi, apice incrassati. Capsulse li mm.
in diametro œquantes globoso-depressie adpresse liirtse, coccis obtuse
carinatis. Semina 1 mm. longa quadrangulo-ovoidea transverse
3-4-sulcata. Planta ab America ad Japoniam introducta.

DiSTRiB. In America boreali a Canada ad Floridam et Texas.
Hab, Prnpe Tokyo, leg. T. ]\[akino, anuo 1894.

18. E. humifusa Willd. (Tab. V. G.) Ledb. Fl.

Boss. III. 557 ; Bolss. in DG. Prodr. XV.-2, 30 ; Miq. in Ann.
Mus. Bot, Lugd-Bat. III. 125 et Prol. 289 ; Franch. et Savat.
Enum. PI. Jap. I. 420; Franch. PI. David. 262; Maxim, in
Mel. Biol. XL 832 ; Debeux, Fl. Shangh. 53 ; Forbes et Hemsl.
Ind. Fl. Sin. II. 414 ; Henry, List PI. Formos. 81 ; L. Diel^^,
in Engl. Jahrb. XXIX. 430.

E. tkymifolia Thunb. Fl. Jap. 196.

Nom. Jap. Nishlkisö, Somokuzusetsu, vol. 9, fol. 24.

EUPHORBIACEARUM ET BUXACEARÜM JAPONICARUM. 79

Cailles prostrati filiformes teniiiter patuli hispiduli v. glabri,

dichotome ramosi. Folia basi valde inœqualia oblonga, oblongo-

elliptica obtusa serrulata, glabra, sœpe subtus hirtula, 6-12 mm.

longa, 3-6 mm. lata, stipulée lineari-lanceolatœ sœpe trifîdœ. In-

volucra solitaria internodiorum brevitate ssepe congesta, turbinato-

campanulata, iiitus et extus glabra, lobis triangularibiis subdentatis,

glandulis transverse ellipticis v. rotundatis substipitatis, appendice

angustiori intégra, bracteolis interfloribus masculinis linearibus.

Capsulée lï mm. diametro œquantes, glabrœ, coccis obtuse carinatis.

Semina 1 mm. longa oblongo-tetragona glabra.

DiSTRiB.: In Mongolin, Siberia et Japonia,

Hab. Hondo : prope Tokyo, leg. T. Marino; prov. Suwô, in traetu Yoshiki,

leg. G. NiKAi, anno 1892.
Shikoku : prov. ïosa, in traetu ïakaoka, leg. T. Making.

19. E. microphylla Heyne, (Tab. V. H.) Hook.

f. Fl. Brit. Ind. V. 252.

E. chmncesyce Eoxb. Fl. Ind. II. 473.

E. serpens o Engelm.; Boiss. in DC. Prodr. XV.-2, 30.

Nom. Jap. Kobano-nishikisd.

Cailles filiformes prostrati dichotome ramosi. Folia 2-3 mm.
longa, 1 è— 2 mm. lata valde inaequalia subcordata ovata v. ovato-
elliptica intégra obtusa, stipulée triangulares apice denticulatœ.
Involuera axillaria solitaria, internodiorum brevitate approximata,
tnrbinato-campanulata extus et intus subglabra, lobis ovatis 2-3-
fidis, glandulis transverse oblongis, appendice 2-3-lobata. Styli
breviores bifidi apice subincrassati. Capsulte le mm. diametro
saquantes, coccis acute carinatis. Semina 1 mm. longa, ovato-
tetragona tenuiter rugulosa.

DiSTRiB.: in Indea orientali.

Hab. Formosa : Tamtsui, leg. T. Making, anno 1896.

80 AET. ?,. B. IIAYATA : RE VISIO

BUXACE.41.

Conspectu.^ Genervm.

Folia alterna 1 . Pachysandra.

Folia opposita 2. Buxtis.

1 Pachysandra Michx.

Endl. Gen. PL 1123; Muell. Arg. in DC. Prodr.

XVI.-l, 21 ; Batll. Hist. d. PL VI. 49 ; Benth.

et Hook. Gen. PL IH, 267 ; Pax, in Nat. Pfl.

Fam. IIL-5, 132.
Flores monoici, apetali. Discus 0. FL J : sepala 4, 2-
seriatim imbricata. Stamina 4, sepalis opposita, filaraeutis liberis
exsertis crassiusculis ; antlierse introrsœ, oblong?e, loculis adnatis,
parallelis, longitudinaliter dehiscentibus. Rudimentum ovarii vertice
truncatum. FL ? : sepala 4, 2-seriatim imbricata. Ovarium
4-loculare ; styli basi subdistincti, erecti, longiusculi, super ne
patentes, indivisi ; ovula in loculis solitaria. Drupa, epicarpio
in sicco laxo in vivo verisimiliter pulposo, endocarpio tenuiter
crustaceo. Semina pauca, obovoideo-oblonga, testa Crustacea nitida
— Herbse basi procumbentes radicantesque, ramis adscendentibus
apice foliatis. Folia alterna, petiolata, latiuscula, grosse dentata,

EUPHORBIACEARUM ET BUXACEARUM JAPONICARUM. 81

triplinervia. Spicœ longiusculœ, floribiis Ç ad basin spicœ pauce
dispositis, cœteris J. Flores sub bracteis alternis solitarii fl. J
sessiles, fl. $ breviter pedicellati.

P. terminals Sieb, et Zucc. (Tab. VI. F.) Sieb, et

Zücc. Fl. Jap. Farn. Nat. Sect. Prim. 142 ; Baill. Mono. Bux.
et Styl. 57 ; Miq. in Ann. ^Mus. Bot. Lngd-Bat. III. 128, et
Prol. 292 ; Muell. Arg. in DC. Prodr. XVI.-l, 21 ; Franch.
et Savat. Eum. PI. Jap. I. 428 ; Forbes et Hemsl. Ind. Fl,
Sin. IL 419; L. Diels in Engl. Jahrb. XXIX. 431.

Xom. Jap. Fuhkiso, Sömokuzusetsn, vol. 20, fol. 24.

Gaules seraipedales, undique foliati, sed demum casu foliorum
basi uudati, cum reliquis partibus glabri. Folia 4-8 cm. longa,
2-3 cm. lata, in petiolum 2-2.5 cm. longum decurrentia, e basi
cuneata, obovata, apice obtusa v. truncata, v. etiam acuta, a medio
sursum grosse remoteque serrata, subcoriacea ; serrae deltoidese,
inœquales, acut?e, adpressse. Spica terminalis, solitaria, minus
multiflora erecta. Flores masc. numerosiores, vulgo bibracteolati ;
bracteolœ late ovatse cum sepalis margine tenuiter ciliolatse
subcoriacese, circ. 2 mm. longse. Sepala orbiculari-ovata, obtusa.
Stamina longe exserta. Rudimentum ovarii truncatum quadran-
gulum concavum. Flores foem. pauci pedicellati ; bracteolse ovatae
acutse, coriacese, magnitudine sensim in sepala abeuntes, circ. 2
mm. longue. Sepala triangulari-ovata acuta. Styli reflexi calycem
longius superantes.

DiSTRiB.: In Japonia, China centrali.

Hab, Houdö : Prov. Suruga, Hakone, leg. Y. Yabe, anno 1901. Prov.

Hida ; prov. Eehigo, in monte Gozu ; j^rov. Iwashiro in tractu Aidzu ;

prov. Sado, in monte Kinpokusan, et monte Haguro.
Yezo : Sapporo.

82 AKT. 3. B. HAYATA : REVISIO

2. Buxus L.

LiNN. Sp. PI. ed.-2, 1394; Endl. Gen. PL 1123;

MuELL. Arg. in Prodr. XVI.-l, 14; P)Aill. Mon.

Bux. et Styl. 58, et Hist. d. PL VI. 48 ; Benïh.

et Hook. Gen. PL III. 19; Pax. in Nat. Pfl.

Farn. V.-5, 133.
Flores nionoici, apetali. Discus 0. FL J : sepala 4, 2-
seriatim imbricata. Stamina 4, sepalis opposita, filamentis validis
subcomplanatis liberis exsertis crassiusculis ; antherse, oblongse
demum recurvse, localis introrsum adnatis parallelis longitudinaliter
dehiscentibus. Kudimentum glandulosum ovarii compressum
truncatum v. apice dilatatum 4-lobatum. FL $ : sepala 6, decussatim
biserialia valde imbricata, exteriora minora, cum stigmatibus
alternantia. Ovarium 3-loculare ; styli breves, crassi, inter se
distantes ; ovarii apex inter stylos glanduloso-3-lobatus. Capsula
ovoidea, stylis persistentibus 3-cornuta, loculicide dehiscens, valvis
indivisis stylis fissis 2-cornutis, pericarpio indurato, endocarpio soluto
cartilagineo. Semina oblonga, 3-quetra, caruncula parva, testa
subcrustacea nigra nitida ; albumen oleagineum subcarnosum ;
embryo centralis, cotyledones oblongae radicula non latiores.
Frutices ramosissimi, glabri. Folia opposita, breviter petiolata,
integerrima, coriacea, pennivenia. Spicse in axilla foliorum solitarise
breves, v. terminales. Bracteœ numerosse, sepala similantes. Flores
terminales ssepe Ç, axillares J, breviter pedicellati.

1. B. sempervirens Linn. var. japonica (Muell. Arg.)

EUPHORBI ACE ARUM ET BUXACEARUM JAPONICARUM. 83

Making (Tab. VI. C.) ■Making, in Tokyo Bot. Mag. IX.

(1895) 281 et XV. (1901), 169.

B. japonica Muell. Arg. in DC. Prodi*. XVI. 1, 20 ; Miq.
Prol. 292 ; Franch. et Sa vat. Enum. PI. Jap. I. 428 ; Pax, in
Nat. Pfl. Fam. III.-o, 133.

B. sempervirens Thunb. Fl. Jap. 77 ; Fgrbes et Hemsl. Ind.
Fl. Sin. II. 418.

Nom. Jap. Asama-tsuge, Tsuge, vel Benten-tsuge, Honzôzufu,
vol. 92, fol. 17.

Eamuli glabri. Folia 10-25 mm. longa, 8-12 mm. lata,
obovata, apice emargiuata, v. mucronata, basi acuta v. attenuata,
ad petiolos puberula, petiolis brevibus margine pilosiusculis subtus
convexis, supra obtuse canaliculatis. Inflorescentiee (partes dense
imbricatœ) subglobosse basi bracteis sterilibus ovatis obtusis quam
fertiles duplo v. ultra brevioribus prœditœ. Flores J : antherse
ovatfe V. ovato-sagittatse apice obtusœ, basi emarginatse, post
anthesin arcuato-reflexœ, filamentis calyce duplo-triplo longioribus.
Rudimentum ovarii longum calycem subœquans apice disciformi-
dilatatuni 4-lobatum.

Hab. Shikoku : Prov. Tosa, leg. T. Making.

Hondo : Insula Közusima ; prov. Shimotsuke, in monte Nikko-urayama, anno
1879 ; prov. Musashi, in tractu Tichibu.

2. B. sempervirens Linn. var. microphylla Bl. (Tab.

VI. D.) Blume, in herb. Lugd.-Bat. ex Miq. Prol. 292 ;

HoGK. f. Fl. Brit. Ind. V. 267.

B. microphylla Seib. et Zucc. in Abhandl. Akad. Muench.
IV. 2, 142 ; Baill. Monogr. Bux. et Styl. 64.

B. japonica ß microphylla Muell. Arg. in DC. Prodr.

84 ART. 3. — B. HAYATA : REVISIO

XVI.-l, 20 ; MiQ. Piol. Fl. Jap. 292 ; Franch. et Sav. Enum.
PI. Jap. I. 428.

B. sempervirens Linn. « angustifolia Sieb, et Zucc. I.e.

Nom. Jap. Hime-tsuge.

Folia 13-20 mm. longa, 3-8 mm. lata, spathulato-lanceolata,
lineari-oblonga, obtusa v. subemarginata, basi attemiata, coriacea
nitida glabra. . Styli carnosi, emarginato-bilobati.

Hab. Prov. Awa, Kominyu-iiuu-a, anno 1880; Mama-niura (cult.)

3. B. Wallichiana Baill. (Tab. VI. E.) Baill.

Mono. Bux. et Styl. 63.

B. sempervirens Linn. var. Imkiuensis Makino, in Tokyo
Bot. Mag. IX. (1895), 279, et XV. (1901), 169.

Nom. Jap. Okinawa-tsuge.

Frutices ; rami fusci, ramulis puberuli.s. Folia opposita,
oblongo-ovata, oblonga, obovata, ovata, v. oblanceolata, basi
attenuata, emarginata, intégra, coriacea, viridia, nitida, subtus
pallidiora, li-3cm. lata, 3-6 cm. longa cum petiolis puberulis
brevibus. Inflorescentiae axillares, bracteatœ ; fl. ^ superiores, fl.
J inferiores. Flores J : brevissime pedicellati. Sepala 4, mem-
branacea, concava, margine ciliolata, duo exteriora minora late ovata ;
duo interiora majora, orbicularia, concava, 2è mm. lata. Rudi-
mentum ovarii inclusum brève apice dilatatum 4-lobatum. Stamina
4, multo exserta ; filamenta valida ; antherse angusto-ovatse.
Capsulse ovoidese levés, durse, circ. 1 cm. longœ, stylis persistentibus.
Semina oblonga, nigra, nitida.

Hab. Liu-kiu : Insula Okinawa, in monte Kuzi-magiri, leg. Y. Tashiro/
auuo 1887; Shuri.

■•rs°-:

EUPHORBIACEARUM ET BÜX ACE ARUM JAPONIC ARUM.

85

Distributio Specierum

in Imperio Japonico.

Species et Varietates.

Yezo.

Hondo,
bor.

Hondo,
med .

Hondo. Shi- Kiu-
aust. k(,ku. sliiu.

Liu- j For-
kiu. mosa.

Securinega flueggeoides Muell. Arg.

—

...

...

Flueggea inicrocarpa Blume.

Phyllanthus Niruri L.

—

...

...

...

...

—

„ urinaria L.

—

...

...

...

—

—

„ simplex Retz,

—

„ Matsumuris sp. nov.

— . —

„ liukiuensis Matsum.

„ flexuosus Muell. Arg.

—

—

—

„ reticulatus PoiR,

Glochidion lanceolatum sp. uov. j

„ zeylanicum A. Juss.

,, hirsutum Muell. Arg.

„ bicolor Muell. Arg.

„ obovatum Sieb, et Zucc.

—

... .

...

—

—

„ formosanum sp. nov.

—

Breyuia accrescens sp. nov.

„ rhamnoides Muell. Arg.

—

„ stipitata Muell. Arg. var. for-

mosana.

—

Putranjiva Roxburgh ii Wall.

.

Antidesma japonica Sieb, et Zucc.

—

Bischoffia japonica Blume.

—

Bridelia tomentosa Blume.

Daplmiphyllum niaerojDodum MiQ.

—

—

,, glaucescens Blume.

—

...

...

—

„ „ „ var. Old-

liami Hemsl.

„ himalayense Muell. Arg.

Croton Cumingii Muell. Arg.

„ Tigliura L.

Mercurialis leiocarpa Sieb, et Zucc.

—

—

• • •

_

Mallotus moluccanus Muell. Arg.

„ Playfairii Hemsl.

„ répandus Muell. Arg.

,, philippinensis Muell. Arg.

—

—

„ japonicus Muell. Arg.

—

...

1

—

„ cochinchinensis Lour.

•

86

ART. 3. B. HAYATA : EEVISIO

Species et Varietates.

Alchornea trewioides Muell. Arg.
Cleidion ulnüfolium Muell. Arg.
Macaranga Tanarius Muell. Arg.
Acalypba australis L.

„ „ var. lanceolata n. v,

Ricinis communis L.
Homonoya reparia LouR.
Aleurites cordata Steud.
Excoecaria Agallocha L.

„ japonica Muell. Arg.

„ crenulata Wight.
Sapium sebiferum Roxb.
Euphorbia lathyris L.

,, dendroides L.

,, Heliosoopia L.

„ adenochlora Morr. et Decne.

„ Jolkini Boiss.

,, pekinensis RuPR.

„ togakusensis sp. nov.

„ Orientalis L.

,, ebracteata sp. nov.

„ Sieboldiana Morr. et Decne,

,, Esula L.

„ Sparmanni Boiss.

,, pilulifera L.

,, bypericifolia L.

„ serrulata Reinw.

,, thymifolia L.

„ maculata L.

,, humifusa Willd.

,, microphylla Heyne,
Pachysnndra terminalis Sieb, et

Zucc.
Buxus serapervirens L, var. japonica
Marino.

,, sempervirens L. var. microphylla

Blume.
„ Wallichiana. Baill.

Yezo.

Hondo,
bor.

Hondo,
med.

Hondo.

aust.

Shi- Kiu-
koku. shiu.

Liu- For-
kiu. mosa.

EUPHORBTACEARUM ET BUXACEARUM JAPONICARUM.

87

INDEX NOMINUM

ET SYNONYMORUM

(SYNONYMA LITTERIS CURSIVIS TRADUNTUR)

Acalypha Linn 50

,, australis Linn,

(Tab. IV. D.) 50
„ ,, var lanceolata

var. nov 51

„ chinensis RoxB 50

„ ge7n,ina Muell. Akg. 51

„ pawc*/?ora HoRNEM.... 51

„ virgata Thunb 50

Adelia nereifolia Roth .54

Alchornea S.W 4G

„ trewioides Muell. Aug.

(Tab. IV. A.) 47

Aleurites Forst. 5b

„ corda ta Steud 55

Andrachne trifoUata Roxb 28

Antidesma Linn 26

„ jciponica Sieb, et Zucc.

(Tab. II. L) 27

BischofRa Blume 28

„ javanica Blume 28

„ Rœperiana Decne.... 28

„ trifoUata Hook 28

Bradleia hirsuta Roxb 18

Breynia Forst 21

„ accrescens sp. nov. form.

a (Tab. LH.) 22

Breynia accrescens sp. nov. form.

ß (Tab. I. L) 22

,, rharanoides Muell. Arg.

(Tab. L J.) 22
„ stipitata Muell. Arg. var.
formosaniim var. nov.
form. «...(Tab. IL A.) 23
„ stipitata Muell. Arg. var.
formosanum var. nov.
form./?... (Tab. IL B.) 24

Bridelia Willd 29

,, ZoMre^r^ Hook, et Arn. 30
„ tomentosa Blume.

(Tab. IIL A.) 30

Buxaceae 80

BuxusL 82

,, japonica Muell. Arg.... 83
„ „ ß. microphylla

Muell. Arg 84

„ microphylla Sieb, et Zucc. 83
„ sempervirens Linn, a ang-

ustifolia Sieb, et Zucc. 84
,, sempervirens Linn. var.
japonica. Making.

(Tab. VI. C.) 82
„ sempervirens L. var.

liukiuensis Making..... 84

88

ART. 3. — B. HAYATA : EEVISIO

BUXUS setnperi-'irens Trv^b 82

„ sempervirens L. var. mi-
crophylla Bl.

(Tab. VI. D.) 83
„ Wallichiana. Baill.

(Tab. VI. H.) 84
Chorizandra pinnafa Wight. ... 6
Cicca flexuosa Sieb, et Zucc. ... 12

Cicca microcarpa, Benth 14

Cleidion Blume 47

„ nlmifolium Muell. Arg.

(Tab. IV. B.) 48

Croton LiNN 34

„ Cumingii Muell. Arg.

(Tab. III. B.) 36

„ jcuponicum Thunb 44

„ montanus^WÄÄi 43

„ poly stacht) US Hook, et

Arn 36

„ rhombifolius Willd 42

„ sehiferum Linn 62

„ Tiglium Linn.

(Tab. III. C.) 36

Daphniphyllum Blume 31

,, Benthaml Baill.... 34

„ glaucescens Blume.

(Tab. II. K.) 33
„ var.

Oldham i Hemsl. 34
„ himalayense Muell.

Arg. (Tab. II. L.) 34

„ humile Maxim 34

„ macropodum Miq.

(Tab. II. J.) 32
„ Roxhurghii Baill. 33

Dryandra cordata Thunb .55

Elœococca cordata Blume 65

EuphorbiaceaB 2

Euphorbia Linn 63

„ adenochlora Morr. et

DECNE.(Tab.IV.H.)67
„ Chamcesyce Boxb.... 79

„ coralloides L 71

„ coralloides Thunb. 69

,, dendroides L 65

,, ebracteolata sp. nov.

(Tab. IV. L) 71

„ Esnla Linn 72

„ Guiliehni A. Gray. 71

,, Helioscopia Linn.

(Tab. IV. G.) QCy

„ hirta Thunb 74

„ linmifusa Willd.

(Tab. V. G.) 78
,, hypericifolia Linn.

(Tab. V. C.) 75

„ japonica Boiss 67

„ lœta Ait 65

,, Lathyris Linn.

^ (Tab. IV. F.) 64

,, Jolkini Boisft 68

„ lassiocaula Boiss.... 69

,, maculata Linn.

(Tab. V. F.) 78
,, microphylla Heyne.

(Tab. V. H.) 79
„ parvißora Linn, . 75

„ pekinensis Kufr.

(Tab. IV. J.) 68

„ pilulifera Linn 74

„ Orientalis L 70

„ ramosissima Boiss. 73

„ serpens 3 Engelm

Boiss 79

„ serrata Keinw 76

„ Sampsini Hange. ... 69

EÜPHORBIACEARUM ET BUXACEARUM JAPONTCARUM.

89

Euphorbia Sieboldiana Morr.... 71
,, Sparmanni Boiss. ... 73

„ thyraifolia Linn. ... 77

„ thyQnifoh'alHvmi.... 78

,, togakusensis sp. nov.

(Tab. IV. A.) 69
„ Vachellii Hook, et

Arnt 76

Exccecaria Linn .58

„ Agallocha Linn. ... 58

,, Camettia Willd. ... .59

„ cochinchinensis

LouR 60

,, crenulata Wight.... 60

,, japon icaMuELL. Arg.

(Tab. IV. E.) 59
„ sehi/eraMuELL. AuQ. 62

Fluggea Willd 5

„ microcarpa Blume. ... 5

„ sinensis Baill 6

Gelonium RoxB 56

,, œquoreum Hange 57

Glochidion Forst 15

,, acuminatum Muell. Arg. 18
Glochidion bicolor (Muell. Arg.)

(Tab. IL E.) 18
„ flexuosum Franch.

etSAV 12

„ formosanuni sp. nov.

(Tab. II. G.) 20
,, hirsutnm Muell.

Arg. (Tab. II. D.) 17
„ lanceolatnm sp. nov.

(Tab. ILIO.) 16
„ obovatuna Sieb, et

Zucc. (Tab. IL F.) 19

,. zeylanicum A. Juss. 17

Gomphia neilgJierrensis Wight. 33

Hemotospermitm, salicinum. Baill. 54

Homonoya Lour 53

,, riparia LouR 54

Kirganelia sinensis Baill 14

Macaranga Thouars 48

„ Tanarius Muell.

Arg. (Tab. IV. C.) 49

Mallotus Lour 39

„ cochinchinensis Lour.

(Tab. III. J.) 45
,, contuhernalis Hange. 42
„ japonicns Muell. Arg.

(Tab. III. I.) 44
,, moluccanus Muell.

ARG....(Tab. IIL E.) 40
,, jjoniculafns Muell.

Arg 45

„ philippinensis Muell.

Arg... (Tab. IIL H.) 43

„ Playfairii Hemsl

(Tab. IIL F.) 41
„ répandus Muell. Arg.

(Tab. III. G.) 42

llappa moluccana Wight 49

,, Tanaria Spreng 49

Melanolepis multiglondulosa Rei-

CHENB 40

Blelanthesa rhamnoides Wig h t. . . . 2 3

Mercurialis Linn 37

„ leiocarpa Sieb, et

Zucc.(Tab.IILD.) 38

Nageia Futranjiva Roxb 25

Nymphanihus Niruri Lour 8

Owataria formosana Matsumura. 57

Pachysandra Michx 80

,, terminalis Sieb, et

Zucc.(Tab.VI.F.) 81
Phyllanthus Linn 6

90

ART.

-B. HAYATA : REVISIO

Phyllanthus anceps Vahl 10

„ hicolor MuELL. Arg.... 18
„ flexnoBus Muell. Arg.

(Tab. I. G.) 12

„ Jmsutus Muell. Arg. 18

,, japonicus Muell. Arg. 4
,, /epidocarpns Sieb, et

Zucc 9

,, liukiuensis Matsumura.

(Tab. I. F.) 11
,, Matsumurfe sp. nov.

(Tab. I. E.) 11

,, microcarpus Muell.... 14

,. Niinamii sp. Jiov 14

„ Nirnri Linn.

(Tab. I. B.) 7

„ o&o^;a^^^s Muell. Arg.... 19

„ reticulatus PoiR 13

,, retustrs Roxb 6

„ rhamnoides Muell.

Arg 23

„ simplex Retz.

(Tab. I. D.) 10
,, „ y. chinensis

Muell. Arg 10

,, sinensis ^i\]EiÄj. Arg.... 14
„ urinaria Linn. (Tab. I.e.) 8

,, virOSUS WlLLD 6

„ vitis-idœa Roxb 23

„ Wigh.tianusM.\]FÄÄ,. kwG. 6

Putranjiva Wall 24

,, Roxburghii Wall.... 25

Ricinus Linn 52

„ communis Linn 52

,, Mappa Roxb 49

„ Tanarius Linn 49

Rottlera aurantica Hook, et Arn. 43

„ cordi/olia Benth 42

,, japonica Sier. et Zucc. 44

„ yapomca Spreng 44

,, panicidcda A. Juss. ... 45

,, scabrifolia A. Juss. ... 42

, , tinctoria Roxb 43

, , triccoca Roxb 42

„ trinervis Zipp 42

Sapium P. Br 61

,, sebiferum Roxb 61

Securinega A. Juss 3

„ fluggeoides Muell. Arg.

(Tab. I. A.) 4

,, japonica ISliQ 4

,, ohovata Muell. Arg. ... 5

Stillinga Agalloclia Baill 59

,, japonica Sieb, et Zucc. 59

„ sehifera Michx 62

„ sinensis Baill 62

Treioia nudißora Hange 42

„ tricuspidata Willd 45

Triadeca sinensis Lour 62

Vernicia montana Lour 55

EUPHORBIACEAKUM ET BUXACEARlîM JAPONICARUM.

91

INDEX NOMINÜM JAPONICARUM.

Aburagiri 56

Akagi 29

Akamegashiwa 44

Amigasagiri 47

Annan-akamegashiwa 45

Asama-tsuge 83

Benten-tsuge 83

Chankanii 3G

Daiwan-kobannoki 23

Eüoki-fuzi 48

Enokigusa 51

Fukkisö 81

Hagikusö 73

Hana-komikansö 11

Hanshiren 65

Hazu 37

Hima 53

Hime-mikansö 11

Hime-tsuge 84

Hiaie-yuznriha 33

Hitotsuba-hagi 4

Horutosö 65

Hosobano-enokigusa. 51

Iriomote-nishikosö 77

Iwataigeki 68

Kakibano-kankonoki 17

Kankomodoki 30

Kankonoki 19

Kekankonoki 20

Kidachi-komi ka usö 8

Kidachi-taigaki 65

Kiirun-kankonoki 16

Kobannoki 12

Kobano-nishikisö 79

Komikansö 9

Konishikisö 78

Kusunohagashivva 43

Marumi-nourushi 71

Miyakogima-nishikisö 76

Miyama-nourushi 69

Nagabano-himeyuzuriha 33

Nanban-yanagi 54

Nankinhaze 62

Natsutödai 71

Nishikisö 78

Nonrushi 67

Öagari-nishikisö 73

Öbaki 49

Obano-kekankonoki 18

Öbatsnge 57

Okiiiawa-tsuge 84

Oshima-kobannoki 23

Riyögiyö 67

Shima-ammaroku 14

Shima-azusa 41

Shima-hitotsubahazi 6

Shima-kobannoki 14

Shima-komikansö 10

Shima-manmaroku 14

Shima-nishikisö 74

Shima-seishiboku 60

Shima-tshiraki 59

92

ART. 3. — B. HAYAÏA

Shima-taigeki 70

Shiraki 59

Takatödai. 69

Takusitsu 66

Tödaigusa 66

Tögoma 53

Tsuge 83

Tsuge-modoki 25

Tsuru-akamesashiwa 42

Uguyoshi 27

Unrin-nishikisö 75

Uraziro-kankonoki 18

Yama-se 38

Yama-liihazu 27

Yanbaru-akamegashiwa 40

Yuzuriha 32

Zokuzuishi 65

B. HAYATA.

REVISIO EüPHORBIaCEARÜM ET BUXACEARUM JAPONICARUM.

TABULA I.

Explicatio tabulae I.

Planta3 ex specirainibus exsiccatis depictte.

A. Securinega fluygeoides Muell. Arg. (p. 4). — 1 Alabastri J cum
bracteolis. mag. auct, — 2 Flos $ per anthesin. mag. auct. — 3 Flos maturus J.
mag. auct. — 4 Stamen alterum a facie, alterum a tergo visum, mag. auct. — 5
Ovarii rudimentum cum glandulis. mag. auct. — 6 Flos ^. mag. auct. — 7
Ovarium, mag. auct. — 8 Calyx cum disco, mag. auct. — 9 Capsula a facie
visa. mag. auct. — 10 Capsula a tergo visa. mag. auct. — 11 Capsula dehiscens.
mag. auct. — 12 Verticalis cocci sectio. mag. auct. — 13 Semen alterum a facie,
alterum a latere, visum, mag. auct. — 14 Verticalis seminis sectio. mag. auct. —
15 Embryo, mag. auct.

B. PhyUanthus Niruri L. (p. 7). — 1 Fragmentum rami, cum lloribus
^. et ^. — 2 Flos ^. — 3 Columnte staminales ; altera a facie, altera a tergo,
visa. — 4 Flos ^. — 5 Sepala et discus, ovarium amortum. — 6 Fructus. — 7
Semen. (1-7, mag. auct.)

C. Fhijllanthus uriuaria L. (p. 8). — 1 Fragmentum rami, cum flori J.
— 2 Flos J . — 3 Fragmentum calycis, glandulfe videntur.— 4 Antherœ starai-
nalis columned, a facie visa3. — 6 Ovarium. — 7 Calyx cum disco.— 8 Fructus. —
9 Verticalis capsulfe sectio, semina videntur. — 10 Semen. (1-10, mag. auct.)

D. Fhyllanth'us simplex Retz. (p. 10). — 1 Fragmentum rami. — 2, 3
Flores J. — 4 Stamina juniora, a latere visa. — 5 Stameu, alteriim Junius,
alterum maturum, a facie visum. — 6 Flos ^ . — 7 Calyx et discus, pistillum
amortum. — 8 Fructus. — 9 Ejus verticalis sectio, semen videtur. — 10 Semen.
(1-10, mag. auct.)

E. Phyllanthus Matsunmrce sp. nov. (p. 11). — 1 Fragmentum rami. — 2
Flores ^ et Ç. — 3 Flos ^, glandulœ videntur. — 4 Ejus verticalis sectio. —
5 Flos ^. — 6 Sepala et glandulse, pistillum amortum. — 7 Fructus. (1-7,
mag, auct.)

F. Phyllanthus liuhiuensis Matsumura. (p. 11). — 1 Flos ^ a latere visus.
— 2 Idem a facie visus. — 3 Stamina circa rudimentum ovarii inserta, et
redimentum ovarii. — 4 Idem juniora. — 5 Discus alter a facie, alter a dorso
visus. — 6 Flos ^. — 7 Fructus, a latere visus. — 8 Fructus a facie visus. — 9
Semen validissime auctum. (1-9, mag. auct.)

Gr. Phyllanthus flexuosus Muell. Arg. (p. 12). — 1 Fragmentum ramuli.
mag. auct. — 2 Folium, mag. nat. — 3 Stipula valde aucta. — 4, 5 Alabastra.
— 6 Flos ^. — 7 Disci et stamina, calyces amorti. — 8 Stamen alterum a
facie, alterum a dorso visum. — 9 Flos ^. — 10 Sepalum. — 11 Pistillum, calyx
amortus. — 12 Apex pedicelli floris ^ , pistillum et sepala amorta. — 13 Fructus.
(bacca). mag. nat. — 14 Semen. — 15 Verticalis ejus sectio, cum dorso embryone.
— 16 Embryo alter a facie, alter a latere visus. (3-12, et 14-16, mag. auct.)

H. Breynia accrescens sp. nov. form a. (p. 22). — 1 Fragmentum rami,
mag. nat. — 2 Alabastrum ^ cum bracteolis. — 4 Flos ^. — 3 Ejus verticalis
sectio. — 5 Flos Ç. — 6 Pistillum, ejus verticalis sectio. — 7 Horizontalis ovarii
sectio. — 9' Calyx fructiger accrescens, mag. nat. — 8 Semen a latere visum.
— 9 Idem a facie visum. — 10 Embryo, (2-10, mag, auct.)

I. Breynia accrescens sp. nov. form. ß. (p. 22). — 1 Flos J. — 2 Ejus
verticalis sectio. — 3 Flos Ç . — 4 Fructus. — 5 Calyx fructiger. (1-3, mag. auct.;
4-5, mag. nat.)

J, Breynia rhamnoides Muell. Arg. (p. 22) — 1 Fragmentum rami,
cum folio et fioribus Ç. mag. nat. — 2 Flos J cum bracteolis et stipulis. — 3
Verticalis floris ^ sectio. — 4 Columa staminalis. — 5 Flos ^. — 6 Verticalis
ejus sectio. — 7 Fructus. mag. nat. —8 Semen a latere visum. — 9 Verticalis
ejus sectio, embryo videtur. — 10 Horizontalis ejus sectio. — 11 Embryo a latere
visus. — 12 Embryo a facie visus. (2-6, et 8-12, mag auct.)

Hayata, Revisio Euphorbi. Bux. Japonic

A. 2.

Jour. Sei. Coll. Vol. XX., Art. 3. PI. I.

lull .1- Imy tit* Tokm t'r

R. HATATA.

REVISIO EUPHORBIACEARUM ET BÜXACEARDM JAPONICARUM.

TABULA IL

Explicatio tabulae II.

Plantée tabulte ex speciminibus exsiccatis depictte.

A. Breynia stipitata Muell. Arg. ya,r. formosana n. v. form. a. (p. 23).
— 1 Fragnientum ramuli cum folio et flore Ç . mag. nat. — 2 Flos ^. (Alabast-
rum.) — 3 Idem a facie visum valde auctum. — 4 Flos J (maturus). — 5 Verticalis
ejus Sectio. — 6 Stamiualis columna a latere visa. — 7 Idem a facie visa. — 8 Flos
Ç . — 9 Idem, calyx fissus, pars amorta, pistillum videtur. — 10 Fructus. mag.
nat. — 11 Semen mag. nat. — 12 Semen, alternni a latere, alterum horizontaliter
visum. — 13 Embryo. (2-9, et 12-13, mag. auct.)

B. Breynîa dipitata Muell. Arg. vs.r . formosana n. v. form. ß. (p. 24).
— 1 Fragmentum rami cum floribus Ç et stipulis, folia amorta. — 2 Flos ^,
alter a facie, alter a latere visus. — 4 Flos Ç. (1-4, mag. auct.)

C. Glochidion lanceolatum sp. nov. (p. 16). — 1 Fragmentum rami, cum
floribus J et stipula. — 2 Flos J. — 3 Stamiualis columna. — 4 Fragmentum
rami cum floribus Ç . — .5 Flos Ç . — 6 Verticalis ejus sectio. — 7 Horizontalis
ovarii sectio. — 8, 9 Fructus, alter a latere, alter a facie visus. — 10 Fructus
dehiscens. — 11 Fragmentum folii. — 12 Bracteola. (1-7, et 12, mag. auct.;
8-11, mag. nat.)

D. Glochidion hirsutum Muell. Arg. (p. 14). — 1 Fragmentum ramuli
cum floribus ^ et folio. — 2 Flos $. — 3 Stamiualis columna. — 4 Flos Ç . —
5, pistillum, calyx amortus. — 6, Verticalis ejus sectio. — 7 Horizontalis ovarii
sectio. — 8 Fructus. — 9 Verticalis fructus sectio. a = Semen abortum. b = Semen
maturum. — 10 Semen a latere visum. — 11 Semen a facie visum, cavae
interlaminoe videntur. — 12 Verticalis seminis sectio, embryo videtur. — 13
Horizontalis seminis sectio. (1 et 8, mag. nat.; 2-7, et 9-13, mag. auct.)

E. Glochidion hicolor Muell. Arg. (p. 18). — 1 Fragmentum folii. — 2
Flos J, a latere visus. — 3 Flos J, a facie visus. — 4 Flores Ç. — .5 Verticalis
floris ^ Sectio. — 6 Horizontalis ovarii sectio. — 7 Fructus. — 8, 9 Semina a
facie visa. — 10 Embryo. — 11 Horizontalis seminis sectio. (1 et 7, mag. nat.;
2-6, et 8-11, mag. auct.)

F. Glochidion obovatum Sieb, et Zucc. (p. 19). — 1 Fragmentum rami»
cum floribus ^ et S. — 2 Fragmentum folii. — 3 Flos $. — 4 Stamen. — 5
Flos Ç. — 6 Fructus. mag, nat. — 7 Semen, a facie visum, cava interlamina
videtur. (1, 2, et 6, mag. nat,; 3-5, et 7, mag. auct.)

G. Glochidion formosanum, sp. nov. (p. 29). — Fragmentum folii. mag.
nat.— 2 Fasciculi florum g et J. — 3 Flos J.— 4 Flos Ç.— 5 Ejus verticalis
sectio, — 6 Horizontalis ovarii sectio. — 7 Fructus. mag. nat. (1-2, et 7, mag.
nat.; 3, 4, 5, et 6, mag. auct.)

H. Putranjiva Roxhurghii Wall, (p, 25). — 1 Fragmentum folii, mag.
nat. — 2 Flores ^. — 3 Flos J, calyx fissus et pars amorta, filamenta videntur.
— 4 Alabastrum ^. — 5 Stamen, alterum a facie, alteram a dorso, visum. —
6 Flos Ç. — 7 Fructus, mag. nat. — 8 Verticalis ejus sectio, albumen videtur
— 9 Verticalis albuminis sectio, embryo videtur. (2-G, et 9, mag. auct.)

I. Antidesma japonica Sieb, et Zucc. (p. 27). — 1 Fragmentum rami
cum spica $ et folio, mag. nat. — 2 Fragmentum spicfe ^. — 3 Stamen Junius
alterum a facie alterum a dorso visum. — 4 Stamina matura. — 5 Stamen
dehiscens et oscillatum. — 6 Fragmentum spicfB Ç. — 7 Flos Ç. — 8
Ejus verticalis sectio. — 9 Horizontalis ovarii sectio. — 10 Flos Ç a facie
visus, stigmata videntur. — 11 Flos Ç, pistillum amortum, discus videtur. —
12 Putamen a latere visum. — 13 Idem a dorso visum. — 14 Embryo. — 15
Diagramma fl. ^. — 16 Dagramma fl. Ç. — 17 Spica cum fructibus. (1 et 17,
mag. nat.; 2-14, mag. auct.)

J. Daphniphyllum macropodum Miq. (p, 32) Planta tabulaä ex specimine
récente depicta. — 1 Flos ^ a facie visus. — 2 Idem a latere visus. — 3 Stamen a
facie visum. — 4 Stamen a dorso visum. — 5 Flos Ç , — 6, 7 Eorum verticalis
sectiones. — 8 Ovula pendula. — 9 ladem in loculo. — 10 Fructus ; exocarpium
amortum. (1-9, mag. auct.; 10, mag. nat.)

K. Daphniphyllum (jlaucescens Blume, (p. 33). — 1 Flos J. — 2 Stamen a
facie visum. — 3 Stamen a dorso visum. — 4 Flos Ç . — 5 Fructus, exocarpium
amortum. — 6 Verticalis fructus sectio, albumen videtur. — 7 Verticalis
albuminis sectio, cum apice embryone. — 8 Embryo. (1-4, et 8, mag, auct.;
5-7, mag. nat.)

L. Daphniphyllum himalayense Muell. Arg. (p. 34). — 1 Fragmentum
folii. — 2 Fragmentum ramuli, stipula3 videntur. — 4 Flos J. — 3 Idem a dorso
visus. — 5 Stamina ; alium a fticie, alium a dorso, alium a latere visum. — 6
Glandula et ejus horizontalis sectio. (1, mag. nat.; 2-6, mag. auct.)

Hayata, Revisio Euphorbi. Bux. Japonic

Jour. Sei. Coll. Vol. XX., Art. 3. PI. II.
J. /Vif^ ■

w''-^

■^

B. Uayatadd

d- hap rhf Tokio /VwrfiiMr ۥ

B. HAYATA.

REVISIO EUPHORBIACEARUM ET BUXACEARUM JAPONICAROM.

TABULA III.

Explicatio tabulae III.

Plantfe tabulée ex speciminibus exsiccatis (lepict?e.

A. Bridelia fomentosa Blume, (p. 30). — 1 Flos ^. — 2 Idem, pars amorta.
— 3 Stamen alternm a facie, alteram a dorso visum. — 4 Flos ^ , pistillum
amortum, disci duplicati videntm'. — 5 Verticalis ejus Sectio. — 6 Flos Ç ,
pistillo majori. — 7 Fructus. — 8 Semen alternm a dorso, alterum a facie
visum. — 9 Horizontalis seminis Sectio. — 10 Embryo. (1-6, et 8-10, mag.
auct.; 7, mag. nat.)

B. Croton Cumingii Muell. Arg. (p. 36). — 1 Fragraentum folii. — 2
Facies subtus cum lepidotis. — 3 Alabastrum. — 4 Idem, sepala amorta, petala
videntur. — 5 Idem, sepala et petala amorta, stamina videntur. — 6 Petalum.
— 7 Stamen raaturum, alterum a facie, alterum a dorso visum. — 8 Flos Ç .
— 9 Ejus Verticalis sectio. — 10 Petalum. — 11 Lepidotus ovarii. — 12 Fructus.
— 13 Alter coccus, alterum semen. (1, 12, et 13, mag. nat; 2-11, mag. auct.)

C. Croton Tiglium L. (p. 36). — 1 Fragmentum folii. — 2 Pilus folii. — 3
Alabastrum ^, cum bracteolis. — 4 Flos ^ apertus. — 5 Idem, sepala et petala
amorta, glandulte videntur. — 6 Stamen alterum a facie, alterum a dorso visum.
— 7 Flos Ç , cum bracteolis. — 8 Pilus ovarii. — 9 Fructus. — 10 Semen. — 11
Ejus horizontalis sectio, albumen videtur. — 12 Verticalis albuminis sectio,
embryo videtur. (1, 9, 10, et 11, mag. nat.; 2-8, et 12, mag. auct.)

D. Mercurialis lelocarpa Sieb, et Zucc. (p. 38). — 1 Fragmentum folii.
— 2 Flos $ apertus. — 3 Stamina juniora ; alterum a facie, altemm a dorso
visum. — 4 Stamina juniora. — 5 Stamina matura ; alternm a facie, alterum
a dorso visum. — 6 Fragmentum inflorescentife, cum flore Ç maturo et
alabastro J. — 7 Flos ^ inmaturus. — 8 Fructus. — 9 Verticals cocci sectio,
semen et embryo videntur. — 10 Semen. — 11 Embryo. (1, mag. nat.; 2-11,
mag. auct.)

E. Mallohis moluccanus Muell. Arg, (p. 40). — 1 Fragmentum folii.
mag. nat. — 2 Facies subtus cum pilis et glandulis. — 3 Glandula, validissime
aucta. — 4 Apex petioli cum multiglandulis. — 5 Stamina : aliud a facie, aliud
a dorso a dorso, aliud a latere, aliud supra visum. — 6 Fragmentum infiores-
centite Ç . — 7 Flos Ç , pistillum amortum, discus videtur. — 8 Verticalis floris
Ç sectio. — 9 Fructus. — 10 Columella cum semine. — 11 Embryo. (1, mag.
nat.; 2-11, mag. auct.)

F. Mallotus Playfairii Hemsl. (p. 41). — 1 Fragmentum folii. mag. nat.
— 2 Pagina folii supra.— 3 Indumentum paginée subtus visum. — 4 Glandula
paginée. — 5 Floris ^ verticalis sectio. — 6 Stamina ; alterum a facie, alterum
a dorso visum. (2-6, mag. auct.)

G. Mallotus répandus Muell. Arg. (p. 42). — 1 Fragmentum folii.
mag. nat. — 2 Pagina folii, subtus visa, pili et glandulte videntur. — 3 glandula
validissime aucta. — 4 Alabastrum %. — 5 Horizontalis ejus sectio. — 6 Stamen
alterum a facie, alterum a dorso visum. — 7 Flos ^ . — 8 Verticalis ejus sectio
(2-3, mig. auct.)

H. Mallotus pMlipjnnensis Muell. Arg. (p, 43). — 1 Indumentum folii
subtus visum, pili et glandulte videntur. — 2 Pilus validissime auctus. — 3 Glandula
validissime aucta. — 4 Flos J. — 5 Stamen alterum a facie, alterum a dorso
visum. — 7 Flos Ç, stigmata a facie visa. — 6 Flos Ç, stigmata a latere
visa. — 8 Fructus. mag. nat. — 9 Horizontalis fructus deliiscentis sectio. (1-7,
mag. auct.)

I. 31allotus Japonimis Muell. Arg. (p. 44). — 1, V Fragmenta foliorum.
— 2 Pagina subtus visa, pili et glandulae videntur. — 3 Glandula validissime
aucta. — 4 Flos J. — 5 Stamen alterum a facie, alterum a dorso, visum. — 6
Fragmentum inflorescentiye ^. — 6' Flos ^. — 7 Fructus alter a facie, alter
a latere visus. — 8 Horizontalis fructus deliiscentis sectio. — 9 Verticalis al-
buminis sectio, embryo videtur. — 10 Embryo. (1, 4, 6, et 7, mag. nat.;
2, 3, 5, 9, et 10, mag. auct.)

J. Mallot7is cochincMnensis Lour. (p. 45). — 1 Fragmentum folii. — 2
Pagina folii subtus visa, pili et glandulfe videntur. — 3 Glandula validissime
aucta. — 4 Fragmentum folii basi supra biglandulis. — 5 Flos ^ verticaliter
pectus. — 6 Stamina : aliud a dorso visum, cetera a facie visa. — 7 Fructus.
mag. nat. — 8 Embryo. (I, 4, ot 7, mag. nat.; 2, 3, 5, 6, et 8, mag. auct.)

Ha3'aüv, Revisio Euphorbi. Bux. Japonic.

Jour. Sei. Coll. Vol. XX., Art. 3. PI. III.

B. Hayata del,

Imp fhr Tokto Prtninui l

n. flATATA.

REVISIO EUPHORBIACEARUM ET BUXACEARUM JAPONICARÜM.

TABULA IV.

Explicatio tabulae IV.

Plantœ tabulée ex speciininibus exsiccatis depict;o.

A. Alchornea trewioides Muell. Arg. (p. 47). — 1 Fraginentum folii.
mag. nat. — 2 Fragmentnm «picas cum lloribus ^ et %. — 3 Alabastriim ^ a
facie visum. — 4 Flos % apertus. — .5 Idem a tergo visus, sepala amorta,
filamenta connata videutur. — 6 Stamina ; aliud a facie, aliud a latere, aliud a
dorso, visum. — 7 Pistillum, ovario verticaliter secto, ovula videntur. — 8
Sepala. — 9 Horizontalis ovarii Sectio. (2-8, mag. auct.)

B. Cleidion ulmifolium Muell. Arg. (p. 48). — 1 Segnjentum folii.
mag. nat. — 2 Segmentum spicke ^. — 3 Flos J. — 4, 5 Stamina. (2-4, mag.
auct.)

C. Macaranga Tanarius Muell. Arg. (p. 49). — 1 Segmentum folii.
— 2 Pagina subtus visa, glandulte videntur. — 3 Segmentum paniculae ^. — 4
Flores J, cum bractea. — 4' Flores J; alius verticaliter sectus. — 5 Stamina.
— 6 Segmentum racemi ^. — 7 Flos ^. (1, 3, et 6, mag. nat.; 2, 4', 5, et
7, mag. auct.)

D. Acalypha austrulis L. (p. 50). — 1 Pistillum cum spica J. — 2 Frag-
mentum spicse $. — 3 Flos J a facie visus. — 4 Fl. ^ a tergo visus. — 5 Flos
J, sepala amorta. — 6 Stamina cum vermiformibus antheris. — 7 Pistillum
Junius. — 8 Sepalum. — 9 Horizontalis ovarii sectio. — 10 Capsula. — 11 Semen,
caruncula videntur. — 12 Albumen. — 13 Embryo. (1-13, mag. auct.)

E. Excœcaria j'aponica Muell. Aug (p. 59). — 1 Flores $ cum bractea.
— 2, 3 Flores ^. — 4 Stamen alteram a facie, alterura a dorso visum. — 5
Flores Ç. — 6 Fractus. mag. nat. — 7 Horizontalis fructus dehiscentis sectio.
— 8 Albumen. — 9 Embrjo. — 10 Diagramma fl. ^. — 11 Diagramma fl. ^.
(1-5, mag. auct.)

F. Eupliorhia LaiJiyris L. (p. 64) Plant», ex specimine récente depicta.
— 1 Involucrum cum floribus $ et Ç . — 2 Verticalis ejus sectio. — 3, 4 Stamioa.
— 5 Stamen a latere visum. — 6 Glandula. — 7 Involucri lobus. — 8 Fructus.
— 9 Coccus. — 10 Carimcula subtus visa. — 11 Semen ciun carnncufe stipiti ;
caruncula amorta est, — 12 Embryo. (1-12, mag, auct.)

G. Euphorbia Helioscopia L. (p. Q%) Planta ex specimine récente dépista.
— 1 Involucrum cum floribus ^ et ^. — 2 Idem verticaliter sectum. — 3
Involucrum fissum et explicatum. — 4 Bracteola floris ^. — 5 Capsula. — 6
Coccus. — 7 Semina ; alterum a latere, alterum a dorso visum. — 8 Albumen.
— 9 Embryo. (1-9, mag. auct.)

H. Euphorbia adnochlora Morr. et Decne. (p. 67) Planta ex specimine
récente depicta. — 1 Involucrum cum floribus ^ et Ç. — 2 Idem, verticaliter
sectum. — 3 Involucrum fissum et explicatum. — 4 Ejus fragmentum. — 5 Ejus
glandula et lobi. — 6 Bracteola floris ^. — 7 Stamina ; aliud a facie, aliud a
dorso, aliud a latere visum. — 8 Stamen, anthera déhiscente. — 9 Capsula
dehiscens. — 10 Coccus. — 14 Semen a facie visum. — 12 Semen a tergo visum.
— 13 Verticalis abluminis sectio, embryo videtur. (1-8, et 12-14, mag. auct.;
8 et 9, mag. nat.)

I. Euphorbia ebracteolata sp. nov. (p. 71). — 1 Involucrum cum floribus
^ et ^. — 2 Involucrum fissum et explicatum. — 3 Fragmentum involucri extus
visum. — 4 Flos ^ . — 5 Fructus. mag. nat. (1-4, mag. auct.)

J. Euphorbia pehinensis Rupr. (p. 68). — 1 Involucrum cum floribus $ et
^ . — 2 Involucrum fissura et exj>licatum. — 3 Ejus fragmentum cum glandula et
bracteola. — 4 Glandula extus visa. — 4' Glandula a facie visa. — 6 Stamina.
— 7 Pistillum a facie visum. — 8 Fructus. — 9 Semen. (1-9, mag. auct.)

Hayata, Revisio Euphorbi. Bux. Japonic
A

Jour. Soi. Coll. Vol. XX., Arl. 3. PI. IV.

B. Uayata del.

I.m. a- Imp. Tit* Tokio frintintf Cn.

B. BATATA.

REVISIO EÜPHORBIACEARUM ET BDXACEARÜM JAPONICARÜM.

TABUJLA V.

Explicatio tabulœ V.

PlantsB talmlœ ex speciminibus exsiccatis depictaB.

A. Euphorbia Sparmanni Boiss. (p. 73). — 1 Fragmentum rami, mag.
nat. — 2 Folium, mag. nat. — 3 Cyme. — 4, 5 Involucri. — 6 Involucruin fissuni
et explicatum. — 7 Bracteola. — 8 Stamina. — 9 Capsula. — 10 Semina ; a, a
latere, b, a facie, c, supra visum. — 11 Embryo. (3-11, mag. auct.)

B. Euphorbia pilulifera L. (p. 74). — Fragmentum rami. mag. nat. — 2
Folium, mag. nat. — 3 Cyme, validius auctus, — 4 Involucri cum floribus ^ .
— 5 Involucrum fissum et explicatum. — 6 Glandula cum appendiculo. — 7
Bracteola. — 8 Capsula. — 9 Semen. — 10 Embryo. (3-10, mag. auct.)

C. Euphorbia hypericîfolia L. (p. 75). — 1 Fragmentum rami. mag.
nat. — 2 Folium, mag. nat. — 3 Cyme. — 4 Idem validius auctus. — 5 Involu-
crum fissus et explicatum a esteriore visum. — 6 Idem a interiore visum. —
7 Bracteola. — 8 Capsula. — 9 Semen. — 11 Semen albume. — 12 Embryo. (3-12,
mag. auct.)

D. Euphorbia, serrulata Keinw. (p. 76). — 1 Fragmentum rami. — 2
Cyme. — 3 Involucrum cum flori Ç. — 4 Involucrum fissum et explicatum.
— 5 Bracteola. — 6 Capsula, — 7 Semen. — 8 Embryo. (2-8, mag. auct.)

E. Euphorbia thymifolia L. (p. 77.) — 1 Fragmentum rami mag. nat.
— 2 Idem cum floribus, mag. nat. — 3 Fragmentum ramuli. — 4 Involucri. —
5 Idem fiissum et explicatum. — 6 Fragmentum involucri. — 7 Bracteol?e, — 8
Capsula. — 9 Semen. — 10 Embryo. (3-10, mag. auct.)

F. Euphorbia maculata L. (p. 78). — 1 Fragmentum rami mag. nat.
— 2 Fragmentum ramuli. — 3 Involucrum. — 4 Involucrum fissum et explicatum.
— 5 Fragmentum involucri. — 6 Verticalis glandules et appendiculi sectio. —
7 BracteoUe, — 8 Flos ^ maturus. — 9 Capsula. — 10 Semen. — 11 Embryo.
(2-11, mag. auct.)

G. Euphorbia humifuso. Willd, (p. 78). — 1 Fragmentum rami. mag.
nat. — 2 Fragmentum ramuli. — 3 Involucrum cum floris Ç . — 4 Involucrum
fiasimi et explicatum. — .3 Glandula et appendiculi. — 6 Capsula. — 7 Semen.
— 8 Ejus horlzontalis sectio. — 9 Albumen. — 10 Embryo. (2-10, mag. auct.)

H. Euphorbia microphylla Heyne, (p. 79). — 1 Fragmentum rami,
mag. nat. — 2 Fragmentum ramuli.- — 3 Involucrum Junius. — 4 Involucrum
fissum et explicatum. — .5 Fragmentum ejus cum lobis et bracteolis. — 6
Stamina. — 7 Involucrum maturum. — 8 Capsula. — 9 Semen. — 10 Ejus hori-
zontalis sectio.— 11 Embryo. (2-11, mag. auct.)

llayata, Revisio Euphorbi. Bux. Japonic
A

Jour. Sei. Coll. Vol. XX., Art. 3. PI. V.

B. HAYATA.

REVISIO EUPHORBIACEARUM ET BOXACEARÜM JAPONICARUM.

TABULA VI.

Explicatio tabulae VI.

PlantîB tabulae ex speciminibus exsiccatis depictte.

A. Euphorbia togahusensis sp. nov. (p. 69). — 1 Involucrum cum floribns
g et J. — 2 Idem a latere visum. — 3 Involucrum fissum et explicatum. — 4
Bracteola. — 5 Pistillum a facie visum, styli amorti. — 6 Fructus. mag. nat.
— 7 Fructus. — 8 Semen a latere visum. — 9 Semen a tergo visum. (1-5,
et 7-9 mag. auct.)

B. EupJiorhia Sieboldiana Morr. et Decne. (p. 71). Planta ex specimine
récente depicta. — 1 Involucrum cum floribus ^ et Ç. — 2 Involucrum
verticaliter sectum. — 3 Involucrum fissum et explicatum. — 4 Fragmentum
involucri. — 5 Stamina dehiscentia ; alterum a dorso, alteram a facie visum.
— 6 Stamina ; alius a facie, alius a latere, alius a dorso, visum. — 7 Fructus.
— 8 Semen alaterum a facie, alterum a dorso visum. (1-8 mag. auct.)

C. Buxus sempervirens L. var. Japonica Marino, (p. 82) Planta ex
specimine récente depicta. — 1 Fragmentum rami. mag. nat. — 2 Axillae foliurum
oppositorum a facie visas.— 3 Idem a tergo visas. — 4 Fasciculus J. — 5 Flos
^. — 6 Idem verticaliter sectus. — 7 Stamina ; aliud a dorso, aliud a facie,
aliud a latere, visum. — 8 Stamen anthera déhiscente. — 9 Kudimentum ovarii
alterum a facie, alterum a dorso, visum. — 10 Flos Ç. — 11 Pistillum validius
auctum. — 12 Horizontalis capsulas dehiscentis sectio. — 13 Semen. — 14 Embryo.
(2-11, mag. auct.)

D. Buxus sempervirens L. var. microplmjlla Bl. (p.83) — 1 Fragmentum.
rami. mag. nat. — 2 Axillae foliorum oppositorum a iacie viste. — 3 Idem a
dorso visfe. — 4 Flos ^. — 5 Flos ^, sepala amorta, rudimentum ovarii et stamina
videntur (2-4 mag. auct.)

E. Buxus Wallichiana Baill. (p. 84). — 1 Fragmentum rami. mag.
nat. — 2 Flos ^. — 3 Idem verticaliter sectus. — 4 Rudimentum ovarii. (2-6,
mag. auct.)

F. Pachy Sandra termirialis Sieb, et Zucc. (p. 81). — 1 Folium, mag. nat.
— 2 Spica cum floribus Ç et ^. mag. nat. — 3 Flos ^. — 4 Idem verticaliter
sectus, stamina amorta, rudimentum ovarii videntur. — 5 Stamina. — 6, 7
Sepala. — 8 Rudimentum ovarii. — 9 Flos ^ cum bracteolis. — 10 Pistillum.—
11 Horizontalis ovarii sectio. — 12 Fructus. mag. nat. — 13 Semen. — 14 Ejus
horizontalis sectio. — 15 Embryo. (3-11, et 13-15, mag, auct.)

llMvalii, Uevisiu Eupliurbi. Bux. Japonic
A

Jour. Sei. Coll. Vol. XX., Art. 3. PI. VI.

JOURNAL OF THE COLLEGE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN,

VOL. XX., ARTICLE 4.

The Gephyrea of Japan.

By
Iwaji Ikeda, RigahisM.

With 4 plates

Only four species of Gephyreans, so far as I am aware, have
hitherto been recorded from Japan, viz., Phymosoma japonicum
Grube, Dendroséoma blandum Sel. & De Man, Phascolion
lucifugax Sel. & De Man and Echiurus unicinctus v. Dräsche.
From my studies of the group during several years, 34 more
species have become known to me, thus giving in all 38 species
to the Gephyrean fauna of Japan as at present known. But of
the four previously known species mentioned above, Phascolion
lucifugax has, till the present day, remained wholly missing in
my collection.

In this paper I propose to give short descriptions of, and
comments upon, all the 37 species directly known to me.

Of the said 37 species, 26 belong to the Sipunculoids under
7 genera and 11 to the Echiuroids under 3 genera. Again, 24
species of them seem to be new ; the remaining 13 species I
have identified with those which other investigators have described
from different parts of the world. It should be mentioned that

Z ART. 4. J. IKEDA :

this identification could in several cases be made only with a
greater or less degree of doubt as to its correctness. This is due,
I think, mainly to the excessive conciseness of terms in which
most of the old species (especially in the Echiuroidea) were
described.

I have not deemed it necessary, for the present at least, to
create a new genus for the reception of any of the new species,
although the thought frequently suggested itself to my mind that
some of the forms, e.g. Thalassema tœnioides n. sp., might well
be made the type of a distinct genus.

At the end of the paper will be given a key to all the
species treated of, which may help students of the same field in
the task of identification.

For the purpose of obtaining my material I have made
frequent visits to the Misaki Marine Laboratory. Further I may
mention that on a collecting trip to the Island of Amani-Oshima
(Province of Satsuma) and the Islands of Riukiu in 1901, and
on others to the Province of Aomori in 1900 and to Gogoshima
(Province of lyo) in 1903, special attention was paid to the
Gephyrean fauna.

To a number of friends, whose names will be mentioned in
the proper places, I am indebted for gifts of several valuable
specimens, some of which I should otherwise have been unable
to obtain.

THE GEPHYEEA OF JAPAN.

SIPUNCULOIDEA.

Genus PHASCOLOSOMA.
PHASCOLOSOMA NIGRUM, n. sp.

(Figs. 1, 25-27).

This new species (Fig, 1) is of a moderate size, the total
length being about 10 cm. The introvert is nearly 2? times as
long as the body proper, than which it is slightly narrower. The
posterior end of the body is always conically pointed. The body
in life is coiled more or less in a spiral-like manner. The
greater part of the body is dark gray or bluish-black in color,
except the tentacular basis of the introvert where it is light
brown. The tentacles, of a clear violet color, are filamentous and
very numerous, arranged in 30-32 longitudinal rows. When viewed
from above, the oral ends of every set of two tentacular rows are
joined by a U-shaped ridge, the mouth being thus surrounded
by 15 or 16 of such ridges with radial grooves between. Small
skin-bodies are very abundant and are distributed over the entire
surface of the introvert and of the body proper. They are largest
on the introvert-basis and on the posterior end of the body, in
both of which parts they appear as low papillie of 0.1 mm. diameter
and 0.05 mm. height (Fig. 25). Each papilla is an elliptical and
slightly elevated granular disc of a brownish black color ; in its

4 AtlT. 4. — I. IKE DA :

centre there is a yellow and radially striated area bearing the
opening of the subdermal gland. On the introvert these papillae
become smaller and smaller anteriorly, but without undergoing a
noticeable change in height (Fig. 26). They show no striated
cental area, the entire surface being finely granulated. Neither
hook nor spine is present on the introvert.

The inner surface of the body-wall is remarkably smooth
and shiny. The longitudinal muscles in the same are continuous.
The retractor muscles are only two in number, forming a ventral
pair (Fig. 27, vm). They originate at the beginning of the
posterior third of the body proper, close to the nerve-cord {n).
In the anterior parts they are fused together into one flat band,
running in contact with the long œsophagus {oe) on the ventral
side. Along the dorsal side of the latter runs the dorsal vessel,
the posterior parts of which are beset with numerous but short
contractile villi {cv). The long intestinal convolution {ic) consists
of about 20-25 double spirals, closely twisted around the spindle-
muscle which posteriorly does not pass out of the spirals for
fixation to the body-wall. Only one, but very long, fixing muscle
{fm) extends between the posterior end of the œsophagus and a
point in the body-wall situated a short distance in front of the
origin of the left retractor muscle on the left hand side. The
rectal intestine {re) carries at its beginning a small blind sac,
the rectal gland (rg). The rectum is fixed to the body-wall
by two broad wing-muscles (wm). The segmental organs {so)
are of a conspicuous length ; they are free from the body-wall
except at the external apertures which are situated a certain
distance in front of the anus. They are colored with a deep reddish
brown pigment. The anus is indicated on the outside by a
prominent papilla. The sexual organs are, as is usually the case,

THE GEPHYREA OF JAPAN. 5

two slender wavy bauds {sx), developed across the points of origin
of the retractor muscles (vm).

The present species agrees in some points with Fhascohsoma
semperi Sel. & De Man, as described by Selenka (28), but it
can be readily distinguished from that species by the characteristic
dermal papillœ. The same distinction may be pointed out as
obtaining between the present species and Fhascolosoma fimbriatum
Sluiïer (37).

Habitat : The species is one of tlie commonest Sipunculoids
met with in the vicinity of the Misaki Marine Laboratory ; it is
also common along the coasts of the Tokyo Bay and of the
Inland Sea. It lives in shallow muddy bottom, especially in small
inlets into which fresh-water finds egress.

FHASCOLOSOMA JAPONICUM, n. sp.

(Figs. 2, 28 & 29).

The length of the body proper is about 5 cm. on an average,
the introvert being about as long as two-thirds of the body proper
(Fig. 2) ; the two regions pass over into each other without a
distinct demarkation. The maximum width of the body proper
is about 7 mm. The body-wall is thin and of a brownish yellow
color, of a deeper brown in male worms than in female. To the
naked eye the skin-surface appears quite smooth, but under certain
magnification there are seen numerous skin-bodies of elliptical
shape, thickly distributed (Fig. 28). These bodies are flat, not
covered with chitinous plates, and are somewhat transparent ; a

6 ART. 4. 1. IKEDA :

group of subdermal glands, situated beneath the skin-body, are
seen to open to the exterior by a small common pore (see p) at
tlie centre. The papillae are of nearly the same size throughout
the whole surface of the body, viz., 0.09-0.06 mm. in diameter.
No hook or spine is present. The tentacles are filamentous and
tolerably numerous, and are arranged in about 12-14 longitudinal
rows around the oral disc, each row consisting of about five
tentacles. The rows, in sets of two, are united at the oral end by
U-shaped elevations as in the preceding species. Two dark brown
eye-spots are present on the ganglion or brain {g, Fig. 29).

The longitudinal muscles of the body-wall are uniformly
distributed. The inner surface of the wall is smooth and shiny
like a pearl. There are present two pairs of retractor muscles,
of which the ventral pair [vm) originate at about the middle of
the body proper close to the nerve-cord {n), while the dorsal pair
(dm) arise from a region a short distance behind the anus (a).
The œsophagus {oe) is fixed by thin membranaceous muscles and
is accompanied dorsally by the dorsal vessel {dv), which pos-
teriorly reaches down to the beginning of the intestinal convolution
(ic). This vessel develops no contractile villi. The intestinal
convolution is made up of 17-23 double spirals and is traversed
throughout by the spindle-muscle which at the posterior end is
not fixed to the body-wall. There exist in all three fixing muscles,
one on the right and two on the left side of the nerve-cord (see
fm, Fig. 29). One on each side is attached to the beginning of
the rectum, while the unpaired and asymmetrical one on the left
side is fixed to the posterior terminal portion of the oesophagus.
A large blind rectal gland {rg) is attached to the middle jiart of
the rectum. The anterior portion of the rectum is fixed by well
developed wing-muscles {tvm). The two segmental organs (so),

THE GEPHYREA OF JAPAN.

the external apertures of which lie slightly in front of the anus,
are toterably long, corresponding in length to nearly one half of
the body proper. They are of a deep brownish color and are
entirely free from the body-wall except at the anterior end. I'he
sexual organs (sx) are developed as two narrow transverse bands
placed across the bases of the ventral retractor muscles.

Habitat : This species is common along the coasts of the
Tokyo and Sagami Bays. It occurs very abundantly in shallow
waters near the Misaki Marine Laboratory. The animals live in
clean sand, mostly in narrow fissures of rocks. No specimens
have thus far been obtained at any other locality than those
mentioned above.

PHASCOLOSOMA MISAKIANUM, n. sp.

(Figs. 3, 30-33).

This species is small and slender. The total length of the
animal with the introvert wholly protruded, measures about 5 cm.,
the thickest part of the body proper rarely exceeding 2 mm. in
width. The introvert corresponds in length to about four-fifths
of the total length. The skin is smooth, thin and somewhat
transparent ; it is of a yellowish pink color (Fig. 3 ). Under
high magnification there can be detected numberless small papillary
bodies distributed all over the body-surface. In the posterior
region of the body, Avhere these bodies are most closely aggregated,
they measure 0.03 ??^w. in both height and diameter. Here they
take the shape of a small rounded cone, on the summit of which

8 ART. 4. — I. IKEDA !

numerous small chitinous granules of a yellow color are so arranged
as to form a ring surrounding the external aperture of the sub-
dermal gland (see Figs. 30 a and h). In the middle region of
the body proper, the papillae are quite flat, elliptical in form and
much larger than in the posterior region. The chitinous granules
of such papillae are much smaller than those referred to above.
In the introvert, the papillae are very slightly elevated (nearly
0.015 mm. in height) and so small that they might appropriately
be called tubular with " Mündungskörper " (Fig. 31). The
introvert carries about 60 ring-rows of hooks of a small size.
The hooks in a way alternate in position with the tubular
papillae, there being about five of the former between every two
of the latter. The hooks are 0.025 mm. high, light yellow in
color, and are provided with a sharp and strongly curved apical
tooth and 4-5 short accessory teeth (Fig. 32). The accessory
teeth gradually decrease in number in the hooks forming the
more posterior rings finally to disappear altogether. The tentacles
are filamentous and nearly always 8 in number, surrounding the
oral aperture. The eye-spots are seen as two small, brown specks
lying on the lateral corners of the ganglion.

The longitudinal musculature of the body-wall is continuous.
Of the two pairs of the retractor muscles, which are all slender,
the ventral pair {ym, Fig. 33) originate at the middle of the
body proper, while the dorsal {dm) do so far more anteriorly.
The anus lies at a position nearly midway between the origins of
the two pairs of muscles. A feature peculiar to this species is
presented by the sexual organs (so;). These are narrow band-like
structures situated at a considerable distance behind the origin
of the ventral rétracter muscles. The dorsal vessel {dv) is very
short, extending posteriorly for only a few millimeters beyond

THE GEPHYREA OF .TAPAX. 9

the ganglion, and is not provided with contractile villi. The
intestinal convolution (ic) is formed of 6-10 double spirals which
are more or less irregularly twisted around the fine spindle-
muscle {sm). The latter is fixed to the body-wall at both its
anterior and posterior ends. Its anterior end is at a considerable
distance anteriorly from the anus. There is only one fixing
muscle (fm), connecting the hind end of the œsophagus (near the
commencement of the convolution) with the body-wall on the left
side of the nerve-cord and a short distance in front of the orio-in
of the ventral retractor muscles. No blind diverticulum is present
on the rectum. The latter is supplied with wing-muscles {w7n).
The segmental organs (so) exhibit a peculiar structure unknown
in other Sipunculids. Each of them consists of two long lobes
equal in size and shape, the one being directed anteriorly and
the other posteriorly. The two lobes are directly continuous with
each other. At their boundary lies the external aperture, the inner
aperture or the funnel being found at the basis of the anterior lobe.
The organs have a reddish yellow color. Their wall is very thin.
Habitat : The animals are discovered under sea-weeds cover-
ing a tufaceous sandstone between the tide-marks. They have
hitherto been collected only on the shores near the Misaki
Marine Laboratoi'v, where however thev are rare.

PHASCOLOSOMA OKINOSEANUM, n. sp.

(Figs. 4, 34-38).

In March, 1901, a single specimen of the species here dis-
cribed as new under the above name was dredged at a depth of

10 ART. 4. 1. TKEDA :

400 hir(f' (near the Okinosé bank) in the Sagami Sea. The
following description is liased on that specimen which is preserved
in formalin.

The worm is of a moderate size and of a brownish yellow
color. The body proper is about 4.5 cm. in length and 7 7nm. in
maximum width, it being posteriorly narrowed into a small and
slender tail-process (Fig. 4). The introvert, which had partly
retracted, is apparently about as long as the body proper, but
much narrower, averaging 3 mm. in width. The skin-surface
appears smooth to the naked eye, but under certain magnification
numberless prominent papillœ may be discovered. They are largest
and tallest in the posterior region of the body, reaching up to
0.08 mm. in height and 0.03 mm. in thickness (Fig. 34). They are
of a brownish yellow color and are composed of very small, closely
aggregated, chitinous granules, 'i'he papillse become gradually
shorter and shorter towards the base of the introvert, in which
parts they are represented by flat bodies separated from one another
by irregular narrow grooves (Fig. 35). The latter form a sort of
mesh-work, in the meshes of which is contained a papillary body.
Each body is formed of a small granular ring of a brownish yellow
color and of a colorless transparent central area containing the
small external aperture of the subdermal gland. The granular
ring measures about 0.03 mm. in diameter. Towards the middle
and anterior parts of the introvert, the papillae again become
nearly as large and long as in the posterior parts of the body.
In the regions of the introvert mentioned, the chitinous layer of
the skin shows fine striations intersecting at right angles. No
hook or spine, nor eye-spot, is present. Peculiar structures are

* 1 liiro= about 5 feet.

THE GEPHYREA OF JAPAN. 11

found OU the skin-surface of the body proper ; we have here to
do with tolerably large, elongated sacs with a thin, colorless and
transparent wall. Fig. 36 represents one of the structures greatly
magnified. The organ is hollow ; its cavity communicates with
the body-cavity by a small aperture which is situated just at the
basis of the organ and is perfectly round in contour (Fig. 37).
As Fig. 37 shows, the wall contains a number of longitudinal
muscle- fibers [mf) running parallel to one another. A closer
microscopical examination shows that most of these muscle-fibers
are proximally continuous with the circular muscles of the body-
wall. From the structure and the relation of the organs to the
body-cavity, it seems very probable that they serve as the respira-
tory apparatus. They are found here and there all over the
surface of the body proper, but are most numerous in the posterior
region.

The longitudinal muscles are everywhere continuous. The
inner surface of the body- wall is of a shiny appearance. There
exist two pairs of the retractor muscles, which are fairly long
and very slender. The ventral pair [vm, Fig. 38) spring from
the posterior border of the anterior one-third of the body proper,
while the dorsal pair arise about 7 cjii. in front of that border.
The sexual organs (sx) are developed as two narrow wavy bands
attached to the base of the ventral retractor muscles (vm). The
intestinal convolution {ic) is long and conspicuous, consisting of
about 20 double spirals closely twisted around the spindle-muscle
{s)/i) which posteriorly is not fixed to the body-wall. Three fixing
muscles {fm) are present. One of them originates inside the base
of the left ventral muscle and ends at the posterior free part of
the oesophagus. The other two are rooted each on one side of
the nerve-cord and are attached to the beginning of the rectal

12 ART. 4. — I. IKEDA :

intestine. Just opposite these two points of attachment, there is
found a small roundish rectal diverticulum. The segmental organs
(so) are relatively short and small, measuring about 10 mm. in
length. They are transparent and almost colorless tubes, unac-
companied by a mesentery. Their external oj^enings are situated
nearly on the same level as the anus (a). The presence of the
dorsal vessel could not be ascertained. The tentacles, surrounding
the mouth, are filamentous and very numerous.

PHASCOLOSOMA OWSTONI, n. sp.

(Figs. 5, 39-44).

The single specimen (Fig. 5), on which this species is based,
was dredged in 1899 by Mr. Owston of Yokohama from a depth
of 180 hiro m the Uraga Channel. It has been preserved in
formalin.

The total length of the animal is 8 cm.., one-fourth of which
appertains to the introvert. The posterior half of the body
proper is wider than the anterior and measure 8-10 mm. across.
The entire introvert is nearly of the same thickness as the
anterior and narrower part of the body proper. The ground
color of the skin is a light blue, suffused with pink ; the middle
portion of the introvert is tinted a deep l)rown, and the posterior
portion of the body-proper a light brown. The body-wall is in
the body proper relatively thin, smooth on the external surface
and more or less transparent, while in the introvert it is thick,
opaque and has a closely wrinkled surface. All over the sur-

THE GEPHYREA OF JAPAN. 13

face of the body there are sparsely distributed papilke. Iii the
posterior region of the body proper, these are very large and
long, measuring 0.125 iiiiii. in height and 0.054 mm. in thick-
ness. Each papilla is covered with a number of small chitinous
granules of a reddish brown color ; the apex is bored through
by a small pore leading into the subdermal gland (Fig. 39). To-
wards the introvert-basis, the papilke become gradually shorter
and shorter, until in that region they are very short and assume
the shape of a round disc of a yellowish color (Fig. 40). Fine
radial striation is found in the central chitinous plate containing
the aperture of the subdermal gland. The more anteriorly
situated papilke of the introvert are once again more prominent
and those close to the hooked region have the form of a moder-
ately tall (0.04-0.07 mm.), round-pointed process covered with
small chitinous granules of a yellowish color (Fig. 41). The
hooks (Fig. 42), without being arranged in ring-rows, are ex-
tensively distributed behind the tentacular region of the introvert.
They are 0.22-0.13 mm. high, slender and almost straight ; and
of a brownish yellow color.

The longitudinal muscles are quite continuous and the inner
surface of the body-wall is shiny like a pearl. The retractor
muscles {dm., vm., Fig. 43) are three in number, the right ven-
tral retractor being absent. The left ventral retractor {v7n)
arises at the beginning of the middle third of the body proper,
close to the nerve-cord, while the two dorsal retractors {dm)
originate immediately behind the anus {a). The two muscles of
the left side are entirely coalesced into one band, and anteriorly
are again joined to the dorsal retractor of the right side. In
spite of the absence of the right ventral retractor, the sexual
organs {sx) are symmetrically developed as two slender transverse

14 AKT. 4.- — I. IKEDA I

bands stretcliiûg along the bnsnl line of the left retractor, as
well as in the corresponding position on the right side. Tlie
œsophagus [œ) is connected by a membraneous muscle with the
inner side of the left ventral muscle. The dorsal vessel (dv)
runs along the entire dorsal side of the oesophagus ; it is not
j)rovided with contractile villi. The intestinal convolution [ic)
consists of about 30 double spirals around the spindle-muscle
{sjii). The latter does not posteriorly pass out of the spirals to
insert itself in the body- wall. The small diverticulum attached
to the beginning of the rectum is comparatively short. Two
fine fixing muscles {ß/i, Fig. 43), are attached to the fourth spiral
of the convolution ; they are rooted on the body- wall at points
lying just behind the bases of the dorsal retractor muscles.
The segmental organs (so), of a deep reddish brown color, are
of a moderate size. Their external apertures are situated on the
same level as the anus (a). Tentacles are tolerably numerous
and encircle the mouth in a ring. No eye-spot can be detected
on the ganglion.

A peculiar reticular structure (r^, Fig. 43) is found lining
the inner surface of the body-wall in a zone bounded anteriorly
by a line joining the external apertures of the segmental organs
and posteriorly by a line running midway between the roots of
the dorsal and ventral retractor muscles. It appears as an
irregular, small-meshed network of trabecuhe, forming a layer
over the region indicated. One small portion of the structure
is shown in Fig. 44, as seen in surface view under a low mag-
nifying power. Examined on sections, the trabecuhe are seen to
be composed of very fine connective-tissue fibers in bundles and
of a few muscle-fibers ; further it can be made out that the
reticular layer is joined to the body-wall by numerous short

THE GEPHYREA OF JAPAN. 15

bridges which mostly arise from the nodal points of the reticnlnm.
It scarcely needs to be pointed out that the bridges as well as
the trabeculee are covered by the peritoneum. I am at present
quite at a loss to explain the physiological significance of the
structure. Possibly it is nothing but a certain pathological
product.

Genm PHASCOLION.

PHASCOLION RECTUS, n. sp.

(Figs. 45-49).

This new species is based on a single specimen which was
obtained, September 1903, from a depth of 400 h'lro in the
Sagami Bay by means of the long-line (Dabonawa). It has been
carefully preserved in alcohol.

The worm inhabits the dead shell of a Dentaliuin, the in-
trovert partially protruding out of the Larger aperture. The
body not being cemented to the shell lies free. The entire
length of the worm is about 35 mm., nearly one-third of which
constitutes the partly retracted introvert (Fig. A^)). The body
proper is nearly straight, gently swollen in the middle where it
measures about 2 mm. across, while the introvert is relatively
thick and about 1 myn. broad. The skin is of a grayish yellow

16 ART. 4. T. IKEDA :

color, more deeply yellow in the middle portion of tlie body
proper. The greater part of the body-surface appears nearly
smooth to the naked eye, except on the introvert-basis which is
rough on account of the presence of tall papillie and prominent
chitinous folds. Under the microscope, it is seen that the ap-
parently smooth part of the surface is beset with skin-papillte
which vary much in shape and size in different portions of tlie
body. In the anterior region of the introvert, the papillae are
very small, transparent, oval bodies, measuring 0.025 mm. in
height and 0.014 mm. in basal diameter. From that region
posteriorly, they become somewhat larger and taller, but less
closely arranged, and in the middle parts of the introvert they
measure about 0.028 mm. in height and 0.02 mm. in maximum
breadth (Fig. 46). On the introvert-basis, the papilhe are rather
prominent, showing a club-like shape and measuring about 0.1 mm.
in height and 0.05 mm. in thickness in the broadest part.
They have a thick wall, are yellow in color and are perforated
by a narrow and likewise club-shaped canal which proximally
communicates by a small pore with the cavit}?^ of the subdermal
gland (Fig. 47). The minute external aperture on the apex
of the papilla can scarcely be brought into view. Intermixed
with the papillœ are present large, irregularly shaped thicken-
ings of the chitin (see Fig. 47) ; these appear finely granulated
and are of a deep brown color. In the anal region, the papillae
abruptly become short (0.04 mm. high) and exhibit a conical
form with a basal diameter of 0.046 mm. on an average. The
apex of such a papilla is occupied by a small mamelon-like
body which is pierced through by the papillary canal. Most
of these papilhe are situated amongst closely set, large and
irregular chitinous thickenings of coarsely granular appearance

THE GEPHYREA OF JAPAN. 17

and of a deep brown color. Behind the anal region, the papillae
become gradually smaller until in the middle region of the
body proper they acquire the form of an almost flat body,
measuring 0.02-0.03 mm. across at base (Fig. 48). Structurally
these flattened papillae agree with those of the anal region.
The skin-surface here also is covered with small chitinous
thickenings of various forms and sizes peculiar to this region
(see the above figure). At the posterior end of the body the
papillae are once again of a somewhat conspicuous size : here
they measure on an average 0.05 mm. in height and 0.10 mm.
in diameter. These, in both their external and internal struc-
ture, are quite similar to those found in the middle region of
the introvert. No hook or any spiny structure is present on
the introvert.

The longitudinal musculature forms a continuous sheath.
Only a single and very slender retractor muscle {rm, Fig. 49)
springs with two short roots from the posterior end of the body.
The intestinal convolution consists of about 20 irregularly twist-
ed spirals and is not accompanied by a spindle-muscle. The
intestinal spirals are joined to the body-wall by means of four
fine fixing muscles {fm), two of which are attached to the begin-
ning and the two others near the posterior end of the convolu-
tion. The dorsal vessel is about 15 mm. long ; it is not provided
with contractile villi. The segmental organ {so) is situated im-
mediately behind the anus as a single elongate sac lying on the
right side of the nerve-cord ; it is unaccompanied by a mes-
entery. No eye-spot is found on the ganglion. The tentacles,
about 10 in number, are filamentous.

PhasGolion rectus agrees with Phascolion mancei^s Sel. &
De Man, in wanting the '' Haftpapillen " and the hooks and

18 ART. 4.— I. TKEDA :

also in having only one retractor muscle, but differs from it,
amongst other less important points, in the shape and structure
of the papillœ, in the absence of contractile villi on the dorsal
vessel and in the jiosition of the anus in relation to the introvert.

PHASCOLION ARTIFICIOSUS, n. sp.

(Figs. 6, 50-55).

Two specimens of the new species here described under the
above designation, were collected, July 1903, from a depth of
400 hiro in the Sagami Bay. They were both occupying small
Gastropod shells. One of them was observed by me in the
fresh state and is shown in Fig. 6, enlarged about 10 times.

The aperture of the dead shell inhabited by the worm is
plugged with a conical mass of cemented sand having at the
apex a pore, out of which the introvert of the worm is partly
extruded (Fig. 50). When extracted from the shell, the worm-
body is found to be spirally twisted in accommodation to the
spiral of the shell. The skin is of a bluish gray (in the fresh
state), but bluish black on the closely set papillœ. The body
proper is about 1.5 cm. long and 3.-3.5 mm. thick, while the
introvert is about half as long and somewhat narrower (2 mm.
wide). The papillœ are closely distributed over the entire sur-
face, and vary much in size and shape in different j^arts of the
body. The largest are found in a narrow zone around the
middle of the posterior half of the body proper. Here they
are of a low conical shape and appear like the so-called " Haft-

THE GEPHYREA OF JAPAN. 19

Papillen " (Fig. 51). The dark-colored apical parts of these
papillae are more or less laterally compressed so as to be flattened.
The larger ones measure 0.1-0.2 mm. in basal diameter and
0.8-0.15 mm. in height. In the remaining regions of the body
proper, except at its very anterior end, the papillœ are in general
very much smaller and of a bluntly apexed conical shape. In
the middle region, the papillie commonly do not exceed 0.05 mm.
in basal diameter and in length. In the posterior end they are
somewhat larger, measuring 0.1 mm. in average length, and take
the shape of a pear-shaped sac with a pointed apex (Fig. 53,
PI. III). In the extreme anterior region, the papillae are of a
shape quite different from those of all other regions. Here they
are no longer conical, but are represented by finger-like processes
of about 0.07 7nm. height and 0.04 mm. thickness (Fig. 54).
The chitinous wall of such a papilla is, especially on the apex,
thick and of a yellowish brown color. On the introvert-basis,
there again appear conical papillae similar to those of the body
proper. They gradually decrease in size towards the introvert-
tip. Fig. 52 represents some papillai from the middle region of
the introvert ; here they are seen as transparent, thin-walled
processes, 0.04-0.07 mm. high and conical in shape but with
the apex rounded. No hook or spine can be found on the intro-
vert. The tentacles are simple and not numerous, probably 10

in number.

The longitudinal muscles form a continuous sheet. There
are two retractor muscles, one of which is very small and the
other (the ventral ?) relatively large. The latter arises with two
short roots from near the posterior end of the body [rm and
nil, Fig. bo). The intestinal convolution {ic) consists of a few
irregular spirals unaccompanied by muscles of any sort. There

20 ART. 4. 1. IKE DA I

is a single segmental organ (so) situated on the right side of,
and close to, the nerve-cord {n). It represents an elongate
sac, arising freely from the body-wall.

This species seems to be most closely allied to Phascolion
lucifugax Sel. & De Man, a young specimen of which is said to
have been obtained by Hilgendorf in this country. The
two species resemble each other especially in having large papillae
in a zone around the middle region of the body proper. But
the present species certainly differs from Phascolion lucifugax in
possessing differently shaped papilla and in being devoid of
hooks on the introvert.

Gmius PHYMOSOMA.

PHYMOSOMA SCOLOPS, Sel. & De Man.

Selenka, (28), 1883, 2^. 75.

From Selenka's description the specific character of this
species may be summed up as follows :

Entire body-length 4 cm. on an average. Color reddish
brown, often yellowish brown. Introvert bears on the dorsal
side numerous pigmented bands of a reddish brown or brownish
black color. Papilhe black, tolerably large, but little elevated
in the middle region of the body proper ; they consist of brown
plates which become gradually smaller towards the peripheral

THE C4EPHYREA OF JAl'AX. 21

border. On the iiitrovert-bnsis and the posterior end of body,
the papilke are conspicuously large and elevated. Hooks about
0.07 mm. high, provided with two sharp teeth at the apex and
with warts (*' Runzeln ") at the base ; hook-rows about 15-17 in
number, between them small hook-papillœ (" Hakenpapillen ").
Eye-spots distinct. Tentacles, about 12, arranged in a semicircle
above the mouth.

Longitudinal muscles 20-21, rarely anastomosing. Four
retractor muscles arising from the middle of the body-proper :
the dorsal pair from the lst-2nd and the ventral from the 5th-
6th longitudinal muscles. The intestinal convolution consists of
12-14 spirals, and has a fine spindle-muscle fixed at both ends.
The one fixing muscle present is bifurcated towards the intestinal
attachment. The segmental organs are half as long as the body
proper ; the anterior third or fourth of their total length is
fixed by a mesentery to the body-wall.

In Japan this species has been collected from the Hokkaido
in the north to the Hiukiu Islands in the south. As in many
other cosmopolitan forms, the species shows remarkable variations,
especially in color, according to difterent surrounding conditions.
Specimens from the Riukiu (Loo-choo) Islands and the Ogasawara
Islands exhibit all the essential characters described by Selenka
(28) and other authors, — as, for instance, the black crescent-shaped
markings on the introvert, papillae and hooks of the same form
and size, &c. Those from the main island of Japan are in several
respects slightly different from the above mentioned form. Thus,
the individuals living among Ascidian colonies or eroded tufaceous
sandstone, show an extremely wide range of variation with regard
to the o-eneral color of the bodv and the size and color of tlie

22 AßT. 4. J. IK EDA :

papilke. The specimeus I collected on the shores of the Aomori
Bay, are enth*ely devoid of the characteristic markings on the
introvert, while the skin-papillse are in general unusually
tall. A form similar to the above is also frequently obtained
on the shores of Misaki.

PHYMOSOMA JAPONICUM, Grube.

Selenka, (28), 1883, p. 76.

The main features of this species are embodied in the following
extract from Selenka 's description :

Body proper of larger specimens, about 5 cm. in length ;
twice as long as introvert. Color, gray or grayish brown. Papilhe
visible as reddish brown specks ; they are formed of polygonal
plates arranged in concentric circles. About 25-30 hook-rows,
— in the younger individuals, sometimes even as many as 100,
— are present on the introvert. Hooks 0.07 mm. long, very flat
laterally, provided with two sharp teeth at apex, and with warts
(" Bunzeln ") at base. Tentacles, about 28 in number. Two brown
eye-spots present on the ganglion.

Longitudinal muscles are divided into about 30, frequently
anastomosing bundles. Four retractor muscles arise from the
middle one-third of the body proper, the dorsal pair being rooted
in front of the ventral pair. Dorsal vessel tolerably long, without
contractile villi. Intestinal convolution consists of 10-12 double
spirals, traversed by a spindle muscle fixed at both ends to the
body-wall. A single fixing muscle is attached to the first intestinal
spiral. Two long and moderately large segmental organs extend

THE GEPHYREA OF JAPAN. 23

posteriorly to the roots of the dorsal retractor muscles ; about
the anterior half of the organ is connected with the body-wall
by a mesentery.

A large number of specimens from Enoura in the Province
of Suruga, referable to this species, are preserved in the Zoological
Museum of the Science College, Imperial University of Tokyo.
I myself have obtained three specimens from a shallow bottom
(5 hiro deep) near the Misaki Marine Laboratory. According
to Selenka, Hilgendoef collected the species at Enoura and
also at Hakodate. It seems highly probable that this species
will be found in more southern localities than Enoura, for the
fauna of Enoura, as is admitted by all Japanese zoologists,
comprises many subtropical forms.

In dissecting specimens of the present species, it almost
always happened that a moderately large globular body was
found attached to the dorsal side of the œsophagus nearly
always in the same position. I have ascertained by a study of
sections that the said body is a sac w^hich contains a number
of small parasitic worms ; the parasites are imbedded in a
dense mass of leucocyte-like cells. Each parasitic worm is
externally enveloped by a rather thick cuticula of a bright yellow
color ; it is provided with a large and a small sucker, both situated
on the flattened side of the body. Spengel described quite similar
worms in the general body-cavity of Balanoglossus kapferi ; he
thought them to be the young of a Distoma species. As to
the body-form, the structure and position of the two suckers,
etc., I see no noticeable difference of importance between the
parasites described by Spengel and those found by me in Phymo-
soma japonicum.

24 AKT. 4. — I. IK EDA :

PHYMOSOMA ANTILLARUM, Grübe and Oersted.
Selenka, (28), 1883, p. 57.

From Selenka's description the specific characters of this
species may be summed up as follows :

Length of body-proper 3 cm.-, length of introvert 18 mm.
Skin yellow, thick, covered with low blackish brown papillœ. On
the introvert, the papillae are of a tall conical form, sparsely
distributed. Around the basis of filiform tentacles, of which there
are 50-80, there is present a collar-like skin-fold. Two eye-spots
on the ganglion.

Longitudinal muscles of the body-wall are divided into bundles,
which very frequently anastomose with one another. Intestinal
convolution consists of about 20 spirals. Rectal intestine long.
Spindle muscle is fixed to the body- wall at both ends. A single
fixing muscle joins the first intestinal spiral. Dorsal vessel carries
very numerous, finger-shaped contractile villi. Segmental organs
are about half as long as the body proper, their entire course
being accompanied by a mesentery.

Five specimens of this species were collected on the Islands
of Amani-Oshima (Koniya) and of Okinawashima (Naha, Itoman,
Chinensaki). The general external features as well as the ana-
tomical characters agree well with Keferstein's original descrip-
tion of this species. The Japanese specimens however are much
smaller than those collected in the West Indies (St. Thomas L).
Of the former the largest in my possession does not exceed in

THE CEPHYREA OE JAPAN. 25

total length 22 irwi., of which about 1/3 mm. appertain to the
introvert. The number of longitudinal muscles in the anterior
half of the body proper, is on an average 14, while in the posterior
half each longitudinal muscle divides twice successively before
reaching the posterior end of the body. The tentacles, filamentous
in form and about 30 in number, are arranged in one row,
ventrally and laterally to the mouth. Each tentacle bears in the
fresh state, as was mentioned by Keferstein, two narrow band-
like marking's of a bluish black color.

The anus and the external apertures of the segmental organs
lie at the same level ; the latter perforate the body-wall between
the second and the third longitudinal muscles. The roots of the two
retractor muscles spring from the 2nd-4th longitudinal muscles.
Eectal diverticulum is not present. Two distinct eye-spots are
found on the ganglion.

Specimens are obtained by breaking coral rocks.

PHYMOSOMA PACIFICUM, Keferstein.

Selenka, (28), 1883, ^j». 63.

The main features of the species as embodied in Selenka's
description are as follows : —

Body proper, 8 cm. in length and 14 mm. in maximum
breadth. Introvert longer than bod}^ proper, so that the whole
length may reach 29 cm. Color grayish, blackish, or reddish
brown ; often with small black specks ; introvert generally bears
on the dorsal side irregular pigmented bands. Body-wall thick.

26 ART. 4. 1. IKEDA :

hard, and with very rough surface on account of the presence
of tall (0.3 mm. high) papillse which are closely set. Hook-
rows 80-100, consisting of tall (0.10 mm.) hooks which are
provided with a sharp and strongly curved apical tooth and a
small blunt accessory tooth. Tentacles 35-40 in number,
arranged in a semicircle above the mouth.

Longitudinal muscles, divided into separate but frequently
anastomosing bundles, number 30-35 in the posterior half of
the body proper. In young individuals, they are sometimes found
in the continuous state. Of the four retractor muscle, the
ventral pair are rooted in the middle third and the dorsal pair
in the anterior third of the body proper. Intestinal spirals,
12-14 in number. Rectal intestine short. Spindle-muscle fixed
at both ends. Only one fixing muscle is present. Segmental
organs very long, reaching nearly down to the posterior end
and accompanied by a mesentery along the wliole length.

Numerous specimens of this species were collected by me,
in April 1901, on the Islands of Amami-Oshima (Nase) and of
Okinawashima (Chinensaki and Naha). They burrow in fissures
of coral rocks along the tide marks.

PHYMOSOMA ONOMICHIANUM, n. sp.

(Figs. 7, 56-58).

The present species is based on a single specimen which I
discovered in a collection made by Mr. Izuka, in 1901, in the
inlet of Onomichi (Province of Bingo). The specimen is pre-

THE GEPHYREA OE JAPAN. 27

served in formalin and therefore keeps the natural color almost
unchanged.

The animal is of a moderate size (Fig. 7), the body proper
measuring about 30 mm. in length and 9 iniii. in maximum
breadth. The introvert, which is entirely retracted, is much
more slender than the body proper, being only about 2 mm.
thick. The body- wall is yellowish brown and is covered with
large, variously shaped papillae of a deep brown color. In the
middle region of the body proper, the papillœ are but slightly
elevated bodies of 0.25 mm. basal diameter, each consisting of
])olygonal chitinous plates of a deep brown color and various
sizes (Fig. 56). The small lighter colored area (marked with
X in the fig.) in the centre of the papilla, is perforated by the
minute external opening of the subdermal gland. The non-
pa pillated parts of the surface, in this region of the introvert,
is covered with sparsely scattered chitinous plates which seem
to be of the same character as those of the papillœ. On the
introvert-basis as well as in the posterior region of the body
proper, the papilhe are conspicuously tall, taking the shape of
an obtusely apexed cone (Fig. 57«, taken from the region of the
introvert-basis). It measures on an average 0.20 7]wi. in height
and 0.25 mm. in basal diameter, and is constructed of thick
chitinous plates of irregularly polygonal shape, closely set to-
gether in a pavement-like manner ; the plates at the papillär
base are the largest and the deepest in color (brown). A small
pore {p, Fig. 576) opens at the apex of the cone, surrounded
by a circular area. No hook or spine is present on the intro-
vert ; nor can eye-spots be detected on the ganglion. The
tentacles are filamentous and are remarkably numerous, entirely
surrounding the mouth.

1^8 ART. 4. — I. IKEDA :

The longitudinal muscles are divided into separate, but very
frequently anastomosing, bundles ; in the region of the segmental
organs they are 16 in number and in the posterior region of
the body about 22. The retractor muscles (Fig. 58) occur in a
single ventral pair, the dorsal pair being Avanting. The pair
present are attached to the body-wall at the beginning of the
posterior third of the body proper. The intestinal convolution
consists of about 15 double spirals, twisted around the spindle-
muscle {sm) which is fixed at its posterior end. The intestine
is suspended by two fixing muscles (fin), both of which lie to
the left of the nerve-cord [n) ; one of them arises just behind
the root of the left retractor muscle, and is joined to the pos-
terior part of the intestinal convolution, while the other arises
at a point of the body-wall about 7 /ii/it. in front of the origin
of the first and ends at the first spiral. There is no blind
diverticulum on the rectum. The anus (a) is situated near the
anterior end of the body proper, at the same level as the ex-
ternal apertures of the segmental organs. These (so) are long,
being nearly half as long as the body proper ; throughout their
whole length they are attached to the body-wall by a thin mes-
entery. The contractile villi of the dorsal vessel are very
numerous and tolerably long, reaching up to 2 uini. in length ;
some of them show a slight ramification.

The most remai'kable characters of the species consist (1)
in the total absence of hooks and (2) in the retractor muscles
being present in a single pair. No form combining these two
characters seems to have been hitherto recorded.

Habitat : The sandy shore of the Inlet of Onomichi in
the Inland Sea.

THE GEPHYREA OF JAPAN. 29

PHYMOSOMA NAHJENSE, n. sp.

(Figs. 8, 59-62).

I collected two specimens of this species, in April 1901, at
Naha (Naminouye) in Okinawashima, one of the Islands of
Riukiu. In external features they closely resemble Phymosoma
scolops Sel., for which species they were at first mistaken by me.

In the larger of the two specimens, the introvert is entirely
protruded, the total body-length measuring 45 mm. and the
maximum breadth about 8 mm. (Fig. 8). The skin is of a
brownish yellow color, more deeply tinted at the introvert-basis
and at the posterior body-end. Over the entire surface there
are distributed tolerably large, but very slightly elevated papillœ.
In the middle region of the body proper, these are nearly flat
and elliptical in shape, measuring on an average 0.13 mm. by
0.10 vim. across (Fig. 59). They are thickly covered with
numerous round chitinous plates w^hich gradually become larger
towards the central area (denoted by x in the fig.) containing
the small aperture of the subdermal gland. The papillse gradu-
ally grow taller towards both ends of the body proper. In the
introvert-basis they are on an average 0.2 7um. high and towards
the posterior end of the body 0.15 mm. Hook-rows are numerous
and consist of small hooks, about 0.04 mm. high, which are
provided with a sharp and strongly curved apical tooth (Fig.
60). In the side view of the hook there are seen a transverse
bar [b) across the base and two rows of warts (rz) (" Kunzeln ").
Both tlie hooks and the warts are of a reddish yellow color.

30 AEÏ. 4. 1. IKEDA :

Between every two rows of the liooks, there are further found
numerous perforated papillae or " Miindungspapillen " (Fig. 61),
a small paipllary structure which is so slightly elevated as to
measure only 0.02 mm. in height. On the apex and around
the small aperture of the suhdermal gland, there occur in the
papilla-wall a few chitinous granules disposed in a small ring.

The longitudinal musculature of the body-wall is divided
into 20-30, separate but anastomosing bundles. The retractor
muscles are represented by the ventral pair only, the dorsal pair
being wanting. The ventral retractors {vm, Fig. 62) spring
from the middle of the body proper. The dorsal vessel, runniug
along the dorsal side of the oesophagus, is not provided with
contractile villi. The intestinal convolution {ic) consists of 10
double spirals around the S23indle-muscle, which is fixed at both
ends to the body-wall. There is only a single fixing muscle
(/>«), which, arising on the left of the nerve- cord, attaches it-
self to the first spiral. The rectum is tolerably long; it is with-
out a rectal diverticulum. The segmental organs {so) are about
half as long as the body proper; their external apertures are
situated at nearly the same level as the anus. The organs, for
the greater part of their length, are connected with the body-
wall by a mesentery. The tentacles are about 22 in number ;
they are relatively short and finger-shaped.

According to Selenka's Monograph, there should be known
only one species belonging to the genus Phymoioina, viz., Ph.
Rüppelü Grube, that is characterised by having hooks and only
two retractor muscles. In these respects the present species
agrees with Ph. Püppelli, but in the form and structure of the
papillae, in the number of tentacles, etc., it differs from it sufii-
ciently to justify specific distinction.

THE GEPHYREA OF JAPAX. 31

Habitat : Nalia on Okinawashima (one of the Islands of
Kiukiu), among dead coral masses.

Genvs SIPUNCULUS.

SIPUNCULUS NUDUS, Linné.

Selenka, (28), 1884, p. 92.

The main characters of this species are as follows :
Specimens from the Mediterranean Sea, according to Grube,
may be 2 feet in length, about 8 inches of which appertain to
the introvert. The latter is covered with an immense number
of large and very high papillse. The body proper shows on
the surface 30-33, mostly 32, longitudinal grooves, which by
intersecting the numerous circular grooves form small rectangular
areas. The mouth is surrounded by a tentacular membrane which
is divided into the dorsal and the ventral lobe, the former being
subdivided into large and the latter into small folds — the tenta-
cular flaps.

The longitudinal muscles are divided into 30-33 bundles
which very rarely anastomose. Four retractor muscles arise at
the same level from the 4tli — 8th longitudinal muscles. On the
rectum are found a small diverticulum and two glandular struc-
tures, the " büschelförmige Körper." The spindle-muscle arises
slightly in front of the anus, and is posteriorly free from the

32 AKT. 4. — T. IKEDA :

body-wall. Very numerous fixing muscles are attached to the
intestine. Segmental organs are so long that their posterior
ends reach the roots of the retractor muscles ; the anterior one-
fifth of their length is fixed to the body-wall by means of
mesenteries. Their external opening lies between the 4th and
the 5th longitudinal muscle. Eye-spot absent.

Specimens of this species are frequently obtained in the
muddy inlets near the Misaki Marine Laboratory. They are
very common along the muddy shores of Tateyama (Province of
Böshyü) and of many localities along the Inland Sea. I have had
the opportunity of comparing the Japanese specimens with two
well preserved specimens of the same species from Naples. In
external as well as internal aspects the Japanese specimens agree
well with Selenka's description, except in the number of the
longitudinal muscles of the body-wall. In the Mediterranean
specimens the said muscles are, as was pointed out by Selenka,
32 in number, while in the Japanese forms they number con-
stantly 30.

SIPUNCULUS CUMANENSIS, Keferstetn.

(Fig. 63).

Selenka, (28), 1884, j). 104.

The main characters of this species may briefly be stated
as follows :

Body large, elongate ; introvert about one-third of the total
length. Skin thick, quite opaque, yellowish or reddish brown in

THE OEPHYREA OF JAPAN. 33

color. Skill- bodies small and flat, contaiiining a few hook-like
structures (?) with the points directed outwards. At introvert-
basis and near the posterior end of body, the skin-bodies though
but slightly elevated above the surfece are so large (0.38-0.4 mm.
in diameter) as to be easily recognized with the naked eye.

Longitudinal muscles average 20 in number ; they are broad
and anastomose but rarely. Circular muscles are likewise divided
into distinct but narrower bundles. Four retractor muscles arise
a short distance behind the anus, the ventral pair between the
7th and the 8th longitudinal muscle. Rectal diverticulum present.
Spindle-muscle fixed to the body-wall at both its ends, anteriorly
giving off two lateral branches. The single fixing muscle has
two roots. Dorsal vessel with numerous but short contractile
villi. Segmental organs short, connected with the body-wall for
the anterior one-fourth of their length by a mesentery ; their
external apertures lie between the 3rd and 4th longitudinal muscle.
A large number of crescent-shaped dissepiments are present on
the inside of body-wall in the region between the anus and the
posterior body-end.

This cosmopolitan species is very widely distributed in Japan.
Numerous specimens have been obtained from various localities ;
as for instance, Kamo (Province of Eikuzen) ; coasts of the Tokyo
and Sagami Bays and of the Inland sea ; Kagoshima (Kiushiu) ;
Koniya in Amami-( )shiina ; Naha and Itoman (Kiukiu). The
specimens from various localities in Japan proper, including
those from Kagoshima, are 20-30 (?;>?. in total length and 8-12
mm. in breadth, the introvert and the body proper being nearly
uniformly broad. The color is yellowish brown in the male and
reddish brown in the female. On the other hand, the specimens

ö4 ART. 4. T. TKEDA :

from Naha and Koniya are much smaller, while the body-wall,
which is rough on the surface, is thicker and more deeply brown
in color, without however showing any recognizable difference in
internal anatomy. But as compared with foreign specimens, all the
Japanese forms seem to show some points of structural deviation
as will be gleaned from the following account.

The longitudinal muscles are divided into 18-21 (mostly 20)
separate bundles which very rarely anastomose with one another ;
the circular muscles are also separated into more or less regular
bundles. Of the four retractor muscles, the dorsal pair arise from
the 7th-9th longitudinal muscles ; the veiitral pair from the lst-4th
(mostly the lst-2nd). The roots of all these muscles are situated
nearly at the same level. The two lateral branches of the spindle-
muscle are attached to the 9th or the 10th longitudinal muscle,
close to the root of the dorsal retractor muscle; the main bundle does
not end posteriorly in a single root but divides into 3-5 branches
which give off still finer ones in order to join the body-wall.

The * Ovale Gebilde ' which was first descril)ed by Keferstetn
from German specimens, is present also in all the Japanese speci-
mens I have dissected. As was already noticed by other authors,
these structures are seen as small oblong bodies closely aflhering to
the inner surface of the body- wall. Their longer diameter measures
0.5 miîi. on an average. Sometimes they are found lying on the
longitudinal muscles and sometimes between them. They are most
abundant in the anal region. When seen in the fresh state,
they appear as colorless transparent spots. A study of the bodies
on sections has revealed the fact that they are of quite a different
structure from those which were described by Keferstein. Ac-
cording to that author, the bodies should present an appearance
of skin papillae and should inclose a peculiar kind of hook, 0.08-

THE GEPHYREA OF JAPAN. 35

0.10 mm. high and made up of a substance which is not acted
upon by acetic acid or soda. Now in the Japanese specimens,
neither hook nor any other solid body can be detected in the
interior of the organ ; the organ is evidently a saccular gland
opening by a short narrow duct to the exterior. Fig. 63 represents
a portion of a cross-section through the body-wall, passing through
one of the organs under consideration. As may be seen from
that figure, the organ incloses a cavity and thus represents a
thick-walled sac, the wall being composed of two distinct layers.
The external layer [el) again consists of two layers, the j)eritoneal
covering and a connective-tissue layer, both of which are continuous
with the corresponding layers on the inside of the integument.
The far thicker internal layer (gl) of the organ is formed of an
epithelial row of very tall and large cells with granular contents.
The granulation of the latter is in some cells coarse, in others
much finer. Staining reactions show that the granules are closely
allied to mucin. The nucleus is mostly found situated close to
the inner end of the cells. The outleading canal of the gland
runs through the thick muscular body-wall to open by an external
pore. (The section of Fig. 63 did not pass right through the
length of the duct lumen, but the course of this duct is partially
indicated by its fibrous investment.)'''

* Lately I have had the opportunity of reading Hermann Augenek's paper " Beiträge
zur Kenntniss der Gephyrcen" (Berlin, 1903), in which tlie author gives a detailed account
of the histological nature of the " Ovale Gebilde " or his " Kef erstein' sehe Bläschen." The
author doubts the correctness of Keferstein's representation of the structures as papillary
bodies containing hooks. He considers the wall (exclusive of the peritoneal and the
connective-tissue covering) of tlie body as consisting of two cellular layers : an inner layer,
directly lining the iumen of the body and showing a distinctly epithelial character, being
composed of radially arranged spindle-like cells with similarly shaped nucleus; and an
outer layer of a more connective-tit<sue-like appearance containing large lacunar spaces and
spindle-shaped nuclei. As the author himself makes the statements with much reserve, I
am led to doubt their validity; probably the appearances described by him were due to
some inappropriate methods of preparation. According to my experience, fixing reagents

36 ART. 4. — I. IKEDA I

SIPUNCULUS AMAMIENSIS, n. sp.

(Figs. 64-65).

The species is of a tolerably large size, the body proper
measuring 100-120 mm. in length and 10-15 mm. in thickness
and the length of the introvert being about one-fourth that of
the body proper. The color is yellowish brown in the body
proper, and a deeper brown in the introvert. The skin-surface
is rough, especially at the introvert-basis and at the posterior
end of the body, both these parts being covered with the largest
skin-bodies. In the middle of the main region of the body,
the skin-bodies are nearly flat and of a circular form, measuring
about 0.075 mm. in diameter (Fig. 64). They are each covered
with a thick and finely granular chitinous layer of a yellow
color. The central part of this covering is occupied by a rela-
tively large, circular and colorless area which is composed of
larger, closely aggregated chitinous granules. Between the papillae
the chitinous layer of the skin forms irregular and interrupted
wrinkle-like folds, which divide the surface into irregular mesh-
like areas. Near the anterior end of the introvert, immediately
behind the tentacles, the chitin folds wholly disappear, and in-
stead of them there appear narrow grooves intersecting one an-
other at right angles so as to bring about a tesselated pattern.

make the structure in question much too hard and brittle for the preparation of good
sections. However that may be, I have no doubt that the two layers described by
AuGENER collectively correspond to the single epithelial layer I have seen and the glandular
nature of which seems to be plain from the reaction shown by the granular contents of the
cells towards stains.

As to the physiological significance of the bodies, AuGENER too lias suggested the view
that they may be regarded as "drüsenartige Gebilde" {p. Ç>?>), though this was not proved
by him from the histological standpoint.

THE GEPHYllEA OF JAPAN. 37

No hook or spine is present on the introvert. The tentacles
are filamentous and about 80 in number ; they are arranged in
16 rows, the oral ends of eacli set of two rows being joined by
a U-shaped ridge.

The longitudinal muscles are divided into 15 broad bundles
which rarely anastomose with one another in the region of the
introvert. Of the four slender retractor muscles, the ventral
pair are the larger and spring I'rom the 2nd — 3rd longitudinal
muscles in the middle region of the body proper, while the
dorsal pair arise far more anteriorly from the fourth longitudinal
muscles on both sides (Fig. 65, vm and di/t). The dorsal vessel
[dv) is seen running along the whole posterior course of the
œsophagus and gives off very numerous short contractile villi.
The spindle-muscle {sm), fixed at both ends to the body-w^all,
gives off tW'O lateral branches (sm') at the very beginning of tlie
intestinal convolution. These branches end on the 6th longi-
tudinal muscles, nearly at the same level as the roots of the
dorsal retractor muscles {dm). There are, besides, two stout
fixing muscles {ßa) : one arises from the 4th longitudinal muscle,
close to the roots of the right retractor, and is attached to the
beginning of the rectum ; the otlier springs with two roots from
the first longitudinal muscle and is attached to the œsophagus
near the intestinal convolution. The segmental organs (so) are
small sacs of a deep reddish brown color; the external apertures
lie at the same level as the anus. The organs are entirely free
of a mesentery. On the dorsal lip of the funnel or the internal
opening, we constantly see two or more small processes, some
of which may subdivide into 2 or 3 branches. They seem to
be made up simply of the connective-tissue and of the external
peritoneal covering. I am at present unable to make any sug-

38 AET. 4. 1. IKEDA :

gestion as to the physiological nature of the structure. There is
no rectal diverticulum. Nor is there an eye-spot present on
the ganglion.

The present species seems to be most nearly allied to Sipun-
culus rolumaiius Shipley and S. funafuti Shipley (32). But
those two species may both be distinguished from the present
by the peculiar form and color of their papillse as well as by
the absence of fixing muscles. Besides, S. funafuti is so pecu-
liar in color and body-form, that it bears no resemblance to the
present species in external aspects.

Habitat : Numerous specimens were collected by me, in A^yW
1901, at Koniya (Amami-Oshima), and a single one at Naha
(Okinawashima). All were found near the shore in coral sand.

Genus ASPIDOSIPHON.

ASPIDOSIPHON TRUNCATUS, Kefekstein.

(Figs. 66-67).
SelexVka, (28), 1883, p. 118.

The main characters of this species are indicated in the
following extract from Selenka's descrij)tion :

Color brown all over, but deeper on the anal and caudal

shields, both of which arc radially grooved. Papillae small,

THE GEPHYREA OF JAPAN. o9

slightly larger in the region of the caudal shield. Very numer-
ous rows of very small hooks with incompletely bifurcated apex
and bearing warts ("'Runzeln") at base. Small spines are scattered
to a wide extent behind the hook-rows. Longitudinal muscles
partially divided into irregular, frequently anastomosing bundles,
which are however more or less regularly developed on the in-
side of the anal shield. Retractor muscles spring with two roots
from the inside of the caudal shield. Intestinal double spirals
are travei'sed by a spindle-muscle fixed at both ends to the
body-wall. Rectal diverticulum distinct. Segmental organs long
and large, attached to the body- wall by mesenteries throughout
nearly their entire length.

Several specimens of this species were collected by Mr.
MiYAJiMA (1900) and by myself (1901) in Okinawashima (Naha)
and Amami-Oshima (Koniya). The largest specimens in the fully
extended state measure 5 cm. in total length. In the preserved
state, the larger ones have a yellowish brown color and the smaller
a light reddish yellow color. The anal and caudal shields are
of a deep brown.

The external features and the anatomical characters agree
well with the description given by Selenka. What slight dif-
ferences seem to exist, are as follows :

On the concave side of the hook, just at the base of the
small accessory tooth, a few very fine, thread-like processes of
chitin are produced inwards from the spot mentioned (Fig. ^^).
Between hook-rows there are scattered small tubular papillœ
(Fig. 67). Each papilla rests on a low basal prominence which
is covered with relatively large chitinous granules {x).

The worms live in dead coral masses or in eroded coral rocks.

40 APwT. 4. — I. TKEDA :

ASPIDOSIPHON STEENSTRUPII, Diesinc.

Selenka (28), 1883, p. 116.

The main characters of this species, taken from Selenka's
description, are as follows :

Introvert and body proper nearly equally long. Color light
yellowish brown, nearly black at both ends of body proper.
Skin-bodies formed of closel}^ aggregated polygonal plates, four
or more of AYhich lie in the center ; each of the latter plates
with a small plate. The two shields are of a nearly flat hemi-
spherical form, lightly grooved radially and circularly ; formed
of small oval or polygonal bodies, those lying in the periphery
being largest. Introvert with small double-toothed hooks ar-
ranged in complete rings and with a great number of relatively
tall perforated papillœ (" Ausmündungspapillen "). Longitudinal
muscle, in the posterior part, separating into about 25, frequent-
ly anastomosing bundles ; anteriorly these gradually diminish in
number till they all coalesce into a continuous layer. Retractor
muscles spring from the beginning of the posterior fourth of the
body j)roper, the line of their origin extending from the 3rd to
the 7th (sometimes to the 8th) longitudinal muscle. Spindle
muscle fixed to body-wall at both ends. Rectal diverticulum
present. Fixing muscle said to be absent, except in specimens
from the Philippines, from which Selenka has described a
single fixing muscle attached to the 1st or the 2nd intestinal
spiral. No contractile villi. Segmental organs long, posteriorly
reaching to the roots of the retractor muscles ; their anterior

THE GEPHYEEA OF JAPAN. 41

half acconipanied by a mesentery. Two eye-spots present on
the ganglion.

This species is one of the commonest Sipimculoids met with
in the Rinkiu and the Amami-Oshima Islands.

As in the Philippine specimens described by Selenka, a
fixing mnscle is distinctly present. But it may be pointed out
that unlike them the muscle terminates on the intestine not
with one end, but with two branches ; one of these is attached
to the hindmost part of the œsophagus, while the other is joined
to the spindle muscle just at the beginning of the intestinal
conv^olution. The hooks on the introvert are provided with two
distinct rows of small warts lying at the hook base. This fact has
not been mentioned in the case of foreign specimens of the species.

ASPIDOSIPHON MISAKIENSIS, n. sp.

(Figs. 9, 68-72).

This species is a small form (Fig. 9). The body proper
rarely exceeds 20 mm. in length and 5 mm. in width; the introvert
is unusually long, being at least three times as long as the body
proper. The color in the fresh state is a light brownish pink,
the introvert-basis as well as the anal and caudal shields being
shaded a deeper brown. The anal shield is oval in shape and is
composed of minute calcareous granules of a deep brown color ;
only the periphery of this shield exhibits short radial grooves
(Fig. fj8). In the anal region, the skin is covered with large
and closely set papillie of a deep brown color which are composed

42 ART. 4. T. IKEDA :

of small granules hardened more or less by the deposition of some
calcareous matter. In the middle region of the body proper, the
papillae are nearly flat, measuring 0.15-0.10 mm. in diameter.
Over their surface there are aggregated numerous irregularly
polygonal chitinous plates of various sizes and of a light brownish
yellow or true yellow color. One roundish plate in the center
is perforated by the opening of the subdermal gland. The caudal
shield is of a conical shape and a deep brown color. Like the
anal shield, it is constructed of small calcareous granules, and
the margin alone is slightly folded radially. Just behind the
tentacular region of the introvert, there follow about 35 complete
hook-rows. The hook, 0.032 mm. high and yellowish brown in
color, is suj^plied with two sharp teeth (Fig. G9). Behind the
hook-rows, there are distributed numerous gently curved spines
0.045 mm. in height (Fig. 70). The papilhe on the introvert
are represented by a small colorless hyaline tube standing on a
tubercle-like base (Fig. 71).

The longitudinal muscle in young specimens is quite continu-
ous, while in the adult it is partially divided into indefinite
bundles anastomosing very frequently. There are only two re-
tractor muscles, the ventral pair, which arise very close to the
caudal shield [vm, Fig. 72). For the greater part of their course,
they are fused into one band. The dorsal vessel {dv) is very
short (about 2-3 mm. in length) and has no contractile villi.
The intestinal convolution {;ic) consists of irregular and loosely
twisted spirals indicating no definite number of turns. The
spindle muscle is, fixed to the body-wall at botli ends. A small
blind sac [rg) is attached to the middle part of the long rectum.
Only one slender fixing muscle (/m) is present, attached to the
first spiral of the intestinal convolution at one end and at the other

THE GEPHYREA OF JAPAN. 43

to a point on the left of the nerve-cord and cdose to it, far more
antei'iorly than the root of the retractor nuiscle on the same
side. The segmental organs {so) are remarkably long, being
equal to two-thirds of the length of the body proper. The organs
are fixed throughout their whole length to the body-wall by a
mesentery. The tentacles are short and thick, numbering 8-10.
Two eye-spots are present on the ganglion.

Among the hitherto known Aspidosiphon species, A. elegans
Cham. & Eisenh. seems to be most closely allied to the present
species. However the two species in question may be distinguished
from each other by different features of the anal and caudal
shields and by the difference in the size and form of the hooks
and spines.

Habitat : I have collected the species only at Misiiki, along
rocky shores in the vicinity of the Marine Laboratory.

ASPIDOSIPHON UNISCUTATUS, n. sp.

(Figs. 10, 73 — 77).

The unique specimen (shown in Fig. 10) on which this
species is based, was found in Mr. Miyajima's collection made
at Kataura in the province of Satsuma (1900). In its present
state, preserved in alcohol, the body proper measures 13.5 mm.
in length and 3.5 mm. in maximum width. The skin, which
is relatively thick, is yellowish brown becoming a deeper brown
towards the anal shield. The latter, deep reddish brown in
color, is of a tall helmet-like shape, obliquely truncated at the

44 ART. 4. — I. IKEDA :

basal border. The wall of this shield is unusually thick aud
hard on account of the deposition of a calcareous substance. Its
surface is sculptured into numerous longitudinal ridges and
grooves that converge towards the rounded apex. When examined
under a certain magnification, the ridges are seen to be covered
with small chitinous bodies of a deep brown color (Fig. 73), and
at the hindmost end of each ridge there is situated a large
papilla consisting likewise of chitinous l)odies similar to those of
the remaining part of the ridge. The center of this papilla
shows a circular space formed of small and light-colored plates
and surrounded by 2-3 rings of very dark-colored plates. The
apical parts of the shield are stripped of the ej)idermis, thus
exposing the underlying wdiite calcareous layer. The skin-surface
adjacent to the shield is closely beset with papillae which
appear as deep brown spots. These latter are structurally the
same as those of the papilhe lying at the l:)asal border of the
anal shield, though they are somewhat smaller. In the middle
and posterior regions of the body, the papilke are quite flat, and
are formed of many polygonal chitinous plates separated from
one another by narrow grooves (Fig. 74). These plates are smaller
towards the center of the papilla, where the pore of the subdermal
gland is situated. The caudal shield is not definitely marked
off, but is simply indicated by a darker brown color. The aperture
of the introvert lies just below the ventral border of the anal
shield, so that the introvert as a whole stands out nearly at
right angle from the body proper. The introvert is about as
long as the body ])rope]', and is of a light yellowish bi'own color.
The anterior portion carries numerous complete ring-rows of
hooks which are 0.04 vim. long and of a reddish yellow color
(Fig. 75). Between these rows are scattered tubular papilkie

THE GEPHYREA OF JAPAN. 45

(Fig. 76) which are hyaline and colorless and measure about
0.015 uiin. in height.

The longitudinal muscle is, in the middle of the main part
of the body, divided into frequently anastomosing bundles, but
is quite continuous in the region of the introvert-basis and in
the posterior fourth of the body proper. The sexual oi'gans
{sx, Fig. 77) are developed as two slender wavy bands at the
roots of the retractor muscles {vm). The intestinal convoluti()n
{ic) consists of about 12 spirals, and is traversed by the spindle
muscle (sm) which is fixed to the body-wall at both ends. There
is no fixing muscle at all. The blind diverticulum {rg) is attached
to the middle part of the rectum. The segmental organs (so)
are in length equal to about two-thirds of the body proper, and
are of a pale reddish brown color. The organs are for the whole
length accompanied by a mesentery. Their external apertures
are situated at the same level as the anus [a). The anterior
terminal portion of the i-ectum is provided with well-developed
wing-muscles {wm). Two In-own eye-spots on the ganglion ig).
Tentacles short, 10 in number.

Habitat : Kataura in the province of Satsuma. The worm
was found in a small hole bored through a small piece of dead coral.

ASPIDOSIPHON ANGULATUS, n. sp.

(Figs. 11, 78-80)

Only one specimen of this species has Ijeen collected by me
(1901) in the Inlet uf Katsuyoki in the Amami-( )shima Island.
It lived in a dead coral that was exposed at low tide. The

46 ART. 4. 1. IKEDA :

animal (Fig. 11) is small, measuring 15 mm. in length of the
body proj^er. The introvert is nearly as long as the body proper,
but only half as thick. The skin is ver}^ thin, smooth, partly
transparent and of a light reddish brown except at the anterior
end of the body proper, where it is a deeper brown and of a
rough granular appearance. The anal shield may be said to
consist of an obliquely conical main part and of a short, cylindrical,
basal part. The former has the surface marked with distinct
radial grooves. The basal part also shows grooves which alternate
in position with those in the conical part (see Fig. 11). The
entire shield is formed of small, densely aggregated chitinous
jDlates. The caudal shield is in the shape of two cones juxta-
posed at base ; it is radially grooved, the grooves on the two
cones alternating (see the above figure). The greater part of the
body proper shows no papillary structure ; only by using a high
power of the microscope can the numberless small openings of the
subdermal glands be discovered. Just in front of the caudal
shield there are seen some slightly prominent papillary bodies
measuring 0.06-0.10 mm. in diameter. They are formed of small
polygonal chitinous plates of a light yellowish brown color (Fig.
78). Tlie plates are somewhat smaller in the central part of the
papilla than in the periphery ; the most centrally situated plate
bears the poi-e of the subdermal gland. Around the base of the
anal shield, the papilhe are larger (commonly 0.10 mm. in dia-
meter) and of a deeper brown than in the posterior region. Vei-y
numerous hook-rows are present on the introvert sui-face. The
hooks (Fig. 79), 0.02 7nm. high on an average, have a blujitly
pointed apex; they are of a light yellow color. Between these hook-
rows there are distributed small tubular papillae. No spines can
be detected behind the hook-j'ows.

THE CIEPHYREA OF JAPAN. 47

The longitudinal muscles are incompletely divided into fre-
quently anastomosing bundles. Two retractor muscles, the ventral
pair (mn), arise slightly behind the middle of the body (Fig. 80
PL IV). The sexual organs (so) are situated at a short distance
behind the roots of these muscles. The intestine is convoluted
into about 15 spirals. The spindle muscle {sm) is fixed to the
body-wall at both ends. No fixing muscle and no rectal divert-
iculum are present. Peculiar to this species is the fact that the
œsophagus {œs) is attached to the retractor muscle at a point so
far from the fusing point of the right and left retractors, that it
appears to be entirely free from the muscle. The segmental
organs {so) are about one-third as long as the body proper ; their
posterior halves are free of the mesentery joining them to the
body-wall. On the dorsal lip of the internal opening or funnel,
there is present a very small sac-like body. Eye-spots cannot
be discovered. The filamentous tentacles are probably 10 in
number.

ASPIDOSIPHON SPINALIS, n. sp.

(Figs. 12, 81-85)

The animals (Fig. 12) belonging to this new species are of
small size, the body proper being generally about 10-\ 5 mm., long
(28 mm. in one specimen) and about 3 mm. in breadth. The
skin appears smooth, and has a light yellowish l)rown color in
the main part of the body proper, but is rough and of a deeper
brown on the anal shield and the dorsal side of the introvert.

48 ART. 4. T. IKEDA :

As in Aspidosiphon angulatus, no distinct papilla or skin-body
is present, except the small pores of the subdermal glands. The
anal shield is an elliptical and nearly flat body, covered with
large brown papilhie. These are composed of minute chitinous
granules. The caudal shield is not so thick, nor so distinctly
contoured as the anal shield. Its peripheral portion alone is
marked with small chitinous granules of a brown color. Confined
to the vicinity of the two shields, are the papillary bodies ; they
are very slightly elevated, elliptical in form, brownish in color, and
measure on an average 0.15 mm. by 0.09 mm. in the two diameters.
The papilhie (Fig. 81, PI. HI) show the same structure ns those of
Aspidosiphon angulatus, except in the one unimportant resj^ect
that the chitinous phites are in general larger than in that species.
The introvert is nearly as long as the body proper ; i'ts posterior
half is deep brown on account of the presence of spines of that
color. The spines (Fig. 82) are straight, sharply pointed, broad
at base and are formed of a thin chitinous lamina folded in a
funnel-like manner ; they are 0.04 mm. high in the anterior region,
but become gradually taller towards the introvert-basis where
they measure 0.07 mm. in height. These spines are distributed
over almost the whole surface of the introvert, l^ut are most
closely aggregated on the dorsal side of the posterior half of the
part, thus giving to this region a deeper brown color than else-
where. In the anteriormost parts of the introvert, there are
found numerous complete rows of brown colored hooks. The hook
(Fig. 83) is 0.05 mm. high and is provided with two sharp teeth
at apex. In the interspaces between hook-rows there occur small
colorless and transparent tubular papillse (Fig. 84) about 0.03
'mm. high.

The longitudinal muscles are entirely continuous. The two

THE GEPHYREA OF JAPAX. 49

(ventral) votraetov muscles (vvi, Fig. 85) arise just in front of the
caudal shield. The intestinal convolution (vc) consists of 10-1")
spirals, traversed by the spindle nuiscle {sm), which is fixed to
the body-wall at both ends. No fixing muscle is present. The
rectum is long and lacks a diverticulum. Tlie anus is situated
just behind the anal shield. The two segmental organs (so) of
a reddish brown are about as long as the body proper and are
fixed to the body-wall by a mesentery along their entire length.
Their external apertures lie a short distance behind the anus.
The two eye-spots appear as brown specks on the ganglion.

Habitat : Numerous specimens of this species were found
(April 1901) in dead corals, at Koniya in Amami-( )shima
Island.

Genm CLŒOSIPHON.

CLŒOSIPHON JAPONICUM, n. sp.

(Figs. 13, 86-89).

This species is one of the commonest Sipunculoids met with in
the Eiukiu and Amarai-Oshima Islands. The general external
and internal characters seem at first glance to agree well with
those of Clœosiphon asperg ilium Quatrefages, C. mollis Selenka
& BüLOW or C. javanicum Sluiter, but careful examinations
bring to light some important points characteristic of each of the

50 ART. 4. 1. TKEDA :

species mentioned, especially in regard to their external features.
As to its internal characters, the present species may be said to
be almost identical with C. asj:)ergilh(m and C. javanicum, except
in the facts that the longitudinal muscle is not separated into
bundles and that the sexual organs are not developed in the same
position as in C. javanicum.

The species has long cylindrical body which may reach about
7 cm. in length and 7 mm. in width. The posterior end of the
body is rounded in the fresh state. As was noticed by Shipley
(31) in G. javanicum, the external appearance of the s})eeies is
subject to considerable individual variation, but as a. general rule
two types of characters are distinguishable. In the one the skin
is thick and of a deep brown and the calcareous shield, composed
of calcareous papilloe, is as wide as the body proper ; while in
the other the skin is thin, nearly transparent and of a reddish
brown and the shield, consisting of very small papillœ, is narrower
than the body proper. According to my own observations, how-
ever, these differences are in a great measure those of the
sex. The first type referred to being seen mostly in the male
and the second in the female. Fig. 18 is taken from a large
male in the living state.

The anal shield, which shows a structure peculiar to the
genus, presents varying shapes according to the state of contrac-
tion, a fact previously noted by Sluitek (35) in G. javaniciim ;
it assumes all sorts of forms varying from that of a depressed
disc to that of a cone. The shield is made up of numerous milky-
white calcareous papilLie which are spirally arranged and become
gradually smaller towards the posterior border (Fig. 86). The
papillœ are mostly of a hexahedric form, but their external facets
present rather irregular shapes, being quadrate, rhomboidal,

THE GEPllYREA OF .lAl'AN. 51

pentagonal and rarely polygonal. A condition similar to the
above has also been noticed by Selen k a (28) in C. mollis. The
edges of the external facets are not sharp, but bevelled and are
pigmented a deep brown in the fresh state. The center of each
external facet, Avliich is perfectly flat, is occupied by a small,
deep brown, circular pore of the underlying papillary gland.
Skin-papillte are scattered over the entire surface of the body
proper, and are so large that they may be detected with the
naked eye ; they are especially densely aggregated near the base
of the shield. Under the microscope each papilla appears as a
slightly prominent body of a brown or yellowish brown color.
Its peripheral portion is formed of two irregular but concentric
rings of small chitinous plates, while the central portion is roofed
by a thill chitinous layer which is often incomplete as shown in
Fig. 87. The subdermal gland opens at the center of the papillae.
The papill« measure, in the middle region of the body proper,
about 0.08 }nin. in diameter. They become gradually shorter but
stouter towards the posterior end, while on the other hand they
become taller and more slender towards the base of the calcareous
shield. On tlie introvert-basis, they are remarkably tall, measur-
ing 0.10 m)n. in average height ; they appear as cylindrical
l)rotuberances of a coarsely granular appearance (Fig. 88). Chitin-
ous granules are distributed also over the non-papillated part of
the skin. There are present about 20-25 complete ring-rows of
hooks. These are 0.10 inni. high and of a brown color, and are
provided with two sharp curved teeth (Fig. 89). As was figured by
ÖLUITER from G. javaiiicuiii, there are a few small processes situated
just below and inside the accessory tooth. At the hook-basis
there lie two small transverse Ijars and two rows of warts (see
the above figure). In the interspaces between hook-rows there

52 ART. 4. — I. IKEÜA :

are distributed uumeruus tabular or perforated papill* (Fig. 89);
the upper or free edge of these is guarded by a row of very
small chitinous teeth.

In the individuals which I regard as the female, not only
the shield itself, but also the calcareous papillae on it, are much
smaller than in those which I regard as the male. Besides, in
the former case all the jDaj^illoe are nearly of the same size. Some
individuals are characterised by a peculiarly modified shield ; the
calcareous papillœ found at the apex of such shields are elongated
into long processes of a deep brown color. Quite a similar feature
was described by Sluiter in connexion with the shield of C.
javanicum.

Clœosiphon japoiiiciüii must be regarded as a distinct species
differing in several important points from the hitherto known
species, viz., C. aspergillum, C. mollis and (J. java/iiaum. It
differs from C. aspergillum in the irregular forms of its calcareous
papilhe, in the structure of its skin-papillse (of the introvert-basis
as well as of the body proper) and in having the warts and the
perforated papilhe. Neither C. mollis nor C. javanicum have the
two structures last mentioned. C. mollis may be distinguished
from Ö. aspergillum by the presence of very small hooks and the
absence of eye-spots, — points of difference which exist also between
C. mollis and the present species. Lastly, the differences between
C. javanicum and the present species are most remarkable ; in
the former species one edge of the external facet of the calcareous
papilla Ijeing said to be considerably raised, and the aperture of
the papillary gland on the facet, elliptical in shaj)e. Besides
these two points, the characteristic broad form of the hooks and
the formation of the longitudinal muscles into bundles are quite
peculiar to 6! javanicum.

thp: gephyrea of japan. do

Tiie internal anatomy of the present «pecies may he briefly
outlined as follows :

The longitudinal inuscle is quite continuous. The retractor
muscle arises with two roots at the beginning of the posterior
fifth of the body j^rojier. Double intestinal spirals number about
12. Spindle muscle fixed at both ends. Only one fixing muscle
attached to the beginning of rectum. Segmental organs connected
with the body- wall along their whole length by mesentery. No
rectal gland. Two distinct eye-spots are present on the ganglion.

Gejius DENDROSTOMA.

DENDROSTOMA BLANDUM, Selenka à De Max.

(Figs. 14, 90 & 91).

Selenka, (28), 1883, p. 85.

The main characters of this species are summed up in the
following extract from Selenka's description :

Body of a stunted form ; introvert one-third as long as body
proper. Color brownish or reddish gray. On the middle region
of the introvert are scattered numerous large and sharply pointed
black hooks 0.4iu/ii. in height. Papillae in the middle region of
the body proper, low, oval, 0.07-0.09 m})i. in diameter ; in the

54 ART. 4. 1. IKEDA :

central part of each papilla there is a clear area surrounded by
small horny granules. Papillse on introvert-basis and on the
posterior end of body, slightly larger ; most prominent just behind
the hood-region of introvert. Small papillse also present in the
hooked region of introvert. Tentacles arise from several (5-6)
main stems. Two retractor muscles arise from the middle thii'd
v)f the body ]3roper. Two segmental organs about half as long-
as the body proper ; mesentery entirely lacking. Posterior
end of spindle muscle not attached to the body-wall. Three
fixing muscles ; one of them attached to the rectum and the
others to the œsophagus. Contractile villi remarkably long, but
few in number.

Two alcoholic specimens of this species were found in a
zoological collection made in the Hokkaido in 1900 by Me.
IzLTKA ; one of them is from Wakkanai (near Cape Soya) and
the other from the harbor of Aomori. The latter specimen (Fig.
14) is the better preserved, measuring 2.5 mw. in length of the
body proper. Both specimens are of a brown color in the pre-
served state. The introvert-basis shows on the skin-surface line
circular, and somewhat coarser, longitudinal furrows.

As to the skin-structures and the general anatomy, the
specimens agree within certain limits with Selenka's description
of Dendrodoma blanduiii which was many years ago obtained by
HiLGENDORF at Euoshima. But careful studies of my specimens
disclose some important points which either were omitted by
Selenka, or are irreconcilable with, his description. However,
I do not feel fully justified in establishing a new species on the
strength of these aj)parcnt ditïerences, since the specimens at my
disposal seem too few to be conclusive.

TTTE GEPHYREA OF JAPAN. 00

The papillîTe of tlie body proper and of tlie introvert, are
nearly of the same form, size, and structure as in Hilgendorf's
specimens. The scattered hooks also show the same form and
color as in those specimens, but are very much smaller (0.20 m???.
high), being only half as high as Hilgendoef's (OAOmm.).
The tentacles are reddish brown. In my specimens, they are
given off from the free margins of the eight arms produced by
the dichotomous branching of each of the four main stems (Fig.
90). Some few short tentacles are also present along the inner
margin of each main stem. The above mode of branching of
the tentacular stems is not described by Selenka, who on the
other hand states that the main stems are 5 or 6 in number.

In one specimen — that from Aomori — there are four fixing
muscles to the intestine, two on each side of the nerve-cord.
On either side, one muscle is attached to the œsophagus near
the posterior extremity of the dorsal vessel and the other to the first
intestinal spiral ; both arise from the body-wall close to each other,
far in front of the roots of the retractor muscles and widely separated
from the ventral nerve-cord. In the other specimen — that from
Wakkanai — one of the fixing muscles of the left side is wanting,
the single muscle present on that side dividing into two short
branches (Fig. 91). Both of the latter are attached to the first
spiral of the intestinal convolution (one to the œsophageal^ and
the other to the rectal, section of the intestine. Close to the
entrance to the convolution, there is present on the rectum a distinct
blind diverticulum {rg, Fig. 91), an organ which has not been
mentioned by Selenka from Hilgendorf's specimens. The
fixins; muscles are stated bv Selenka to be onlv two in number.

Öß ART. 4. 1. IKEDA

DENDROSTOMA SIGNIFER, Selenka & De Man.

(Fig. 15).
Selenka, (28), 1884, p. 86.

The main features of the species ave embodied in the follow-
ing extract from ï-^elenka :

Body of a stunted form, tapering posteriorly to a sharp
point ; introvert about one-fourth as long as the body proper.
Color brownish or yellowish red, wâth a dark blue band at the
middle of the introvert. Skin-bodies very small (0.06 //rni. in
diameter), quite flat ; represented by oval areas, from the center
of which a narrow canal leads down to the granular layer below.
Hooks generally wanting. Tentacles very numerous, springing
from 5 or 6 main stems. Eye-spots present.

Longitudinal muscle continuous. Two strong (ventral) re-
tractor muscles arise from the beginning of the posterior third
of the body proper. Intestinal spirals lo-20 in number, tra-
versed by a spindle muscle not fixed to the body-wall at its
posterior end. Three fixing muscles : one attached to rectum and
two to œsophagus. Contractile villi very numerous, long, and
branching.

Four specimens (Fig. 15) of the above species were collected
in the Riukiu Islands. They were found in holes bored through
dead corals. All measure about 30 mm. in total length. They
tally well with Selenka's description of the Phillippine specimens,

THE GEPHYREA OF JAPAN.

57

except in the facts that the main tentacular stems are constantly
four in number (instead of 5 or 6) and that the hooks are entirely
wanting in all the specimens. There occurs on the rectum a
distinct blind diverticulum such as has not been noticed in any
species of Dendrostoma yet described.

DENDROSTOMA MINOR, n. sp.

(Figs. 16, 92-95).

This species is a small form, not exceeding 15 mm. in length
and 3 mm. in breadth. The introvert is one-third as long as the
body proper. The worms in the fresh state are of a yellowish
brown color, only the anterior portion of the introvert being of
a light violet tint. The whole surface of the skin is covered
with small papillœ which are nearly flat in the middle main
region of the body proper. But in the anterior and posterior
regions, the papillœ are somewhat raised and appear as oval
tubercles of 0.02-0.03 mm. height and 0.10 mm. long diameter
(Fig. 92). They are formed of numerous polygonal chitinous
plates which are larger towards the center of the papillae ; one
plate in the center contains the opening of the subdermal gland.
In the hooked region of the introvert, the papillœ are much
smaller, but are of nearly the same height, appearing like a
hemispherical tubercle resting upon a broader basal elevation
(Fig. 93). Hooks, about 50 in number, are found scattered in
the anterior parts of the introvert (Fig. 94). They are 0.04 mw.
high, blackish brown in color, slightly curved and bluntly pointed
at apex. Specimens from Funakawa in the province of Ugo (on

58 ART. 4. 1. IKEDA :

the Japan Sea) differ from those from other localities in having
fewer but larger hooks, 20-30 in number and about 0.06 mm. in
height. The tentacles are given oif from four stems : these latter
split each into two branches ; each branch again divides into
4-6 long tentacles which are of a violet color except at the
bright yellow tip.

The longitudinal muscle is continuous. There exists only
one pair of retractor muscles (the ventral), originating in the
middle of the body proper or at a somewhat more posterior
point. The dorsal vessel, running along the dorsal side of the
œsophagus, bears numerous, tolerably long, contractile villi [cv,
Fig. 95). The intestinal convolution, consisting of 7-12 spirals,
is loose and irregular, especially in young specimens. The spindle
muscle is not attached to the body proper at the posterior end
of the body. The rectal portion of the intestine is tolerably long
and bears at its beginning a small blind diverticulum {rg). Of
fixing muscles there are present sometimes two, sometimes only
one and in still other specimens, none at all. When present in
a pair, both are attached to the first intestinal spiral and end
on both sides of the nerve-cord ; when present singly, the muscle
is on the left side. The segmental organs {so) are about one-
third as long as the body proper, and are entirely devoid of a
mesentery. Their external apertures lie at the same level as the
anus. Eye-spots are present on the ganglion.

The present species in the manner of the branching of its
tentacles and in their color resembles Dendrosto7na alutaceum
Grube, but is quite different from that species in the form, size,
and structure of the papillœ, in the size of the hooks, in the
number of contractile villi and in the total size of the worm-body.

Habitat : Specimens of the species have been collected at

THE GEPHYREA OF JAPAN. ^^

FunakHwa in the Province of Ugo, Misaki, Habu on Oshiina
(Izu Province), and Tanegashima, a small island near Kagoshima.
The species lives under alga^ covering rocks along the tide-marks

ECHIUROIDEA.

Genui ECHIURUS.
ECHIURUS UNICINCTUS, Von Dräsche.

Selenka (29), 1885, 1). 385.
RiETSCH (26), 1886, 11. 188.
Shipley (31), 1899, 77. 344.
Embleton (8), 1900, p. 77.

Shipley's diagnosis runs as follows :

" Length of body averages 8.5 cm. ; length of proboscis, ^-^

when contracted. Color, bright yellowish brown. Papillœ
uniform in size, only arranged in transverse rows in special places,
e.g., near the hooks. Single circlet of bristles, usually 11, but
any number from 9-13 has been found. Circular muscles consist
of 200 bundles frequently anastomosing. Nephridia, two pairs,
with spirally coiled internal openings."

This species is the most abundant Echiuroid in Japan, and
is the only representative of the genus in this part of the world.
The worms are known to live in shallow sandy inlets along the

mm

60 ART. 4. 1. IK ED A :

Pacific coasts of Japan proper, south of the Tokyo Bay. As
far as I know, they are exceedingly abundant in shallow parts
of that bay, for instance, at Inage (Province of Shimosa), Haneda
(Province of Musashi), and at Tateyama (Böshiu). From the
last mentioned locality specimens of a large size (25 cm. or even
30 cm. in length and 3-4 cm. in breadth) have been obtained.
Tomo (Bingo) and Hiroshima on the Inland Sea may also be
mentioned as localities where this species is known to occur in
great abundance.

As to the systematic characters and the microscopical anatomy
of the species I have nothing to add to the exact and detailed
studies of Selenka and Embleton. Of much interest to me is
the observation by the latter author that in the present species there
is no true vascular canal at all — a fact which I have also observed
independently.

Genus THALASSEMA.
THALASSEMA KOKOTONIENSE, Fischer. (?)

(Fig. 17).

Fischer (g), 1892, p. 2.
Shipley (34), 1899, p- 337.

During a trip in 1901 to the Amami-Oshima and and Riukiu
Islands, I collected in all 10 specimens which I provisionally

THE GEPHYREA OF JAPAN. 61

refer to the above species. In the preserved state they pre-
sent nearly all the essential characters attributed to the species
by Fischer. But in the fresh state the worms were found to be
of a characteristic coloration, quite different from that known from
typical T. kohotoniense. Fischer's description of the internal
anatomy is so brief, that it does not help me much and so
I must say I have considerable doubt as to the correctness
of my identification of the specimens.

The body in the fresh stete attains 'o.b cm. in length and
15 mm. in width, the proboscis being much shorter and narrower
than the body (Fig. 17). The proboscis is straw-coloured, edged
with deep green on the free margin ; the body is of a grayish
yellow color, its posterior end being suffused with green. When
the circular muscles contract, there appear a series of longitudinal,
deep red, dotted lines over the body-surface (see Fig. 17). The
papillae are larger in the posterior end of the body than in other
regions, where they are arranged in transverse rows or rings on
the skin.

The longitudinal muscle is always divided into 18 bundles.
The segmental organs are present in three pairs ; they are colorless
and transparent and are all provided with two short spiral lobes.
The first pair lie in front of the ventral hooks. The anal glands
are one-third as long as tlie body-length and are fixed at their
bases to the body-wall by 4-5 small muscles. Over their whole
surface are scattered numerous small funnels which have very
short stalks. As has been noted by Fischer, the terminal part
of the rectum bears a large rectal gland which is connected with
the ventral nerve-cord by a short delicate mesentery.

Habitat : The species was found burrowing in coral sand at
Naze (Amami-Oshima) and at Naha (Kiukiu).

62 ART. 4. — I. IKEDA :

THALASSEMA OWSTONI, n. sp.

( Thalassema diaphanes Sluiter. ?)
(Figs. 18, 96 & 97).

The single specimen, on which this species is based, was
dredged by Mr. Owston together with the type specimen of
Phascolosoma owstoni, from a depth of 180 fathoms in the Uraga
Channel. It is small, measuring 17 mm. in length and 5 mm. in
breadth. The proboscis, about 7.5 m.7n. long, is deeply furrowed
longitudinally. The color, as preserved in formalin, is a pale
reddish brown. The skin is soft and smooth, except at both
extremities of the body proper, where the papillae are largest and
most closely aggregated. In the main middle region of the body
the papillte cannot be detected with the naked eye. There are
two comparatively large ventral hooks of a bright yellow color.

The longitudinal muscles of the body-wall are thoroughout
continuous. Only a pair of small sac-like segmental organs are
situated just behind the ventral hooks (Fig. 96, so). They have
no spiral appendage, but are provided with a large ciliated funnel
(Fig. 97, fn). The erector (or the radial) muscles of the hooks
are well developed {em, Fig. 96) ; the interbasal muscle is very
long. Unfortunately the specimen is torn near the posterior end
of the body, so that the greater part of the alimentary canal
and of the anal glands has been lost. Inspite of this defect it
can be ascertained that there are present small ciliated funnels

THE OEPHYREA OF JAP AN. 63

on the last mentioned organs. The funnels are large, situated
close together and borne on short stalks. The specimen is a
female and the segmental organ of the right side is filled with
ripe eggs.

The present species seems, as far as I can judge, to be most
closely allied to Thalassema diaphanes Sluiter. I regret to say
I have been unable to get access to Sluiter's original description
of that species.* From what Shipley (34) says of it in his
revision of Echiuroids, it seems that the two species agree in
having only one pair of segmental organs, in the continuous
longitudinal muscle, and in wanting spiral lobes at the internal
openings of the segmental organs. But the specimen in my
possession shows in other points considerable deviations from
Shipley's description {l.c, p. 336 and 346) and figure {I.e., PL
XXXIII, 2) of T. diaphanes, which fact has led me to base a
new species on it. The most important points of difference
consist in the smaller size of the body, and the thicker body-
wall in the present species, and in the manner of distribution of
the papillae.

THALASSEMA TiENIOIDES, n. sp.

A full description, with figures, of this most remarkable new
species has been prepared for publication in a separate paper.f
Here only the specific characters will be given in brief.

The species may be said to be of gigantic size, the body
proper reaching 40 cm. in length and 2-3 cm. in breadth, while

* Naturk. Tijsch. Nedorl. Ind., Vol., XLVIII, 1888.

t " On Three New and Remarkable Species of Echiuroids," soon to appear in this Journal.

64 ART. 4. T. IKEDA :

the proboscis, when fully extended, is as long as 100-150 cm.
The body proper is of a deep reddish brown color ; the proboscis
bears narrow transverse stripes of brown. The papillae are ir-
regular in shape, and are largest and most closely crowded together
at both ends of the body proper. The ventral hooks are relatively
small and brownish yellow in color. In the fresh state, there
are seen on the body proper five narrow longitudinal stripes of
a pale yellow color, corresponding to special thickenings of the
longitudinal muscle. The bulbous bases of the paired hooks are
joined by a slender interbasal muscle. The alimentary canal
pursues a course which is peculiar to this species. The pharynx is
accompanied by two large wing-like muscles. The anal glands are
6-7 cm. long, fixed at the apex by a long muscle ; they appear
to be hairy on account of the presence of very long funnels.
No rectal gland is present. As the heart lies in a very posterior
situation, the dorsal vessel is remarkably long. The segmental
organs are extremely abundant ; in the larger specimens, they may
number not less than 200 on one side of the ventral nerve-cord,
arranged not in pairs but apparently without order. They are
small, measuring 10-12 mm. in length ; the internal opening
si terminal in position.

Habitat : The species inhabits shallow muddy bottoms in
deep vertical holes, from the opening of which the long worm-
like proboscis is extruded. It has hitherto been obtained at
Tsushima ; Bay of Amakusa (Kiushifl) ; Tomo and many other
localities in the Inland Sea ; Tateyama (Bôshiù) ; Haneda (Tokyo
Bay) ; and Misaki. In the last locality the species is common.

THE GEPHYREA OF JAPAN. 65

THALASSEMA ELEGANS, n. sp.

This highly interesting species will also be fully described
in a separate paper together with the preceding. Here I restrict
myself to giving only the most salient characters of the species.

Externally the species looks very much alike Thalassema
tœnioides, though there exist remarkable differences in internal
structure. In size, shape and color it shows a far reaching
agreement with that species except as regards the proboscis.
This in the present species is 30-40 cm. long and 1.7-2 cm. wide,
and is of a bright yellow color which becomes deeper towards
the mouth. The ventral surface of its anterior portion, for a
length of 6-7 cm. is marked with a fine and irregular net-work
of a bright green color. As in T. tœnioides, there are found
on the body-surface longitudinal stripes of a pale yellow color,
which however number 10, instead of 5. The ventral hooks
show nearly the same size, form, color and position as those
of T. tœnioides ; but internally they are not joined by an in-
terbasal muscle. The circular muscle of the body-wall is
separated into more or less regular bundles. The segmental
organs are numerous (in one specimen 27 were counted) and
are distributed generally in groups of which there are 7 pairs,
each consisting of 1-3 segmental organs. Aside from the pair-
wise arrangement indicated, there seems to obtain no definite
order in the manner of the occurrence of single segmental organs.
They represent large elongate sacs, each bearing two compara-
tively short spiral lobes. The pharynx is joined to the body-

Q6 ART. 4. 1. IKEDA I

wall, but is not accompanied by a mesenterial muscle. The
anal glands are long slender tubes, about 20 cm. in length, and
are provided Avith small, short-stalked funnels. No rectal gland
is present.

Habitat : The species inhabits shallow muddy bottoms, in
deep, somewhat obliquely sunk pits. It occurs together with T.
tœnioides in the inlet of Moroiso, close to the Misaki Marine
Laboratory.

THALASSEMA GOGOSHIMENSE, n. sp.

(Fig. 19).

The present species (Fig. 19) most closely resembles Thalas-
sema mucosum both in color and external shape, with the excep-
tion of the facts that the body is much larger and the papillary
spots more deeply colored with a dirty green pigment. The
dorsal surface of the proboscis is covered with small green
spots, while its free margin is colored yellow. The total length
is, in the fully extended state, about 15 cm. of which 10.5 cm.
appertain to the proboscis ; the maximum width, about 2 cm.
Over the surface there are closely strewn moderately large
papillae which are slightly larger and more crowded towards
both extremities of the body proper. The two ventral hooks
of a golden yellow color are found in the usual situation just
behind the mouth.

THE GEPHYREA OF JAPAN. 67

Inspite of its external resemblance, however, this species
differs remarkably from Thalassema mucosum in internal anatomy.
In this resj)ect it shows, contrary to expectation, an essential
agreement with Thalassema elegans. Indeed, the agreement may
be said to be complete, the only difference being that all the
visceral organs in the present species are developed on a smaller
scale in proportion to the smaller size of its body. It therefore
seems unnecessary to give a description of the anatomy. How-
ever, certain points with regard to the segmental organs of the
species require special mention. In the females, the said organs
as far as they have come under my observation, are invariably
present in three pairs, all situated behind the ventral hooks.
On the other hand, the males show these organs in varying
numbers and arrangement, a fact which seems to be of interest
in relation to the polynephridism I have observed in T.
tcenioides and T. elegans. In one male specimen of the species
I have found them in eight groups arranged pairwise, each
group consisting of four organs, making thirty-two in all.
In some other male individuals, the organs were present in
three paired groups, some of which comprised four and
others only one or two organs. As to the shape of the organs
in question, there exists scarcely any difference from those of
T. elegans.

Habitat : I have found the species very common in the
shallow waters around the islet of Gogoshima (Province of lyo).
The presence of the worms is indicated by the large and pecu-
liarly shaped faecal masses scattered around small elevations of
the bottom exposed at low tide. Lately I was presented by
Mr. Izuka with a specimen of this species which he obtained
in the inlet of Moroiso near the JMisaki Marine Laboratory.

68 AßT. 4.— 1. IKEDA :

THALASSEMA MUCOSUM, n. sp.

(Figs. 20 & 98).

Three specimens of this new species have been obtained
from the mudd}^ shore of Moroiso near the Misaki Marine
Laboratory.

The cylindrical body in the live state reaches 7 em. in
length and 1.8 cm. in width ; the tubular proboscis is much
shorter and narrower, and has a truncated apex. The body-
wall is thin and is of a light green color, which becomes deeper
at both extremities and also along the margins of the proboscis.
The papillae, large enough to be visible with the naked eye, are
also deep green ; they are closely aggregated, especially at both
ends of the body proper. In the fresh state, the body-surface
always appears dirty owing to thick mucous spread over it.
The two ventral hooks are small, slightly recurved and bright
yellow in color. They are provided with rather short radial
muscles {em, Fig. 98), but are devoid of an interbasal muscle.
The longitudinal and circular layers of the dermal musculature
are not divided into separate bundles. There are three pairs of
comparatively short small segmental organs {sg), all of which
lie behind the ventral hooks and have long spiral lobes. The
pharynx (ph) is accompanied by two weakly developed mesen-
teries {wm) arising from the body-wall, close to the nerve-cord-
The mesentery on the right side is much longer than that on
the left. The neuro-intestinal vessel (ni), instead of being

THE GEPHYREA OF JAPAN. 6*J

single, appears as two vessels, which unite into one to communi-
cate with the ventral vessel. There are two long anal glands,
the anterior ends of w^hich extend to the middle of the body ;
a few fine muscular strands join their posterior portions to the
body-wall. The ciliated funnels of these glands are small, their
stalks being very short. No rectal gland is present.

This species seems to be closely allied to both Thalassema
vegrande Lampert and T. Mœbii Greeff, as judged from the
agreement in internal structure, but it may readily be dis-
tinguished from either. Unlike the present species, 2\ vegrande
is said to be without proboscis and also without funnels on the
anal glands. Of the proboscis, Shipley (32) says : " Like
Lampert I had only a single specimen, and, like his, mine
had no proboscis. There was no trace of one and no scar to
indicate that there had ever been one, and I am inclined to
think that this species may be without a proboscis."

In contrast with Thcdassema 3Iœbii the present species
exhibits considerable difference in external features as a com-
parison of Greeff's (14) figure of the former with mine of the
latter (Fig. 20) will show. The difierences mostly concern the
size, form and coloration of the proboscis as well as of tlie
body proper.

THALASSEMA FUSCUM, n. sp.
(Figs. 21 & 99).

In external features the present species closely resembles
Thalassemia mticosum, except in color. The long cylindrical
body proper is about 7.5 C7u. long and 15-18 mm. thick, while

70 ART. 4. r. IKEDA :

the proboscis is about 2 cm. long and truncated at the apex.
The skin is of a slightly pinkish orange-yellow which becomes
reddish on the edges of the proboscis. Large papillse are scat-
tered irregularly over the entire body-surface, more thickly at
both ends than in the middle. The two golden-yellow ventral
hooks are relatively small in size and are supplied with well
developed radial muscles {em, Fig. 99). The longitudinal dermal
muscle is perfectly continuous. There exists only one pair of
comparatively long (about 20 m??^.) segmental organs (so), attached
to the bod}'- wall just behind the ventral hooks. The opening of
the organs appears as a large funnel {fn) unprovided with spiral
lobes. The pharynx as well as the œsophagus are entirely with-
out a mesentery or muscular strands. The anal glands are about
as long as the body proper ; the ciliated funnels, sparsely dis-
tributed over the surface, are relatively small and short-stalked.
No rectal gland is present.

There have hitherto been known four Thalassema species
which, like the present, have a continuous longitudinal muscle
and only one pair of segmental organs, viz., T. diaphanes
Sluiter, T. fœx Selenka, T. gigas Müller and T. Lankes-
teri Herdman. The first further agrees with the present
species in one other point, viz., that the segmental organs lack
spiral lobes ; nevertheless, the former shows several important
characters which are entirely wanting in the latter. Of the three
remaining species mentioned, T. gigas seems to be nearest to the
present species in respect of general internal anatomy ; but, that
species, as figured and described in detail by Greeff (14),
should be easily distinguishable from the new species here
described, owing to the differences in size, color and general
features of the entire worm-bodv.

THE CEPHYREA OF JAPAN.

71

THALASSEMA INANSENSE, n. sp.
(Figs. 22 & 100).

A single specimen of this new species was obtained by me
(1901) on a coral reef, called Inanse, about three miles off Naha
(Okinawashima). It was found under a large and living Mad-
reporite mass. The body in the fresh state was about 4.2 cm.
long and 13 mm. thick, while the proboscis was about 23 mm.
long. The latter appeared to be tubular but laterally expanded
at the apex so as to present a bilobed appearance. The body
proper was of a bluish violet color, spotted with small, uniformly
and closely set papilla of a dirty green color. The proboscis
was yellowish gray, edged with brown on the lateral margins.
The two very small ventral hooks are bright yellow.

The longitudinal muscle is continuous. There were three
pairs of segmental organs {so, Fig. 100) filled with eggs, the
most anterior pair lying in front of the ventral hooks {hk).
The organs appear like elongate sacs and are provided with well
developed spiral lobes. The anal glands (ag), of a brown color,
are remarkably long and reach up to the level of the second
pair of segmental organs. Very close to the anterior end they
are joined to the body-wall by a slender muscular string. Small
funnels with short stalks are sparsely distributed over the surface
of the glands. The posterior terminal parts of the rectum are
fixed to the body-wall by two muscles, one on each side. No
rectal gland is present.

72 ART. 4. — I. IKEDA :

Genus BONELLIA.

BONELLIA MINOR, Marion.

(Figs. 23, 101 & 102).

ElETSCH (26), 1886.
Shipley (34), 1899, p. 340.

About forty specimens of the above species were collected
by me, in April, 1901, in the Kiukiu Islands. The body
reaches about 2 cm. in length and 7 mm. in maximum width ;
the proboscis is 2-2.5 times as long as the body and much
narrower (Fig. 23). The color is deep green on the body and
a lighter green on the proboscis ; the anterior ventral portion of
the body shows through the body-wall a reddish color caused by
the sexual cells in the oviduct. Thus, in the color of the female
the species approaches closer to Bonellia pumicea Sluiter (36)
than to the Mediterranean specimens of B. minor ; but the
anatomy of the female as well as the general structure of the
parasitic male, essentially agrees with the description and figures
of the latter species given by Rietsch except in the one point
that the anal glands in the female Japanese specimens bear short
branches, from each of which arise about 5-7 funnels which
are relatively longer than in the Mediterranean specimens (see
Fig. 101).

Fig. 102, drawn from a stained preparation, represents a
male which was found in the pharynx of a female. The males

THK OEI'HYREA OF JAPAN. 73

i^enerally measure 1 mm. in length and O.l-") nun. m maximum
breadth, tapering towards the ])osterior end. As was pointed
ont by RiETscii, tiie skin-surface is thickly ciliated except on
tlio dorsal side where the cuticula is comparatively thicker than
in other regions. Two long and strongly curved hooks (/»/•) are
present in an anterior position ; behind tliem is seen a large
spermatic reservoir {sv) provided with a small funnel. The duct
of the reservoir begins just l)ehind the hooks and proceeds
anteriorly as a narrow tube to open at the apex of the body.
'Ulis canal {vd) is surrounded by a nerve-ring lying just in front
of the roots of the venti'al hooks.

iral)itat : The species is common and readily discoverable in
shallow waters in the vicinity of Naha (on Okinawashima, one
of the Eiukiu Islands). In May, 1900, a few specimens were also
collected by Mr. jNTiyajtma at Kataura in the Province of
Satsuma. The worms live in holes which they bore in dead corals.

BONELLIA MIYAJIMAI, n. sp.

A tull description of this species together with figures, will
be given in another paper which will appear in this Journal.
Here will be given only tlie more important characters whicli
serve to distinguish the species.

The l>ody nieasures about 2 cm. in length and 1 cin. in
maximum l)readtli, the proboscis being six times as long as the
body. The color is grayish brown. The ventral hooks, instead of
being present in a single pair, number 29 in all ; they are ar-
ranged on the ventral surface in no definite order. The internal

74 Am. 4. — r. tkeda :

opening of the unpaired oviduct is found almost at the free end
of its thin-walled portion. The main canal of the anal glands
divides twice Ix-fore ending in funnels.

The male worm is of an unnsually large size, measuring
nearly 28.5 nfm. in length and 2 j/tm. in breadth. It l)ears no
ventral hook. The alimentary canal is found broken up into
innumeral)]e separate pieces which float free in the body-cavity.
The vas deferens is internally provided with four large funnels.

Habitat : A single female of the species was discovered on
breidsing a dead coral, on a coral beach near Tomari, a village
two miles from Naha (Okinawashima).

BONELLIA MISAKIENSIS, n. sp.
(Figs. 24, 103-10Ö).

In duly, 1(S96, Mr. Tokt^vacja obtained the unique type
specimen of this new sjiecies on the southern shore of Jöga-
shima,, Misaki. It was found under a large stone near the beach.
At the time some sketches and notes were made of the worm
by the collector. jNTr. Nishikawa also made a colored sketch
of the worm, which is reproduced in Fig. 24. To both gentle-
men I am indebted for the material which they j^laced at my
disposal.

According to ÄIr. Nishikawa's observation on the living
worm, the body proper measured 40 w?w. in length when con-
tracted and 60 mw. when fully extended; the proboscis and its
arms measured respectively 50-90 mm. and 25-30 m7n. in length.
The proboscis as well as its arms were found to be deeply

THE GEPHYREA OF JAPAN. 70

I'liiTowed along the midventral lijie. The form of the body
])roper was veiy changeable, but assumed an elongate sac-like shape
when preserved. Tlie ground color was a light grayish brown ;
the l)ody proper and the greater part of the proboscis were
densely marked with bluish black sports, while the lateral arms
of the proboscis were of a dirty yellow coloi'. As shown in
Fig. 24, a number of slender spine-like bodies 2)rojected from
the entire external surface of the worm.

In the preserved state I have been unable to find the spine-
like bodies just mentioned. They had probably fallen oif during
tlie killing process ; and judging from a figure of the same,
drawn by Mk. Tokunaga, it seems highly probable that they
Avere nothing but parasitic organisms.

Peculiar to this species is the fact that no ventral houk is
present and, consequently, no muscular sheath of the same.

Fig. 103 represents the anatomy of the specimen. The
oviduct {od) is a sac lying on the left side of the nerve-cord,
and is nearly as long as the body proper and is fully distended
by the ripe eggs contained in it. Anteriorl}^ it is greatly narrowed
and its wall highly nuiscular, while the remaining parts are, on
the contrary, much swollen and very thin-walled. The internal
opening or funnel, unlike those in all other previously known
Bonellia species, is not situated near the anterior end of the
organ, but is found in the posterior terminal portion. The
anal glands {ag) are comparatively short and are branched at
least three times before ending (Fig. 104). Ten to twenty-five
funnels a.)e attached to each of the tertial branches — of course
some others being attached directly to the secondary branches.
Most of the In-anclies have a fine fixing nuiscle joining them to
the bod V- wall.

76 ART. 1. r. IKEDA I

In the ulimeiilary cuiiul aiul tlie vascular system, there is
110 point requiring special mention.

A male worm was discovered in the anterior muscular portion
of the oviduct. It is shown in Fig. lOo under low magnifica-
tion {"Xoß). Length about o.'S //u/i ; maximum breadth 0.20 /m//a.
The body narrows towards both ends. The entire surface is
thickly ciliated. No ventral hook is present. There is an
unpaired vas deferens {vd), about one-third as long the body-
length. It is thickest (0.07 ?nm. across) in the posterior two-
thirds of its length, ending with a relatively small funnel if''")-
The anterior one-third is slender, running forwards througli
the body-cavity finall_y to o})en to the exterior. The exterual
opening is situated not at the very tip of the body, but at a
])oint a short distance behind it on the ventral side (x. Fig.
105). The alimentary canal is found in a state of degeneration
similar to that observed in the male of Bonellia »iiyajimai. It
does not form a continuous tube, l)ut is broken up into about
ten small pieces of different lengths and sizes [pa, Fig. 105).
These pieces are scattered in the body-cavity, being apparently
in no definite relation to one another. The blood corpuscles and
sperm-masses are found in abundance, Ijotli floating free in the
body-cavity.

As to further details of the histology of the male, I have
been unable to obtain definite results, owing to the imperfect
preservation of the specimen.

THE GEPHYKEA OF JAPAN. 77

Key to the Genera and Species of Gephyrea treated
of in this Paper.

SipuncuJoUJea,

I. The longitudinal muscle of the body-wall is divided into J.)-ol
.se|)arate bundles. Two or four retractor muscles present.

A. — Body covered with well develi:)[)ed papilla?. Hooks usually
present on the introvert. Filamentous tentacles arranged in
a single semicircle above the mouth. Retractor muscles com-
monly in two })airs (except in two s|)ecies, PJuiniosonn' oiio-
mlchianum and Fh;/inostuua iKiI/aciisr, both which are cliarac-
terised by having only two muscles). No rectal gland, no

ventral ves'-el. Two distinct eye-spots on tlie ganglion

Geiiits Phymosoma. R 78.

B. — Body devoid of papill;o. Tentacles tilamentous or flap-like,
in the latter case they spring from the tentacular membrane.
Hooks usually absent. Rectum with one or more cteca.
Usually a ventral vessel in addition to the dorsal vessel.
No eye-spot Genus Sipunculus. R 79.

II. The longitudinal muscle of the body-wall is perfectly continuous.

A. — Two segmental organs and a si)indle muscle present.

a. — Numerous tentacles. »Spindle muscle is sometimes
fixed posteriorly to the body-wall. T^sually two

pairs of retractor muscles

Genus Phascoi.osoma. R. ~i\).

b. — Tentacles s[)ring from foar main stems. Hooks, if
present, large, but few and scattered. Spindle
muscle posteriorly free from the body-wall. Onl}^

one ]»air (ventral) of retractor ujuscles

(reiius Denduostoma. I'. 8U.

B, — Only one segmental organ ; spindle muscle wanting.

c. — Numerous tentacles arranged in a circle round the
mouth. Two retractor muscles. The worms be-
longing to this genus inhabit Molluscan shells

Genus 1'hascoijox. R. 80.

78 ART. 4. — I. IKEDA :

HL At butli t'iid« ul' the body [)roper, a hurny ur ciilcareoiis s^hield.
Hüok:s |iieseiit. Longitudinal muscle continuous. Only two retractor
inuscles (fused together generally from their origin).

c:7. — A horny (rarely more or less calcareous) shield at
both ends of the body proper. The introvert is
excentric, arising i'rom the ventral side of the an-
terior or anal sliield. Tentacles few and arranged in

a single semicircle Geiiu-s Aspidosu'Hon. P. 81.

It. — A spiral series of calcareous papilh«3 forms the anal
shield ; the posterior, or caudal, shield is wanting.
The introvert is extruded from the centre of the
anal shield Genus Clœosiphon. P. SI.

PHYMO^OMA.

I. 4 retractor njuscles.

A. — Dorsal vessel devoid of contractile villi. No collar-like skin-
fold behind tentach^s.

a, — A single 1)i furcated lixing ujuscle. Tentacles ab(nit 12.
The longitudinal muscle divided into 20-21 bundles.
Crescent-shaped markings on the dorsal side of

introvert R svolops. F. 20.

b. — Fixing muscle single and simple.

1. — Tentacles about 28. The lougitudiual muscle
se[)arated into about 30 bundles. Papillae pro-
minent, tegmental organs long and accom-
panied by a mesentery along their entire

length P. Japonu'um. F. 22.

2. — Tentacles 35-40. Longitudinal muscles, 30-35.
Papillœ remarkably prominent. »Segmental or-
gans short and accompanied by a mesentery

along tlie anterior half of their length

F. pucijivuiR. F. 25.

B. — Dorsal vessel with contractile villi. A collar-like fold be-
hind tentacles.

Tentacles very numerous (50-80). PapilUe
small and low. Ijongitudinal UJUscle varying in
the nundier of l)undles P. aidillari'iii. F. 24.

THE fJErUVREA OF .TATAX.

79

II, 2 nitraetor innsclcs.

f(, — Hooks provitU'd with one apical tooth. Dorsal vessel

devoid of contractile villi 7'. nahuouHc. P. 20.

ft. —Hooks wantiii'j;'. Papillae prominent. f 'Ontractile

villi numerous and ramified

T. oiiomichinjium. P. 2(!.

SIPUNCÜLÜS.

j^,_Tentacles flap-like. The chitin over the body proper folded
into rectangular areas. Introvert with very tall, posteriorly
directed papilla>. Longitudinal muscle separated into ?>\

Imndles '^'- i""^"^-- I'- •>'•

B.— Tentacles filamentous. Body-surface shows no rectangular
areas. Introvert and body proper covered with skin-bodies.
«7.— Longitudinal muscles 18-21. Crescent-shaped dis-
sepiments on the inner side of the body-wall. With
a bifurcated fixing muscle.... 6'. cumanensis. P. 32.
/,. —Longitudinal muscles 15. Devoid of crescent-shaped

dissepiments. Fixing muscles 2

,S'. (irnii niicn^i-'i. P. 3S.

PHASCOLOBOMA.

I. 2 retractor muscles.

No hook. Introvert long. Body as a whole coiled
into a spiral P- nif/mm. P. 'X

II. 4 retractor muscles.

^,__With hooks and 1-2 fixing muscles.

a, Hooks provided with 4-5 small accessory teeth and

arranged in ring-rows. A single fixing muscle.
Segmental organs consist of two elongate lobes

directed oppositely P. mhal-ianum. P. 7.

/;. A somewhat larger form. Hnoks dispersed, almost

straidit, and devoid of accessory teeth. 2 fixing
muscles. Segmental organs of an ordinary shape....

7^. (wsfoiii. P. 12.

80 ART. 4. r. IKE DA I

B. — No hooks; witli 3 fixing muscles.

a, — Eye-spots présent. No respiratory apparatus on the

body ]->roper. lÀi\.orSi\ïox\n. ...P. japonicum. P. 5.

1). — Eye-spots wanting-. Respiratory apparatus present

over the surface of the body proper. Deep-sea

form P. okinoseanu'in. P. 9.

DENDROSTOMA.

I, Hooks generally wanting.

A dark-blue marking at the middle of the introvert. 3 fix-
ing nuiscles. Contractile villi numerous, long and branched.

Spindle muscle posteriorly free from the body-wall

,,D. signifer. F, 5G.

II. Hooks present.

A. — Tentacles short but very numerous. Hooks very large
(0.20 mm. higli). Contractile villi few, but remarkably long.

3 or 4 fixing muscles D. hlamlum. P. ."53.

B. — Tentacles relatively long, aboiit S-12 of which branch off
from one of the 4 main stems. Hooks small (0,04-O.OG m)n.
high) and about .10 in number. Contractile villi short and

niimei-ous. Fixing muscles indefinite in number (0-3)

I), nnnar. P. ;">?.

PHASCOLTON.

I. Body straight. Papillary bodies are in general small and the middle
part of the body proper seems almost smooth. A single

retractor muscle with two short roots. 4 fixing muscles

F. rcchiK. F. ir).

II. Ijody coiled into a spiral. The middle part of the body proper cover-
ed with especially large papillre. 2 retractor muscles, one
of \vhich is nnich more slender than the other. No fixing
muscle /'. artificiosus. P. 18,

TFir: (lEPirviiEA of japax. 81

ASPIDOSIPHOX

I. Skin-papilla3 distinct. A single fixing muscle and a rectal gland
]iresent.

A. — Both anal and caudal shields present.

(I. — The two shields radially grooved.

I-. — Longitudinal nniscle separated into very irre-
gular bundles. Retractor muscles arise from the
inside of the caudal shield.... J. truncatuf<. P. .38.
2. — Longitudinal muscle separated into more or less
regular bundles. Retractor muscles arise far an-
terior to the caudal shield.... J. sfeenstrup/i. 1\ 40.
/>. — The peripheral border of the anal shield only is
lightly grooved.

Longitudinal muscle continuous. Retractor
muscles arise just in front of the caudal shield.

A. ')iti.'^al-/rnms. F. 41.

B. — Caudal shield indistinct.

The anal shield is of the shape of a tall, ob-
liquely directed cone ; it is very thick and
calcareous, and its siuface roughened by the
presence of distinct radial ridges and grooves.,,.

A. nm'i^r/itdfns. P. 43.

II. No papillîo over the greater part of the body pro])f-r. No fixing
muscle or rectal gland.

ff. — The two shields show a jirominent circular ridge, on
both sides of which are |)resent longitudinal ridges
and grooves. Hooks have a bluntly |)ointed <ipex.

No fixing muscle or rectal gland

A. aiignlatm. P. 4.5.

h. — The two shields show a granular appearance.
Hooks two-toothed at apex. The greater part of

the introvert-surface covered with scattered spines

A. sj>nia//s. I'. 47.

CLOŒSIPHON.

Only one species is known in this coiuitry C. Japoiiicinn. P. 49.

.S2 ART. 4. T. IK EDA

FA'hliivoiilcu.

I. Proljoscis not bifurcated at apex, l>ut truncated transversely. Sexes
of similar appearance.

A. — Anal hooks arranged in 1 or '2 r-ircles. 2 pairs of segmental

organs Geiiits Echiuuus. F. 82.

B. — No anal hook. 1-3 pairs or more of segmental organs

Genus Thalassemia. P. 82.

II. Proboscis bifurcated at apex. Sexu.il dimorphism remarkably pro-
nounced,

Xo ana] hook. Unique segmental organs

Genm Bonellia. P. 83.

ECHIURUS.

Anal hooks arranged in a siugle circle. No definite blood vessel.
E. wit'cinctus. P. 59.

THALASSEMA.

I, Longitudinal muscle continuous.

A. — Segmental organs in only one pair ; the internal openings
of the organ lack spiral lobes.

a. — Body of a small size. Ventral hooks provided in-
ternally with a well developed interbasal muscle

T. owstoni. P. 62.

b, — Skin-papillie prominent over the whole surface. No

interbasal muscle T. fuscum. P. 69.

B. — Segmental organs in 3 pairs ; ])rovided with spiral lobes.

a, — The anteriormost pair of segmental organs lie in front

of the ventral hooks T. inansense. P. 71.

h, — All segmental organs lie behind the ventral hooks.
T. mncosurii. P. GS.

THE GEPHYREA OF JAPAN. OO

(J, — Segmental organs indeßnite in number, always more than
3 pairs.

ff^ — Segmental organs very numerous, small and not
provided with si)iral loLes. Prohoscis remarkably

Ions: T. iœnîoldes. P. 63.

h. — Segmental organs grouped commonly into 7 pairs,
each group consisting of 0-3 large organs ; they
are [)rovided with spiral lobes. Proboscis nearly

as long as body T. fleyaiis. P. 6.5.

c, — Segmental organs, in the female, in 3 pairs, while
in the male they are found in 6 or 8 groups ar-
ranged pairwise, each group consisting of 1-4.
The shape of the organs is like that of P. elegans.

T. gogoshimense. P. 66.

II Lonoitudinal muscle divided into 18 broad bundles.

Segmental organs in 3 pairs, the anteriormost pair be-
ing situated in front of the ventral hooks; they are
provided with short spiral lobes. A large rectal gland
present T. koJwtoniense{?) P. 60.

BONELLIA.

I. Very numerous ventral hooks in the female and no hook in the

male.

The female wi>rm bears a grayish brown color ; a large

form. Parasitic male of a remarkably large size

B. 'iniijajiiitai. P. 73.

II. No ventral hook in either sex.

The female of a grayish brown color, spotted with a

deep bluish black pigment B. viisaJciensis. P. 74.

III. Two ventral hooks present in both sexes.

Color bright green on body and pale green on proboscis.
A small form, living in dead corals.... i>. minor. P. 72.

84 ATIT. 4. 1. IKEDA :

Lisb of reference-papers which have been avaible
to me in working out the present study.

1. Andrew, J. — Zur Anatomie des Sipimculus iinJus L. (Zool. Anz., 4

Jahrg., p. All, 1881).

2. Andke.k, J. — Beiträge zur Anatomie und Histulogie des Sipuncuius

nudus L. (Zeit. Wiss. Zool., Bd. 36, p. 201, 1882).

3. Andrews, E. — Reproductive organs of Phascolosouia Gouldii (Zool.

Anz., No. 302, p. 140, 1889).

4. Andrews, E. — Notes on the Anatomy of Sii)uncuhis Gouldii B. (^tud.

Biol. Labor. Johns Hopk. Univ. Baltm., Vol. IV, No. 7, p. 389, 1890).

5. 1-)A1RD, W. — Monogra])h of the Species of Worms belonging to the

Subclass Ge^ihyrea (Proc. Zool. Soc. Bond., 1868, p. 11).

6. BiiUMPT, E, — Quelque Faits relatifs a l'Histoire du Phascolion Strombi

M. (Arch. Zool. Exper., 3e 8ér., Tom. 4, 1897).

7. L^ANiELSEN, D. and Koren, J. — Gephyrea (Norslc. Nordh. Exped.,

m, Zool., 1881).

8. E>[BLETON, A. — On the Structure and Affinities of Echiures uniciuctus

(Trans. Lin. Soc. Lond., 2nd Sen, Zool., vol. VIII, Part 3, 1900).

9. FiscHEii, W. — Uebersicht der von Herrn Dr. Fr. Stuhlmann auf

Sansibar und an der gegenüber liegenden Festlandküste gesammelte
Gei)hyreen (Jahrb. d. Hamb. Wissensch. Anstalt. Bd. IX 2, 1892).

10. FiscTiER, W. — Die Gephyreen des Naturhistorischen Museum zu Ham-

burg (Abband, a. d. Gebiete d. Naturwiss., Bd. XIII, 189.5).

11. Fischer, W. — Gephyreen (Hamb. Magalh. Samml., Hamb., 1896).

12. Fischer, W. — Gephyreen (Zool. Forsch, in Austr., 1896).

13. Fischer, \V. — Weitere Beiträge zur Anabjmie und Histologie des

Sipunculus indiens Peters (Jahrb. d. Hamb. Wiss. Anst., Bd. 10, 1892).

14. (lîiîEEFF, It. — Die Echiuren (Gephyrea Armata) (Nova Acta d. ksl.

lA'Oi).-Carol.-Deutsch. Acad. d. Naturw., Bd. XLI, Pars II, Nr.
l, 1879).

15. (JuKEFF, It. — lieber den Bau der Echiuren (Sitzungsber. d. Ges. z.

Marb., Nr. 4, 1879).

16. Ukrdman, W. — Note on a New British Ijchiuroid Gephyrean, with

Remark on the Genera Thalassuma and Haniiitgia (Quart, Jourii.
Micr. Sc, Vol. XL, p. 367, 1897).

THE C4EPHYEEA OF JAPAN. 85

17. Höhst, K. — I)i(î Uephyrea (Kiedcii. Arch. i'. Zuui.j buppl.^ vol. i, iÖ8i).

18. Jameson, L. — Contribiitious to the Anatomy and Histology of Tlialas-

MMiui neptmii G. (Zool. Jahrb. Abt. Anatom., Bd. 12, Heft 4, 1899).

19. Jamkson, L. — Thalasseraa papillosum (Delle Chiaje), a forgotten

Echiuroid Ge[)hyrean (Mitth. Zool. tStat. v. Neapel, Bd. 13, Heft
4, 1899).

20. Kkfeiistein. — Beiträge zur anatomischen und systematischen Kennf-

niss der Sipunculoiden (Zeitschr. f. wiss. Zool., Bd. XV, p. 404, T86.5).

21. Keferstein. — Untersuchungen über einige amerikanischen Sipunculiden

(Zeitschr. f. Wiss. Zool, Bd. 17, p. 44, 18G7).

22. liAMPEKT, K. — lieber einige neue ïhalassemen (Zeitschr. f. wiss,

Zool, Bd. 39, p. 334, 1883).

23. Laxkester, K. — On Thalassema neptuni ll. (Zool. An/,., 4 Jahrg.,

jK 3.30, 1881).

24. MiCH^ELSEN, W. — Die Gei^hyreen von Süd-Georgien, etc (Jahrl). d.

Hamb. wiss. Anst., Bd. 6, 1889).

25. Mktalnikoff, S. — Sipunculus nudus (Zeitschr. i'. wiss Zool, Bd. 68, 1900).

26. EiETSCH, M. — Etudes sur les Gephyriens armes ou l^chiuriens (These

d. Dorct., Paris, 188G).

27. itoui.E, L. — Notice préliminaire sur les es[)eces des Gephyriens recueilles

dans les Explorations sous-marines du Travailleur et du Talisman
(Bull d. Mns. d'Hist. natur., Ann. 1898).

28. Selenka, E. — Die Sipunculiden (1882-1883).

29. Selenka, E. — Report on the Gej»hyrea (Cliallenger Report, vol.

XIII, 188.-)).

30. Siiii'LEY, A. — Notes on the Genus ►Si[)unculus (Froc. Zool. Soc. Lond.,

p. 326, 1893).

31. Shipley, A. — A Re[)ort on the Sipunculoidea, etc. (Willey's Zool

Results, Part II, 1899).

32. Shipley, A. — Report on the Gephyrean Worms, collected by ]\Ir.

Stanley Gardiner at lu^tuma and Funafuti. (Proc. Zool. Soc. Lond.,
Part III, p. 468, 1898).

33. Shipj-ey, a. — Notes ou a collection of Gephyrean Worms found at

Christmas Island by Mr. (J.W. Andrews (Proc. Zool. Soc. Lond.,
Part I., p. ::>ô, 1899).

34. Shipley, A. — On a collection of Echiuroids from the Loyalty Island,

etc. (Willey's Zool. Results, Part III, 1899).

35. SnjrrEK, C. — Beiträge zu der Kenntniss der Gephyreen (Naturk.

Tijdschr. v. Nederl Ind., 1883-1886).

S6 ART. 4. 1. IK EDA :

36. Sluiter, C. — Die Evertebraten aus der .Saiiiinliing des königlichen

Naturwissenschaftlichen Vereins in Niederländish Indien in Batavia
(Naturk. Tijdschr. v. Nederl. Ind , Bd. L, 1900).

37. Sluiter, C. — Die Sipunculiden und Echiuroiden der Siboga-Expedi-

tion (Sibüga Expeditie, Monographie XXV, 1902).

38. »Sluitep.j C. — Gephyreen von Süd-Africa, nebst Bemerkungen fiber

Si[)uncnlns indiens Peters (Zool, Jahrb., Abt. {Syst., Bd. XI, 1898),

39. Sluiteu, C. — Ge[)hyriens (tSipunculides et Echiurides) provenant des

Campagnes de l'Hirondelle et de la Princesse Alice (Resultats d.
Coming. tScient. d. Pr. de Monaco, Fase. XV, 1900).

40. Si'ENGEL.— Ueber die Organisation des Echiurus Pallasii (Zool. An/,.,

Bil. 2, p. 542, 1879).

41. Spengel. — Beitrage zur Kenntniss der Gephyreen. I. Die Eibildung,

die Entwicklung und das Männchen der Bonellia (Mitth. a. d. Zool.
Stat. z. Neap., Bd. 1, p. 3.57, 1879).

42. »Sfengel. — Beiträge zur Kenntniss der Gephyreen. II. Die Organisa-

tion des Echiurus Pallasii (Zeit. wiss. Zool., Bd. 34, p. 640, 1880).

43. QuATREFAGES. — Histoire des Anneies (Vol. 2, p. 563, 1865).

44. Vejüovsky. — lieber die Eibildung und die Männchen von Bonellia

viridis (Zeitschr. f. wiss. Zool., Bd. XXX, p. 487, 1878).

45. The List of the Gephyrean Worn)s of Funafuti (Australian Museum,

Sydney, Memoire III, 1899).

Postscript ;
Augenek, H. — Beiträge zur Kenntniss der Gephyreen, Berlin, 1903.

THE GEPHYREA OF JAPAN.

87

List of Abbreviations used in the plates.

a., K\\\w.

*i(.f., Anal g];ui<l.

A., Transverse liar of lioolc.

re. Contractile villi.

dm., Dorsal muscle.

dv.. Dorsal vessel.

em., Erector (or radiating;) muscle

of ventral hook.
/■///., Fixing muscle.
fa., Ciliated finint^l.
<j., Ganglion.
Id:., Hook.
U., Heart.

/>',, Intestinal convolution.
i'ni., Interbasal muscle of ventral liook.
m.. Mesentery.
inf., Muscle-fibres.

nc., Xerve-cord.

ni., Neuro-intestinnl vessel.

i)d.., Ovid net.

Of'.. Œsophagus.

pa., Pieces of digestive canal.

p/i , Pharynx.

rg., Rectal gland.

f\.. Reticular tissue.

/■;., Warts or ' Rnn.a'ln.'

.s'///., Spindle muscle.

M)., Segmental organ.

sr.. Spermatic reservoir.

SX., Sexual organ.

vd.. Vas deferens.

vm., Ventral muscle.

œm., Wing-like muscle.

I. IKEI»A.

THE GEPHYREA OF JAPAN.

PLATE L

Flg.

1

Fig.

2

Fig.

3

Fig.

4

Fig.

5

Fig.

6

Fig.

7.

Fig.

8.

Fig.

9.

Fig.

10

Fig.

11

Fig.

12.

Fig.

13

Fig.

14.

Fig.

15.

Fig.

16.

Fig.

17.

Fig.

18.

Fill-.

19.

Fig.

20.

Fig.

21.

Fig.

22.

Fig.

23.

Fig.

24.

-Phascolosoma nigrurii, n. hsp. Natural size.
-Phascolosoma jcqwnicum, n. sp. Natural sise.
-Phascolosoma misahianum, n. sp. About 7 times enlarged.
-Phascoloso'/ua okinoseanuvi, n. sp. Natural ûze.
-Phascolosoma oiostovi, n. sp. Natural size.
-Phascolion artlßctosus, n. sp. About 10 x .
-Phymosoma onomiclilanmii, n. sp. Natural size.
-Pltijmosoyna nahaense, n. sp. Natural size.
-Aspîdosiphon inisoldensis n. sp. 2 x .
-Aspidosiphon umsadatus, ii. sji. (Ireatly juagnified.
-Aspidoslphon aiigidatvs, n. sp. 10 x .
-Aspidosiphon spinalis, u. sp. 3 x .
-Clœos'iphon japoiiicmii, n. sp. Natural size.
-Dcndrosfoma blanduiit, »Sel. and De Man. Natural size.
-Dendrostoma signifer, Sel. and De Man. Natural siae.
-Dendrosto7ua minor, n. sp. f) x .
-Thalassemia kokotoniense , Fischer {?). Natural size.
-Thcdassema owstoni, n. sp. 2 x .
-Thalassema gogoshimense, n. sp. Natural size.
-Thalassema mucosum, n. sp. Natural size.
-TItalassenta fascuni, n. sp. Natural si;je.
-Thalassema inansense, n, sp. Natural size.
-Boiiellia miiior, Marion. Natural size.
-Bonellia tnisahiensis, n. sp. Natural size.

Jour. Sei. Coll. Vol XX. Art. 4: PI I

Fis-l

Fig. 24-

Fig. J.

Fijf. ZO

I.lkeda;The Gephyrea of Japa

\.\i\\ V. K.>ahjl>n Tnkvu .lapdJi

r. IKKOA.

THE GEPHYRKA OK JAPAN.

PLATE II

Phascolosoma niyruin .

Fig. 2r). — A papilla from the posterior Lod}^ region. 180 x .
Eig. 26. — A papilla from the introvert. 180 x .
Fig. 27. — A specimen dissected.

riid coloKoriKi jdpoiiicii III.

Fig. 28. — Skin with chitinous bodies from the middle body region. 2f).) x .
Fig. 20. — A specimen dissected.

Phnxcolnsomo misYihiniifim.

Fig, :X). — A papilla from the [)osterior body end ; a in longitndinal optical

section and h in surface view. 180 x.
Fig. .31. — A ])erf()rated papilla from the introvert. 410 x.
Fig. ?>2. — A hook from the introvert; side view. 410 x.
Fig. 33. — A speeituen dissected.

Fhascoloftoma oki/iosecmum.

rig. 34. — A pa]nlla from the posterior body end. Zeiss Oc. 2 and Obj.

Innn. 7)2 •
Fig. 3~). — Papillfe from the introvert basis. 180 x .
Fig. .'>G. — A sac-hke respiratory organ on the skin. 65 x .
Fig. .'^7. — Basal part of the same. Zeiss Oc. 2 and Obj. 1mm. '/i--
Fig. 38. — A specimen dissected.

F/i ascolosoma owstoni.

Fig. 39. — A papilla from the posterior body region. 180 x .

F^'ig. 40, — A jiapilla from the introvert basis. 180 x .

Fig. 41. — A papilla from the introvert. 180 x .

F'ig. 42. — A hook from the introvert. 180 x . "

F'ig. 43. — A specimen dissected.

F'ig. 44. — Magnified surface-vit^w of the pecnliar reticular structure on the
inside of the body-wall. Zeiss Oc. 3 and Obj. a,_,.

Fhascolion rectus.

Fig. 4;"). — The whole worm-body. 2 x .

Fig. 46. — Papillte from the middle |)arts of tiie introvert. 330 x .
Fig. 47. — Papillaî and chitinous thickenings of the skin at the introvert-
basis. 200 X.
F'ig. 48. — Pa])ill;\^ from the posterior end of body. .')30 x .
Fig. 49. — Specimen dissected.

Flidscolion (irtijiciosus.

Fig. :"30. — Gastropod shell containing the worm. 2 x .

Fig. .")1. — Large papilUe seen from above. 200 x .

Fig. r)2. — Small ])apilUe from the middle region of the intr.)vert. 330 x .

Jour. Sei. Coll. Vol XX. Art. 4-. PI II

I.Ikeda^The Gephj-rea of Ja pa

ilhE Koi-tiib» fpViyp J«piir

I. IK Kl» A.

THE GEPHYREA OF JAPAN.

PLATE III.

FJinficoh'i))) (iiiijtci()s>/s (ci^utiniird).

Fig. 53. — Papilla3 from the posterior end of body. 200 x .

Fig. 54. — Finger-shaped papillfe from the basis of the introvert. 200 x .

Fig. 55. — A specimen dissected.

Flijlinof^onin onoiiilchidniivi.

Fig. tyCj. — A papilla from the middle body region. 180 x .

Fig. 57. — A papilla from the introvert-bisis ; a side view; h surface

view. 1 80 X .
Fig. 58. — Specimen dissected.

F/i(/ijwsoiiia nal/ae/i.se.

Fig. 59. — A papilla from the middle body region. 265 x .

Fig. 60. — A hook from the introvert; side view. 265 x.

Fig. 61. — A perforated papilla of the introvert ; side view. 265 x .

Fig. 62. — A specimen dissected.

SipuncvJus evmanensis.

Y\a^ GP,. — Portion of a transverse section through tlie body-wall, showing
the structure of the ' ovale Gebilde.'

Sipuncvlufi awamiem^is.

Y\,^^ f;4, — \ papilla from the middle Iwdy region ; surface view. 294 x .
I'^ig. (55. — A specimen dissected.

A»pidosiphon truncatus.

Fig. G6. — A hook from the introvert; side view. 410 x.

Fig. fiT. — A perforated papilla from tiie introvert; side view. 410 x.

Aspidosiplion misahievsis.

Fig. G8. — Anterior end of body, showing the anal shield. 20 x.

Fig. 69. — A hook ; side view. (jOO x .

Fig. 70. — One of the spines found scattered on the introvert ; side view.

600 X.
Fig. 71. — A perforated papilla lying between hook-rows. 294 x .
Fig. 72. — A s]iecimen dissected.

Axpi((()sip]ton /rn/'scuidfns.

Fio. 73. — Magniiled view of the b:isal parts of a longitudinal ridge of

the iinal shield. 9;") x .
Fig. 74. — A ])api]la from the middle body region. 410 x .
l^'ig. 7"). — A hook from the hook-rows; side view. G80 x.
Fig. 76. — A perforated pai)illa lying between the hook-rows. 680 x .
Fig. 77. — A specimen dissected.

Aspidosiplton (tv(/ulaim.

Vi^. 78. — A papilla from the posterior body-region ; surface view. 26:") x .
Fig. 79. — Hooks and perforated papillas from the introvert. 410 x.
Fig. SO. — A specimen dissected. (This figure has been inserted in PI. HI).

Af^pidosiphon spinalis.
Fig. 81. — A papilla lying near the caudal shield; surface view, 265 x.

Jour. Sei. Coll. Vol XX. Art. 4-. PI

Fig. 6.

I.IkedaiThe Gephyrea of Japan.

i. IKKUA.

THE GEPHYREA OF JAPAN.

PLATE IV.

(For explanation of Fig. 80, see under PI. III.)

Aspiikmplioii KpinaHi^ (continued ).

Fig, 82. — One of the spines scattered on the introvert. 21)4 x.

Fig. 83. — A hook from the rino;-row,s on the introvert; side view. 294 x .

Fig. 84. — A perforated papiHa from the introvert ; side view. 294 x .

Fig. 8.3. — »Specimen dissected.

Chi'osiplioK j'aponicimi.

Fig. 86.™ Magnified view of the calcareous shield. Zeiss Oc. 2 and

Obj. aa.
Fig. 87. — A pai)illa Ironi the middle hody region ; surface view. 4Jü x .
Fig. 88. — A papilla from tlie introvert-basis ; side view. 180 x .
Fig. 89. — A hook and a }ierforated ))apilla from the introvert. 410 x.

Dendrostoma bland u m.

Fio;. 90. — (Side view ol' the main stems of tentacles. 20 x ,
Fig. 91. — Magnified view of a portion of the digestive canal, showing
the rectal gland {rg), the fixing muscles {fin), efr. 5 x .

Di'itdrcMoiiia mi nor.

Fig. 92. — A i)apilla from the iK)sterior body end; surface view. 294 x.
Fig. 93. — Side view of a pa{)illa from the hooked region of the introvert.

265 X .
Fiu'. 94. — One of the scattered hooks of the introvert. 410 x .
Fig. 95. — A specimen dissected.

Thn/<i,sse»/a o/cdoni.

Fig. 9G. — Specinjen dissected ; the greater part of the digestive canal is

cut ofi". 3 x .
Fig. !t7. — JMagnified figure of a segmental organ.

Tludatiseiiui ihaicosuik.

Fig. 98. — Magnified view of a dissected si)ecimen, showing only the an-
terior body end.

ThalanHenta fasvLiin.

Fig. 99. — Magnified view of a dissected specimen, showing only the an-
terior body end.

Tli'dostivhia iiiausensc.

Fig. 100. — Specimen dissected ; the greater part of the digestive canal is
cut olf.

Bonellia minor.

Fig. loi. — Distal portion of an anal gland of the female. 20 X.

Fig. J 02. — Male worm, drawn from a stained toto preparation. 131 x.

Ji()U(/lia misal:i('utiis.

Fig. 103. — Specimen dissected, all internal organs In .situ. 2 x .

Fig. 104. — One of the secondary branches of the anal gland. Highly

magnified.
Eig. 105. — Male worm, drawn from a stained toto preparation. 56 x .

Jour. Sei. Coll. Vol XX. Art. 4-, PI. IV.

Fig. 9Z

Fi,,. .<)!)

Fi 9.80

Fig. 81

I.IkedatThe Gephyrea of Japan.

JOURNAL OF THE OOLLE&E OF SOIEHOE, IMPERIAL UNIVERSITY,
OKYO, JAPAN.

VOL. XX., ARTICLE 5.

Mesozoic Plants from Nagato and Bitchu.

By

Matajiro Yokoyama, Rigakuhakushi.
Professor of Palaeontology, Imperial University of Tokyo.

With 3 plates.

1. Plants from Nagato.

Some fourteen years ago, I wrote a brief paper^^ on fossil
plants discovered by Dr. T. Kochibe at Yamanoi, in the Province
of Nagato, in which I advanced the opinion that they were most
probably Khsetic in age. This view is now entirely confirmed
by my recent study of a later collection made by Mr. K. Inouye
at the same place. This collection, although not rich in the
number of species, contains specimens which are valuable both
as supplementing those previously found only in an imperfect
state, and as making possible a determination of the exact age
of the layer in which they were obtained.

The plant-bearing formation of Yamanoi is below the so-
called Inkstone Series with which it was formerly incorporated.
The lower part of the latter has been recently proved by me to
be Liassic^^ from the Ammonites entombed in it. The former

1) On Some Fossil Plants from the Coal-bearing Series of Nagato. Journal of tlie Col-
lege of »Science, Imperial University, Japan, Vol. IV, part, II, 1891.

2) Jurassic Ammonites from Echizen and Nagato. Ibid., vol. XIX, art. 20, 1904.

2 ART. 5. — M. YOKOYAMA.

contains anthracite layers in its lower portion, and is underlaid
by a marine limestone filled with Fiisulina.

The species of clearly determinable plants which I can now
mention from Yamanoi are the following :

1. Cladophlebis nebbensis (Brgnt.)

2. Cladophlebis yamanoiensis Yok.

3. Dictyophyllum Nathorsti Zeil.

4. Dictyophyllum japonicum Yok.

5. Dictyophyllum Kochibei Yok.

6. Podozamites lanceolatus (Lindl. et Hutt.)

7. Nilssonia Inouyei Yok.

8. Baiera paucijMriita Nath.

Among these, Cladophlebis nebbensis and CI. yamanoiensis are
what I formerly called As])lenium Rœsserti Presl, and Asp.
Rœsserti var. whitbiensis Brgnt. respectively, while Dictyophyllum
Nathorsti is a species which I formerly compared to D. acutilobum
Braun.

It may be here added, that among the plants brought back
by Mr. Inouye, there are fragments which seem to be referable
to the genera Pinus and Phœnicopsis, but which are at present
still undetermined.

A glance at the above list shows the indubitable Rhsetic
nature of the florula. The three species, Clado2jhlebis 7iebbensis,
Dictyophyllum Nathorsti and Baiera paucipartita, are forms hither-
to known only from the Rhsetic. Podozamites lanceolatus is in-
deed a form of wide occurrence in the Jurassic, but its first
appearance seems to be in the Rhœtic. The remaining species
being those found only in Japan are not available for the deter-
mination of the age. But we must remember that a form like
Dictyophyllum japonicum exhibits a great resemblence to one

MESOZOIC PLANTS FROM NAGAïO AND BITCHU. 3

already found in the Rhœtic of Europe. Therefore we may
safely conclude that at least the upper portion of the plant-bearing
series of Yamanoi belongs to the uppermost Keuper, or E-hietic
as it is generally called.

DESCRIPTION OF THE SPECIES.

1. CLADOPHLEBIS NEBBENSIS (Brgnt.).

PI. I. Fig. 1-3.

Cladophlebis nehbensis Nathoest, Bidrag till Sveriges fossila
Flora, p. 16, pi. II, fig. 1-6, pi. Ill, 1-3. Möllee, Bidrag till
Bornholms fossila Flora, Pteridofyter, p. 29, pi. II, fig. 22, ^/.
///, fig. 1. Zeillee, Flore Fossile des Gîtes de Charbon du
Tonkin, p. 45, pi. IV, fig. 2-4.

Pecopteris nehbensis Beongniaet, Histoire des Vêgêt. Foss.,
p. 299, 2^1- XGVIII, fig. 3.

Äsplenium Rœsserti Yokoyama, On Some Fossil Plants from
the Coal-bearing Series of Nagato, p. 241, pi. XXXII, fig. 1-5.

This fern which I formerly took for Cladophlebis [Asplenium]
Rœsserti (Peesl.), a well known Rhœtic form, proved to be a
different, but still very closely related s]3ecies of the same form-
ation, viz. Cladophlebis nehbensis (Begnt.), as has already been
pointed out by Zeillee in his excellent work on the fossil flora
of Tonkin, above cited. This author who studied specimens of
both species occurring in Tonkin says that the pinnules of Cla-
dophlebis nehbensis, besides being denticulated at least in the
posterior and middle portions of the frond, are generally larger
and broader, with apex more rounded and margins more parallel
than those of Cladophlebis Rœsserti. He adds moreover that in

4 ART. 5. — M. YOKOYAMA.

the former the two basal pinnules of every pinna, one in front
of the rachis and the other behind it are different either in shape
or in mode of attachment to it, a character never observed in
the latter. Lastly it is said that the lateral veins of the pinnules
of Brongniart's species are more divergent, further apart, and
less divided than those of the pinnules of Presl's species.

Although none of the specimens found at Yamanoi show
denticulated pinnules, yet the other characters, especially the
peculiarity in the two basal pinnules, being observable in them
the plant is now referred to the species tirst created by Brongniart.

Very frequent, but mostly fragmentary.

2. CLADOPHLEBIS YAMANOIENSIS Yok.

Äsplenium Rœsserti var. whitbiensis Yokoyama, On Some
Fossil Plants from the Coal-bearing Series of Nagato, p. 242, pi.
XXXII, fig. 3, 3a, 4.

Frond bi-or tripinnate. Principal rachis moderately strong
and rigid. Pinnse elongated, linear, alternate, slightly inclined
forward, rather widely separated so that there is more or less
space left between them. Pinnules falcate, those on the posterior
side of the pinnae being especially so, and also more strongly in-
clined forward than those on its anterior side, alternate, close
together, but separate to the base, bluntly pointed, thin in texture.
Midrib distinct, but diffusing into branches toward the end.
Lateral veins going off' obliquely at a very wide angle, divergent,
forking twice, except near the end where they are forked only
once.

Although I formerly identified this fern with Cladophlebis
whitbiensis Brgnt., the strong forward inclination of the pinnules

MESOZOIC PLANTS FROM NAGATO AND BITCHU. 0

of the posterior side of the pinnae compells me to treat it now
as a new species. In the form and venation of the pinnules, our
specimens show a close resemblance to those figured and described
by Heer as Asplenium whitbiense var. tenuis {Beitr. z. Juraflora
Ostsib. It. d. Amurl, p. 39, pi III, fig. 3) from Siberia, which,
however, have the pinnœ so close together as to partly overlap
one another. A plant described under the same name from the
Jurassic of China {Schenh in Richthofen's China, vol, IV, pi.
LII, fig. 1) has the pinnae further apart, but the veins are some-
what denser, and moreover there is not the peculiarity of the
posterior pinnules before mentioned.

Specimens of this plant are not found in the collection of
Mr. Inouye. Therefore the only ones which are at hand are
those formerly figured by me in the work above cited. For this
reason the figures are not given here.

3. DICTYOPHYLLUM JAPONICUM Yok.

PI. II. Fig. 3.

DictyophylliLm japonicum Yokoyama, On Some Fossil Plants
from the Coal-bearing Series of Nagaio, p. 243, pi. XXX.

Among the specimens collected by Mr. Inouye, there is a
fragment of a basal portion of a frond in which the more or
less palmate nature of the pinnae is well shown. As may be
seen from our figure, the primary rachis is divided apparently
into two very divergent branches, each carrying on its upper or
front side several pinnae which are separated from one another
by a short space and assume a more less fan-shaped appearance.

Very frequent.

6 AKT. 5. — M. YOKOYAMA.

4. DICTYOPHYLLUM NATHORSTI Zeiller.

Dictyophyllum Nathorsti Zeiller, Flore Fossile des Gîtes
de Charbon du Tonkin^ p. 109, pi. XXIII, fig. 1, XXIV, 1,
XXV, 1-6, XXVI, 1-3, XXVII, 1, XXIII, 3.

Dictyophyllum cf. acutilobum Yokoyama, On Some Fossil
Plants from the Coal-bearing Series of Nagato, p. 242, p/. XXXlIy
fig. 6.

No specimen of this plants being found in the Inouye col-
lection, I am again obliged to refer to the one previously figured
by me in the work above cited.

At that time, I had called attention to the fact that the
teeth in the frond of the Japanese specimen are closer together
than in most of the figures of D. acutilobum (Bkaun), given by
Schenk and others. And now I am quite convinced that our
plant is identical with D. Nathorsti, a species recently created
by Zeiller for specimens from Tonkin which he also formerly
took for those of Braun's species. A glance at plate XXV of
Zeiller's work will show to a certainty that the Japanese and
the Indo-Chinese plants belong to one and the same form.

5. DICTYOPHYLLUM KOCHIBEI Yok.
PI. I. Fig. 5, 7. PL II. Fig. 1, 2.

Dictyophyllum Kochibei Yokoyama, O71 Some Fossil Plants
from the Coal-bearing Series of Nagato, p. 244, pi. XXIV, fig.
1, la.

The diagnosis formerly given of this species may be revised
as follows :

MESOZOIC PLANTS FROM NAGATO AND BITCHU. 7

Primary pinnœ elongated, deeply pinnatifid. Secondary pinnse
elongated, mostly inclined a little forward, but sometimes going
off from the rachis nearly at right angles, rigid or slightly bent
forward or even a little flexuous, mostly separated from one an-
other by a greater or less interval, alternate or subopposite,
gradually diminishing in length toward the front, until at last
they become mere lobes of the primary pinnœ, lobed. Lobes in
the posterior pinnœ ovate or ovately lanceolate with very deep
incisions between them, crenate at margin, obtusely pointed at
apex, while those in the anterior pinnœ become shorter, with
shallower incisions, entire and more blunt, so that in the most
anterior ones they change into mere crenations and then finally
disappear. Eachis of the primary as well as of the secondary
piunse slender. Midrib of the lobes distinct, but weak, somewhat
zigzag, evanescent. Lateral veins forming polygonal nets within
which there are still smaller ones. Veins in the lobed wings
similar to those of the lobes of the pinnse. Fertile pinnœ like
sterile ones, with numerous, crowded, more or less rounded sori
on the finer veins.

When I first described this species, I had only a single
piece of stone on which was preserved a part of two consecutive
pinnae which I then took for the primary ones. But now on
examining the excellent specimens collected by Mr. Inouye, I
find them to be the secondary pinnse which belong to the
posterior portion of the frond. Figs. 5 and 7 in pi. I re-
present secondary pinnse probably belonging to the middle
portion of a primary pinna, while fig. 1, pi. I shows its terminal
portion.

This plant seems to be not so rare, as was at first sup-
posed.

ö ART. 5. — M. YOKOYAMa.

6. PODOZAMITES LANCEOLATUS (Lindl. et Hutt.)

PL I. Fig. 6.

Podozamites lanceolntus Yokoyama, On Some Fossil Plants
from the Goal-bearing Series of Nagato, -p. 245, p/. XXXIV^ fig.
3, 4.

Podozamites distans Zeiller, Flore Foss. d. Gîtes d. Cfiarb.
d. Tonkin, p. 159, pi. XLTI, fig. 1-4. Nathorst, Beitr. z. foss.
Flora Schwedens, p. 23, pi XIII, fig. 1-6, XV, 20.

Zamites distans Schenk, Flora der Grenzschichten, pi. XXXV,
fig. 10, XXX VI.

Specimens belonging to this well known form of Podozamites
are quite frequent at Yamanoi, though mostly in a fragmentary-
state. The one shown in our figure has the leaflets attached to
the rachis. Judging from their shape, the plant seems to belong
to the variety genuina of Heer.

Zeiller is of opinion that the Rhsetic Podozamites distans
(Presl) should be kept separate from P. lanceolatus (Lindl.) of
the Middle Jurassic, on the ground that they belong to two dif-
ferent geological horizons. Indeed it is quite true that, similar
as they are, they may possibly be two different species. But in a
case like this, in which the determination must be based upon the
leaves alone in which the two species present no marked differen-
ces, their union is not at all unjustifiable, especially when we
consider that some of the plant fossils are remarkable for their
longevity. One instance of this is found in our Onychiopsis
elongata (Geyl.) which is not only abundant in our Lower Oolite,
but also in our Neocomian. Recently it has even been suspected
that it occurs also in the Lias.

MESOZOIC PLANTS FROM NAQATO AND BITCHU. 9

7. NILSSONIA INOUYEI Yok.

PI. I. Fig. 4. PI. II. Fig. 4.

Leaf elongated, narrow, sides nearly parallel, 8-12 mm broad,
entire, bluntly pointed at apex. Midrib more or less straight,
moderately strong. Lateral veins fine, straight, parallel, dense,
slightly inclined forward.

This species first found by Mr. Inouye is represented by
only a few specimens. Fig. 4, pi. II shows one which, though
not quite complete, illustrates the general character of a leaf. It
is about 12 mm at the broadest part, tapering very slightly
toward front and back. Fig. 4, pi. I represents a part of a
narrower leaf.

The species exhibits a great resemblance to Nilssonia ozoana
Yok. {Jurassic Plants from Kaga, Hida and 'JEchize7i, p. 42, pi.
JT, ßg. 2b, 11-14). But the veins are coarser, there being 2-3
to a millimeter, while in the latter we can count 4 such in the
same space.

8. BAIERA PAUCIP ARTITA Nath.

PI. II. Fig. 5.

Baiera paucipartila Nathorst, Om Floran i Skanes Kol-
fôrande Bildningar I. Floran vid Bjuf, p. 94, pi. XX, fig. 7-13,
XXI, XXII, 1-2.

Baiera ? sp. Yokoyama, On Some Fossil Plants from the
Coal-bearing Series af Nagato, p. 246, pi. XXXIV, fig. 6.

What I had formerly suspected to be a Baiera proved,
not only to belong to that genus, but also to a form already

10 ART. 5. — M. YOKOYAMA.

known from the Ehsetic of Sweden. The specimen shown in our
figure is the best we have. It is a leaf composed of several,
narrow, parallel-sided lobes which are obtuse at apex and arranged
in a fan-shaped manner. The breadth of the lobes reaches 5 mm,
in which space we can count 7 or 8 fine, parallel veins.
Not rare, but mostly in small fragments.

2. Plants fVom Bitchu.

The neighbourhood of Nariwa, a little town in the province
of Bitchu, has long been known as one of the localities of Pseu-
domonotis ochotica Keys., a bivalve peculiar to the Alpine faciès
of the Upper Triassic.

The order of rock-layers observed at this place and believed
to belong to the Mesozoic formation, when counted from below, is
as follows :

1. Sandstone, grey, fine-grained, often argillaceous and then
darker-coloured, containing innumerable remains of Pseudomonotis
ochotica which is almost the only fossil ever found in this layer.

2. Sandstones and Shales, alternating with each other. The
former is similar in character to No. 1, while the latter is often
coaly and black, and contains impressions of vegetable remains.
In this complex, there are two anthracite layers near Nariwa,
each a foot in thickness.

3. Schalstein, red to dark red, very fine-grained and looking
like shale. In the lower part of this stratum, there is intercalated
a layer of conglomerate consisting of pebbles of a light grey
Fusulina-limestone and of red and green tufiites, cemented by a
green tufaceous matter, while in its upper part we find a stratum

MESOZOIC PLANTS FKOM NAGATO AND BIÏCHU. 11

of limestone-breccia between, the limestone being similar to that
of the conglomerate. The whole bed is pierced here and there
by dykes of porphyrites and quartz-porphyries, often causing con-
tact metamorph ism.

The whole complex of strata from No. 1 to No. 3 rests on
a Palaeozoic formation whose upper part at least is attributable
to the Carboniferous.

The Pseudomonotis bed (No. 1) in Japan is now generally
accepted as an equivalent of the Noric stage of the Alpine Keuper.
Therefore I once thought that the plant-bed which is immediately
above it at Nariwa might belong also to the same stage ; but
a closer examination of the fossil plants, though fragmentary and
not clearly determinable, gives one the impression that they be-
long to a somewhat higher horizon. The species of plants which
I have been able to distinguish in them are as follows :

1. CLADOPHLEBIS sp.
PI. Ill, Fig. 2, 4, 7.

Fragments of the pinnae of a Cladophlebis with pinnules
which show a close resemblance to those of C. nebbeîisis (Bkgnt.),
occurring in the Rhsetic of Nagato. The fine lateral veins of
the pinnules are not always clearly observable, but when distinct,
they are forked only once.

2. SAGENOPTERIS sp.
PI. III. Fig. 3.

A segment of a Sagenopteris lacking the apex. It has an
elongated shape, broadest near the middle and tapering toward

12 ART. 5.— M. YOKOYAMA.

both ends, being cuneate at base. The broadest part measures
21 mm ; the length may be estimated at about 55 mm. The
midrib is rather weak, becoming very weak towards the apex.
The lateral veins are very fine, rise at acute angles from the
midrib and are divergent, although anastomozing with one an-
other to form elongated nets.

That this fragment belongs to the above named genus can
hardly be doubted. It is very likely that it represents a form
which, if not quite identical with, is at least closely akin to,
Sagenopteris Fhillipsl Presl (Möller, Bidray till Bornholms
Fossila Flora, Pieridofyter, pi. V, fiy. 1-7) and S. rhoifolia Presl
(Schenk, Die Fossile Flora der Grenzschichten des Keupers and
Lias FranhenSf pi. XII), both from the Rhsetic of Europe.

3. ARTHROPHYOPSIS ? sp.
PI. III. Fig. 6.

A small fragment of a fern-frond (?), having distinct, equal,
and more or less parallel veins which here and there unite to
form long nets. According to this mode of venation, it bears a
close resemblance to what Nathorst has given the name of Ar-
throphyopsis Nilssoni {Floran vid Bjuf, pi. VII, fig. 5, VIII, 6.
Floran vid Höganäs und Helsinghorg, pi. I. fig. 6) from the
Rhsetic of Sweden.

4. NILSSONIA SP.
PI. III. Fig. 1, 8.

There are two specimens of a Nilssonia, which, though
somewhat different in appearace, probably belong to the same

MESOZOIC PLANTS FKOM NAGATO AND BITCHU. 13

species. Fig. 1 shows the back-side of a leaf while fig. 8 shows
the upper or front side of the same. The midrib, though not
very strong, is rigid with lateral veins, either at nearly right
angles to it, or slightly inclined forward. The leaf seems to have
been either entire or only partly segmented.

The already known form of Nilssonia which can be compared
to the present one is Nilssonia polymorpya Schenk (Nathorst,
Beitr. z. foss. Flora Schivedens, pi. XI) from the Ehsetic of
Europe.

5. PODOZAMITES LANCEOLATÜS (Lindl. et Huit.)
PL III. Fig. 5.

A leaf lacking the apical portion. The general outline and
fine parallel longitudinal veins, numbering about 25, show that
it belongs to the above named species of Podozamites, widely
distributed in the Rhœtic as well as in the Jurassic.

Although the above five plants are by no means decisive
in determining the age of the bed in which they occur, yet
their general character indicates that they belong most likely to
the Rhœtic. This is supported, not only by the occurrence of
anthracite layers as in the Rheetic of Nagato, but also by the
presence of a schalstein above, a rock which exhibits a great
similarity to the so-called Inkstone (schalstein) of the Inkstone
Series of the latter place. And as the lower part of this series
has already been proved to be Liassic, the schalstein of Nariwa,
if not entirely, at least its lower part may also be looked upon
as belonging to the same age.

Bf. TOKOYAMA.

MESOZOIC PLANTS FROM NAGATO AND BITCHÜ.

PLATE I.

Plate I.

(Plants from Nagato).

Figs. 1, 2, 3. Cladoplilebîs nehhensis (Bkgnt.).

Fig. 4. Nilssonia Inouyei Yok.

Figs. 5, 7. DicUjophijllum Kochibei Yok.

Fig. 6. Podozamites lanceolatus (Lindl.).

Jour. Sei. Coll. Vol. XX, Art. 5. PI. 1.

M Yokoyama : Mesozoic Plants.

TOKOTAMA.

MESOZOIC PLANTS FROM NAGATO AND BITCHU.

PLATE II.

Plate II.

(Plants from Nagato).

Fig. 1. DicUjophyllum KocUhei Yok.
Fig. 2. do. Fertile pinna.

Fig. 3. DicUjophyllum japonicum Yok.

Fig. 4. Nilssonia Inouyei Yok.

Fi». 5. Baiera paucipartita Nath.

Jour. Sei. Coll Vol. XX., Art. 5. PL II.

M. Yokciyama : Mesozoic Plant

Plate ni.

(Plants from Bitchu).

Figs. 1, 8. Nilssonia sp.

Figs. 2, 4, 7. CladophleUs sp.

Fig. 3. Sagenopteris sp.

Fig. 5. Podozamites îanceoîatus (Lindl.).

Fig. 6. Arthrophyopsis ? sp.

Jour Sei. Coll Vol. XX., Art. 5. PL III.

Fier. I.

F:o: 6.

Fi ST. 2.

F'^S- .?•

Fig. 4.

Pig- 5-

Fig. 7.

FiiT. 8.

M. Yokoyama : Mesozoic Plants.

JOURNAL OF THE COLLEaE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 6.

On the Magnetization and the Magnetic Change

of Length in Ferromagnetic Metals and

Alloys at Temperatures ranging

from -186°C to +1200°C.

By

K. Honda, Rigakuhakushi.

and

S. Shimizu, RigakusM.

With 4 plates.

In this journal Vol. XIX., Art. 11, 1903, Professor H.
Nagaoka and one of us published the result of experiments on
the magnetization and magnetostriction of nickel steels containing
different percentages of nickel. The present experiment was un-
dertaken, on the one hand, to extend the above investigation to
different temperatures, and on the other hand, to form a con-
tinuation of our former experiment.*

The experiment was made in three separate stages. In the
first experiment, which extended from February 21 to July 2,
1903, the magnetization and the magnetic change of length at
ordinary and liquid air temperatures were measured ; in the
second experiment extending from January 17 to February 2,

*) K. Honda and S. Shimizu, this Jour. Vol. XIX, Art. 10, 1903.

2 ' ART. 6. K. HONDA AND S. SHIMIZU.

1904, the same measurements were extended so as to include dif-
ferent intermediate temperatures between the ordinary and liquid
air temperatures ; lastly in the third experiment, which extended
from March 10 to May 17, 1904, the magnetizations at high
temperatures were measured.

Our specimens consisted of five ferromagnetic metals, and
twelve specimens of nickel steels, kindly placed at our disposal by
M. Ch. Ed. Guillaume. They were all examined in the form
of ovoids (major axis = 20cm and minor axis = 1 cm).

In the first experiment, the specimens were first annealed
for about 4 hours at 1000°C— 1100°C in charcoal fire, after they
were well wrapped in asbestos, and then gradually cooled. These
annealed specimens were tested at the ordinary, then at liquid
air temperature, and lastly again at ordinary temperature. In the
second experiment, the measurement was always commenced with
freshly annealed ovoids. Lastly in the third experiment, all
specimens were cooled in liquid air for about 15 or 20 minutes.
The measurements at the ordinary, and then at higher tempera-
tures, were carried out ; the measurements at different descending
temperatures down to the ordinary were also made. Thus all the
measurements, when set down in order, form a complete cycle
with regard to temperatures, whose limits lie between liquid air
temperature and 1200°C. The methods and the results of the
experiments are given in the following pages.

I. FIRST EXPERIMENT.

The apparatus for measuring the change of length was similar
to that used in our former experiment above referred to. The
ends of the ovoid to be tested were soldered to two short brass

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH. 3

rods, each end of the ovoids entering about 2 mm mto the rod.
The upper piece is connected with a wire
stretched vertically, and the lower piece is
screwed to the bottom of the specimen-holder,
as shown in the annexed cut. The axis of the
ovoid can be adjusted by three small screws.
The holder was made of a copper tube, with
— " three long slits, equally distant from one ano-

ther, along its axis. These slits permit the
adjustment of the specimens to the axial line of the tube. The rest
of our apparatus was exactly the same as in the former experiment.
In the present experiment, the magnetization was, at the
same time, measured by the magnetometric method. The mag-
netometer consisted of a bell-shaped magnet suspended by a quartz
fibre in a thick copper case. A magnetizing coil (length =40 cm,
4/T?z= 394.4) and a compensating coil of nearly the same dimensions
were placed respectively due magnetic east and west of the magneto-
meter. The magnetometer was placed in such a position that the
specimen exerted the maximum effect upon it. The vertical com-
ponent of the earth's field was compensated for. The deflection of
the magnetometer was measured by means of a scale and telescope.
In measuring magnetization, the following precautions were
taken. The verticality of the two coils was tested by means of
a level. The line of the magnetometer, the compensating coil
and the magnetizing coil, was then tested by a compass needle.
The compensation of the earth field and then that of the mag-
netizing coil were next effected ; lastly the scale and telescope
w^ere placed in correct positions.

The precautions above enumerated were especially necessary,
as the magnetization and the magnetic hysteresis in strong fields

4 ART. 6. — K. HONDA AND S. SHIMIZU.

were to be studied with the ovoids placed in vertical positions.
Though these precautions were taken with the utmost care, the
magnetizations by opposite currents of equal strength were not
exactly equal in absolute amounts, so that a small asymmetry of
the hysteresis curve was also observed in the strong fields. This
difference was, in the most unfavorable case, not greater than l"/o of
the total magnetization for a field of 700 C.G.S. This probably
arose from a slight deviation of the coils from the vertical line.
If the lines of force at the center of the magnetometer be not
vertical, but be in the meridian plane, the field due to the coils
may slightly affect the horizontal component of the earth's field,
without producing any deflection of the magnetometer. If the
horizontal component be increased by a current in one direction,
a current in the opposite direction will diminish it. In the first
experiment, therefore, two magnetizing curves for opposite currents
were taken. These two curves almost coincided wdth each other
below a field of 200 C.G.S. , but slightly deviated above that field.
Since the disturbing force is proportional to the magnetizing
current, the correction for the intensity of magnetization can
be derived from a pair of the opposite magnetizations by equal
and opposite currents ; hence in the second and third experiments,
the magnetizations by equal and opposite currents of the maxim-
um strength were taken, and the correction was found, and
applied to magnetization by currents of one direction.

The current was measured by a Si em en-Hals kï: ammeter,
which was occasionally compared with a Kelvin ampere-balance.

The experiment was conducted in the following order. The
adjustment of the magnetizing and compensating coils having been
completed, the specimen-holder containing the ovoid was fixed ver-
tically in the correct position in the magnetizing coil. The ovoid

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH. 0

was then vertically stretched upwards by means of a copper wire
with a spiral spring, special care being taken to stretch the copper
wire in tlie direction of the axis of the ovoid. The magnetic change
of length was then measured in the usual way. The magnetization
and the magnetic hysteresis were next observed. Liquid air was
next gently poured into the DEWAR'tube in the magnetizing coil,
until the tube was nearly filled with the liquid ; then the exposed
parts above the magnetizing coil were carefully protected with
cotton wool. Owing to the boiling of the liquid, a small oscilla-
tion of the image in the field of the telescope was at first observed ;
but after some ten minutes, the image became almost steady.
The change of length was then taken. Next, adding more liquid
air to that in the Dewar tube, when necessary, the magnetiza-
tion and the magnetic hysteresis were measured. Lastly when
the specimen was heated to the temperature of the room, the
change of length and the magnetization were again noted. After
the experiments, the compensation of the magnetizing current was
always tested, and found perfect, except in a few cases.

(a) IVIagnetization of Ferromagnetic IVietals.

In Table I, the magnetization in different fields is given,
where I is the intensity of magnetization, and H the internal
field (external field — demagnetizing force). The temperature of
liquid air has been assumed to be — 186°C; but owing to the
fractional evaporation of nitrogen, as the experiment proceeds,
the actual temperature may, according to circumstances, be greater
or less by few degrees than the above value. The figures of the
last row in each column are the residual magnetism, when the
external field vanishes.

ART. 6. — K. HONDA AND S. SHIMIZU.

TABLE I.

SWEDISH IRON.

/=27.5°C

t= —

i86°C

t=2\

•5°C

H

I

H

I

H

I

1.77

81

2.2 x

77

1.76

86

3-55

243

3-39

247

2.85

250

4.10

582

4-55

453

3-39

431

7.40

890

5.00

578

441

^12

16.3

1138

154

1085

7.00

858

30.3

1251

44-7

1302

14.1

I 102

53.4

1330

I35-0

1464

lyl)-7

1264

103.7

1419

221.8

1543

108.8

1424

234.6

1545

283.9

1587

225.4

1536

349

1609

385

1636

390

1623

473

1647

478

1665

485

1648

554

1 62 1

543

1681

547

1660

33-5

39-7

37-6

TUNGSTEN STEEL.

/=23.5°C

/= —

i86°C

/=30.7°C

H

I

H

I

H

I

2.22

22

1-37

13

2.16

21

8.36

119

5-89

64

4-38

49

10.90

275

1975

130

6.91

104

12.10

528

12.40

254

9.25

193

157

695

16.8

661

10.35

376

20.1

928

28.8

1035

25.8

1064

28.5

1062

62.7

1235

I II. 2

1326

96.6

1292

1 40. 1

1371

216.2

1413

185.1

1381

272.4

1468

333

1463

282.1

1432

402

1523

397

1483

416

1477

469

1535

483

1499

574

1503

570

1557

557

1509

123

136

127

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

NICKEL.

^■=1:

5.o°C

t=-

i86°C

/=i7

.o°C

H

I

H

I

H

I

I.I I

14

1.42

9

0.53

7

3-26

88

3-34

26

3.63

9f

6.96

177

11-35

187

7.92

192

1 1. 12

-^17

21.95

298

18.60

309

19.85

322

34-1

357

38.1

400

37-94

394

80.4

418

63.8

439

64.1

438

152.8

443

1 17-3

469

153.6

478

240.3

462

182.5

483

247.6

490

342

477

1>Z2

494

359

496

493

490

484

503

493

503

586

506

639

505

710

510

704

518

703

506

CAST COBALT.

/=20.0°C

t= —

i86°C

/=i9.5°C

H

I

H

I

H

I

5-1

72

3-90

28

2.58

22

10.7

170

10.57

122

14

14

223

139

247

22.65

277

20

03

302

22.8

371

39-3

427

24

85

374

35.7

497

74-7

608

33

I

454

54-5

628

101.9

690

50

6

594

85.9

749

206.6

848

77

3

eje

147-7

865

281. 1

911

121

*>
0

780

244-3

953

357

956

275

3

941

392

1027

449

999

406

lOIO

467

1047

570

1038

543

1053

610

1088

668

1063

659

1083

8

ART. 6, — K. HONDA AXD S. SHTMTZU.

ANNEALED COBALT.

/=I^

5.6°C

t=-

i86°C

t=i\

.o°C

H

I

H

I

H

I

4-37

6

7.80

9

6-94

8

13-53

26

16.23

20

18

39

29

25-25

^7

39-0

64

29

23

56

36.59

115

57-8

102

45

07

108

52.5

171

80.0

148

65

0

164

86.7

257

119.6

22 1

95

7

232

122.3

320

167.4

282

147

5

313

221.8

433

248.4

358

167

3

415

354

532

347

426

364

508

445

583

496

504

484

572

527

613

561

543

542

603

699

687

702

589

684

664

From these numbers, we see that in Swedish iron, tungsten
steel and nickel, the cooling- by liquid air diminishes the mag-
netization in low fields, but increases it in the strong. In Swedish
iron and tungsten steel, the cliange is very small, amounting at
most to 2 or 3 per cent ; but in nickel, the initial diminution is
considerably larger, amounting to about 10 per cent in the
maximum. The field, in which the effect of cooling changes its
sign, is 115 C.G.S. for Swedish iron and tungsten steel, and
580 C.G.S. for nickel. Fleming aud Dewar'^' who first thorough-
ly studied the magnetization of iron and steel in liquid air, did
not observe an increase of magnetization ; perhaps the field was
too weak to indicate such an increase.

In cast and annealed cobalts, cooling always diminishes
magnetization ; the effect is rather greater in cobalt when annealed
than in the cast state.

*) Fleming and Dewar, Pm, Roy. Soc, 60, 81, 1896.

MAGNETIZATION" AND MAGNETIC CHANGE OF LENGTH. \f

In Swedish iron, tungsten steel and nickel, the magnetizations
before and after the cooling coincide with each other for all
fields. In cast and annealed cobalts, there is a considerable re-
sidual change of the magnetic condition.

(à) IVIagnetization of Nickel Steels.

The intensities of magnetization at the temperatures of the
room and of liquid air are given in Table II.

TABLE IL

NICKEL STEEL 70.32?^.

i=2i

.i°C

t= — ]

86°C

/=30

.5°C

H

I

H

I

H

I

0.24

9

0.28

6

0.93

35

1.52

120

2.02

96

1.52

127

2.20

310

2.61

22 1

1.65

233

391

500

3.62

455

2.75

374

1 1.20

721

10.94

662

6.07

593

17.20

794

20.5

835

II. 17

712

32.5

903

314

911

16.20

784

63.3

969

45-5

971

36.9

869

91.4

988

81 4

1031

52.9

952

232.1

IC04

169.3

1063

101.3

987

381

1008

253-3

1072

186.0

996

492

1008

337

1078

283.6

999

576

1008

439

1088

438

999

672

1009

22 2

601

1098

628

999

10

ATIT. 6. — K. HONDA AND S. SHIMTZU.

NICKEL STEEL 50.72?^.

/=24.o°C

t= —

[86°C

^=32.o°C

H

I

H

I

H

I

1.86

58

1.22

33

1-54

55

2.92

207

3.80

197

2.24

112

3-35

324

413

333

2.74

191

5.16

569

5-55

509

4.44

498

8.80

727

8.27

712

8.77

736

17.12

946

21.9

1003

19.52

961

36-4

1076

27.2

105 1

33-4

1066

514

1137

41.6

1133

49-5

1126

73-0

1181

65.8

1207

84.8

1 184

146.3

1225

148.0

1285

168.0

1220

251.4

1 241

230.6

1302

255-3

1229

328

1244

315

1308

375

1230

426

1246

439

1310

518

1235

604

1247

573

1312

602

1237

NICKEL STEEL 46?^.

/=20.0°C

I S6°C

^ = 21

.o°C

H

I

H

I

H

I

0.29

7

1-59

38

1-54

47

2.30

79

6.00

359

2.79

116

3-02

175

7.90

554

374

228

4.08

279

10.85

770

5-55

490

4.92

362

19.90

974

10.49

754

5.88

473

26.60

1059

15 30

883

8.22

684

39-6

1 147

29.8

1029

II. 10

793

49 -o

1187

66.5

1 145

17-15

909

72.8

1246

166.4

1206

33-8

1054

100.9

1295

311

1219

91. 1

1171

166.6

1338

411

1221

266.7

1214

242.5

1360

489

1222

372.1

1219

455

1377

608

1222

532.9

1 222

631

1384

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

11

NICKEL STEEL 36?^.

^=1:

,.o°C

/= — ]

86°C

/=i9.o°C

H

I

H

I

H

I

1.27

3^

2.18

30

1.48

36

2.45

73

6.00

207

3-39

138

4.00

176

8.16

426

5-37

--> -> r*

jo5

7-45

472

11.70

621

9.14

550

950

524

16.60

801

12.60

658

11-55

624

26.3

982

16.95

741

16.05

714

r 2,2

1153

29.4

863

30.4

853

72.8

1210

55-9

939

65.1

953

121. 5

1279

102.2

985

110.5

989

169.0

1310

187.4

1002

203.5

1007

248.3

1332

341

1014

350

1012

326

1344

436

1015

473

1013

450

1354

528

1016

662

1014

636

1367

657

1017

NICKEL STEEL 29.24?^'.

/=23

.o°C

^= — ]

[86°C

/ = 36.5°C

H

I

H

I

H

I

1.58

41

5-32

0.69

5

2.46

66

12.97

146

4.00

27

4.41

92

24.48

382

8.45

63

22.40

138

29-38

464

18.14

180

28.5

144

43-2

573

26.45

316

35-3

147

590

058

38.4

394

43-7

150

71.0

696

54-1

404

83.6

156

lOI.I

793

91.5

544

210.5

171

152.4

886

117. 8

592

324

181

2350

981

210.4

677

440

189

507-4

1030

320

740

504

193

403.1

1080

463

789

592

197

519

1 121

551

808

657

201

591

1135

623

887

192

171

12

ART. 6. — K. HONDA AND S. SHIMIZÜ-

NICKEL STEEL 29^

t=\:

>.o°C

/= —

i86°C

t=\i

).o°C

H

I

H

I

H

I

I.OO

1 1

2.24

9

2.20

8

2-95

Ï31

10.13

93

4-52

29

7.60

207

13-39

155

9-35

75

19.6

272

20.74

321

12.92

123

67.1

307

29.5

505

1595

180

1437

315

37-6

608

24.72

355

243.6

321

48.0

708

44.2

582

400

328

75.0

858

74-5

708

515

zr:.

139.6

1036

105.7 ■

779

532

333

182.9

1104

174.9

874

738

341

356

1247

279-3

954

424

1276

378

1004

517

1306

547

1052

631

1330

664

1079

/=22.0°C

t=-

186X

/ = 26.o°C

H

I

H

I

H

I

0.29

6

2.97

14

0.85

4

1.61

70

6.85

35

8.91

77

4.20

156

1 1.42

106

13.12

128

13-70

242

18.28

201

17.01

193

29.20

278

27.82

378

22.38

325

46.2

292

30.8

427

33-0

453

87.0

305

39-5

549

48.1

566

159-3

314

66.7

730

79,6

658

233-3

319

178.9

991

III. 8

746

320

321

235-3

1060

207.7

851

406

327

316

1098

307

911

472

328

375

1153

457

966

627

333

461

1191

543

987

672

1 T r-

523

1227

662

IIIO

17-3

226

182

MAGNETIZATION Ax\D MAGNETIC CHANGE OF LENGTH.

lo

NICKEL STEEL 28.32?^.

NICKEL STEEL 26.64?^.

/=22.0°C

/= — ]

86°C

^=27.7°C

H

I

H

I

H

I

23.2

2

3-57

18

3-87

2

71.6

5

8.72

45

9-03

49

124.3

8

13.90

105

16.47

129

225.0

1 1

18.10

194

23.89

267

355

14

28.15

369

38.7

488

524

17

48.5

605

50.5

613

635

18

57-2

671

65.2

721

744

20

739

764

89.2

836

107.6

893

145-7

982

169.8

1023

230.0

1085

246.0

1118

307-2

1161

383

1219

422

1231

466

1260

501

1255

559

1293

621

1287

14

ART. 6. — K. HONDA AND S. SHIMIZU.

NICKEL STEEL 25^.

i=i:

i.S°C

i86°C

/=i5

.o°C

H

I

H

I

H

I

—

32.9

i.i

—

—

—

—

531

0 0
-0

58-3

0.6

—

—

91.8

37

94-9

1.4

—

—

128. 1

5-4

155.8

2-3

354

0.4

209.5

8.0

344-8

2.9

—

313

II.O

401

3-2

516

2.0

498

15.2

529

3-3

605

17-5

749

33

752

20.3

NICKEL STEEL 24.40?^.

/=Ig

rc

/= —

[86°C

^=23.5°C

H

I

H

I

H

I

29.5

1.9

6.22

23

1.03

3

71.0

34

12.54

60

1 2.22

78

126.2

5-4

28.15

241

19.16

153

230.4

7-8

40.4

359

26.2

260

412

9-3

62.7

529

37-1

408

541

10.6

93-6

675

47-6

508

653

10.8

120.0

747

63.1

630

769

II. 5

1835

860

83.7

731

238.1

9:6

119. 6

839

282.8

989

1747

933

370.4

1034

272.8

1024

448

1067

438

1141

510

1107

530

1174

575

II 37

644

1210

2.0

251

232

MAGNETIZATrON AND MAGNETIC CHANGE OF LENGTH.

15

NICKEL STEEL 24.04%^

^=i6.oX

/=-i86°C

/ = 30.o°C

H

I

H

I

H

I

9-5

2.6

0.99

0
J

3-32

1 1

19.6

7.2

4-35

14

8.28

36

33-8

14. 1

II. 14

44

16.58

lOI

52.8

22.2

27-35

191

29.8

295

89.8

33-2

32.9

262

54-4

575

141. 3

42.4

61.3

526

67.6

657

226.6

53-1

83.4

643

83.9

741

417

68.4

107.0

725

1243

861

536

75-6

183.8

870

169.5

942

626

79-9

260.6

968

222.8

1012

7^1>

85.4

376

1053

322

1090

466

IIOI

418

1155

540

1 143

512

1191

591

1152

579

1212

9-3

280

256

As in the case of iron, tungsten steel and nickel, the mag-
netization of the alloys of nickel steels containing percentages of
nickel greater than 26.64:% is decreased in weak fields and
increased in strong, by cooling them in liquid air. In alloys
containing lower percentages of nickel, the magnetization always
increases on cooling. The amount of the change of magnetization
by cooling is considerably large ; witli the exception of 28.73'?^
nickel steel, it increases as the percentage of nickel decreases, up
to 26.64^0. The magnetization in liquid air of 25^^ nickel steel,
which is almost non-magnetic at ordinary temperatures, is also
very small. As tlie percentage of nickel further decreases, the
change of magnetization by cooling again increases. With 26.64?^
nickel steel, which undergoes the greatest change of magnetization

16

ART. 6. — K. HONDA AND S. SHIMIZU.

in liquid air, tlie intensity of magnetization increases from 16 to
1275 for H = 600 C.G.S., or by about 80 times.

The magnetizations, before and after cooling, of reversible
nickel steels containing greater percentages of nickel than 36^,
nearly coincide with each other. If, however, irreversible nickel
steels be once cooled in liquid air, the recovery to the initial
value becomes less and less, as the percentage diminishes ; in
24.40^^ and 24.04^ nickel steels, the magnetization after cooling
is even greater than that in liquid air. Some of the above results
had already been obtained by Hopkinson'^ OsMONDf, and DumasJ.

(c) Hysteresis-loss in Ferromagnetic Substances.

The hysteresis was studied at the temperature of the room
and at that of liquid air. The areas of the hysteresis-loops were
carefully measured by a planimeter with the results given in
Table III. and in Fig. 1 a, b.

TABLE III.

FERROMAGNETIC METALS.

(

Ordinary tempei

'ature.

Liquid aii

temperature.

Swedish iron.

H
B
W

4.0 6.0
4504 10660
2236 S195

15.8
17160
20960

H
B
W

1-9
4792
2091

6.4 63.7
10680 17400
85 1 S 23070

Tungsten steel.

H
B

W

10.7 16.S
3493 10030
4610 25790

91-5
16460
56520

H
B
W

12.5
3223
4266

18.3 91.5
966S 16520
28320 69750

*) Hopkinson's original paper.s, Vol. II, p. 227.

t) Osmond, C.R. CXXVIIl, p. 30G and i:!96, 1899; CR. CXVIII, p. 532, 1894.

t) Dumas' Recherches sur les Aciers tni Nickel, p. 49 — 67, 1902.

MAQiVETIZATION AND MAGNETIC CHANGE OF LENGTH.

17

Ordinary temperature.

Liquid air

temperature.

Nickel.

II
B
W

6.3 20.9 120.5
1996 4405 5931
806 3341 3915

H
B
W

I.I

1569
1589

26.8 122.2
4117 5518
6763 10240

Cast cobalt.

H
B
w

6.6 12.6 92.4
1107 2753 9259
481 2391 21220

H
B
W

7-4
960

441

14.8 96.3
2425 8676
2272 218S0

Annealed cobalt.

H
B
W

43.7 117. 2 2180
1802 3702 550S
5669 22680 35770

H

B
W

4S.3
1115

3244

122.7 224.3
2893 4466
19970 38710

Here H, B and W denote respectively the internal field, the
magnetic induction, and the hysteresis-loss, all expressed in C.G.S.
units. In Swedish iron, the hysteresis-loss is decreased in weak
inductions and increased in the strong, by cooling it in liquid air.
Fleming and Dewar* found no effect of cooling on the hysteresis-
loss of iron. In tungsten steel, nickel, cast and annealed cobalts,
the hysteresis-loss is always increased by cooling. These changes
are briefly expressed by saying that the cooling in liquid air
magnetically hardens the specimens.

In comparing the hysteresis-loss of different metals, it is to
be observed that the hysteresis-loss in nickel is the smallest among
them and that those for tungsten steel and cast cobalt are greater
by about three times than the hysteresis-loss of Swedish iron, and
the same relation also holds between annealed cobalt and tungsten
steel.

From the courses of the curves in Fig. 1, it is evident that
Steinmetz's formula giving the relation between the hysteresis-loss
and the maximum induction holds for nickel and annealed cobalt
up to an induction of 3000 C.G.S. ; it holds for cast cobalt and

*J FiiEMiNG and Dewak, loc. cit.

18

ART. 6. — K. HONDA AND S. SHIMIZU.

tungsten steel up to 8000 C.G.S. ; and lastly for Swedish iron,
it fails beyond an induction of 18000 C.G.S. If, however, the
specimens are cooled in liquid mr, the applicable range of the
law for induction is notably extended, as may be seen from
Fig. 1.

As regards the residual magnetism, the cooling always increases
it in a marked degree.

(d) Hysteresis-loss in Nickel Steels.

The hysteresis of nickel steels was studied at ordinary tem-
perature, and seven of them at liquid air temperature. The
results are given in Table IV and in Fig. 2 a, b.

TABLE IV.
NICKEL STEELS.

Ordinary

temperature.

Liquid air

temperature.

H

2.1

3-4

31.2

H

2.5

4-9

30.2

70.32^

B

2675

7069

11.^70

B

2628

6850

"550

W

886

3510

5353

W

9S41

4140

7494

H

3-9

4.6

38.9

50.72^

B
W

4104
2015

6790
4300

13913
11160

H

2.9

4-9

7.0

H

2.6

4-9

6.8

460/0

B

1350

3845

77S3

B

919

3493

7677

W

202

1969

5S36

W

1026

1970

6867

H

4.0

9-5

39-7

H

5-8

9-5

31.0

360/a

B

2782

6910

1 1520

B

2581

694S

12800

W

1264

5290

9404

W

1654

8coo

19303

H

0.7

6.9

78.0

29.240/0

B
W

749
14.7

2774
109

419

155

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

19

Ordinary

temperature.

Liquid air

temperature.

H

'•3

97

33-7

H

5-0

14.5

22.6

290/

B

1099

2807

3654

B

507

2132

5188

W

106

395

393

W

79

2271

13120

H

0.S9

4.96

53-3

2874^^

B
W

1261
58

305 +
189

4662
322

H

46.5

218.6

—

H

1 1.8

19-3

57-2

28.320,^

B

673

1184

—

B

1938

3964

8394

W

4-5

16

—

W

1888

7439

28060

H

25-3

62.3

148.7

H

14.5

23.2

60.2

26.649^

B

64

173

354

B

1552

3406

8076

W

51

276

703

W

1396

7864

3'9oo

H

35-6

85.0

126

24.40 ^0

B
W

94

153

215

5«!

295
731

H

30.3

102.7

244.6

H

17. 1

29.6

89.4

24.04^^

B

331

60S

9:6

B

1 180

3212

8783

W

mo

2586

4108

\V

1309

10520

63550

The hysteresis-loss of nickel steels at ordinary temperature
is generally small compared with other ferromagnetic metals. The
values for reversible alloys are, however, comparable with those
of nickel ; but for irreversible alloys, they are all very small.
Especially nickel steel of 28.32 ?« does not almost enclose any
area, giving only 16 ergs for an induction of 1200 C.G.S. If
the alloy has a greater value for induction, it will be very useful
for the construction of transformers. Thus the magnetic state
of the irreversible nickel steels corresponds to that of ferromagnetic
metals at high temperatures. As seen from Fig. 2 a, the Stein-
metz's formula does not apply, except for very weak inductions.

If the alloys be cooled in liquid air, the hysteresis-loss
increases. With irreversible alloys, the increase is enormous.

20

ART. G. K. HONDA AND S. SHIMIZU.

The applicable limit of Steinmetz's formula becomes greatly ex-
tended. Thus we may say, as in the case of the pure metals,
that the cooling in liquid air hardens the specimens magnetically.
As regards the residual magnetism, the cooling considerably
increases it.

(e) Length Change of Ferromagnetic IVIetals.

Table V and Fig. 3 a, b give the observed changes of length
at the temperatures of room and of liquid air. Here -y- denotes
the elongation per unit of length.

TABLE V.

SWEDISH IRON.

^ = 24.5°C

/= —

i86°C

^=29.o°C

H

y X 10«

H

'/ X 10«

H

T ^ ÏO'

4.0

0.21

3-0

0.9

4.0

0.7

8.5

27

5-3

1-5

5-1

1.6

21.7

0 r'

20.0

3-2

II. 7

3-4

72.4

00

66.5

3-2

37-0

4.2

123-5

2.4

165.0

0.9

104.0

1 '^

217-5

0.13

274

— 2.0

214

I.I

326

-^■3

396

-4-5

313

-o.S

4B3

-4.8

565

-7.6

478

640

-6.2

—

—

642

-4-5

792

-7.0

814

— 10.2

839

-51

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

21

TUNGSTEN STEEL.

/=2V2°C

i86°C

/--=307°C

H

— - X r 0''

H

y X lo'--

H

lOO

0.2

17.6

0.9

13-7

1.0

13-9

1-5

30.0

31

20.7

3-r

31.0

3-5

91.0

4-3

45-7

4-3

81.0

4.6

174

4-3

112. 3

4.8

209

44

287

3-4

1742

4.5

393

3-4

366

3-1

363

3-4

507

2.6

525

2.3

527

2.6

671

2.4

587

2.1

640

2-3

818

2.1

827

1-5

848

1.9

NICKEL.

/=i9.5°C

t—

i86°C

/=i4.o°C

H

y '^ ^^'

H

'/ ^ 10"

H

y X 10«

5-4

-2.3

3-^

-0.8

5-0

-1-5

lO.O

-6.5

9.6

-2.5

7ß

-4.0

18.3

— 12. 1

16.0

-7-4

13-4

-8.8

32.2

-19.4

31.2

-15-9

29.2

-18.0

75.0

-29.5

749

-24.4

39o

— 26.4

145-8

-34-7

138.4

-28.5

1 10.4

-32.2

243.6

-36.7

252.6

-317

215.5

-36.0

353

-37-3

441

-34-9

387

-37-2

492

-37-7

662

-37-9

485

-37-4

796

-38.0

800

-39-4

77^

-37-7

22

ART. 6. — K. HONDA AND S. SHIMIZÜ.

CAST COBALT.

t=z

o°C

t= —

i86°C

t = 2(

D.O°C

H

y X 10"

H

^/ X 10«

H

y X 10«

14-3

— I.I

94

-0.3

15.6

— I.I

33.2

-4.2

30.0

-l-l

^ 1 p*

-3-5

62.3

-7-5

64.2

- 7-5

63-4

-6.4

106.5

-9o

126.4

— 10.2

124.3

-7-7

167. 1

-9-3

2347

-11.7

231-3

-7-1

217.7

-8.5

350

— 10.9

449

-5-7

348

-6.1

465

-9.8

465

-3-8

498

-4.0

616

-'^■1

616

-1-5

762

— 0.2

749

-7.2

755

— 0.1

ANNEALED COBALT.

t=\

3.5°C

/= —

i86°C

t = 2_

5.5°C

H

'/ X ^0"

H

^/ X lO"

H

y X 10«

31

— 0.1

95

-0.3

59

— 0.2

46

— 0.2

167

— I.I

98

-0.9

86

— I.O

312

-2.7

153

-1-7

142

-1.8

422

-34

268

224

-2.8

512

-4-5

444

-6.1

347

-4-3

635

-6.1

538

-7.6

498

-6.1

755

-7-4

717

-10.3

647

-7-9

772

-jß

869

-12.4

800

-9.8

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH. 23

In Swedish iron and tungsten steel, cooling by liquid air
decreases the elongation of the metals ; the change in tungsten
steel is very small, but in Swedish iron, it is relatively large. In
nickel, the contraction is diminished by cooling to a field of 670
C.G.S. ; but it is increased in stronger fields. In cast cobalt,
the contraction is considerably increased, except in weak field,
where a slight decrease of contraction is observed. In annealed
cobalt, the contrary is the case ; the contraction is always dimin-
ished.

With tungsten steel and nickel, the magnetic changes of length
before and after the cooling coincide with each other. But in
Swedish iron, the elongation after cooling becomes greater than
that before cooling. In cast cobalt, the magnetic contraction after
cooling slightly decreases, as compared with the elongation before
cooling ; but with annealed cobalt, the contrary is the case, except
in weak field.

The above results for Swedish iron, nickel and annealed
cobalt agree with those obtained by us"^ with rods of these metals.
The change of elongation of tungsten steel in high fields does not
coincide with that of our former experiment. But the change
being very small, the discrepancy may be accounted for by taking
into consideration the demagnetizing force, with which our former
experiment was not concerned.

(/) Length Change of Nickel Steels.

The observed changes of length are given in Table VI and
in Fig. 4 a, b, c, d, e, f, g, h, i.

*) K. Honda and S. Shimizu, loc. cit.

24

ART. 6. K. HONDA AND S. SHIMTZÜ.

TABLE VI.
NICKEL STEEL 70.32%.

^'=22.0°C

t= —

i86°C

^=26.5°C

H

;' X 10«

H

y X 10"

H

y X 10"

3-0

0.6

3-0

0.7

2.9

0.9

8.3

3-2

19.0

5.2

20.2

5-3

21. 1

6.1

32.8

8.6

67.9

10.4

45-9

9.2

85-3

II. I

121. 2

11.6

109.4

1 1.2

106.8

12. 1

280.1

11.9

232.6

11.7

228.7

12.8

422

12.0

392

11.8

334

12.9

499

12.0

546

11.9

473

13.0

639

1 1.9

663

II. 7

631

12.9

802

11.9

791

11.6

1^1

12.6

NICKEL STEEL 50.72?«^

/=26.5°C

t=-

i86°C

r^2

7.5°C

H

y X lo'''

H

~x 10"

H

y X 10

3-8

0.6

Z-^

0.4

50

1.0

6.8

2.6

6.8

1.9

9.9

4-3

II-3

5-5

13.0

5-1

44.2

14.5

23-8

10.2

41.0

12.8

102.7

19.8

56.6

16.0

106.0

20.7

236.0

22.8

182.8

22.1

207.2

24.8

363

23-5

280.2

23.0

351

26.4

482

24.1

449

24.0

506

26.7

673

24.7

561

24.2

665

26.7

783

24.9

695

24-3

835

26.8

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

LJ->

NICKEL STEEL 46?^.

/=20.0°C

/= —

i86°C

/=2]

.o°C

H

y ^ ^°"

H

y X 10«

H

'/ '^ ^^'

4.0

0.3

5.0

0.9

4.0

0.6

7.0

2.8

7-7

2.8

4.8

1.2

15-5

6.2

26.4

9.2

94

4.0

76.3

18.5

820

20.4

28.1

10.5

168.0

22.7

113,0

22.0

88.7

19.0

255-5

23.8

310

28.6

207.9

22.8

440

24.8

522

29.5

326

23-7

608

25-3

728

30.7

442

24.0

760

25.4

603

746

24.4
24.4

NICKEL STEEL 36?^.

^=20.0°C

[86°C

/'=i4.o°C

H

y X 10«

H

-^xio''

H

'[ X 10«

5-1

0.3

5.6

0.6

5-3

0.3

10.2

2.7

104

1.4

130

3-3

37-8

8.0

28.2

5-1

27.9

6.6

101.4

135

76.3

14.0

71.9

II. 8

205.9

16.0

186.4

23.8

174.8

15-5

363

17.4

289.8

26.5

298

16.8

503

18.4

428

28.8

441

17.6

616

19.0

587

29.8

592

18.6

756

20.3

738

30.5

717

19-3

26

ART. 6. — K. HONDA AND S. SHIMIZU.

NICKEL STEEL 29.24?^.

/=i9.5°C

t= —

86° C

t=2l

,.6°C

H

y X 10"

H

'/ ^ ^0"

H

-^XIO«

i6.o

0.6

18.5

0.3

20.0

0.3

55-0

I.O

26.0

0.9

52.6

1-7

93-3

17

66.8

2.7

101.9

3-2

149.8

3-4

142.2

6.0

171.8

4.9

291.8

3-6

264.0

9.6

304

7-7

441

5-3

422

I3-I

444

9.8

591

6.7

524

15.0

565

II. 5

699

7.8

675

16.5

708

I3-I

897

9-7

837

17.8

801

13-9

NICKEL STEEL 29^.

^=11

■5°C

t= —

[86°C

/=i4.o°C

H

^/ X 10"

H

H

y X 10«

15.4

i.i

22.9

O.I

18.8

0.3

36.9

1.9

49-3

2.5

31.8

I.I

83.6

2.7

105.7

7-7

59.6

3-2

147.6

3-4

206.9

12.6

122.2

5.6

323

4.9

313

16.4

349

I I.I

472

6.9

412

18.0

470

12.8

692

9.0

555

20.1

633

14-5

850

10.3

681

21.6

770

15.8

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

27

NICKEL STEEL 28.74$

/=225°C

t= —

[86°C

^ = 21

•5°C

H

y X 10«

H

y X 10"

H

'/ X 10'

11.4

0.6

35-0

I.O

16.0

0.2

46.7

1-7

81.0

4.0

37-0

1.6

92.6

2.4

138.7

6.7

76.0

3.5

1557

2.7

184.7

8.2

133-2

5.6

281.0

4.2

248.0

10.3

250

8.5

419

5.6

327

II. 8

337

10.3

681

7.2

519

154

457

12.4

782

8.6

687

17.7

14.7

822

19.4

820

16.7

NICKEL STEEL 28.32?^.

/=20.0°C

t= —

[86°C

t=2\

.5°C

H

-^XIO«

H

'/ ^ ^°'

H

y X 10«

21.3

0.2

27.0

0.8

15.0

0.2

50.2

0.6

390

1-5

415

1-5

85.3

0.8

66.2

2.7

92.0

30

169.0

1.2

lOI.O

44

144.4

4.9

281.6

2.0

231

94

275

8.3

378

3-0

368

12.9

420

II. 2

455

3-6

454

143

599

14.0

601

4.8

618

16.2

788

16.4

810

6.1

826

18.3

878

17.4

28

AET. 6. K. HONDA AND S. SHIMIZU.

NICKEL STEEL 26.64?^.

/=24.5°C

/= —

i86°C

^=2

.o°C

H

4^x,o«

H

'/ X 10«

H

y X 10«

28.7

0

19.6

0.2

20.0

0.3

81.3

0

42.0

1.6

47.2

2.4

118.2

0

92.6

4.8

94.0

5-7

180.0

0.08

145.0

6.9

168

9-3

302

0.13

^55

10.8

300

12.8

489

0.15

353

12.9

408

14.9

680

0.17

490

154

530

16.9

890

0.21

593

16.9

668

18.2

774

17.7

778

18.8

935

195

NICKEL STEEL 24.40?^.

t=2S

.o°C

^--

[86°C

/=20.5°C

H

y X lo'-

H

y X 10«

H

y X 10«

—

—

22.7

0

I

19.0

0.3

—

—

50.8

I

4

.32.0

0.9

—

—

91.0

3

I

78.2

3-4

118

0

156.5

5

2

149

6.8

236

Ü

295

9

0

240

9-5

367

0.02

405

II

2

362

12.2

494

0.08

544

13

I

484

14.6

705

O.IO

640

13

9

621

16.3

886

O.I I

839

16

9

795

18.4

MAGNETlZATlOxX AND MAGNETIC CHANGE OF LENGTH.

29

NICKEL STEEL 24.04^

^ = 21

.o°C

/= —

i86°C

^-23.o°C

H

^/ X io'->

H

'1 X 10«

H

y X 10"

22.9

0

34-3

07

o5o

I.O

76.5

0.1

657

2-5

66.7

2.9

196.6

0.2

124.0

59

1 12.0

5-4

281.7

0.3

238.5

99

253

10. 0

429

0.35

o:)4

^3Z

398

12.7

600

0.6

468

17.8

508

14.4

808

07

622

21.6

695

159

835

25.6

858

16.9

The efîëct of cooling on the magnetic elongation in nickel
steels is exactly parallel to the same effect on magnetization. In
nickel steels containing percentages of nickel greater than 28.74^,
the elongation is diminished in weak fields and increased in the
strong, by cooling them in liquid air ; with other nickel steels,
the initial decrease of elongation vanishes.

The ratio of the elongation in liquid air to that at ordinary
temperature increases in strong fields, as the percentages of nickel
decreases. In 36?^ Ni, it amounts to about 1.6 in H =500
C.G.S. ; and in 28.32^/^ Ni, to 3.7, and in 24.40^/ Ni to 160 for
the same field.

For reversible nickel steels, the elongations after and before
cooling coincide with each other. The elongation of other nickel

30

ART. 6. K. HONDA AND S. SHIMIZü.

steels, once cooled in liquid air, is always greater than that
before cooling. With 26.64^ and 24.40?ö alloys, the elongation
is even increased, by heating it to the ordinary temperature.

259^ nickel steel does not sensibly elongate at ordinary tem-
perature nor in liquid air.

(g) Change of Density by Cooling.

The density of the irreversible nickel steels at ordinary tem-
perature suffered a permanent change, if they were once dipped in
liquid air. This singular fiict was first observed by Hopkinson*.
The following table contains the observed values of density : —

TABLE VII.

Alloys

28.329'^Ni

26.649^ Ni

24.409^ Ni

24.04^ Ni

I3efore cooling

8.15

8.16

8.13

8.06

After cooling

8.01

7-99

8.06

7-94

Thus the density is diminished, by cooling them in liquid
air ; M. Ch. Ed. Guillaume'I* specially investigated this point,
by measuring the coefficient of thermal expansion at low tem-
peratures. He found that the irreversible nickel steels expand
on being cooled in solid carbon dioxide and again expand when
heated to ordinary temperatures. Hence the effect of cooling is
to doubly diminish the density of the alloys.

*) Hopkinson's Original Papers Vol. 11, p. 240.

t) Guillaume, Bulletin de la Scciété d'Encouragement, mars 1898, p. 273.

MAGNETIZATrON AND MAGNETIC CHANGE OF LENGTH. 31

II. SECOND EXPERIMENT.

To obtain a constant low temperature lying between the
ordinary and the liquid air temperatures, a method of slow cooling
was applied. The specimen-holder in the former apparatus was
water-tightly covered with a brass cylinder, and a suitable amount
of liquid air was poured into the interspace between the cylinder
and the Dewar tube. The temperatures above — 15°C were,
however, obtained by dipping the specimen directly into a freezing
mixture (snow and common salt) contained in the Dewar tube.
The experiment was commenced with the specimen in the annealed
state, and the measurements at successively decreasing temperatures
were made. During one set of observations, which usually required
7 or 8 minutes, the temperature was fairly constant and its change
did not exceed one degree in the most unfavorable case. Since
the cooling was very slow and the specimen was doubly enclosed
in copper and brass tubes, the temperature of the specimen may
be regarded as constant throughout its entire length.

The temperature of the specimen was measured by a thermo-
electric couple of platinum and german silver. The wires were
insulated with a thin caoutchouc tube. One of the junctions was
brought in contact with the specimen at its middle, while the
other was insulated with asbestos papers and inserted in a copper
tube. This tube was dipped into the water bath, and its tem-
perature was observed with a thermometer placed in the bath.
The thermoelectric current w^as measured with a low resistance
galvanometer. The calibration of the galvanometer was made
by using a mercury thermometer and a petroleum-ether ther-
mometer.

32

ART. 6. K. HONDA AND S. SIIIMIZU.

Since the character of the pure metals and the reversible
nickel steels were not much altered by cooling them in liquid
air, the measurements of the magnetization and the magnetic
change of length at the intermediate temperatures were confined
to only the irreversible nickel steels, that is, those, whose per-
centage-contents of nickel were less than 29.24?^ (excluding
25?^ Ni).

(«) IVIagnetization of Nickel Steels.'

The observed values of the intensity of magnetization are
given in Table VIII. Here H and I have the same meaning as
before.

TABLE VIII.

NICKEL STEEL 29.24^.

t—

i.9°C

^=-:

,2.5°C

/=-62.5°C

H

I

H

I

H

I

CIS

10

0.16

8

0.16

15

0.40

41

0.23

48

066

24

0.79

79

0.49

105

2.26

76

1-34

no

1. 10

167

5.40

152

3-19

172

2.19

232

9.01

227

6.88

226

3-89

298

17.07

341

^^■33

264

11.22

381

22.65

391

22.81

287

12.49

-153

33.8

450

46.2

297

67.6

530

61.9

561

75-4

301

1209

565

120.0

670

'SrS

306

188.0

583

234-7

758

313

311

323

590

329

811

448

315

369

596

408

812

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

NICKEL STEEL 29?^.

/=I2

.4°C

^—:

>7-5°C

/=- I

2r.8^C

H

I

H

I

H

I

0.2I

18

0.24

24

0.42

8

0.61

69

0.52

52

1-93

53

0.98

109

1. 01

107

3-93

107

1.84

164

2.30

203

9.87

215

3.00

214

4.19

287

16.30

309

5-34

273

9.20

405

22.58

376

13-17

352

15-50

475

45-5

506

26.62

375

311

521

77-3

614

66.4

382

62.2

539

160.5

740

161. 2

387

126.9

544

280.8

813

302

390

261.5

546

395

845

438

392

416

548

3.5

5

47

NICKEL STEEL 28. 74?^^

/= —

[0.4°C

/=-54.5°C

/= — I

o8.o°C

H

I

H

I

H

I

0.22

16

0.25

13

1.24

12

0.43

42

0.63

62

4.29

64

0.95

95

1.83

153

9.05

153

2.09

154

2.95

205

14.64

261

3-69

210

5-17

275

21.40

359

9.68

317

10.99

382

29.83

424

19.82

383

25.02

481

49.2

536

32.0

402

69.9

542

66.9

607

64.8

415

189.0

550

131-2

741

189.0

419

325

550

218.0

827

298.8

420

420

551

310

874

445

423

380

904

2.4

3-8

91

34

ART. 0. — K. HONDA AND S. SOIMIZÜ.

NICKEL STEEL 28.32?^.

t=-l

2.5°C

^=-45.o°C

/=-79.5°C

H

I

H

I

H

I

0.31

24

0.16

18

0.45

16

0.86

55

0.50

62

0.96

40

2.63

lOI

1.41

121

2.79

86

473

128

3-33

186

6.24

142

14.98

159

7.00

252

10.98

203

32.9

170

15.46

301

17.85

266

75-3

180

28.78

315

30.2

336

180.6

193

73-4

321

63.2

417

332

205

168.9

325

90.0

451

450

211

345

327

191-4

502

438

328

305
419

526

542

0.5

0.4

19.5

NICKEL STEEL 26.64?^.

t=-(

5.6°C

t=-l

6.5"C

i'=-8o.o°C

H

I

H

I

H

I

lO.O

2.8

3-95

9

2.15

12

20.2

7.0

11.38

24

5.96

39

39-3

17.2

24.95

58

11.60

92

94.6

32.3

36.6

82

16.43

142

1957

47.8

55-9

122

31.0

290

312

60.0

129.8

168

50.1

386

470

71.9

222.0

212

63.0

437

332

247

97.2

524

446

275

169.5
224.6
321
392

623
726
736
770

15-3

64

166

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

35

NICKEL STEEL 24.40%

t=6

•5°C

t=-

i3-9°C

/=-68.7°C

t=-\

o5.5°C

H

I

H

I

H

I

H

I

11.4

1-9

5-1

2.1

3-0

7

2.6

9

16.9

3

2

II-5

6.1

10.4

28

9.2

38

27.2

5

4

19.1

13.0

16.8

57

18.5

108

38.5

7

7

28.2

21.2

26.0

119

25.1

176

55-9

II

I

43-2

00-5

35-8

187

32.1

268

989

16

I

64.5

437

57-3

283

45-6

339

193-9

24

5

102.7

58.4

69.6

335

75-7

481

291.4

31

5

170.4

75.8

104.9

401

106,8

561

391

37

0

255-5

90.5

177.0

488

166.2

651

323

99.2

297.9

569

226.4

717

407

618

320
396

786
826

7-1^

11-^

192

226

NICKEL STEEL 24.04j

t=-

6.2°C

/= —

[3-2°C

/= — (

5o.8°C

/-^-99.o°C

H

I

H

I

H

I

H

I

5-9

^3-2

4-4

4

4.1

9

2.6

8

I3-I

9.4

11.6

1 1

1 1.2

32

7-3

24

26.2

26.0

25.2

17

20.6

82

16.1

66

43-7

46.8

40.5

71

34-1

191

21.4

123

74-1

72.8

79.4

123

49 I

285

293

223

1 34- 1

95-3

159-5

175

66.2

363

49.2

367

228.8

118.8

295-4

219

lOI.I

461

62.8

452

342

137-3

393

239

184.8

580

80.4

533

457

148.3

450

250

255-6

661

1 12.5

629

400

713

240.8
282.2
323

807
842
870

50.4

87

231

266

36

ART. 6. — K. HONDA AND S. SHIMIZU.

In weak fields, the intensity of magnetization gradually in-
creases, as the temperature falls, till it reaches a maximum, and
then gradually decreases. As the field is increased, this maximum
recedes towards lower temperatures, and beyond 50 C.G.S., the
maximum altogether disappears. These changes are common to
nickel steels of 29.249^ to 28.32?^ Ni. In 24.04?^, 24.40?-^ and
26.64^0 Ni, the maximum does not appear from the outset, i.e.,
as the temperature falls, the intensity of magnetization at first
rapidly increases and soon approaches to an asymptotic value for
every magnetizing field.

{b) Length change of nickel steels.

The magnetic change of length of nickel steels is given in
Table IX and in Fig. 4 c, d, e, f, g, h, i, from which the curves
of the change of length for constant fields are obtained and
drawn in Fig. 5 a, b, c, d, e, f, g.

TABLE IX.

NICKEL STEEL 29.24?^'. NICKEL STEEL 29?^.

t=-

o.3°C

t= —

70.o°C

t=-

38.o°C

t=: — l

23.3°C

H

^/ X 10«

H

^/ X 10«

H

'/ >< ^0«

H

^xio«

4-7

0.3

4.2

0.2

3-4

0.7

10.8

0.2

i8.6

1-5

^5-3

0.8

12.3

2.8

40.5

1.8

65.1

2.0

46.1

2.2

36.9

4-7

1 16.5

5.6

1 18.6

2.7

61. 1

3-8

7^-Z

5-6

241.2

9-3

251.6

4.0

177.7

7-7

175-8

6.5

321

10.6

■^ '■i -y

4.8

287.6

9-8

302

7.0

394

II. 8

469

57

404

II. 8

415

7-7

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

37

NICKEL STEEL 28.74%',

NICKEL STEEL 28.32%.

/ = -

8.3°C

t= —

53-0°C

/=-45.2°C

/=-77.5°C

H

'/ X 10"

H

y X 10«

H

'/ X Ï0'

H

y ^ ïo"

6.8

0.8

2.7

0.3

15.8

1.4

II. I

0.2

17.2

2.1

9.9

1.6

56.0

2.2

25.7

I.O

72.9

3-5

53-^

51

149-3

3-2

349

1-4

^17-3

48

175-3

6.7

248.3

4.0

79-5

2.6

316

5-7

299

7-7

353

5-1

193-3

3-9

430

6.8

418

8.6

443

i

5-9

286.6
417

5-1
6.4

NICKEL STEEL 26.64%.

NICKEL STEEL 24.40%.

^-=-39•5°C

/=-78.4°C

/=-69.2°C

t=-gy.fC

H

y X lO*^

H

'/ ^ ^0"

H

4^x10«

H

y X 10«

45-7

0.3

26.6

0.5

21.4

0.1

28.8

0.5

127.9

1.0

67.8

2.0

47.8

0.2

67.0

1.9

268.0

2.0

166.7

4.8

152.9

2.2

177-5

3-8

442

3.8

266.5

7.0

240

3-5

281

6.6

390

9-5

336
411

4-8
5-3

389

7-9

NICKEL STEEL 24.04;

t—

2.0°C

/==-64.o°C

t=-

-96.o°C

11

4^X10«

H

y X 10«

H

^/ X 10"

40.9

O.I

43-5

0.5

20.5

0.1

124.4

0.2

1 14.0

2.0

44.2

1.0

266

0.6

^^7-?>

3-7

125.4

3-4

463

I.I

402

6.6

216.8

318

412

5-6

7-7
9-3

38 ART. 6. K. HONDA AND S. SHIMIZU.

From these numbers, we find a parallelism between the change
of magnetization and that of the length-change. In weak fields,
the change of length gradually increases as the temperature falls,
till it reaches a maximum, and then decreases. As the field
becomes stronger, the maximum elongation is displaced in lower
temperatures, and at last vanishes. These changes are common
to nickel steels of percentages higher than 28.32^0 ; for percent-
ages lower than 26.64 ?o, the elongation for a constant field at
first increases gradually and then rapidly, soon approaching an
asymptotic value, as the temperature falls.

III. THIRD EXPERIMENT.

In the third series of experiments, the magnetization was
measured at different stages of ascending as well as descending
temperatures, the measurement of the change of length by mag-
netization being left for future experiments.

The heating w^as effected by means of an electric current ;
a porcelain tube (external diam. = 1.7 cm, internal diam. = 1.05 cm,
length =47 cm) was covered with a few layers of asbestos paper,
and the lower part (36 cm) was wound anti-inductively with a
platinum wire 0.4 mm thick at the rate of 2 turns per cm. It
was then wrapped in asbestos papers to a thickness of about
5 mm. To the upper end of the porcelain tube, a brass flange
was fixed, while to its lower end, a short porcelain cylinder
was inserted tightly, so as to arrest air currents. The length of
this cylinder was so chosen tliat when the tube was placed in
the right position in the central line of the magnetizing coil,
the ovoid occupied the central position of the coil. The mag-

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH. 39

netizing coil was provided with a water-jacketed arrangement,
and a coil for the compensation of the vertical component of
earth field.

The temperature of the ovoid was measured with a platinum
rhodium-platinum junction. One of the junctions was placed
in contact with the specimen at a point a quarter of the distance
from the upper end of the ovoid, the rest being well insulated
with asbestos paper. The interspace between the lead wire and
the wall of the porcelain tube was tightly filled with asbestos
fibres, and thus protected as much as possible from the convection
current. The other junction was arranged as in the second ex-
periment. The thermoelectric current was measured with a d'AK-
SONVAL galvanometer from Keiser and Schmidt, the reading of
which was corrected by the authors with a mercury thermometer
containing nitrogen below 5oO°C, and by Professor Nagaoka
and Mr. S. Kusakabe with the melting point of sodium chloride.
A low resistance galvanometer was, at the same time, employed
to measure the temperatures lower than 200°C. A simple con-
nection permitted us to pass the thermoelectric current through
the d'AESONVAL or the low resistance galvanometer, as the case
might be.

The experiment was conducted in the following order. The
adjustments of the magnetometer and the coils, as described in
the first experiment, were effected ; the heating coil with the
specimen was then placed in the right position. The magnetiza-
tion at the temperature of the room was first determined ; then a
current from a dynamo was passed through the heating coil, till
the temperature of the specimen became constant. The direct
effect due to the current in the heating coil was tested by break-
ino; or reversinsf the current. The small deflection of the mag-

40

AKT. 6. K. HONDA AND S. SHTMIZU.

netometer, when there was any, was completely eliminated by
altering the form of the lead wires. The demagnetization by
reversals, while the heating current was passing, showed no trace
of residual magnetism, which indicates that the magnetization due
to the heating current was insensibly small. When the tempera-
ture became constant, the magnetizations at gradually increasing
fields were measured. Another stronger current was next sent
through the heating coil, and the same processes were repeated as
before. In this way, we measured the magnetization in the stage
of ascending temperature, and then that in the descending stage.
During each set of observations, the temperature was fairly con-
stant, and even in very unfavorable cases, it did not exceed 2
degrees. The temperature was always noted both before and after
each experiment, and the mean was taken. When a series of
experiments was finished, the specimen was taken out of the
coil, and the compensation tested. Excepting in a few cases,
we found the compensation undisturbed ; when, however, the
disturbance was such as to require a correction, it was uniformly
distributed.

In the present experiment, the strength of the heating cur-
rent and the temperature thereby caused were as follows : —

Current

1.8 amp

2.9

3-8

47

5-5

Temperature

ioo°C

300

600

900

1200

The heating of nickel steels protected in the manner above
described showed only a trace of surface-oxidation, if the tempera-
ture did not exceed about 800°C. If, however, the temperature
was raised to 1200°C, the surfece-oxidation became considerable, so
that the heating of some specimens was stopped at about 800°C,

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

41

if there was nothing of special importance to be gained by heating
them above that temperature.

(a) IVIagnetization of ferromagnetic metals.

The ovoids were first cooled in liquid air ; the observations at
ordinary and then at higher temperatures were taken ; the results
are given ia Table X.

^=7i4°C

TABLE X.
SWEDISH IRON.

^=736°C

^=757°C

^=76i°C

H

0.13
0.26
0.50

1-39
2.48
22.46
258.6

377

49
115
338
533
597
744
829

834

H

0.13
0.79
18.42
115. 6
258.0
398

I

H

40
220
373
493
542
561

0.05
3-29
2939
125.8

305
425

I

50
121

166
211
241
250

H

8.1

75-5
188.9

n o ^

437

I

11.8
39-0
64.5

85-3
97.6

t=yy2°C

/=849°C

il =ioog°C

/=I2I4°C

H

I

H

I

H

I

H

I

52
140
281
444

6.3
15-4
23.1
29.0

128

300
443

2.0
3.4
3-9

99

280

439

1-7

3-4
3-9

92

286
442

1.4
2.8
3-1

42

ART. 6. K. HONDA AND S. SHIMTZU.

NICKEL.

^=205°C

/ = 3i9°C

/=362°C

^=4io°C

H

I

H

I

H

I

H

I

1.58

5.62

18.83

857
207.4

334

447

71
190

314
390
398
401

403

1.77
9.06
3I-0
124.5

285
464

99
176
204
210
213
215

32.8
182.4

363
481

0.8

5-3

9.0

10.8

55
179

345
479

i.i

4-4
6.5
8.1

^=5

i8°C

^=678°C

^=874°C

/=ii49°C

H

I

H

I

H

I

H

I

183

3.8

182

3-5

236

4.6

129

2.8

366

6.3

343

5-4

363

5-4

349

5-1

478

7-3

477

6.4

479

5-9

481

5-7

ANNEALED COBALT.

/=i85°C

/=307°C

/=428°C

/=546°C

H

I

H

I

H

I

H

I

13-45
19-52
32.4
63-5
94-3
1 84. 1
279.2
376

49

96

198

325
411
568

677

758

3-87

7-43

14.69

37-6
92.8

195-7

319

370

45
112

319

570
770

913

998

1018

3-22
8.23

13-41
21.46

57-7
1 19. 1
240.1
379

22

83
185

514
662
812
906

4-27

13.62

22.43
59-9
126.7

253-2
374

57
141

406

547
710
832
928
978

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

43

^ = 6

i9°C

/=:77o°C

t=9

ï9°C

t= io6o°C

H

I

H

I

H

I

H

I

2.48

32

2.40

50

1.26

46

4.14

56

4-79

108

4-45

151

2.55

141

22.03

84

7. 88

250

13.21

411

13.83

399

51. 1

98

31-31

573

28.15

531

42.20

518

90.0

106

72.1

716

76.0

674

94-4

591

212.5

126

128.5

806

134.0

749

184.9

639

333

136

271.4

909

262.9

^^-3

291.6

665

439

140

378

951

386

856

400

674

/=io66°C

t= io74°C

^=:II09°C

2'=I2I9°C

H

I

H

I

H

I

H

I

2.25

7-54
46.8
149.6
244.1
436

25
44
62

87
100
107

7-5

41.7

172. 1

343
446

10.6
19.4
32.8
41.9
45.0

142

324
446

3-0
6.1

7-7

130
284
443

2.2
3-6
4-3

The magnetization of iron and nickel at high temperatures is
so well known that it is superfluous to give all the numerical
data obtained by our experiment. Hence in the above table, the
numbers for iron and nickel are limited to those at very high
temperatures, in which they become of interest.

Swedish iron. The magnetization in constant temj)erature
was measured at 20 different temperatures in ascending as well
as descending stages ; the curves of magnetization were then
plotted against the internal field. These curves were cut by an
ordinate of constant field. The curves of magnetization in a

44 ART. 6. — K. HONDA AND S. SHIMIZU.

constant field plotted against the temperature were thus obtained,
and are given in Fig. 6 a.

The change of magnetization of Swedish iron by temperature
rise was found to agree well with the results obtained by previous
investigators*. The weak magnetization beyond the critical point,
as first observed by Curie, was also noticed. Here the magnetiza-
tion at different temperatures ranging from 800°C to 1200°C
diminishes very slightly as the temperature rises. Thus the
meaning of the critical point becomes vague ; H. Du Bois de-
fines this temperature to be a point of inflexion in the curve of
magnetization to temperature ; but it is more convenient to define
the temperature as the point of the maximum curvature. The
critical temperature so defined is, in the case of Swedish iron,
780°C for H =400 C.G.S. It is also to be observed that the
critical temperatures for ferromagnetic metals and alloys depend
more or less upon the strength of the field.

The magnetization in a stage of descending temperatures falls
a little short of the magnetization in ascending temperatures for
the same field and temperature. But at ordinary temperatures,
they coincide with each other. Combining the above results
with those of the magnetizations at the liquid air temperature,
we obtain a hysteresis curve with regard to temperature, whose
lower range is considerably extended by the present experiment.

Annealed Cobalt. As in the case of Swedish iron, the curves
of magnetization to temperature were obtained, and are given in
Fig. 6 b.

*) J. HoPKiNSON, Phil. Trans. CLXXX, p. 443, 1889; Proc. Roy. Soc. XLIV, p. ;517,
1888. LYDALLand Pocklington, Proc. Roy. Soc. LIT, p. 228, 1893. D. K. Moekis, Phil.
Mag. XLIV, p. 213, 1897. Ledeboer, C.R. CVI, p. 129, 1888. TomlinsOxV, Proc. Phys.
Soc, IX, p. 181, 1888. CuKiE, C.R. CXV, p. 805, 1892; CXVIII, p. 796 and 859. Wilde,
Proc. Roy. Soc. L, p. 109, 1891. Kunz, Elekt. Zeits., XV, p. 194, 1894. Wills, Phil.
Mag. L, p. 1, 1900. Xagaoka and Kusakaee, Jour. Coll. Sei., XIX, Art. 9., 1904.

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH. 45

The magnetization of annealed cobalt at high temperatures
was first observed by Professor Nagaoka and Mr. Kusa.kabe'^.
The present results generally agree with those obtained by them,
but in our case, the cooling in liquid air slightly altered the
the magnetic property. In Fig. G b, the point corresponding to
the magnetization in liquid air is also included.

As the temperature rises from — 186°C, the magnetization in
a constant field increases at first slowly and then rapidly, till
it reaches a maximum at about 300°C, after which it decreases.
The magnetization reaches a small minimum, and theii begins to
increase, and after passing through another maximum, rapidly
decreases, reaching its critical point at 1090°C for H = 400. The
descending branch of the curves cuts the ascending branch at
about 850°C from downward to upward ; but its general course
is similar to that of the ascending curve. The minimum point in
the ascending branch is about 4o0°C, and nearly coincides with
the singular temperature observed by us in the change of length
by magnetizationf ; at this temperature the sign of the length
change is reversed for all fields.

It is also to be noticed that the course of the curve beyond
the critical point is nearly parallel to the axis of temperature.

Nickel. The specimen, which was first cooled in liquid air,
was heated and the magnetizations at ten different ascending
temperatures were observed ; since the dynamo stopped, when the
temperature attained 1150°C, the magnetizations at decreasing
temperatures were not taken as in tlie other cases, excej^t for the
maximum field only.

The curves of the magnetization to the temperature are

■•^■) Nagaoka and Kusakabk, loc. cit.

t) Honda and Shimizu, Jour. Coll. Sei., XIX, Art. 10, 1903; Phil. Mag. VI, p. 392, 1903.

46 Art. 6. — k. honda and s. shimizu.

drawn in Fig 6 c, in which the results of the first experiment
were also included. The character of the change of magnetization
by heating coincides with the results obtained by the former in-
vestigators*. Here the range of the temperature is considerably
extended on the negative side of zero temperature. It is remarkable
that though the magnetization falls very rapidly near the critical
temperature 360°C, its further decrease is very small, and even at
1200°C, a magnetization of about 6 C.G.S. for H=400, is still ob-
served. This important phenomenon was first observed by Curie.

(6) iVIagnetization of Nickel Steels.

In nickel steels, the magnetic state after cooling in liquid
air slightly changes as the time proceeds. In some alloys, it
does not return to its initial state, when they undergo a cyclic
change of temperatures between — 186°C and 1100°C. This change
of character is greater in the irreversible alloys than in the
reversible.

The magnetization of the alloys at different temperatures
presents a striking contrast between the reversible and the irre-
versible alloys. Some of the interesting results had already been
obtained by previous investigatorsf.

The manner in which the magnetization of the reversible
nickel steels changes with the temperature is similar to that of
nickel, as given in Table XI.

"■■) J. HoPKiNsoN, loc. cit. Curie, loc. cit. Nagaoka and Kusakabe, loc. cit.

t) H. Eecqueeel, C.R. XCIII, p. 794, 1881 ; J. Hopkinson, Pro. Roy. Soc, XLVII,
p. 23, 1890, and XLVIII, p. 1, 1890 ; H. Le Chateijer, C.R., CX, p. 283, 1890 and CXI,
p. 454, 1890; H. Tomlinson, Pro. Roy. Soc, LVI, p. 103, 1894; F. Osmo>-d, C.R, CXVIII,
p. 532, 1894, and CXXVIII, p. 304, 1396, 1899; Ch. Ed. Guillaume, CR., CXXIV, p.
176, 1515, 1897; CXXV, p. 235, 1897; CXXYI, p. 738, 1898; Les aciers au nickel, Paris
1898; E. DuMONT, C.R., CXXVI, p. 741, 1898; L, Dumas, C.R., CXXX, p. 357, 1900.

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

47

TABLE XI.

NICKEL STEEL 70.32?^.

t=l

[.7°C

/=i04°C

t=4

[o°C

/=658°C

H

I

H

I

H

I

H

I

0.34

9

0.40

16

0.41

256

35

2.8

1.07

42

1.08

69

1.89

368

132

9.1

173

232

1.25

164

8.92

485

276

18.6

4.28

570

1.60

374

31-4

536

476

29.2

12.29

716

2.86

516

64.1

545

25.09

864

29-5

867

153-5

552

41. 1

947

68.3

944

294.6

565

76.4

1000

166. 1

970

434

573

1937

1029

264.6

980

393

1043

405

989

/ = 733°C

^'=900°C

^=753°C

^=ii.5°C

H

I

H

I

H

I

H

I

48
100
248
469

3-9

5-6

14-3
26.7

125

305
491

3-8

7-9

10.3

40

304
488

17

5.0

8.5

I I.O

0.29

1.08

4.22

21.64

56.3
204.4

349
405

28

171

491

908

1008

1034
104 1
1045

48

ART. G. — K. HONDA AND S. SHIMTZU.

NICKEL STEEL 50.72^

t=\:

2.0°C

t=\

26°C

t=2

96°C

t=A

io°C

H

I

H

I

H

I

H

I

0.54

21

0.95

31

0.76

42

0.40

34

1.91

79

2.90

336

1.27

133

0.70

154

270

189

4-30

536

4-44

548

1.67

314

3-29

334

9-54

697

21.39

791

8.81

A77

22.68

971

27.40

972

41.9

^77

2443

536

47.5

1123

52.1

1077

91-5

898

57-6

558

108.0

1218

115.9

1 146

213-7

913

1594

565

186.6

1250

21 5.1

1166

417

921

305

567

309

1263

398

II 76

443

579

386

1267

/=639°C

/=8o7°C

^=634°C

/=495°C

H

I

H

I

H

I

H

I

21

0.7

31

i.i

31

0.7

30

0.6

113

3-9

206

6.9

136

4-7

98

34

267

8.2

487

II-5

322

9,0

345

10.7

490

II-3

484

11.6

482

13-6

^=43i°C

t = 2

63°C

z'=i92°C

t=\

2.2°C

H

I

H

I

H

I

H

I

127

4-7

0.06

46

0.07

34

0.20

26

34-7

II-5

0.20

169

0.18

432

0.49

^7

101.6

31.2

0.45

586

0.70

649

0.77

355

245.1

46.2

85-3

941

6.53

951

0.94

582

402

58.7

269.8

949

24.20

1048

45-7

1235

47S

65.1

408

952

72.0

1069

143.0

1259

253.5

1078

261.7

1264

390

1082

378

1267

MAGNETIZATION AND MAGNETIC CHANGE OP LENGTH.

49

NICKEL STEEL 46?^.

/=i2.3°C

^=i3i°C

t=2efQ

/=345°C

H

I

H

I

H

I

H

I

0.78
1.83

3 -40

7-49
9.87
17.70
42.9
80.3
195.4
375

22

60

190

549
706

913
1113
1201
1260
1276

0.78
2.56
3.21
5. II
9.71
22.80
46.6
106.6
264.4
381

28
144

319
521

751
943
1047
1114
1140
1146

0.59
1.29
2.71
5.40
15. II
34-8
78.9
213-4

317
409

31
96

403
561

704
778

834
851

852

854

0.42
0.82
2.51

4-59
13.46
40.2

135-3
253-9
431

38
112

227
380

495
562
588
593
597

/=4

ii°C

^'=456°C

t=6

24°C

/=488°C

H

I

H

I

H

I

H

I

17

2.8

47

0.6

56

2.0

87

4-5

50

6.2

105

2.8

240

7-3

234

8.7

137

14.6

305

1 0.0

387

11.6

385

12.5

477

40.8

483

14.2

479

13-4

475

14.2

/=37i°C

^=3i8°C

^ = 2I0°C

^=33.2°C

H

I

H

I

H

I

H

I

12.3

18.9

45-6

190.6

347
459

30

45

91

143

168

183

0.59
1.51

5-24
II. 41
45.2
198.0

329
425

31
209
400
490
592
626
631
638

0.72

1-35

2.14

5.86

50.95
169.4
284
399

27

80

204

527
887
964
974
979

1.62
2.90

5-94
17.62
69.6

161.8

280

369

49

132

512

827

1 1 20

1217

1243
1252

50

ART. 6. — K. HONDA AND S. SHIMIZU.

NICKEL STEEL 36^.

^=io.o°C

t=i6i°C

^ = 225°C

/=259°C

H

I

H

I

H

I

H

I

1.24

3-51
4-32
7-79
32.9
64.2
176.2
299
419

30
119

232

499
863

954
1021

1035
1039

1.06
2.23

3-99
12.89

54-9
1351
276.5

437

55
191
308
461

547
563
572
578

5-7
17.8
48.2

259.0

478

33
45
54
84
102

31.6
152.2

355
497

0.7
6.0

13-2

16.6

^=447°C

f=S

28°C

/=423°C

/=303°C

H

I

H

I

H

I

H

I

96

3-3

105

4.7

31

1.8

79

4.2

251

8.4

251

10.5

103

5-7

232

10.5

491

13-5

488

17.1*

283

12.9

363

14.7

484

18. 1

482

17.8

^=2i8°C

t=i

82°C

t=i

45°C

/=io.o°C

H

I

H

I

H

I

H

I

30.5

14.7

0.26

54

0.21

42

0.31

32

77-4

23-9

1.54

198

0.46

127

0.71

118

160.2

35-5

8.08

335

1.51

242

1.66

314

320

55-3

20.58

364

6.46

459

4.88

556

481

70.6

71.4

379

49-6

582

16.15

801

250.7

391

151.4

592

1 12. 1

1013

447

401

292.3

597

247.9

1027

428

602

302

375

1028
1029

*) In this case, the compensation of the magnetizing coil was found to he disturbed
and therefore the disturbance was equally distributed over the whole set. This small in-
crease of magnetization is probably due to the impared compensation.

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

51

Comparing the above values for ordinary temperature with
het corresponding values in the first experiment, we notice that
except with Sß^o nickel steel, the magnetizability of these alloys
had slightly changed by the repeated heating and cooling, which
the alloys underwent, since the end of the first experiment.

From these results, the curves of magnetization to tempera-
ture are obtained and given in Fig. 7 a, b, c, d. In these figures,
we have also included the results obtained in our first and second
experiments. As seen from the figures, the diminution of mag-
netization, after the critical point is reached, is very slight ; and
to judge from the course of the curve, it seems probable that the
magnetization does not altogether vanish, till the melting points
are reached.

The curves of magnetization at a constant field in the ascend-
ing and descending stages of temperature do not exactly coincide
with each other when the range of temperature is large, the two
curves thus enclosing a small area between them.

As the critical points of these nickel steels for H =400, we
give the following values : —

Alloys

70.32^

50.72^

46^

36^/0

Ascending branch

66o°C

490° C

4i2°C

255°C

Descending branch

46o°C

395°C

240°C

These values nearly coincide with those of M. Osmond and L.
Dumas. Thus the critical point decreases with the percentage
content of nickel.

The manner, in which the magnetization of irreversible nickel
steels changes with temperature, is very striking. The observed
values are given in Table XII.

52

ART. 6. — K. HONDA AND S. SHIMIZU.

TABLE XII.

NICKEL STEEL 29.24?^.

t=l

i.6°C

t=i

40°C

/ = 2

io°C

/ = 352°C

H

I

H

I

H

I

H

I

3-21

30

5-83

31

1 1.27

57

9.87

56

9-33

103

12.57

109

18.51

162

18.10

161

14.20

200

29.8

299

29.1

276

31.3

275

22.35

355

74-3

433

68.0

428

69.4

399

48.3

507

169.5

566

164.7

566

167.7

518

79.0

604

329

678

280.5

646

308

595

150.5

723

449

729

442

713

445

639

244.2

810

345

870

417

902

^=499°C

^=547°C

^=799°C

^=466°C

H

I

H

I

H

I

H

I

20.3

6.2

51

1.8

51

2.0

52

2.2

56.0

18.9

168

6.1

140

4.8

152

5.8

183.2

57-3

357

10.9

308

9-3

311

12.8

304

80.4

495

139

494

12.3

494

13-7

410

94.6

487

102.5

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

53

7=299°C

i'=i92°C

/=93°C

/=I2.0°C

H

I

H

I

H

I

H

I

31
105
256
376

497

I.I

4.9

10. 1

12.8

15-5

31
103

254

375
494

1.4

5-1
10.5

13-6
16.2

45
107
272
494

2.2

5.8

13.2

20.6

0.21

1.63

5-15
27.91

95.6

188.7

334

477

46
107
182
250
265

275
286

295

NICKEL STEEL 29?.'.

^'=i4.3°C

t=\

68°C

^=268°C

^ = 383°C

H

I

H

I

H

I

H

I

4-35

3

5-95

23

7-43

21

9-85

44

5.31

71

12.60

%6

14.55

71

26.24

153

12.31

214

19.61

176

23.40

173

49-3

230

20.76

367

30.5

254

36.1

257

109.7

314

29.1

432

64.5

353

1154

407

231. 1

387

57-5

585

208.7

503

281.5

512

441

446

125.2

717

323

561

434

561

238.0

815

436

600

416

889

^=465°C

^=5i5°C

/ = 644°C

/=799°C

H

I

H

I

H

I

H

I

16.9

45-4

93-3

212.4

321

448

16

50

96

161

192

217

45.7
161. 3
340
473

2.3
10.8
22.9
29.9

54
280

475

2.0

7-7
10.3

94
282

477

2.5
3-6
8.9

54

ART. 6. K. HONDA AND S. SHIMIZU.

^'=6oi°C

/=405°C

t=2o6°C

t=i.

2.3°C

H

I

H

I

H

I

H

I

65

2.3

94

3-4 •

86

3-4

0.41

78

247

7-3

232

7-4

229

8.0

1.36

182

479

10.4

344

9.0

358

10.7

4-54

302

476

10.8

473

12.2

17.32

138.4

369-9
466

381
412
426
428

NICKEL STEEL 28.74?^.

/=i:

2.3°C

/=87°C

t=\

43°C

t=2

28°C

H

I

H

I

H

I

H

I

3-09

22

3-99

21

5.66

32

3-63

14

7.20

67

8.26

52

10.47

92

10.53

61

14.04

192

12.79

114

15-30

177 ,

16.50

163

21.16

340

20.17

244

23-95

310

25.82

316

51.9

569

27.27

331

47.5

426

54-9

468

127.0

777

52.0

460

75-9

518

126.8

611

225.7

896

82.6

564

203.1

701

249.0

722

307.2

955

185. 1

716

318

784

333

766

422

1015

313

429

820
882

436

841

450

813

/=327°C

t=4

i8°C

/=497°C

^=634°C

H

I

H

I

H

I

H

I

8.01

38

6.45

41

23.6

5-9

83

2.5

16.46

167

13-19

1 1 1

52.4

14.6

214

6.9

25-33

298

22.50

193

97-1

28.4

239

10.2

59-9

461

33-6

256

158.9

42.6

507

12.6

109.9

556

68.5

352

364

79-4

224.1

659

II7-3

428

506

93-9

365

726

266.7

527

443

752

446

584

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

55

/=795°C

/=399°C

^=ioi°C

/=24.2°C

H

I

H

I

H

I

H

I

60

157
342
511

2,1

5.2

9.6

12.0

98
262
511

4.2

9-7
14-3

24
107
258
507

I.I

6.6
14.6
24.2

O.IO

3.85
23.07

96.1

257-5
490

41
207
285
303
319
334

NICKEL STEEL 28.32^

if=i2.5°C

/=i

3o°C

/=300°C

^=453°C

H

I

H

I

H

I

H

I

3.28

21

S.16

26

7-43

33

12.45

28

6.67

75

9-36

77

17.01

156

28.82

72

14.58

183

17.05

233

25.74

293

45.6

114

24.56

328

25.68

348

47.2

434

89.1

185

41.4

438

40.8

432

95.6

582

189.2

287

65.6

555

81.7

579

194-3

697

314

339

132.1

713

183.5

734

318

778

469

401

266.6

861

315

833

439

824

420

962

423

872

/=504°C

^=649°C

/ = 8

i8°C

^=58i°C

H

I

H

I

H

I

H

I

22

1.4

114

4-5

120

3-9

109

3-9

52

3-9

217

7-2

263

7-9

246

7-9

276

24-3

516

12.5

513

11.4

385

10.8

506

39-5

510

13-7

56

AKT. 6. — K. HONDA AND S. SHTMIZU.

/=387°C

^=200°C

52°C

I2.8''C

H

I

H

I

H

I

H

I

105
216

367
512

4.8

8.2

11.7

14.2

105

215

341
511

5-1

9.1

12.5

15.9

46
III
388
513

3-7

8.4

23.1

28.8

3.19

18.38

37-8
125.2
289
492

20

36
46

72
102
126

NICKEL STEEL 26.64^.

i=i

5-3°C

r=i

2I°C

t = 2

39°C

^=392°C

H

I

H

I

H

I

H

I

2.55

10

5.20

25

6.38

27

5-03

55

10.22

65

9.02

59

13-88

93

9.42

138

16.87

154

16.46

163

19.21

209

13-93

226

25.32

292

24.45

297

26.04

341

24.14

364

37-4

402

33-2

384

39-1

482

60.6

553

57-4

581

50.2

533

54-1

613

1 1 1.9

665

105.3

737

104.9

738

100.9

755

226.4

774

1 80. 1

876

179.4

867

192.3

882

420

849

271,1

976

290

969

330

982

406

1064

401

1035

395

lOOI

^=463°C

/=5i7°C

^=665°C

2'=828°C

H

I

H

I

H

I

H

I

7.27

15-03
28.18

40.5

lOI.I

216.1

352
439

28

63
121
167
297
417
480

538

50
204

484

2.4
10.3
20.2

104

248
484

3-9
8.8

13-9

79
301

481

3-5

9-8

13.6

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

57

t = S

I4°C

/ = 40i°C

/ = 227°C

/=8o°C

H

I

H

I

H

I

H

I

103

4-5

123

5-4

124

5.2

113

5-6

301

10.3

303

10.8

301

10.5

301

12.4

480

14.4

479

14.8

480

15.0

479

16.6

NICKEL STEEL 24.40?^.

^=1.

5.o°C

/=ioi°C

t=\

90°C

t=z66°C

H

I

H

I

H

I

H

I

3.89

14

8.42

37

1 2. Il

71

9.00

58

12.35

66

19.06

155

19.50

184

16.20

187

20.60

171

27.62

282

27.16

304

23-75

331

27-75

297

48.4

450

49-5

508

40.5

475

56.09

497

81.7

618

61.3

579

62.9

589

126.5

744

152.4

775

148.3

791

132.0

733

184.9

837

229.5

864

261.2

903

213-3

805

296

941

308

926

420

988

312.0

856

400

1009

419

991

427

895

^=466°C

^=490°C

t=^

92°C

/=207°C

H

I

H

I

H

I

H

I

5.25

35

8.48

25

12.42

30

10.39

37

10.72

89

18.47

60

15.66

73

23.25

96

25-72

234

39-2

138

40.5

125

37-78

167

44-7

338

73-2

226

80.3

230

76.7

302

II3-7

521

145.4

343

165.8

362

138.5

438

244.9

645

296

456

279.4

460

272.2

575

442

724

458

516

455

542

447

661

58

ART. 6. — K. HONDA AND S. SHIMIZU.

/=I4.2°C

t = 2

37°C

/=4

io°C

/=490°C

H

I

H

I

H

I

H

I

5.78

32

8.17

28

12.55

33

14.36

29

13-39

86

16.48

61

23.55

68

37-3

90

20.82

155

31.36

126

38.2

126

90.4

209

47.2

375

74-9

267

73-2

201

184.1

327

77-0

543

155.6

425

105.2

270

342

423

164.8

722

309

569

230.6

413

434

465

260.0

797

449

631

334

489

412

880

447

538

^=542°C

^=59i°C

/ = 85i°C

^=iooo°C

H

I

H

I

H

I

H

I

29.2
98.8

333

472

13

50

128

146

56.7
166.5
308.2
483

2.0

5-2

8.2
10.5

57.1
152.0

308.3
486

1.7

4-1

7.4
Z.7

105
305
483

2.5
6.0
7-3

/=I200°C

/ = 86i°C

/=2I2°C

/=r4.6°C

H

I

H

I

H

I

H

I

290
448
480

3-4
3.6

3-7

105.3

304

481

2.1

4-7
5-4

64.4
165.4
323
485

5-1

8.9

11.9

13.8

19.6

46.2

92.8

193.6

351
464

25
83
124
176
214
233

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

59

NICKEL STEEL 24.04^.

/=io.i°C

^=i82°C

/=287°C

^=452°C

H

2.44
1074
27.60
35-6
58.2
80.6
II 5.0
185.6
276.2
393

H

42
230
306
520
641
756
888

985
1060

3-68

1 1 43

20.48

26.97

39-1

58.9

90.7

174-7
284.5
395

I

H

13

54
138
270
386
552
703
875
982
1046

348
12.36
18.63
23.06

51.8

95-7
183.4
275.8
38-,

I

H

14

17

193

313

427

592
760
886

954
1002

3.81
12.27
24.91

477
71.9

164.3

271.2

447

I

27
123

254
372
452
602

673
745

^=5

3i°C

/=586°C

/=7

i7°C

^=6o4°C

H

I

H

I

H

I

H

I

41

1.9

lOI

34

124

4.6

64

34

184

7-9

268

7-7

348

1 0.0

236

7-7

370

13-7

487

II-5

487

12. 1

411

II-3

490

16.2

489

12.9

/=502°C

H

I

^=274°C

H

I

^=i93°C

H

I

^=ii.3°C

H

79
250
392
487

30

9.0

1 1.9

13-8

50
151
352
487

2.8

7-7
14. 1
16.6

34
102

310

494

1.8

5-7
12.9

16.5

57
165
286
484

2.8

8.9

13.8

18.9

60 ART. 6. — K. HONDA AND S. SHIMIZU.

From these values, the curves of magnetization to temperature
are obtained and drawn in Fig. 7 e, f, g, h, i, j, k. In these
figures, we have included the results obtained in the first and
second experiments.

Here we also notice that except with 28.74^, the magnetiza-
bility of these alloys had considerably changed by the heating and
cooling, which the alloys underwent since the first experiment.
Hence in some of the figures, the portions corresponding to the
first experiment were displaced parallel to themselves so as to
form closed curves. Thus the displaced portions are given in
dotted lines.

As the temperature gradually rises from — 186°C, the mag-
netization of 29.249^ Ni diminishes at first slowly, then rapidly,
and after passing through an inflexion point, the diminution
becomes slow. The curve, passing through a second inflexion
point, begins to descend very rapidly, as the critical temperature
is approached. If this temperature be passed, the diminution of
the magnetization by heating is very small, so that the curve is
nearly parallel to the axis of temperature. From the course of the
curve, it seems probable that the magnetization does not altogether
vanish, till the melting point of the specimen is reached. As
the temperature is next gradually reduced, the increase of mag-
netization is very small ; this state continues, till the temperature
falls to about 100°C ; then the increase becomes very rapid. For
example, in H=400 C.G.S., the intensity of magnetization at the
descending temperature is only 20 C.G.S. for a temperature of 80°C,
but it amounts to 200 for 20°C, and at — 60°C, it increases to
790. Thus the magnetization of the specimen displays a remark-
able diflerence between the ascending and descending branches of
the curve.

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH.

61

The above manner, in which the magnetization is changed
by temperature, is common to all other irreversible nickel steels.
As the percentage of nickel decreases, the concave portion of the
ascending branch becomes fainter and fainter ; and with 24.409^
and 24.049^ Ni, it almost vanishes for strong fields. Apparently,
the forms of the two curves for nickel steels of 29.249^ and
24.04?^^ Ni, are widely different from each other ; but if we
compare the forms of the curves of two consecutive nickel steels,
we can trace transition stages from one form to the another.

The critical temperatures of the alloys for H =400 C.G.S.
are given in the following table : —

Alloys

29.24°^

^9%

28.74?^

28.32^^

26.64^^

24.40?^

24.04^

Ascending branch

53o°C

53o°C

530°C

5io°C

5io°C

580^0

52o°C

Descending branch

70

140

80

50

lO

130

40

Thus in the ascending branch, the critical temperatures of
these irreversible nickel steels are nearly equal, except with the
last but one. The above numbers fairly coincide with those ob-
tained by M. Osmond, except with 24.40^ Ni. AVith this alloy,
the critical temperatures are greater, in our case, by about 50°C
for the ascending branch and 100°C for the descending, than in
the experiment by Osmond. The values given by L. Dumas for the
first four of these alloys are considerably less than those obtained
by us ; but for the remaining alloys, the contrary is the case.
These discrepancies may probably be due to the previous history
of the alloys.

It remains to mention a singular phenomenon. If at a point
in an ascending branch of the temperature-cycle, the temperature
be reduced to the ordinary, the path is utterly different from the

62 ART. 6. K. HONDA AND S. SHIMIZü.

ascending one. If, however, the temperature be again increased
to its former value, the path nearly coincides with the former one ;
the further increase of temperature diminishes the magnetization
in such a manner that the magnetization is not interrupted by
the cooling process. An instance is seen in Fig. 7 j. Hence in
irreversible nickel steels, the magnetization at ordinary tempera-
ture can have any value ivhatever within a given limits, if the
cooled specimens be heated to a suitable temperature. Becquerel
who first studied the magnetic properties of irreversible nickel
steels, found that in the alloy, there were two states of stable
equilibrium ; but according to our results, there are an infinite
number of such states, a fact which may possibly prove to be
important in the theory of molecular magnetism.

Comparing the magnetization at different temperatures in
these nickel steels, we notice that the critical temperature in the
descending branch of the temperature-cycle generally becomes
less, as the percentages of nickel decreases. As the content of
nickel diminishes from 70.329^ to 26.64?^, the critical temperature
falls from several hundred degrees to the ordinary temperature.
It is then highly probable that 25 9^ nickel steel, which is feebly
magnetic both at ordinary and liquid air temperatures, would
become strongly magnetic, if the cooling should be pushed still
further. If it once become strongly magnetic by cooling, it may
preserve this property, after the alloy is heated to the ordinary
temperature. It will be interesting to investigate, whether other
non-magnetic alloys, which consist of a magnetic and a non-magnetic
metal, would display a similar phenomenon on being cooled to a
sufficiently low temperature.

The fact that the two strongly magnetic metals form a non-
magnetic metal is then nothing more than the lowering of the

MAGNETIZATION AND MAGNETIC CHANGE OF LENGTH. 63

critical temperature of the alloy to the ordinary temperature.
Owing to some changes occuring in the molecular configuration
during the process of fusion of the constituent metals, the critical
temperature of the alloy in the descending branch of the tem-
perature-cycle falls to a low temperature, and therefore the alloy
behaves as a weakly magnetic or non-magnetic alloy at ordinary
temperature. The same remark will also apply to a non-magnetic
alloy which consists of a magnetic metal and a non-magnetic ones.
The above view is also favored by the fact that in irreversible
alloys, the hysteresis-loss at ordinary temperature is markedly
small, which corresponds to the hysteresis of iron or nickel at
high temperatures, but its value at a low temperature considerably
increases, corresponding to the hysteresis of the same metal at
ordinary temperature.

Jour Se/.

Coll. Vol XX. Art. VI. P

/.

1 j

A 1

__

-

-

-

-

—

_.,,

-

-

...

-

-

—

-

-'

—

--

F„

1-1

a..

1

/■i

r. /

2.

fL 2,

b.

JV6

Joe.

xteeU

1

/

j ■

^

w

W

,

1

/

i

/ /

/.

--

■~

-

-

j

^

^

W

k

'

__

1

>//

i

1 1 ;

j

1

//

/4

j

1

1^

d'

^

yij

i 1 1

/

If

y>

^**

r

/

j

;

/

1

i

^

//

/

W

1

f

/

/v

/

\

,

1

/

'/

>

>«^

~ier

tper'

itta

'■

\

Zi^uJfl a|w-.

1

v^

k

'

A

„

^-

/,

^

SX

•o^

i

M-

/

1

"

""

"

1

!■/

.f

f

^/

/

^

r/

/

i

/

1

'

V

i\

^ /

1

/

J ' _ 1

s

1

/

1

/

/

-

-

-

-

1 ^

/

1

'

1 1 1

1

Fi

7.2

tz..

/v

.cité

Isl

eef.

1

■7

/

1 i

t

1

"•

L

1

1

f

1

1

/s

^A

9'

1

1

'

/

3„

■'

?

":;^

1 f

/

If

\ /\S'

Or

iùi

■tl-l/

ten

pei

liu.

'•e..

1

'i/

/

-

]

/

'/

i^5

^

!>■'

1

II

/

//

1

^'.

^«

é

,

»,

y

II

'

/

u /

h

7

/

/

^

^y\

^

^

//

r'

^^

..

%

^^

.,«

■■2',,

-

é

£.

»

^

,^o.

32%

1

Jâ

'

/y

^

^

-^

=s=S

^

1

^

■^

-^

//,

'//AC

--

- -

-

Â-

€^

s^

f

>l

^

^

».7»

ïf. ^/

%m

^

s^

^

-^

-

"

ro*-

"

ld''~

°

"

»•

-

~-

-

"

—

1

1

1

t

J

Jour. Sei. Coll. Vol. XX. Art. VI. PI. II.

' ; 1

>

/■"/

_\\m_

ilour. Sei. ColL Vol. XX. AH. VI. PI. I

i

1

;

1

'

'

1

1

1

-

-

-

~f%

r-s.

a,j

>B.i

f°A

A^.

Fi.^

r.S,

b,i

9%

PH.

Fii

.S,

^,2

».7i

%.

fi-

Fiif.S,

d,i.\S^^9^i.

i

i

4

N

i

f

~iV

-

-

\

m

aoo

\

Vt

' s

N

~^

^

~Ä

Ï»?

~>

^x

\

--J,

-s^

,r-i

»_

^

-^

-^

*«

^

^-~

-

-

■ —

>

^

^

\

rr-ä

-^

'

^

~~

~~-

"'"

5^

H

—

—_

<

-26

fc

--

—

-1

°°

-/

°

"

''

o-c

"

^i.

7 s

ej.

6.6

f%

Vi,

^

V6^,.2

â.é

\%.

Vi.

Fh

-.5,

7,^.o\%

Ifi

i

I

--

^

■

^f

-

^

\

\

1.,,-rf

:

~-

-

N

~

!^

>~

■^

'=

\

\

,

\

S

'-

~

"x,

\

_

1 ,

\

~~;

^?5j~~~-.

^

\

V.

^

i^

^

V

^

~ts

"■'

-1

"

"'

0-c

''•

°

°''

°

—

-

-

—

-

1

—

-^

i^

Pi

y. 6

a..

-

-

„

00

S

V

Sw

'dta

& Ù

Fl

?.«,

b.

-~

\

Ca

bay,

*o

alec\

—

-

1

^

Fig. 6

o.

■ "■

-

«

w

^

^

too

NicktiV.

\

t

7

\

^

'*"'

^

^

\

S

-

"

B--

T'

Nl

\

1

/

^

^

K"

^^

—

.^

\

1

r^

J

.

^

^

'-

W

1

/

/

^

■-<<

\

\^

^

\

—

.__

1

^

/

\

/

■^

V*^

*

-y

-

-1

-

1

■A

/

\\

1

/

V

—

k-

-

1

V

"^

-

1

^

1 )..!__

--

-

-

4 1

1

1,

1

V

_

■*

"

'

t

+»

'

"

„

'

o

'

'"

*

10

«

>o

«

w^

-^

"

s

"

°

■*

"

1

1

L

._!

1

Jour. Sei. Coll. Vol XX. Art. W. PI. IV.

1 —

V '.

:

T-

^^

~

^~

^

<

"

-

... .

\

1

—

Fii

1

-

'-

-

nr

i I

-'

.7,i

i. i m-^'

.:■>

1.3

r

VÙ

>

Fe^.7,

i,6

1.71

%

n

-

-

\

V

^^

■7,

>,*!'%

Vi

-

\

\

—

,3t

%fi

'"" \

1 1 1

'

-

"^

\

~

1

' •--.

<

^-.

^ ~

—

-

-

-

feN^

--

-

-

f.-x

-

—

h

"

-i\v

1

«i^v't

1 1

~-

^i

-

-'

-j

KO^-

j

—-

s?

r-^

-

"b'

;r^

s

1

^

">! ^

s

'\

^\

\

N^

—3

go

^

.....

-

-

-

1

-

\

--

-

-

1

—

-

-

\

1

—

\

i*^

t

^

t

^r

t

l

1

-i]i>a '

1

'

.0

'

'

'

gov

0Û

'

*

"

"

o'o

■'

'

"

*

'f

*

ote

"'

'

'

'

«fc

"

V

-,.

■"

-

-

N

i

-

1

\

1

N

-

-

-

—

^

\

Fii

1-7.

',2

9.3

i%

ffv

\'^

N^

\

Fii

■zj

\2i

%.

Vi

\

fh

■7,

jid

.7^

%|w;

--

-

__^

-

I

E

V

\

\

\

,-.

--

-

1

\ \

\

\

\

V

^^

s«,-

^

^^-1

f-.

\

"*'

't^

\

i_

\

N

\

■s-

-

\

-

! "

\

\

\

^

\

\

i 1

:\

-

\

\

A

\

V

>,

\

^ ü

1

"i-s

\

\

V

\

\

\

—

-

'

1 ^

\

\

^

\

\\

1

1

v

V

\

\\

'-

t

-

-

-

1

\

i 1 ■ :

V-j

\

t

V.

\

\

V

^

e

"'

oa

"

'"

°

IC

'

-

°

II

»>'c

"-

■'

X

i

°

'

■"

°

"

T

'

.

-

-

-

-

\^

«^

^<

.7X;^.0i

"

■.. ,.__.!__..___

.^

\Fii

."„ Xf

^

^:

_Ph

.7,%,2i1.32

%.

Vi

Fis

7,i

,2t

et

%

Vi

L>

Fi,

fZ

i.,2

..*

>%.

Vi

-

—

—

^

1
1 1 - 1- -

\,

---»

,

\

y

^

^S

s

—

-

~

^

1

\

\

\

.\

°°~

<

^ 1

\

^

-

—

'

\

\

\

\

--

£:i

i-.

A

P

\l ■

__.J_J

_

_^

>t7.

^

'

\ '

..,

V

\

—

-

~

/

\?

\

\

u

\

\

>,

k-L

\

-

--

1

««,

—

U--li J

,

^

—

/

1

~

—

~

-

/.

--

-l-

\

-r

■^■

:

11

''n

\

\

f

^

-

—

i

-

X

\

'

i.

\

\

^

\

I

'^ ! 1

\ V

t

\

f

\V

V

e

-. ' '

' : »f" f » «

■'

f

■''

" ! "" \

—

-ö^

"

5iE

-

-=i

IS—

" 1

—

"^

—7.

ö—

-^

ö^

,0,0

,100-C\

.„1

"

■■ 41,

- \

1

1

1

1 1

1

JOURNAL OF THE COLLEGE OF SOIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN,

VOL. XX., ARTICLE 7.

On a New Cestode Larva Parasitic in Man

(I^lerocercoitles pt'olifer).

By

Isao Ijima, Rig., Ph. D., Rig. -Hah,
Professor of Zoology, Imperial University, Tokyo.

With one plate.

The remarkable Cestode larva to be described in this paper
probably belongs to the Bothriocephalidae. It may be called a
Plerocercoid and the name, Plerocercoides prolifer, is given simply
because it is desirable that there should be a distinctive name
for practical use.

For the opportunities of studying this most interesting parasite
I am indebted to Dr. J. Kondô, Professor of Surgery in the
College of Medicine, Tokyo Imperial University, and also to his
assistants, Drs. H. Shioda, M. Yamamuea and K. Sugi.

The patient who harbored, especially in the subcutaneous
tissues, enormous numbers of the worm in question, was a woman,
Yae Tanaka by name, and resident either in Tokyo or in its im-
mediate vicinity. Her occupation had been that of a weaver until she
was married to a dealer in old furniture, occupations which place
her decidedly in the lower class of society. In the spring of 1904,
she came to the University hospital for the treatment of an in-

2 ART. 7. — I. IJIMA.

guinal hernia. Dr. Shioda tells me that the cause of the hernia,
which was on the left side, was traceable to the presence of tlie
pnrasites in the region of the Ligamentum Poupartii. The patient
was then thirty-three years of age. Aside from the affliction
complained of, what at once attracted the attention of the medical
attendants was the peculiarly swollen state of the integument
which bore scattered spots of acne-like appearance. This abnormal
state of the skin extended over nearly the entire body, except, so
far as I could roughly observe, on the face and the upper ex-
tremities, but was most strikingly noticeable on the left thigh.
This was excessively swollen and had very much the appearance
of elephantiasis, though the skin and the underlying tissues were
quite soft so that they hung down by their own weight and could
be grasped in a flaccid mass by the hand.

Of the statements made by the patient I will refer to only
a few points. When twenty-five years old, she had had a tape-
worm ; but that surely had nothing to do with the present para-
sitic disease. The dermal flPection was said to have been first
felt in her thirty-first year, i.e., about two years previously to her
coming to the hospital. It had apparently given the patient no
trouble beyond that imposed upon motion by the swollen thigh
and the itching of the skin in parts where a pimple-like hardening
made its appearance. Scratching with the nails in order to ap-
pease the itching had led to breaking of the skin at the spot,
from which could then be pressed out a soft whitish mass together
with some fluid. A number of small scars, seen more especially
on the breasts, attested to abrasions thus effected apparently not
long before Details of clinical and pathological observations
will doubtless be given in a forthcoming report by Professor

KONDÖ.

ON A NEW CESTODE LAKVA PARASITIC IN MAN. Ö

In the examinât ion of preparations made of the skin taken
from the left thigli, I was taken in consultation by Mr. Yama-
MURA. It Avas not long before I was convinced of the presence, in
the connective tissue, of numerous encapsnled worms, the Cestode
character of which was evident from the calcareous bodies con-
tained in the parenchyma.

Subsequently on two occasions, July 9th and 24th, 1904, I
was present at surgical operations undertaken to relieve the patient
of the superfluous tissues and to procure fresh material for study
and experiments. Each time a very large piece of the skin and
the underlying tissues was excised from the left thigh. I should
think that altogether several pounds' weight of them was removed
during the patient's stay in the University hospital.

Of quite an unusual appearance were the subcutaneous tissues
when freshly taken from the affected limb. At places several
centimeters thick, they were moderately rich in panniculus
adiposus and extraordinarily so in lymph. The latter swelled
the connective tissue between the panniculi, giving it a slimy or
gelatinous appearance and consistency. The slimy character seemed
to be more manifest in the deeper parts. The lymph exuded
copiously from the cut surfaces. The numerous capsules with the
worm within were observable as whitish objects isolated or in
clusters, in all parts of the tissues.

In fig. 1 (PI.) I have shown, in natural size, a vertical slice
of the skin and subjacent parts taken from the lateral side of
the thigh and hardened in spirit. In the large piece, from which
the slice was taken, is represented nearly the entire thickness of
the tissues between the skin surface and the underlying muscles.
I may say that this thickness measures 30-60 nun., notwithstand-

4 ART. 7. — I. IJIMA.

ing the fact that the hardening process has contracted the sub-
cutaneous connective tissue, through loss of the lymph, into dense
fibrous bundles, so that it no longer bears a semblance of what
it was in the fresh state. The corium in the same piece may be
said to be 3-6 mm. thick ; it seems to be on the whole consider-
ably thicker than in the normal state. In the figure above
referred to, the lightly shaded subcutaneous spaces represent the
adipose tissue, which, when blackened with osmic acid, stood out
in sharp contrast to the connective tissue bundles and the cap-
sules containing the worm.

The Worm Capsule, — The worm capsules of various sizes
occur in abundance in all parts of the subcutaneous tissues and
less abundantly in the corium. They w^ere also observed in some
numbers in the intermuscular connective tissue but not in the
muscles themselves, so far as such observation could be made on
parts incidentally exposed during the surgical operation. In the
corium the c;ipsule may be situated so close to the epidermis that
the latter is externally raised into an acne-like prominence. On
a piece of the preserved skin about 2 inches square I find at
least four such prominences, which, as seen on the surface, appear
smooth and less pigmented than the surrounding parts. Capsules
in so superficial a position might without difficulty be caused to
break through the epidermis by force applied from the exterior.
The " soft whitish mass " which the patient pressed out after
scratching certain itchy spots, could have been nothing else than
the parasite which was present in the ruptured capsule.^^

1) The escape of the worms from the host, in the manner indicated, is to be hioked
upon as merely an accidental occurrence dependent upon tlie exceptionally superficial sitna-
lon of their capsules; it was surely not an event natural to the species' life-history.

ON A NEW CESÏODE LAßVA PARASïIC IN MAN. 5

In shape the capsules are generally subspherical or ovoid.
While the smallest of them are considerably less than 1 ram. in
diameter, others measuring 1-2 mm. or more across are of quite
common occurrence. One of the largest I have taken was of an
elongate shape, 2h mm. broad and 8 mm. long. Another measured
3 mm. by 6 mm. The larger capsules are found only in the sub-
cutaneous parts, not in the corium. In the fj-esh state I could
isolate the capsules without much trouble by tearing and cutting
them off from the surrounding loose tissues.

The capsular wall (fig. 23), consisting of a dense feltwork of
connective tissue fibers of the host, may reach nearly i mm. in
thickness. In sections the capsules may appear not unlike a
transversely cut blood-vessel on account of the tough and compact
looking wall. Seen under the hand-lens, the internal surface of
the wall looks smooth. In some large cysts the internal cavity
was seen to be traversed by a branching pillar-like trabecula, the
presence of which may be explained by supposing an enlargement
of the cavity to Jiave taken place around a strong connective
tissue bundle or bundles. Under the microscope the internal
surface of the wall either shows no special limiting structure or
is lined with a deposit of what may be considered to be a granular
coagulum or some tissue debris, the same as that which is found
free in the cavity together vvith the worm.

Abundance of the Parasite, — An approximate idea of the
enormous numbers of the parasite which infested the patient may
be obtained from the fact that in the vertical section of fig. 1,
presenting a surface of about 11 sq. cm., there can be counted
nearly 60 capsules which had been cut by the knife. So far as
concerns the most thickly infested parts of the patient's left

6 ART. 7. 1. IJIMA.

thigh, I should consider it a moderate estimate to allot on an
average one capsule to every 20 sq. mm. of such cut surface or to
every 100 cu. mm. of the infested tissues. This would give 1000
capsules to a mass of 100 cu. cm. of the same, in other words, to
a mass of the subdermal tissues represented on the skin-surface by
an area of 25 sq. cm., assuming the depth to be 4 cm. At any
rate, I believe it to be in no way an exaggeration to say that
there must have been considerably over ten thousand capsules in
the left thigh alone.

Worms tvithout Capsule, — Here I may mention that not
all the worms were found in the encapsuled state. During the
surgical operation a number of them were discovered free, so to
say, in the connective tissue. This mode of occurrence was
placed beyond doubt by observations of cases in which some
worms, with moderately extended body and without a trace of an
enclosing capsule, were found lodged right within a ßlm of the
fresh connective tissue when the latter was carefully distended
between the fingers. It is noteworthy that these frec^ worms were
always small and of the simple slender shape, — typical Plero-
cercoids such as those represented in fig. 2 a-c ; they were never
so plump-bodied as are many confined within capsules. This fact
indicates that it is only the comparatively younger individuals
that are sometimes found free in the connective tissue.

Movements of the Worm; the Head. — Living worms lib-
erated from the capsule by cutting this open and left in the warm
lymph, showed slow movements, alternately extending and con-
tracting, but effecting little or no change in position. On cooling
the movements subsided, leaving the worm in a more or less

ON A NEW CESTODE LARVA PARASTIC IN MAN. 7

contracted condition. So likewise with all the worms which I
have thrown into the cold (summer temperature) physiological
fluid. However, for some time afterwards, motion could be restored
by imparting a little warmth to the medium they were in. With
such worms as I had kept in the physiological fluid, after about
four hours had elapsed, my attempt at reviving them into motion
was only partially successful ; many of them then seemed to be
nearly or quite dead, showing on the surftxce pustule-like swellings
due to imbibition of the fluid into the parenchyma.

Observing the moving worms under the microscope, the nar-
rower end of the body — undoubtedly the head — showed most
motion. This consisted chiefly in evagination and invagination
of the extreme end as well as in a shortening and lengthening
of the parts. In the more active individuals the head-end was
seen now and then to evaginate and elongate to the fullest extent,
exposing the rounded extreme tip of the head (as in fig. 18). At
other times there was perceivable on the tip a small depression
or concavity, which, especially when the terminal parts of the
head was somewhat broadened after a narrow neck-like part as
was frequently the case for the time being, reminded me of tlie
single terminal sucker known from the fish Bothriocephalid Oy-
athocephalus Kessl. But that depression was evidently simply
due to invagination of the extremity of the head, for I have
frequently seen it gradually but totally disappear as the head-end
prolonged itself into a simple finger-like shape, losing the termijial
broadening which might be observed so long as the invagination
was present. The completely everted, slender head-end moved
about as if feeling for something and at the same time generally
started a rather lively peristaltic motion which traveled back-
wards for some distance. Altogether the movements are such

8 ART. 7. 1. IJIMA.

as would eifectually assist the worm in penetrating into and
shifting through tlie tissues of the host. In several cases I have
succeeded in killing the worms with the head completely or
nearly completely everted and protruded, by suddenly pouring on
hot corrosive sublimate (figs. 18, 21).

The broad posterior parts of the body showed at most slow
vermiform movements, keeping up nearly all the time a more or
less strong indentation or invagination at the extreme hind end.
Attempts to bring fresh worms into full extension by applying
pressure between glasses, invariably failed.

From what I have seen of the moving head I must say
that this is entirely destitute of any definitely formed and per-
manent organ of attachment. Examinations of well extended
heads laid out into serial cross-sections have failed to reveal
anything like bothria or suctorial grooves characteristic of the
Bothriocephalidse. It can certainly not be denied that the ter-
minal invagination, an inconstant and temporary condition though
it be, may under circumstances play the part of a sucker.

Division of the Worm tvithin the C(ti)sule» — With the
smaller capsules it is the rule that they contain each a single
worm. The same may sometimes be the case with the larger
capsules also, but these more frequently contain two or more
worms or pieces of the worm. From one of the largest capsules
I have seen, five worms were obtained, and in another case seven
(fig. 2). This increase in number is explained by the propensity
of the worm to divide by transverse fission at a certain stage of
its growth. Very frequently the worm body shows a constriction
or constrictions at which places it is easily torn apart. In many
cases I could not prevent this separation of the body taking place

ON A NEW CESTODE LARVA PARASITIC IN MAN. 9

during the process of removal from the capsule, however carefully
this was done. In oilier cases a slight pressure between glasses
sufficed to effect the separation at the constricted place. After
all I have observed, there can be no doubt that the worm has
the power of dividing and multiplying on its own account within
the capsule. Thus the worm shown in fig. 5 may be looked
upon in the light of one which is about to divide into two in
the middle. I return to this point again soon.

Size and Shrtpe ; Biidding. — While many of the worms
are very small and filiform, measuring not over 0.3 mm. in breadth
and 3 mm. in length, others (such as those of figs. 4 and 5)
in a moderately contracted state are nearly 12 mm. long and 21mm.
broad. The latter, when fully extended, would probably double
in length but diminish considerably in breadth.

To illustrate the various shapes presented by the worms I
have given in figs. 2-15 a selection of specimens, all killed with
sublimate solution while living and moving in a watch-glass. In
some specimens the body is distinctly flattened, without doubt in
the dorso-ventral direction (figs. 4 and 5). In other cases this
is not so apparent or is even scarcely perceptible ; but then the
body in transverse sections generally presents a more or less oval
outline (figs. 16 and 17). There exists no clue whatever to
decide which of the flattened sides is dorsal and which ventral.

The shape of the worm in the simplest case is that of a
typical Plerocercoid (fig. 2 a-c ; figs. 3 and 13). It is simply
thread-like in the fully outstretched state or may consist, in a
moderately extended state, of the narrow, highly retractile head-
end and of the broader hind region, or when strongly contracted
may be of an irregularly cylindrical form, transversely folded on

10 ART. 7. — I. IJIMA.

the sides and broad at both ends which show each a depression
due to invagination of terminal parts (figs. 2d, 9, 12). When
the head is withdrawn to the utmost degree it may be difficult
to distinguish that end from the hind end, and moreover it may
happen that such a worm is but a fragment constricted off from
the hind parts of the Plerocercoid and is thus devoid of the parts
which may be called the head.

The simple Plerocercoid I regard as the original form of
the larva, which, as will soon be shown, may acquire a widely
different shape in a later larval stage. I regard it also as the
form the larva would be in just before it begins strobilation after
introduction into the final host.

Now a remarkable fact about the present Plerocercoid is
that it is capable, at a certain advanced stage of its intra-
capsular life, of proliferating by a process of budding coupled
with the before mentioned division of the body. We seem to
have here to do with a ]3rocess of which counterparts are known
in the well known Echinococcus and Cœnurus as well as in
Cysticercus longicollis and certain Cysticercoids, but which presents
some peculiar features standing probably in connection with the
Plerocercoid form of the proliferating larva. So far as my know-
ledge goes, a budding Plerocercus or Plerocercoid seems not to
have been known before, at least not with certainty.

Among a large number of the specimens freed from capsules,
those bearing from one to several buds or supernumerary heads
in addition to the original terminal head of the simple Plerocer-
coid, are of quite common occurrence. The buds are found only
on the larger-bodied specimens and therefore on those taken
from the larger capsules. This shows that the budding takes
place only after the Plerocercoid has grown to a certain size.

ON A NEW CESTODE LARVA PARASITIC IN MAN. 11

Facts observed warrant the statement that the worms in the early
period of encystment, and therefore when still small and solitarily
confined in the cyst, are simple Plerocercoids with the terminal
head only. The budding may occur in any region of the broad
posterior parts, but not on the narrow retractile head.

The specimens of figs. 4, 6 and 14 bear each a bud in dif-
ferent states of contraction and looking much like a lateral branch.
The specimen of fig. 5 is in possession of two unmistakeable su-
pernumerary heads; in this specimen, as also in that of fig. 4,
the buds are situated plainly on lateral edges of the flattejied
body. This situation of the buds may in fact be the rule in the
earlier period of proliferation ; but as they increase in number
and the mother-body assumes an irregular shape, the rule loses
applicability in that they then appear to arise from quite inde-
finite parts of the external surface. Of specimens with numerous
buds I have represented two in figs. 7 and 8. The original of
the latter figure bore considerably over a dozen thread-like buds.

The structures referred to above as buds, difier in no way,
except in situation and perhaps also in size, from the original,
terminally situated Plerocercoid head. They manifest exactly the
same shape and movements as the latter. The numerous buds in
the specimens of figs. 7 and 8, at the time of killing, were almost
all in vermiform motion while the mother-piece showed little or
no movement. I think there can be no doubt whatever that the
buds are to be looked at in the light of heads, secondarily pro-
duced but quite equivalent to the single terminal head. In the
end it is quite immaterial to distinguish this original head from
those formed later by budding.

Since now the larva is capable of transverse fission, it follows
that the heads, irrespective of the manner of their origin, may

12 ART. 7. — I. IJIMA.

separate off from that body part with which they were originally
continuous and may thus represent in themselves new independent
individuals of the typical, though at first comparatively small,
Plerocercoid shape. In this way can be explained the origin of
the two or more Plerocercoids, occasionally found together in the
same capsule.

Division alone or in combination with budding gives rise to
various shapes of the worm-pieces which in various combinations
may occur together in the same sapsule. To give a few illustra-
tions from a number of cases observed :

On several occasions a typical Plerocercoid with head was
found together with a plump, nearly cylindrical and headless
piece or with two such pieces (like fig. 9 or 12). All the two
or three pieces must have been derived by division from an origi-
nal Plerocercoid that at first singly occupied the capsule. In
these cases the division of body had evidently taken place before
the worm had commenced budding.

From a large cyst were once obtained : a Plerocercoid with
a supernumerary head on one side and looking very much like
fig. 4 and two nearly barrel-shaped headless pieces (like fig. 12),
both of which must have originally been continuous with the
hind end of the first.

Another large cyst gave two small Plerocercoids of typical
appearance and a large apparently headless body (like fig. 9). One
of the former I take to be original Plerocercoid head and the
other, an individual which has been budded off.

The specimen of fig. 5, bearing two buds and showing a
strong constriction in the middle, was the sole occupant of a cyst.
Should that constriction give way, what the result would be needs
no explicit mention.

ON A NEW CESTODE LARVA PAEASIÏIC IN MAN. 13

Of tlie seven specimens shown in fig. 2, all from a single
cyst, three (perhaps four, a-d) are siraj)le-shaped Plerocercoids,
while three others {e-g) are quite irregularly shaped pieces bear-
ing each a number of budded heads. All these worms, in my
opinion, must have arisen out of a single original Plerocercoid
after its encystment.

Not infrequently the more plump-bodied specimens show on
the surface rounded wart-like or papilla-like tubercles which are
quite distinct from the buds (figs. 2/, 7, 8, 10, 11, 15). Their
formation is due, partly at least, to the accumulation in the
parenchyma of a peculiar reserve nutritive matter which will be
described further on. The external tubercles, conjointly with
the buds, greatly contribute towards giving the worm a remark-
able irregular shape (figs. 7, 8), which is sometimes met with,
especially among the plump and many-headed — therefore assumably
old — pieces of the worm.

Here a word about the headless and budless pieces (like fig.
12) that are segmented off from the hind parts of the Plerocer-
coid within the capsule. They are probably not in all cases to
be considered as parts thrown off in the manner for exampe of,
the caudal appendage of Cysiicercus. Presumably they are still
capable of producing buds. At any rate, there is no reason
whatever to think that the budding can take place only while
the original terminal head is present and not after this is lost
by fission. Whether at the end of the proliferation the entire
mother-piece would be used up in that process or whether there
would finally remain a residue comparable to the vesicular ap-
pendage of a Cysticercus, and whether or not the Plerocercoid
sheds off the posterior parts before it begins strobilation in the

14 AKT. 7. 1. IJIMA.

final host, are related questions on which my observations offer no
clue to a solution.

The head-buds in the Plerocercoid were observed in various
sizes, but I have not been able to trace them back with certainty
to the earliest stage of their formation. Perhaps from an early
period they are capable of active movements, now everting and
then inverting, so that the little rudiment, when fixed, may take
the form of an elevation or depression, which on sections might
not be easily distinguishable from mere irregularities of the surface.
An accumulation of parenchyma cells, such as might possibly
occur at the spot where a bud is to arise, has not come under
observation.

The enormous numbers in which the parasite occurred in
the patient is explained, in large measure, by proliferation taking
])lace in the host. A young and small Plerocercoid, after separa-
tion from the mother-body which produced it, may be assumed
to find its way out of the capsular wall. That act has not been
actually observed but seems to really take place from the fact
before mentioned that a number of the smaller sized Plerocercoids
were found free in the connective tissue. It is not to be doubted
that these free worms are to a certain extent capable of wander-
ing through the tissues b}^ virtue of the power of movements
with which they are endowed. Probably however the wandering
ceases after the worm has grown to a certain size and then it
would begin to give stimulus to the surrounding connective tissue
to form the capsule around it. Tliis stands entirely in harmony
with the account of the patient that from time to time new
acne-like elevations made their appearance on the skin.

ON A NEW CESTODE LARVA PARASITIC IN MAN. 15

Some Points in the Structure of the Plerocercoid, —

The cuticula, sometimes 8 /i thick but usually much thinner,
presents a homogeneous or finely granular appearance. In contact
with it is the delicate dermal musculature which as usual consists
of the external ring and the internal longitudinal fibers. As regards
the subcuticula and the general parenchyma there seem to exist
no points deserving special notice.

The calcareous bodies {cal., figs. 17, 21, etc.), spherical or
ellipsoidal in shape and measuring 7i — 12 fx across, are abundantly
present in all parts of the parenchyma except in the anterior
body-parts. The head-end is always found to be free of them.
(In fig. 19 the numerous black dots stand for the calcareous
bodies.) Especially numerous are they in the bulky parts of the
larger specimens. It may happen that very small Plerocercoids,
such as seem to have been recently budded off, contain only quite
a limited number of them in the hind parts. In stained sections
the bodies appear sometimes pale and homogeneous, but more gen-
erally exhibit the usual concentric lamination ; they are either
stained or unstained.

Quite remarkable is the occurrence, in the parenchyma, of
peculiar fat-like or yolk-sphere -like bodies of various sizes (r. n.,
figs. 17, 19, 20, 22). They occur isolatedly or several together
in groups, without regularity as to their number and position in
the body. Sometimes they are numerously present and are often
of a considerable size, forming conspicuous objects when the
worms are observed under the microscope and giving the impres-
sion that they represent some internal organs (fig. 19). As a
rule they are constantly found in a greater or less number in the

16 ART. 7. — .1 IJIMA.

broad hind parts of the older specimens. Especially abundant
are they in those irregularly shaped, independent pieces which
bear a number of head-buds. The head or the anterior parts of
the worm are generally devoid of the bodies in question ; but this
is not always the case, as sometimes a limited number of the bodies
occur even as far as a position close to the tip of the head. A
few cases in which a small group of them was situated at the
head-end misled me for a time into thinking that I probably had
suckers before me. Small and young Plerocercoids, evidently re-
cently detached from the mother-body, mostly showed no trace
at all of the bodies under consideration. In all probability they
are something which develops and grows both in size and number
as the worm gets older and which, I may add, begins to undergo
a disintegrating process after a certain period of existence.

In an early stage of development the bodies are small, spheri-
cal or irregular-shaped masses appearing very much like yolk-
granules, — homogeneous, refractive and strongly stainable. The
circumstances of their occurrence and the shape of the larger ones
were often such as suggested the formation of these by coalescence
of several smaller ones. The bodies may grow to a size of 150 n
or more in diameter, retaining the original compact aj)pearance,
though in sections they usually appear more or less broken into
irregular fragments separated by narrow fissures. As yet they
seem to l)e imbedded directly in the parenchyma, exhibiting
neither a vacant space around them nor a special enveloping
membrane. As already mentioned, the same bodies after growing
to a certain size begin to show signs of dissolution in that they,
begining at the periphery, break up into finer or coarser granules
as well as into conglomerate-like spherules that appear much like
the wdiite yolk-spheres of a hen's egg. Thus, the body may now

ON A NEW CETTODE LARVA PARASITIC IN MAN. 17

be, as it frequently is, represented by an unevenly granular sub-
stance inclosing a compact core of varying size, which core is the
remnant of the original body in an unchanged condition. In a
more advanced stage of disintegration the core disappears altoge-
ther and the body is entirely changed into a conglomerate-like
mass consisting of granules and spherules of various sizes (r. n.,
figs. 17, 20, 22). This change in character of the body is ac-
companied by a considerable increase in bulk. Moreover, after
tlie setting in of the disintegrating process, the mass comes to
lie in a vesicular space which is delimited from the parenchyma
by a sharply defined, structureless membrane. The vesicle is
generally roundish or oval, measuring 100-300 // in diameter ;
occasionally it is found drawn out into a canalar shape of con-
siderable length (as on the left of fig. 20). The bounding wall
is at first thin but later so increases in thickness that it comes to
resemble in a remarkable degree the cuticula of the worm. Now
and then I have noticed on its inner surface fine hair-like pro-
cesses, but this appearance I am inclined to regard as having
been caused by a part of the contents sticking to that surface.
The conglomerate-like mass either entirely fills up the internal
space or leaves between it and the wall a peripheral space, which
in sections is quite clear but may sometimes show traces of a
very finely granular and faintly stained coagulum.

As to the significance of the body or substance described
above, the most likely explanation seems to be that it represents
a nutritive matter held in reserve, physiologically comparable to
the fat of higher animals or to the amylum in plants. The
storing up of the matter probably stands in relation with the
high proliferating power possessed by the present worm. The
reactions towards reagents indicate that the substance is not ex-

18 ART. 7. — I. IJIMA.

actly a fat but more probably an albuminoid. Similar, if not
identical, bodies are known from certain other larval Cestodes*
Aubert'^ found in Gryporchynchus jjusillus, a Plerocercus from
Tinea vulgaris, a large number of refractive spherules massed
together in the hind body ; they were held by the describer to
be fat, but this seems to require confirmation. The peculiar
" Schollen," described by Bartels^^ from the hind parts of Oys-
ticerciis fasciolaris, seem to closely agree with the bodies I have
met with in Plerocercoides proUfer. In this relation might fur-
ther be mentioned the clear drop-like spheres which are known
to appear in Cysticercus pisiformis, etc. at the place occupied
later by the internal cavity of the caudal appendage. This may
be held to be an indication that there possibly may obtain a
genetic relation between that cavity and the vesicular spaces con-
taining reserve nutritive-matter in Plerocercoides prolifer.

An extensive system of excretory vessels traverses the entire
body of the worm {ex., figs. 16, 17, etc.). As in the Bothrioce-
phalidse generally, there exist several main vessels running in an
approximately longitudinal direction. These frequently branch
and anastomose with one another, thus bringing about an irregu-
lar network of the vessels. Those of largest caliber are seen in
the posterior parts of the body, where they may be of a con-
siderable width. As in Bothriocephalus mansoni, transverse sections
through that region show them in section numerously in all parts
of the parenchyma except in the peripheral zone (fig. 17). The
external opening at the hind end could never be distinctly
brought into view. Towards the anterior parts the network of

1) H. AuBERT. Zeitschr. f. wiss. Zool. Bd. 8. P. 284,

2) E. Bartels. Zool. Jahrb. Abth. f. Anat. u. Ont. Bd. 16. P. 516.

ON A. NEW CESTODE LARVA PARASITIC IN MAN. 19

the larger vessels divides into two, more or less clearly distinguish-
able lateral tracts, each of which, further anteriorly in the head,
runs out into a single main canal on both sides. I have no ob-
servation on the capillary vessels.

On several occasions I have seen parts of the excretory
vessels excessively swollen for a longer or shorter extent, appa-
rently the result of an unnatural stowing of the fluid contents.
In fig. 18 is shown the head of an individual with an abnormally
swollen vessel, which at the extreme tip formed a loop and
was continuous with another of much smaller caliber. I have
been unable to exactly decide whether the vessels referred to
were the two lateral vessels communicating with each other at
the anterior end, or whether they represented ascending and
descending parts of a lateral vessel of one side. Frequently, as
the worms were observed under pressure between glasses, the
swollen parts of the excretory vessels were seen to be filled with
granules, apparently those of the reserve nutritive-matter that
must have found its way into the vessels by rupture of tissues
at some point.

Of the parenchymal musculature there exists in the first
place a well developed system of longitudinal muscles, to the
action of which should be ascribed the contractility of the body
in length and the great retractility of the head. In them the
fibers form anastomosing bundles, running from the tip of the
head to the caudal end and present in all parts of the parenchyma
except in the periphery (figs. 16, 17, 21, 22). The bundles are
strongest and most numerous in the thick hind parts of the body.
Where a head bud arises from the mother-body they give off
branches into it as do also the excretory vessels.

20 AET. 7. 1. IJIMA.

Another, much less strongly developed system of the paren-
chymal musculature consists of fine muscular fibers which run
singly, leaving wide spaces between, in various directions but
mainly in the transverse plane of the body. In the narrow an-
terior parts (the head), as seen on cross-sections (fig. 16), there
is observable a tolerable regularity in the arrangement of the
fibers in that a set of them runs in an approximately transverse
direction while another takes a dorso -ventral course. In the
thick posterior parts (fig. 17), all the fibers run in quite an ir-
regular manner, which fact is probably largely due to the crowded
occurrence, in this region, of the wide excretory vessels and of
the longitudinal muscular bundles. Transverse muscles in a con-
tinuous layer, dividing the parenchyma into cortical and medullary
zones, do not exist.

With respect to the nervous system, all that I have seen is
limited to the following facts. A pair of nerve trunks {n., fig.
16) run in the usual position ; they were observed with distinct-
ness only in the anterior parts of the body. They seemed to unite
at a position close to the tip of head. In the thick posterior
parts of the body, they were not at all distiuguishable.

Taken all in all. Pier ocei'co ides proUfer shows a far-reaching
structural agreement, especially in the musculature and the ex-
cretory system, with the Bothriocephalid larva Sparganum of
DiESiNG, as well as with Cobbold's Ligula mansoni so-called,
which was provisionally referred to Botkriocephalus by Leuckart.
Indeed, it may be admitted to be nearly certain that the worm

ON A NEW CESTODE LARVA PARASITIC IN MAN. 21

is one of the Botliriocephaliche, notwithstanding the fact that the
head is found to be without bothria. Perhaps it is assumable
that a definite suctorial organ or organs might yet develop at a
later developmental stage of the worm than those I have had for
observation. Equally possible it seems, on the other hand, that
the head keeps up the observed simple condition even to the
mature stage, — that we have here to do with a Bothriocephalid
which has lost all trace of bothria. Perhaps these are wanting
in the so-called Ligula mansoni also ; at any rate they have never
yet been seen with certainty in that larval Cestode. Accordingly
it may possibly turn out that " Ligula mansoni " and Plerocer-
Goides prolifer represent very nearly related forms, perhaps
generically identical. A more exact systematic determination of
the larval forms can not be made with certainty until they have
been traced up to the sexually mature stage.

I consider it futile to enter here into speculations concerning
the past and future life-history of the larval Cestode described
in this paper. Great interest is of course attached to feeding
certain animals {e. g., cats, dogs, or pigs) with the larvse for the
purpose of raising the mature worm, and also to determining if
they can be operatively transplanted into the connective tissue of
animals and there made to proliferate. Experiments in these
directions I have caused to be made by Professor Kondö's assis-
tants, but unfortunately they have borne no fruit. As the patient's
return to the hospital is held in prospect, I am in hope of
obtaining a new supply of the material with which to renew the
experiments.

I. IJI9IA.

ON A NEW CESTODE LARVA PARASITIC IN MAN.

PLATE.

Explanation of figures.

Plerocercoides p'olifer.

Fig. 1. A vertical slice of the skin and subdermal tissues taken from the
left thigh of the patient, showing numerous encysted Plerocercoides
proiifer in situ. Hardened in alcohol. Above, the epidermis. From
some of the cysts the worm had fallen out. Natural size.
Fig. 2 a-g. Seven separate pieces of the worm taken from a single large
cyst. Magnified li times. Photographed after fixing with corrosive
sublimate, a-c, simple Plerocercoids. d, a strongly constricted piece
of the worm (with involuted head ?). e-g, irregularly shaped pieces
budding out heads.
Figs. 3-15. Worms in various shapes; all drawn from fixed specimens,
magnified 4 times.
Fig. 3. A specimen of simple Plerocercoid shape, with the extreme

head-end invaginated.
Fig. 4. Plerocercoid bearing a branch-like supernumerary head on

one side.
Fig. 5. A similar specimen, bearing two supernumerary heads and

strongly constricted in the middle.
Fig. 6. A specimen with a branch-like bud ; the terminal head, either

not present or strongly withdrawn.
Figs. 7 and 8. Irregular-shaped specimens with numerous heads formed

by budding.
Figs. 9 and 10. Contracted specimens, either without the head or

with the same strongly wrthdrawn.
Fig. 11. A specimen, irregularly coiled and with tubercle-like pro-
tuberances.
Fig. 12. A piece constricted off from the hind parts of a Plerocer-
coid. Invaginated at both ends.
Fig. 13. A Plerocercoid greatly stretched out, but with the extreme
head-end still retracted.

Fig. 14. A Plerocercoid with the terminal head either lost or strongly

retracted, but with a greatly outstretched head-bud.
Fig. 15. A Plerocercoid moderately stretched out and with irregularities
of contour in the anterior parts.

Fig. 16. Cross-section through the anterior part (head region) of a Plero-
cercoid. Magnified 100 times. %., lateral nerves, ex., excretory
vessels in section. The black dots represent partly nuclei and partly
longitudinal muscular fibers in section.

Fig. 17. Cross-section through the posterior part of a Plerocercoid. Mag-
nified 100 times, cal., calcareous bodies, ex., excretory vessels.
mus., bundles of longitudinal nuiscular fibers, which, in many other
parts, are represented by the larger black dots. r. n., reserve nutri-
tive-matter in capsule.

Fig. 18. Head-end of a Plerocercoid fully stretched out, showing the simply
rounded tip. Drawn from a specimen clarified with glycerine, 30
times magnified. Excretory vessel in part strongly swollen on account
of the stowing of the liquid contents.

Fig. 19. A Plerocercoid pressed under glass; over-stained with carmine
and afterwards bleached with caustic potash. Black dots represent
well-stained calcareous bodies, which are absent in the head region.
Keserve nutritive-matter (r. îz.) in the form of numerous balls. A
pair of excretory vessels (ea;.) in the anterior parts. Magnified 30
times.

Fig 20. A section through an irregular-shaped piece bearing a number of
buds or heads, parts of which are seen in two places (A.). Other
lettering as in fig. 17. Magnified 50 times.

Fig. 21. A horizontal section through the nearly fully evaginated head-
end of a Plerocercoid. Lettering as in fig. 17. Magnified 50 times.

Fig. 22. A horizontal section through the hind parts of a Plerocercoid.
Lettering as in fig. 17. Magnified 50 times.

Fig. 23. Section of a worm-cyst lying in the subdermal connective tissue.
About 8 times magnified.

Ijima : Plcrocercus prolifer

Jour. Sei. Coll.. Vol. XX., Art. 7.

10

.^

./^^v

fm

JOURNAL OF THE COLLEGE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 8.

Mesozoic Plants from Korea.

By

H. Yabe, Eigakislii.
Univorsity Hall, Imperial University of Tokyo.

With 4 Pkifes.

Introduction.

Palseontologically speaking, Korea'^ is the least known of all
the countries of Eastern Asia. The first notice on Korean fossils
is found in Gottsche's " Geologische Skizze von Korea ", in which
he mentions the occurrence of a Cambrian fauna in a limestone

1) Having arrived at Chemulpo (Inchliyön), Korea, in the end of August, 190.'], the
present writer was engaged in field work for about two months up to the beginning of
November, when he took his way home from Fusan (Pusan). During the month of Sep-
tember, he made a journey along one of the main roads from Seoul to Fusan, through
Tyo-ryöng on the boundary of northern Kyöng-syang-Do and Ciihung-chhyöng-Do, in order
to obtain a general knowledge of the various geological formations in which there was
probability of the occurrence of organic remains, and especially to make a collection of the
Palaeozoic plant fossils at a locality close to Mun-gyöng, known since Gottsche's explora-
tion. From the 3rd to the 15th of October, he made a second tour from Fusan to Yöng-il
along the eastern coast, making there a collection of plant fossils in the younger deposit
of the said district. The existence of a plant deposit of a very recent age in this district
was one of the important discoveries of Prof. Koto. During the remaining days, the writer
was able to traverse Chyol-la-Do, starting from Mok-pho, by the way of Chyang-syöng,
Chin-an ; thence he crossed Yuk-sim-nyöng, and traversed Kyöng-syang-Do for the third
time, passing near An-eui and through Chin-jyu, Sa-chhyön, Chin-hai and Masampho.

The paleeontological collections made during his first two months stay in the peninsula
comprise : —

I) Plant fossils of the Palaeozoic (?) from a locality near Mun-gyöng.

II) Plant fossils of a middle or upper Mesozoic age from Naktong.

III) Plant fossils of a younger Tertiary age (?) from Yöng-il.

AKT. S. — H. YABE.

found near Chho-san and Ui-uön, northern Phyöng-ang-Do, and
of plant impressions in a slate exposed near Mun-gyöng, northern
Kyöng-syang-Do.

He also mentions indeterminable plant remains as occurring
in the slates of Naktong, Northern Kyöng-syang-Do, which are
said to lie unconformably upon gneiss, and to gradually pass
above to a thick complex of sandstones and shales. This forma-
tion was provisionally taken by him as synchronous with the
plant-bearing slate of Mun-gyöng, above alluded to, which he
believed to be of the Carboniferous age from the presence of a
species of Neuropteris which he compared to N. flexuosa Sternb.
But Professor Koto who has recently made extensive travels in
Korea considers these two plant-bearing rocks as belonging to two
distinct geological horizons, for the younger of which he proposes
the name " Kyöng-syang Formation."

The second tour in the southern part of the peninsula was made in September and
October of 1904, for the purpose of tracing the extension of the Mesozoic deposists in tlie
southwestern portion of Chyol -la-Do including Na-jyu and Nam-uön, and in a narrow belt
of laud stretching northeastward from the said district to the southeastern corner of Chhung-
chliyöng-Do, where he had previously no opportunity of observation. Consequently, this
time, he traversed the Chyol-la-Do by the way of Mokpho, Yong-am, Nam-phyöng, Koang-
jyu, Tong-pok, Ok-koa, Nam-uön, Im-sil, Chyön-jyu, Chin-an, Yong-dam, Keum-san and
Mu-jyu, and then entering the Chhung-chhyöng-Do went as far as the Keifu-railway line,
then in the process of construction, at Yong-dong. Thence he made a visit to Ok-chhyön
on tlie west and to Hoang-gau on tlie east, and then went to Syang-jyu, Ham-chhyang,
Po-eun and Chliyöiig-san, returning to Hoang-gan after a week. In this way, the Syong-ni-
san ridge was crossed between Ham-chhyang and Po-eum from east to west. Afterward he
went along the railway southeastward by Chhyu-phung-nyöng, Keum-chhyön and Pu-sang,
to Tol-pa. Next a few days were spent in making a research along both sides of the Nak-
tong-gang between Tol-jja on the south and Nak-tong village on the north, and tlien going
from Tol-pa to Fusan by railway, the journey came to the end.

Pakeontologlcally, no important results were obtained from this second journey of about
50 days. Only in the Mesozoic deposits of Yong-dong and its vicinity, a few plant remains
of an almost indeterminable preservation and consequently of very little or no value, were
obtained in a slaty rock harder, though geologically a little younger, than the shales of
the Nak-tong plant beds. From tlie latter, on the contrary, he got 4 or 5 additional species,
not found in the collection of the preceding year.

MESOZOIC PLANTS FllOM KOREA. 3

Tlie development of this Kyöng-syang formation is princi-
pally found in a rectangular area bounded on the east and south
by the coasts of Kyöng-syang-Do, on the west by the meridian
of 128° and on the north by a line roughly coinciding with
lalitude 36° 30'. The region covers the main portion of the well
populated Kyöng-syang-Do, with exclusions of numerous small
areas composed of Tertiary and younger sediments, and also of
eruptives.

There is another area in which this formation is found, but
it is less extensive, lying in southern Chyol-la-Do. Like the
other it is almost quadrate in outline, being bounded by the lines
connecting Mokplio, Ok-koa, Ku-ryoi and Hai-nam successively,
and is composed of two topographically well marked portions, the
low, well cultivated southwestern plain and the more or less wooded
northeastern highland.

Connecting these two separate areas, there runs a narrow
belt of the same sediments, from northeast to southwest for about
200 km. through Chyol-la-Do. From Mokpho in the southwest,
it runs through Chyang-syong and then north of Im-sil, west
and north of Chin-an and near Mu-jyu where it disappears for a
short distance in a region consisting of gneisses and biotite-schists.
But it reaj^pears at the southwest of Yong-dong in Chhung-
chhyöng-Do which lies due north of Hoang-gan. It is again
intercepted by the gneiss plain of Syang-jyu for a distance of
about 20 km. In general, the belt has a width of about 16 km.
along the road between Ok-chhyön and Hoang-gan, and between
Im-sil and Chyöng-jyu.

East of Chhyön-san, in Chhyung-chhyöng-Do, there is also
an isolated area composed of the same rocks, but its exact extent
is at present not definitely known.

4 AKT. 8. H. YABE.

Ill tliis formation, Gottsche distingiiislied five series of rockö
wliicli enumerated from below are as follows :

1) Dark marly shale, alternating with a fine grained, fragile
sandstone.

2) Bituminous clay, partly discoloured, with small coaly
Hecks and obscure vegetable impressions.

3) Conglomerate, arcose near the base, with numerous, very
compact layers of the same rock.

4) Marl of various grades of colour between violet and
chocolate-brown, with frequent intercalations of compact lime-
stones.

5) Thick banded sandstone, conglomerate-like near the
base.

The total thickness of these series of rocks is estimated by
Gottsche as more than 600m., i.e. 25 m. for the first; 15m.
for the second ; 450 m. for the third ; 70 m. for the fourth ; and
40 m. for the last series.

According to the writer's view, the Kyöng-syang formation
is a much thicker complex, with frequent intercalations of red
tuffs and porphyrite sheets in the upper portion. The study of
many local profiles led him to recongnize by means of the pre-
dominating rocks the following four principal divisions of the
formation. The uppermost part is characterised by porphyrite
sheets and a green breccia associated with them. The next part
contains thick layers of several kinds of red coloured tuffs partly
brecciated, together with a greenish slaty rock. The third part
is mainly composed of a hard sandstone, often passing into a
conglomerate and usually underlaid by a thin shale, a red tuff
and an amygdaloid al sheet. The fourth part is composed of a
series of thick slates, green or black in colour and often sandy.

MESOZülC PLANTS FllOM KOliEA. 5

containing a few plant remains, mostly indeterminable. Below
these four parts, there is a thick conglomerate gradually passing
below into a sandstone Avitli intercalations of shales which are
sometimes coaly.

As to the distribution of these parts in Kyöng-syang-Do,
the uppermost shows a wide extent around Fusan, while the
lowest appears along the boundary of the Mesozoic area against
granite and gnesis in the west ; the other two parts occur in
areas between these two, although their distribution has in many
cases become complicated by dislocations.

The upper conglomerate, No. o of Gottsche, is distinguished
from the lower usually by the occurrence of a red tuff' and
amygdaloidal rock sheet below it and also by being covered by
a thick series of a red tuff* above. A similar rock series also
forms an extensive area in southern Chyol-la-Do and in the
narrow belt between Chang-syöng and Yong-dong. In the latter,
it is underlaid by thick greenish and black slates. Gottsche
differs from the writer in considering this conglomerate exposed
near Chin-an as the equivalent of a similar rock which he calls
No. 3 in his Nak-tong section, and consequently in taking the
underlying complex as corresponding to No. 2 and No. 1. But
the writer, from his own observation, is inclined to consider the
conglomerate as representing the upper horizon, i. e. No. 5 of
Gottsche, while the lower series probably corresponds to his
No. 4.

According to the writer's opinion, the lowest part of the
Kyöng-syang formation is to be distinguished from the rest,
under the name of the Naktong series. It seems exactly to
correspond to the divisions. No. 1, No. 2, and No. 3, of
Gottsche. It is characterized by loose sandstones and inter-

6 ART. 8. H. YABE.

calated seams of a coaly shale as above stated. In thickness, it
is far inferior to the remaining uj^per portion of the Kyöng-syang
formation. Its peculiarity consists in the easy destructibility of
its shales and sandstones by atomospheric agencies ; the hills in
the region of these rocks are usually very low and undulating,
except where capped with a thick conglomerate bed when they
are high and steep. The scarcity of vegetation and the low
mound like hills give the landscape a peculiar aspect. The
upper portion of the Kyöng-syang formation is generally com-
posed of harder rocks, aloso with frequent intercalations of con-
glomerate and breccia, especially near the top. The topography
of the region is tolerably distinct from that described above ; it
consists of mountains or hills of higher elevations and where
the water courses have cut into these rocks, perpendicular
bluffs are presented which afford excellent opportunities for the
study of the order of rock layers, and for searching for fossils,
although few of the latter have been obtained and those only
from the lower series in a very bad state of preservation.
Ripple-marks have frequently been met with in the rocks of the
upper series.

A greater regularity prevails in the arrangement of the rock-
layers of the Kyöng-syang formation. They seem not to have
suffered from any great subsequent disturbances except by the
faulting of the Korean and Hansan systems, therefore they retain
nearly their orginal horizontal position, dipping very slightly
to the east (strictly speaking, the dip varies locally from NE.
to SE.).

The collection of fossils to the description of which this
paper is solely devoted, was made on a slope of a hill named
Pul-tang-kokai, on the way from Söul to Fusan, and at about

SfESOZOIC PLANTS FKOM KOREA.

4 km., northwest of a village called Naktong wliicli is situated
close to the river of the same name. Here the rock of the

Naktong series is typically developed. Numerous seams of coaly
shale, seldom over half a meter in thickness, are found exposed

8 ART. S. H. YABE.

along tlie road ; they are intercalated in a complex consisting of
alternations of shales and sandstones, the latter of which often
pass into a conglomerate, which is nnderlaid by marly shales
and hard sandstones. Some of these underlying sandstones are
coarse grained, and in one of them was found a large impression
of a fern frond, which, liowever, is too badly preserved for de-
termination. Marl nodnles, rarely with Foraminifera, are found
in the sandstones. In the coaly shales themselves and in the
intercalated shales, vegetable remains are everywhere found, but
either as indistinct impressions or as small fragments ; among
tlie former we find sometimes crowded impressions of Nilssonia,
and in one of the latter abundant fragments of Onychiopsis
were also observed. A locality for better preserved plants is
in a valley at the eastern flank of the hill above alluded to,
where a shale of a dark brown colour is very rich in fossils.
This locality promises to yield an abundant harvest to future
collectors.

The fossil flora, so far as the number of species is concerned,
must be called a poor one, although some of the species are rich
in individuals. It contains certain forms by means of which it
is possible to determine with approximate accuracy the age of
the strata in which they occur. There is no doubt that the
flora is of the middle Mesozoic age ;^^ it is therefore of considerable
value, not only for the determination of the probable age of the

]) Once, Mr. Y. Ishii suspected tliat the coal-bearing deposit near Phyöng-yang
belonged to the Cretaceous age, tliougli, as he said, there was neitiier palteontological nor
stratigrapliical data for confirming such an opinion. ]\Iore lately, Biiüno Knochenhauer
has said " Kretacische Gebilde treten nur inselformig an der Oslküste auf" (Korea, p. Ill,
1909) ; but how he came to speak of the Cretaceous deposits, it is not known.

E. V. AiiNERT also reported the occurrence of a Jurassic coal-bearing deposit in an
area along the middle course of the Tu-man-gang and its left braiicli the Bololiotung (/(Q

MESOZOTC PLANTS FROM KOREA. 9

upper K3^()iig-synni]!; fornuition and hence of the dislocation of tlie
Korean and Hansan systems, but also of the geographical dis-
tribution of the middle Mesozoic flora.

The study of the flora, above alluded to, was first suggested
to the writer by Prof. B. Koto, by whose encouragement the
writer had the opportunity to twice visit the territory under dis-
cussion. It was first intended that this note should be included
in the more comprehensive work on Korean geology by this
professor ; but by kind permission, it is now given an earlier
publication.

The writer is also indebted to Prof. ]M. Yokoyama for
assistance in regard to palseontological points and for a painstak-
ing revision of his manuscript.

Acknowledgment is also due to Prof. K. Mitsukurt, Di-
rector of the Science College, and to Prof. I. Ijima, of the
publication committee of the Journal.

DESCRIPTIONS.
CYCADOPHYTA Nathorst.
DICTYOZAMITES Oldham.

T. Oldham'^ speaks of the genus Dictyozamites as " very
interesting as a fossil plant, but of no importance in the determina-

1) T. Oldham and J. ISfoRKis: Fossil Flora of the Rajaniahal Series in the Eajamlial
Hills. 186?..

10 AET. 8. H. YABE.

tion of the age of the flora ; only characteristic of the Rajmahal
group " ; but the subsequent discovery of the genus in the fossil
flora of many other lands has led A. C. Seward^^ lately to make
a short but interesting contribution on the history, time range,
and distribution of the genus, accompanied by an important
article upon the plant geography of the Jurassic age. Indeed,
it was by the occurrence of this plant in the Naktong flora
that the present writer was able to draw the conclusion that the
deposit is of the Mesozoic era and most probably of the Jurassic
age.

The species of the genus hitherto known, according to
Seward, number only three : —

1. D. falcatus Oldham.

Kajmahal series of the Kajmahal Hills, and upper
Gondwana of the Madras coast, India.

2. D. falcatus Oldh. var. distans Yok.

Tetori series of Japan.

3. D. falcatus Oldh. var. grossinervis (Yok.)

Tetori series of Japan.

4. D. Johnstrupi Nath.

Liassic or Rhaetic bed of Born holm.

5. D. Hawelli Seward.

Lower Estuarine of Marske, Yorkshire, England.
As is seen from the above enumeration, this genus is surely
known from the Jurassic. In strata older or younger than the
above, it has not yet been known to occur.

]) A. C. Sewiiid : Occurrence of Dir.lyozamitei> in England, and on European and Eastern
Mesozoic Floras. 1903.

MEÖOZOIC TLANTÖ l-KOM KOllEA. 11

DICTYOZAMITES FALCATUS (Mokiuö).
PL IL, figs. 2-7/^

1863. Dictyopteris falcata and D. falcala var. obtualfolia
Oldham and Morris: The Fossil Flora of the
Rajmahal Series, Rajmahal Hills, Bengal. P. 38,
pi. XXIV., figs. I, 2.

1876. Dictyozamites indiens Feistmantel : Ueber die indi-

schen cycadeen Gattungen Ftilophyllum Morris u.
Dictyozamites Oldham. P. 18, pi. IV., figs. 7, 8 ;
pi. v., figs. 1-4 ; pi. VI.

1877. D. indicus Feistmantel: Jurassic Flora of the Raj-

mahal Group, in the Pajmahal Hills. P. 70, pi.

XLVL, figs. 7, 8.
1877. D. indiens Feistmantel : Jurassic Flora of the

Rajmahal Group of Golapili. P. 180, pi. IL, figs.

o, 6.
1879. D. indiens Feistmantel : The Fossil Flora of the

upper Gondwanas Outliers of the Madras Coast.

P. 24, pis. III-V. ; pi. XVL, figs. 4-7.
1889. D. indiens var. distans Yokoyama : Jurassic Plants

from Kaga, Hida and Echizen. P. 53, pi. X.,

figs. 4-10; pi. XL, fig. 5.
1899. D. yrossinervis Yokoyama : 1. c. P. 55, pi. VIL,

fig. 10.
1903. D. faleatus Seward : Occurrence of Dietyozamites in

England. P. 218, pi. XV., figs. 5-S.

1) All the specimens, tigured in the accompanying plates, are in the Sei. Goll. Mus.

12 ART. S. — H. YABE.

Frond pinnate ; pinnae closely set in two alternating rows,
attached to the upper surface of the rachis by the middle of the
base at an angle of about 70°. Pinnae linear, very slightly fal-
cate, gradually tapering to the bluntly pointed apex ; ranging
between 5.0 x 0.2 cm. and 1.2 x 0.3 cm. in size. End pinna
smaller than others. Venation very distinct, showing numerous
aréoles. Aréoles narrow, elongate and subquadrate, parallel to
the margin in the middle of the pinnae becoming shorter and
diverging gradually near the margin.

This most characteristic species is quite common.

The pinnate fronds obtained in Korea are a little broader
than the Indian examples figured by Oldham and Feistmantel,
but the pinnae of the former are generally narrower than those
of the latter. In other respects, there are no distinctions between
them.

As in I), falcatus from India, our examples also vary greatly
in the shape of the pinnae and the size of the aréoles. Thus,
the basal portion of the frond shown in fig. 2, is shorter and
broader than in others. But there is no example in the writer's
collection in which the [)innae show aréoles so coarse as in some
of the Indian and Japanese specimens. 1). grosmiervis of Prof.
YoKOYAMA,^^ distinguished by having a less number of coarser
aréoles, was reduced by Seward"^ to the rank of a variety of the
latter. Probably the Korean examples might also better be
treated as a particular variety.

Two fragments of detached jnnnae drawn in the same plate
(figs. G, 7) show a somewhat different outline from the others,
being considerably broader and more abruptly attenuated to the

1) YoivOYAMA : ]. <.:. r. 55, pi. VII., iig. 10.

2) ÖKWAKU : 1. c. r. o.

3IEÖ0Z01C PLANTS FUOM KOilEA. 13

pointed jipex. Tliesü are here also treated as belonging to the
same species, altliougli with douljt.

NILSSONIA Brongn.

NILSSONIA ORIENTALIS Heek.

PI. III., figs. 1-5.

1878. Nlksoiiia orlenialis Heer: Beitr. zum Juraflora Ost-
sibiriens u. d. Amurlandes. P. 18, pi. IV., figs.
5-9.
1889. JV. orientalis Yokoyama : 1. c. P. 41, pi. XIV., figs.

4-9.
1897. N. cfr. orlenialis Nathorst : Zur Mesoz. Flora Spitz-
bergens. P. 24, pi. I., figs. 18, 19.
Frond longly oval, entire ; truncated at the apex and at-
tenuated gradually below. Veins fine, dense, o-4 in the space of
1 mm. ; slightly curved and directed forward, especially near the
apex.

None of the fronds of JVilssonia orientalis from Siberia figured
by Heer are complete. Yet, the Korean specimens show in
general a great resemblance to them and also to the Japanese
specimens described by Prof. Yokoyama. The only difference
between the Korean and the Japanese specimens is that the base
of the frond in the former is not so much rounded as in the
latter. The writer, however, thinks it almost certain that Ijoth
belong to one and the same species. Their resemblance to the

14 ART. S. H. YABE.

froncis mentioned by Nathoest as N. cfr. orlentalls from Spitz-
bergen is also striking. N. tenuinervis Nath. from Yorkshire
and Bornholm is no doubt an ally of this oriental species, as
already pointed out by Nathoest, Heee and Sewaed. Very
recently, Nathoest reported the occurrence of a Nihsonia very
much like the European species in a Jurassic deposit at Espérance
Bay, Louis-Philipp land.

NILSSONIA SP.
PI. ILL, fig. 6.

The fragment here figured belongs to a species decidedly
different from the preceding. It shows a lower portion of a leaf,
which is cut into two alternate segments. The segments are
continuous at the base ; the upper margin is nearly straight, the
lower strongly convex, and both make an acute angle between
them. In this respect it closely resembles N. nipponensk Yok.^^
from the Japanese Jurassic, but is distinguished from it by having
very densely crowded veins (about 6 in a milimeter).

DIOONITES, BoENEM.
DIOONITES (?) SP.

PI. III., fig. 7.

There is another fnigment of a pinnate frond, 2.7 cm. in
length, which appears to have been thin in texture. About

1) Yokoyama: 1. c. p. 42, pi. VI., üg. Sd ; pi. VIL, figs. 2-7, 8a; pi. XLI., fig. 6.

MESOZOIC PLANTS FROM KOREA. 15

eleven pinnne which arc close together are preserved on one side
of the rachis, meeting those of the other side along the median
line of its upper surface ; none of the pinnae are perfect ; they
vary in breadth from 1.5 to 2.5 mm. and are parallel sided, ex-
panding a little near the base where they are slightly decurrent.
The veins are fine, simple, and uniformly distant ; 5-8 in a
segment.

This fern is not unlike Dioonites Kotdei Yok., but its im-
perfect preservation renders a strict comparison impossible.

CTENOPHYLLUM, Sch.

CTENOPHYLLUM (?) sp.

PL IV., fig. 7.

A slab of shale with a splendid example of Cladophlelm
koraiensis shows, besides, two fragmental impressions of pinnate
cycadean fronds.

One of them bears about fourteen linear pinnae on one side
of the rachis, and the other four ; none of these pinnae are,
however, perfect. The rachis is stout, about limm. in breadth.
The pinnae attenuate but little toward the broken apex ; they
are very close together and alternate ; the mode of insertion
cannot be satisfactorily made out, the rachis showing only the
dorsal side. The veins are rather thick, 5-6 of them fork shortly
after leaving the rachis. The longest pinnae, also imperfect at
its apex, has a length of 1.2 cm. and a breadth of ca. 3 mm.

16 ART. S. — Tl. YAP.E.

Only two fragments of this plant liave been ol)t:iine(l, so
that exact determination is at present impossible. Though not
unlike Dioonites Kotöei in some respects, their generic distinction
is obvious, for the Japanese species has simple veins while those
from Korea usually fork once.

PODOZAMITES,^) Fr. Braun emend. Schenk.

PODOZAMITES REINII Gfaler.

PI. IV., fig. 6.

1877. Fodozamites Eeiiiii Geyler : Ueber Fossile Pflanzen
aus der Juraformation Japans. Pp. 229-230, pl.
XXXIIL, fig. 4a ; XXXIV., 1, 2, 3b, 4, 5a.

1899. P. Reinii Yokoyama : 1. c. p. 50, pl. III., fig. Ga-c ;
IX., 12a ; XII., 4.

Two detached pinnae of Fodozamites, distinguished from the
following species, by their ovate outline and more numerous
veins, are to be identified with F. Reinii Geyler, peculiar to,
and abundant in, the Japanese Tetori series, first described in
detail by Geyleii and then by Prof Yokoyama. Hitherto it
has not been known to occur outside of Japan ; therefore it is
to be looked upon as a species indicating a close relationship
between the Tetori and the Naktons; floras.

One of the specimens measures 2.0 cm. in breadth and has
forty- two parallel veins with numerous fine interstitial ones.

1) This genus is here provisionally placed in Cycadophyta according to the customary
accoplntion, though its position is not certain.

MESOZOIC PLANTS FROM KOTIEA. 17

PODOZAMITES LANCEOLATUS (Lindl. and Hutton).
PI. IV., figs. 1-5.

1877. P. lanceolntus Geyler : I.e. P. 228, pi. XXXII.,

%s. 1, 4; XXXIII., 1-3, 4b; XXXIV., 8a, 5a.
1889. P. lanceolatus Yokoyama : 1. c. P. 45, pi. IV., figs.
la & c, 2, 3a, 4a-b ; V., 1, 2b, 3, 4, 5a-c1, G, 7, 8, 9 ;
VI., 1 ; VIL, 8b ; IX , 6 ; XII., 18 ; XIV., 12a.
1894. P. lanceolatus Yokoyama ; Mesozoic Plants from Ko-
zuke, Kii, Awa and Tosa. P. 222, pi. XIII., figs.
1, 3, 4, 5, 6, 8, 9; XIV., 1-3, 5.
1905. P. lanceolatus Yokoy'Ama : Mesozoic Plants from
Nagato and Bitchii. P. 6, pi. L, fig. G; p. 13,
pi. III., fig. 5.
For further references see Seward's Jurassic Flora, p. 242.
This widely spread, well known species is also present,
tliougli rare, in the Naktong flora. The few specimens obtained
are all detached pinnae, mostly imperfect either in their apical
or basal portions, or in both. The poor state of their preserva-
tion and the rarity of their occurrence, in contrast with the
richness and excellent preservation of the associated fern fronds,
seem to indicate that they belong to that kind of vegetable re-
mains which have been drifted from a great distance.

Some of the best preserved examples are given in the accom-
panying plates. Figure 2 represents a broad pinna lacking its
apex, but with rapidly narrowing base pierced by about thirty
veins in a space 1.9 cm. in breadth. It reminds one strongly of
var. lati folia Hr, Fig. 1 is distinguished by its more lanceolate
outline, gradually tapering above. It is not unlike var. inter-

18 ART. 8. — H. YABE.

7?iedia Hr., bnt tlie veins are much more crowded. Besides,
there are many other oblong pinnae (figs. 3, 5), smaller in size
and with 22-25 veins. One (fig. 4) is characterized by its rounded
apex and gently tapering basal portion and only 20 veins, the
breadth of the pinna being about 1 cm. Therefore it may better
be compared with P. pulchellus Hr.^^

CONIFERAE.

PINUS L.

PINUS SP.

PL II., figs. 9-10.

The existence of coniferous plants in the deposit of Naktong
is shown by isolated linear leaves much like those of Fîmes Nor-
denskjoldi and of its allied species described from the Jurassic
deposits of Spitzbergen, Siberia, China, Japan etc. They are
found uniformly scattered in the bed, though not numerically
abundant. One can distinguish two types by their larger or
smaller leaves. The larger leaves (fig. 9) usually attain 3 mm.
in breadth and more than 4 cm. in length, with a midrib and
numerous fine longitudinal striae on either side of it, thus being
quite similar to Finus Nordenshjoldi Hr.^^ in every respect.

1) Heer: Beitr. zur Fossilen Flora Spitzbergens, p. 38, pi. IX., figs. 10-14. Nathokst :
Zur Mesozoischen Flora Spitzbergens, p. 14, pi. I., figs. 6-11.

2) Heer: Beitr. z. Foss. Flora Spitzb., p. 45, pi. IX., figs. 1-6. Beitr. z. Juraflora
Ostsib. u. d. Amurl., p. 76, pi. IV., fig. 8e; p. 117, pi. XXII., fig. 4a, b; XXVII., 4. Beitr-
1878, p. 26, pi. Tl., figs. 7-10. Nachtrag p. 28, pi. I., figs. 6b, 8c. Yokoyama : 1. c. p. 63,
pi. IX., fig. 12b.

MESOZOIC PLANTS FROM KOREA. 10

Some bring P. Nordenskjoldi under the genus CyclojyitySj
against which view Prof. Yokoyama'^ has given a short argument
in his description of the Jaj^anese examples, citing the opinion
of Nathorst. Potonie also made a remark on Cyclopitys?^ But
in this place, it is sufficient to mention that the Korean example
lacks the transverse wrinkles, though the longitudinal striae are
quite distinct.

The smaller leaves (fig. 10) are about 1 mm. broad and
more than 2 cm. long with a distinct midrib ; there are no other
particular characters to be mentioned. Finus cfr. prodromus Hr.
from Japan^^ is at least a similar plant, if not specifically identical.

Though these two fossils are not of much value either from
the chronological or the botanical point of view, yet they are
at least worthy of consideration as representatives of the scanty
remains of Coniferae in the deposit.

PTERIDOPHYTA.

FILICINAE.

LEPTOSPORANGIATAE.

POLYPODIACEAE.

ONYCHIOPSIS, YoK. 1890.

(Jour. Coll. Sei. Japan, vol. TH., 1890, p. 26. Cat. Mes.
PI. Dep. Geol. Brit. Mus. Wealden Flora I., p.
40, 1894.)

1) YOKOYAMA : 1.' c. p. 63.

2) Potonie: Pflaiizenreste ;ius der Juraformation. Futterer'a Durch Asien III.
o) YOKOYAMA : 1. c. p. 62, pi. XII., fig. o.

20 ART. 8. — H. YABE.

The genus was first established by Prof. Yokoyama iii 1890
for the reception of Thyrsopteris elongaia Geyi>er from the
Japanese Jurassic, on account of the close resemblance of its
fertile as well as sterile pinnae to those of the living species of
Onychium. Immediately afterward, Nathorst^' pointed out the
great similarity of Velonowsky's figures of the fertile frond of
Thyrsopteris capsulifera to the Japanese species, which fact led
him to bring it under the same genus, and he further proposed
to assign the well known European Wealden species — Sphenopteris
Mantelli Brongn. — to the same genus, which closely resembles
0. elongaia. This view was subsequently verified by Seward,"^'
who examined fertile pinnae of Sphenopteris Mantelli from the
British rocks and confirmed Prof. Yokoyama's view, that the
features of the fossil fern show a closer relationship to Onychium
than to any other living genus.

Saporta^^ however on examining a fertile frond from Portugal
was inclined to bring the position of Sphenopteris Mantelli under
the genus DavaUia or at least into the group of the Davalliae.
But, Seward's words^' about this run as follows : " The compari-
son made by Yokoyama in the case of Onychiopsis elongata and
by myself as regards 0. Mantelli, w^ith the recent genus Onychium,
is, I belive, a much nearer approach to the truth than if we
ado]3t the conclusions of Saporta."^^

The genus has a wide range from Jurassic through Wealden
to Cenomanian ; altogether 4 species being known at present.

1) jSatiiokst: lleitiilge z. Mesozoischen Flora Japaiis. 1S90.

2) Seward : AVealden Flora I. 1894. P. 43.

8) Sapokta : Flora Fossile du Portugal. 1894.

4) Sewabd : Wealden Flora II., p. 229.

5) The fertile fronds of Sphenopteris Manidll described by Satokta from the lower
Cretaceous of Portugal appear to tlie writer somewhat diflerent from those of the same
species from Yorkshire and also of Onychiopsis elon<jula.

MESOZOIC PLANTS FROM KOREA. 21

Of these four, 0. elongaia seems to have been abuiidaiit in the
fossil floras of the Japanese Mesozoic. In the Tetori and Kyoseki
series, a great many specimens are found from many locaHties, and
the occurrence of the same in the somewhat older strata of Nagato
has also been very recently pointed out by Prof. Yokoyama.

Another Japanese species — 0. elerjans Yok. — is known only
by one specimen from the Kyoseki series of Tosa.

Outside of Japan, the genus also possesses a very wide
geographical distribution. Particularly worldwide is the distribu-
tion of 0. 3Iantelli which is a common fossil in the Wealden
rocks of England, and also appears in the deposits of the cor-
responding age in Germany, Portugal, Austria and Belgium.
Besides, the species is reported from the upper Jurassic of Por-
tugal, though the identification of Saporta is doubted by Seward -^
otherwise it is an exclusively Wealden species. In North America,
this species has been described from the Potomac formation of
Virginia,'"^ and in South Africa from the Uitenhage Series^^ of
Cape Town. According to the recent publication of Seward,^'
Trichomanites laxum and T. spinifolium from Australia are mere
synonyms of this species.

From the low^er Cretaceous of Bohemia, another species is
described under the name of Thyrsoj^teris caps.ulifera."'' Besides,
some of the specimens from England*^' and Bohemia'^^ are referred

1) Seward: Wealden Flora II., p. 229.

2) Some of the numerous species of the genus Thyrsopterls described by Fontaine from
the Potomac of Virghiia, are believed by Sewakd to be 0. Maatelli.

8) Sewakd: Fossil Floras of Cape Colony, p. 5.

4) 1. c.

5) Velenoavsky : Die Farne der Böhmischen Kreideformation, p. 10, pi. I., tigs. 6-12.
Krasser : Beiträge z. Kenntniss der fossilen Kreideflora von Kunstadt, p. 121. Ant. Fkic
u. E. Bayer: Perucer Schichten, 1900, p. 44, fig. 12.

6) Seward : Wealden Flora, I., p. 55.

7) Krasser : Beiträge, p. 121, pi. VII., fig. 1.

22 AET. 8. H. YABE.

to 0. elongata. All of these species closely resembling one an-
other, it is sometimes questioned whether 0. elongata is really
distinct from 0. caiisulifera on the one hand or from 0. 3Jantelli
on the other. More recently, Seward'^ has united 0. elongata
with 0. 3Iantelli.

In short, Onychioj)si8 is a genus ranging from the Jurassic
to the oldest part of the upper Cretaceous and is of a wide
geographical distribution, and of the 4 species known, 0. elongata
seems to have the greatest chronological range.

ONYCHIOPSIS ELONGATA (Geylee).
PL L, figs. 9-14; pi. III., fig. 15.

1871. Spheno2:)teris Geopperti Schenk (in part) : Palœonto-

graphica vol. XIX., pi. XXX., figs 2 and 2a.
1877. lliyrsopteris elongata Geyler : Ueber Foss. Pflanzen

a. d. Juraform. Japans. P. 224, pi. XXX., fig.

5 ; pi. XXVI., figs., 4 and Ö.
? 1883. Thyrsopteris elongata Schenk : Pichthofen's China.

Vol. IV., pt. X., p. 263, pi. LIV., fig. 1.
188G. Dicksonia elongata Yokoyama : On the Jurassic Plants

of Kaga, Hida and Echizen. Bull. Geol. Soc. Japan.

Pt. B. Vol. I., no. I., p. 6.
? 1889. Thyrsopteris rarinervis Fontaine : The Potomac or

Younger Mesozoic Flora. P. 124, pi. XX VI., figs.

6, 7 ; pi. XLIIL, figs. 4-6 ; pi. XLIV., figs. 1, 2

and 5; pi. XLIX., fig. 2; pi. CLXIX., figs. 6, 7.

1) Sewakd : Fossil Flora of Cape Colony, 1903, p. 5.

MESOZOTC PLANTS FROM KOREA- 23

1890. Onychiopsis elongata Yokoyama : Jurassic Plants from
Kaga, Hick and Ecliizen. P. 27, pi. II., figs. 1-3 ;
pi. III., fig. 6d; pi. XII., figs. 9. 10.
1890. Oiiychiojms elongata Nathorst : Beiträge z. Mesozo-
ische Flora Japans. P. 4, pi. I., figs. 1-3, pp. 8,
10, 13, 14, pi. VI., fig. 5.
1894. Onychiopsis elongata Yokoyama : Mesoz, Plants from
Kozuke, Kii and Tosa. P. 215, pi. XX., fig. 8 ;
pi. XXI., figs. 1 and 4.
? 1884. Onychiopsis elongata Seward : Wealden Flora I.

P. bb, pi. II., fig. 2.
?1896. Onychiopsis elongata Krasser: Beitr. z. Kenntniss
d. Foss. Kreideflora v. Kunstadt. P. 121, pi. VII,
fig. 1.
This elegant fern is exceedingly common in the plant bearing
beds of Naktong, althongh it is not easy to obtain specimens
sufi[iciently large and well preserved to show the entire frond.
Not seldom, large blocks of shales are found containing innumer-
able fragments of pinnae frequently bearing fructification. On
pi. I. and II., some of the specimens are figured ; all of them
show the characteristic features of the species, quite agreeing with
the description and figures of the Japanese specimens in Prof.
Yokoyama 's works.

Prior to Yokoyama's publication, IMr. S. Matsushima, in
his " Geological Eeport of the Eastern Part of Echizen including
the Upper Tetorigawa Begion " 1883 (MS.), described a fern
under the name of Sphenoptevis 31antelli, much like Thyrosoptevis
elongata Gey'ler but more slender in appearance. This specimen
is reproduced here in pi. I., fig. 12, from which we know that
it is almost indistinguishable from the drawings of European

24 A ET. S. H. YABE.

Sphenoplerls Mantelli given in varions publications. As in the
Japanese and the Korean examples, the foreign specimens seem
equally variable so that Seward once distinguished a form with
broader pinnules from those with narrower.

Seward also referred some of the ferns described as Splieno-
pteris Goepperti by Schenk to Onychiopsls elongata, which was
subsequently verified by Krasser's reexamination of the original
specimen of Scpienk. Besides, Krasser added that there was a
great probability that the specimen referred to the Japanese species
was a narrow leaved form of 0. cajosulifera (Velonowsky).
Seward is also of the opinion that the distinguishing characters
given of 0. MantelU and 0. elongata are not of sufficient impor-
tance to render them separate species.

In the collection from Korea, there are many pinnae with
fructification ; the fertile pinnules agree with the Japanese speci-
mens both in their dimensions and in their mode of arrangement
on the axis. They are arranged on the axis sometimes quite
distant from one another as in the figures of the present work,
sometimes very close as in the Geyler's drawings. Seward
described the fertile pinnules of 0. Mantelli as sessile or short-
stalked, and Velonowsky's illustration of 0. capsulifera also
clearly shows the same character. A specimen of 0. elongata
which Geyler figured, belongs to the upper portion of pinnae
with fertile pinnules which are crowded together and furnished
with short stalks. Specimens from Korea differ from those above
mentioned in having, " the sori placed two at the end of each
pinnule, which is considerably narrowed looking like a w^inged
stalk ", as Prof. Yokoyama aptly expressed it in his description
of the Japanese specimens.

A comparison of the fertile pinnules of 0. Mantelli with

MESOZOIC PLANTS FROM KOREA. 25

those of 0. elongata sliows a slight difference in form. Tii the
former, the apical portion is nioi'e elongated while, in the latter,
it is usually rounded instead of being acuminated.

However great the resemblance may be between the oriental
and the European species, the writer considers it at present proper
to keep them as two distinct species, as there are slight differences
in the form of the fertile pinnules and in the general habits of
the sterile joinnae.

An examination of numerous figures of the species of Thyrso-
pteris described by Fontaine from the Potomac formation of
Virginia,'^ shows many forms moie or less closely allied to the
Japanese species. This fact was already pointed out by the author
of the Wealden Flora who placed T. rarinervis Fontaine among
the synonyms of 0. elo7igata. However, until a direct comparison
of the specimens from both sides is made, the question must be
left unsettled. Sewaed also compared Asplenium dicksonîanum
Heer^' from the Kome-beds of Greenland with the present species,
their external resemblance at least being very striking. Heer
placed the fossil under the genus Asplenvmn, because he was
convinced of its relation to the living Asplenium nigrum L..
But there is no fructification to decide the question.

Schenk"^' determined a specimen from Japan in one of
Richthofen's collections of fossil plants, as Thyrsopteris elongata
Geyler, to which Prof. Yokoyama and Seward also seem to
agree although the pinnules are apparently too broad to allow
this identification to pass unquestioned. It reminds some of

1) Fontaine : Potomac Floni p. 123, pi. XXVI., figs. 6 & 7 ; pi. XLIII., figs. 4-6
pi. XLIV., figs. 1, 2, 5; pi. XLIX., fig. 2; pi. CLXIX., figs. 6, 4, 7.

2) Heer: Fl. Foss. Arct., III., p. 31, pi. I.

3) Schenk: Fossilen Pflanzen (Richthofen's China IV.), p. 26.3, pi. LIV., fig. 1.

26 ART. S. H. YABE.

what Heer, Prof. Yokoyama and others described as Thyrsopteris ;
and it is also in some respects not unlike 0. capsulifera.

In this connection, the present writer figured (pi. I., fig. 12)
a specimen from Shimamura, Prov. Kaga, one of the localities
in Japan which are prolific of Jurassic plants. It shows a
character very much like Schenk 's Thyrsopteris elongata, and at
first it was taken for the same. But on making a thorough
examination of the fertile portion of the frond, it was found not
to belong to the genus Onychiopsis but in all probability to
Coniop)ie7'is. Although the plant will not be described here, the
above fact forces the present writer to doubt Schenk's determina-
tion above alluded to.

CYATHEACEAE.
CONIOPTERIS Bronyniart.

(Tableau Foss. Veg. p. 26).

In agreement with to Sapoeta, Solms Laubach and Seward,
the following fossils are described below under the generic name
of Coniopteris, on account of the external resemblance of their
fructification to that of some recent genera of Cyatheaceae
(especially of Dicksonia and Thyrsopteris). For such forms,
some authors are accustomed to use the generic names of the
living ones, even when the real nature of the sori and sporangia
is unknown.

MESOZOIC PLANTS FROM KOREA. 27

CONIOPTERIS HEERIANUS (Yokoyama).
PI. IL, fig. 8; pi. III., figs. 9, 14.

1889. Adicmtites Heerimius Y ok.: 1. c. p. 28, pi. XII., figs.
1, la, lb, and 2.

Frond bipinnate, ovate ; principal racliis Kslender, but stout
and straight, partly winged ; pinnae oblique (making an angle of
about 40°-50° with the principal rachis), rather distant from one
another, long, rapidly becoming shorter toward the anterior por-
tion of the frond ; pinnules rather distant, alternate, oblongly
rhomboidal, gradually tapering below, and lobed ; lobes small,
acutely directed forward, acute at apex ; venation very fine
and indistinct, median vein sending off alternately into the
lobes branches which fork once or twice. Leaf substance coria-
ceous ?

The pinnae of the anterior part of the frond are simple-
lobed. The sori which are found on these pinnae are reniforra,
fixed to the margin of each lobe, at the end of the vein ; usually
there is one sorus on the inner margin of the lobe, but sometimes
many along the inner as well as the outer margin. Each anterior
portion of the posterior pinnae is also simple-lobed, without being
cut so as to form separate pinnules.

After a careful study, the writer w^as convinced of the identity
of tlie examples with a Japanese species, very rarely found in
the plant-bed of Shimamura, prov. Kaga. The Korean and the
Japanese examples are precisely similar in characters, but it is
necessary to note that the type specimens from Japan consist of
but four fragmental pinnae, for which Prof. Yokoyama gave the
following diagnosis :

28 ART. S. — H. YABE.

'' Piiinue elongated ; pinnules alternate, acutely directed forward,
rhomboidal, attenuated below, acute at apex, acutely lobed ; veins
equal, fine, repeatedly dichotomous."

Prof, YoKOYAMA brought two pinnae, wdiich differ from
each other in the size and form of the last segments, under the
species in question and quite rightly, as tlie Korean material
shows that the pinnae of the posterior portion of a frond possess
broad segments while those of the anterior portion present narrow
ones. As to the identity of the Korean and Japanese forms, a
comparison of a specimen shown in fig. 8, pi. II., with fig.
la in pi. XII of Prof. Yokoyama will place it beyond all
doubt.

In one of the Japanese examples, all of the pinnules show
dark dots around their margin borne on the tip of each lobe ;
the fructified example from Korea, fig. 9, pi. III., being distinctly
the apical portion of a frond, shows pinnae which are simple-
lobed instead being deeply divided into pinnules ; and the lobes
which sometimes bear more than one sorus along the inner
margin are only the posterior ones. From this, it is highly pro-
bable that the fructified example from Japan figured by Prof.
Yokoyama is from the posterior portion of a frond.

The writer has brought this species under the genus Conlo-
pler'is for its fructification obviosuly resembles, at least in its
external appearrance, some of the recoi'ded cases of fossil Dlcksonia
and of the living representatives of the same genus. Yet it must
be admitted that with the material at the writer's disposal it has
been impossible to examine in any way either the real nature of
the sori, indusium and sporangia, or the anatomical characters
of the stem. Prof. Yokoyama placed his species in the genus
Adianliles, although he admitted the resemblance of its fructifica-

MESOZOIC PLANTS FIlOM KOREA. 2U

tiou also to Dicksonia ; Seward/^ however, brought it within the
synonyms of Goniopteris hymenophylloides with a query.

The sterile pinnae of this species are so much like those of
Dicksonia acutiloba,'~^ D. concinna^^ and D. Saportcmci^^ from the
Jurassic Deposits of eastern Siberia, that the writer at first
hesitated to separate it from any of those species. But the dis-
tinctly contracted base of the pinnules and other minute differences
led him to make it separate from the Siberian species. Among
the Potomac plants, there are also many species allied to the
Japanese, such as Thyvmiiieris ellipiica Fonïaine,^^ T. varlaiis
Fontaine*'^ and T. crenata Fontaine,"' none of which however
seems to be quite identical with it."*^

CONIOPTERIS HYMENOPHYLLOIDES (Brongniart) ?
PI. IL, fig. 8 ; pi. III., fig. 8.

1851. Sphenopteris nephrocarpa Bünbury : On Some Fossil
Plants from the Jurassic Strata of the Yorkshire
Coast. Quart. Jour. Geol. Soc. London. Vol. VIL,
ph XIL, figs, la, b, p. 179.

1) Skwabd : Jurassic Flora, p. 100.

2) Heer: Beiträge z. Jura-flora Ostsibiriens vi. d. Amurlandes, p. 92, pi. XVIII., fig. 4.

3) Heer: 1. c. p. 34, pi. XIV., fig. 6; p. 87, pi. XVI., figs. 1-9.

4) Heer : 1. c. p. 89, pi. XVII., figs. 1, 2 ; pi. XVIIL, figs. 1-3.

5) Fontaine: Potomac Flora, p. 133, pi. XXIV., fig. 3; pi. XLVI., fig. 1; pi. L.,
figs. 6, 9; pi. LL, figs. 4, 6, 7; pi. LIV., fig. 6; pi. LV., fig. 4; pi. LVI., figs. 6, 7; pi.
LVII., fig. 6 ; pi. LVIII., fig. 2.

6) Fontaine: 1. c. p. 137, pi. LIL, figs. 2-4; pi. LIL, figs. 1-3 j pi. LIV., fig. 10;
pi. LVII., fig. 2.

7) Fontaine: 1. c. p. 127, pi. XXXIX., figs. 1-2.

8) In a piece of shale with a sterile frond of the species, there is an impression which
looks as if it were a portion of a tree fern stem 3.5 cm. in length and 2.5 cm. in breadth.
It shows an obliterated surface only, the vascular bundles being not quite recognizable.
Tlie petiole bases are arranged in close spirals.

30 ART. 8.— H. YABE.

1863. Hymeno'phyllites Bunhuryanus Feistmantell: Jurassic

Flora of the Kaymahal Group. P. 78, pi. XXXII.,
figs. 5-7.

1864, Tympanopora racemosa et simplex Leckenby : On the

Sandstones and Shales of the Oolithes of Scarborough.

Quart. Jour. Geol. Soc. London. Vol. XX., p. 79,

pi. XL, fig. 2.
1876. Thyrsopteris Murrayana Heer : Beiträge z. Jura Flora

Ostsibiliens u. des Amurlandes. P. 30, pi. L, fig.

4 ; pi. IL, fig. 4 ; pi. VIIL, fig. IIb.

Thyrsopieris Maakiana Heee : 1. c. P. 31, pi. L,

fig. 1 ; pi. IL, fig. 5.

Dichsonia davipes Heee : 1. c. P. 33, pi. IL, fig. 7.
1878. Thyrsopteru Murrayana Heer : Nachträge z. Jura-
flora Sibiriens. P. 6, pi. L, fig. 1.
1889. Dicksonia nephrocarpa Yokoyama : 1. c. P. 25, pi. L,

figs. 1, la.
1895. Thyrsopteris (?) Murrayana Raciborsky : Flora Ko-

I)alna. P. 180, pi. X., figs. 15, 16; XII., 17-21.

Dicksonia [Eudicksonia) Heerii Paciborsky : 1. c.

P. 174, pi, X., figs. 5, 6a, 7-lla, 12-14.

Dicksonia Zarecznyi (par.) Eaciborsky : 1. c. P.

175, pi. XIL, figs. 8, 9, 11, 12.
1900. Coniop)teris hymenophlloidcs Seward : The Jurassic

Flora. Yorkshire. P. 99, pi. XVIL, figs. 6 and 8 ;

pi. XXL, figs. 2-4.

Two fertile pinnae are referred by the writer with doubt to

the well known, and widely distributed species, C. hymenopliyl-

loides, whose fertile pinnae have been described under various

names from different localities, for example, Splienopteris nephro-

MESOZOIC PLANTS FROM KOREA. 31

carpa, Tympanopora racemosa and T. simplex from Yorkshire,
DicJcsonia clavipes, T/iyrosopieris 3Iiirrayana, T. Maahiana from
Siberia and Dlcksonia {Eudichsonia) Heerii, D. Zarecznyi from
Cracow. What are known under these different names have
subsequently been proved to be simple modifications of the fertile
segments of Coniopteris hymenophylloides, and the presence of a
gradual transition in these fertile pinnae from forms with a few
broad and round lobes to those with narrow, deeply dissected
segments, has been traced by Seward in materials found at
Scarborough.

Nathorst united Dlcksonia clavipes Heer, from Kaja in
Siberia, with Sphenopteris nephrocarpa Bunbury ; then, Prof.
YoKOYAMA described a fertile pinna from Japan of precisely
similar nature as Dichsonia nephrocarpa. Sphenopteris {Hymeno-
phyllites) Bunbiirianus^^ from India is also no doubt a close ally,
if not identical.

One of the fructified examples from Korea, being quite
similar to Sphenopteris nephrocarpa figured by Bunbury from the
Yorkshire Jurassic, is a bipinnatified fragment with an almost
straight and slender rachis about 2.5 cm., and alternate pinnae
(4 on one side) attached to the rachis at a wide angle. The
pinnae are but slightly flexuous and are furnished with alternate
pinnules. A round sorus is on the dilated margin of each pinnule
or lobe, which is contracted below and at the end of the vein.

In an other example, the lamina of the segments is much
more reduced ; each pinnule has usually a single cup-shaj^ed (?)
sorus ; some of the lower portion however have the lamina less
reduced and bear 2 or more sori.

1) Oldham and Mokris : Pal. Indica, pi. XXXII., p. 54. Heer took it to be Thyrso-
•pteris Maahiana; and Seward referred it with some doubt to Coniopteris hymenophylloides.

32 AET. S. H. YABE.

Geneva of Uncertaiii Sf/sfenmfie Position,

CLADOPHLEBÎS, Brongniart.

CLADOPHLEBIS cfi. DENTICULATA (Broxgniart).

PI. III., %. 11.

The only specimen obtainable being fragmentary, it has been
impossible to make an exact specific determination. Only the
apical portion of a pinnae is shown, about 4 cm. in length,
bearing seven j)iniiules on one side of the slender racliis. The
pinnules are more or less falcate, rapidly narrowing forward to
the acute apex, and show a few serrations along the margin ;
they are separate, but are set close together, being attached by the
whole base. The venation is fine but quite distinct ; the median
vein is somewhat flexuous, and the lateral veins are distant, op-
posite, directed acutely forward and forked once near the base.

Fragmentary as the specimen is, its resemblance to Glado-
phlehis denticulata (Brongn.) is considerable, all the characters
seen in it being quite similar to those of this well known Jurassic
species.

The full account of 0. denticulata is given in the Jurassic
Flora Pt. I., p. 134.

CLADOPHLEBIS KORAIENSIS sp. nov.
PI. IL, fig. 1 ; pi. III., figs. 12, 13.

Frond tripinnate in the anterior portion and quadripinnate
in the posterior ; main stalk longitudinally striated, rather broad

MESOZOIC PLANTS FROM KOREA. 33

and strong. Pinnae usually linear, subopposite, leaving the raeliis
at an angle of about 45° or more ; or, as is sometimes the case,
standing perpendicular to the rachis ; close together so as to
overlap in the posterior portion of the frond. Pinnules vary
considerably in size and form ; those in the posterior part being
generally very narrow, linear, often finger-shaped and falcate ;
decurrent at base, giving the rachis a slightly winged aspect ;
they are usually close together but sometimes quite remote from
one another ; serrate along the margin, the teeth gradually changing
into numerous small trigonal segments. Anterior pinnules close
together, varying in form from a short and more or less trigonal
one to lanceolate, with the front margin usually straight and the
back more or less strongly covex and with a bluntly pointed
apex ; attached by the whole of the broad base to the rachis.
Median vein well preserved, scarcely reaching to the summit of
the pinnule ; lateral veins very obscure, somewhat distant, oblique
and with single bifurcation. Sori large in proportion to the
pinnules, nearly round, very prominent, appearing as pustular
elevations on the upper surface ; crowded, arranged in a single
row on each side of the median-vein ; fertile pinnules somewhat
rolled up along the margin, with lateral veins quite obscure.
Even by Schulz e's method, it is not possible to determine the
character of the sporangia or the spores.

Among the Japanese fossils, there are many forms, more or
less closely related to this Korean species. First of all, there is
Pecopteris exiliformis^^ of Geyler, who pointed out its close resem-
blance to P. exilis. Subsequently, however. Prof. Yokoyama"^
considered this Scarborough species as itself represented in the

1) Geyler: Pflanzen a. d. Juraform. Japans, p. 226, pi. XXX., fig. la.

2) YoKOYAMA : Jurassic Plants, p. 35, pi. I., figs. 8, 10.

34 ART. S. PI. YABE.

Japanese Jurassic flora and believed P. exilJform'is to be in reality
no other than this species ; bat subsequently, he expressed the
opinion that the Tetori form might possibly belong either to P.
Browniana or P. Geyleriana ; Nathorst also pointed out the
great resemblance existing between P. exiliformis and his P.
Geyleriana 'P and Seward-^ is of the opinion that P. exilis from
Japan is probably identical with, or at least very close, to
P. Browniana ; and P. exiliformis also does not suggest a plant
Avith a well marked specific difference.

According to Bunbury"^^ and Raciborsky^^ who described the
fertile pinnules of P. exilis, the sporangia possess an apical
annulus, and are in a single row on each side of the median vein ;
hence the latter author was led to propose the new generic name
Kluhia for the above fossil, which shows a decided affinity to
Schizaeaceae.

The round dots in the Korean examples are in all probability
of a different nature from those of Kluhia exilis, though there is
a great resemblance between them. The present writer does not
regard this resemblance as an evidence of relationship.

At any rate, fig. 9, pi. I. in Prof. Yokoyama's work ex-
hibits a greater likeness to the Korean form than figs. 8 and 10
in the same plate. The example figured by Geyler also resem-
bles it.

The writer has examined some specimens of the P. exilis
type from Hakogase, collected by Mr. S. Matsushima and now
deposited in the Science College Museum. One of them is pre-

1) Nathorst : Mesoz. Flora Jap;ins, p. 48.

2) Seward : Fossil Flora of Cape Colony.

?>) Eunbury: Quart. Jour. Geol. Soc. 1851. VII. P. 188, pi. XIIL, fig. 5.

4) Raciborsky: Ueber die Osmundaceen u. Schizaeaceen d. Juraformation.

MESOZOIC PLANTS FßOxAl KOREA. ÔO

cisely similar to Geyler's drawiüg, and hence also like figs. 8
and 10 in Prof. Yokoyama's work. The other examples show
much resemblance to fig. 9 of the latter and evidently belong to
the terminal and middle portions of the same fern.

The Japanese fern, so well illustrated by Prof. Yokoyama
possesses oblong pinnules, in general differing from the Korean
by a decidedly obtuse outline, often broader and rising from the
racliis at a wider angle.

A similar fern was described by Prof. Yokoyama from
Japan under the name of Asplenmm argululum ; among the figures
referred to this species, in pi. III., fig. 1, agrees well with the
Chinese examples described by Schenk ; while those represented
in fig. 9, pi. XII r. and fig. 22, pi. XIV., show a great resem-
blance to the Korean. Prof. Yokoyama had already made the
remark that the latter two examples have a great resemblance
to some forms of P. Fhillipsi Brgt. {P. exilis). It is by no
means improbable that some of these figures represent plants
which are really identical with the Korean species. In the
Ryoseki flora of Japan, there is Cladophlehis Browniana^'' a
Wealden species of world wide distribution, associated with C.
Geyleriana Nath.,"^ a hitherto solely Japanese form.

When Nathorst first described P. Geyleriana, he was unable
to trace out the lateral veins of the pinnules, and therefore,
though he possessed the fertile pinnae provided with round sori
on each side of the median vein, he took it to be simply a species
of Pecopterh. Its relation to Lonchopteris which he then suspected,
was subsequently made impossible by Prof. Yokoyama's exami-

1) Yokoyama: Mesoz. Plants from Kozuke, Kii, Awa and Tcsa, p. 218, pi. XXIV.,
figs. 2 & 3; XXVir., 1-4, 5c, d.

2) Yokoyama: 1. c. p. 219, pi. XXI., figs. 1, 2; XXIII., la, 1; XXVIII., 5.

36 ART. 8. — H. YABE.

nation of better material which showed the pinnules with bifur-
cating lateral veins. Therefore, Sewakd^* thinks that possibly
the fertile pinnae described by Nathoest should be kept as a
species distinct from the sterile examples, and referred to Weich-
selia (Lonchopteris) Blantelli. In a more recent publication,
Sewaed^^ has placed the specimens represented in figs. 2-6 of
Nathorst's paper under C. Browniana, considering them as
distinct from the others for which he retains the name of P.
Oeyleriana.

The present writer has mentioned above that the P. exilis
of Prof. YoKOYAMA (P. exiliformis Geyler) shows some slight
differences from C. koraiensis, on account of wdiich the Japanese
species approaches to, or more likely is identical with, C. Brow-
niana, as YoKOYAMA and Seward had already pointed out.

The fertile pinnae of P. Browniana figured by Prof. Yoko-
YAMA and those of P. Ditnlcerl of Fontaine"" from the Potomac,
strongly suggest those of the living Aspidium in the form and
mode of attachment of the sori, so that the latter author even
went so far as to bring his P. Bunkeri under the genus Aqyidium,
although according to the writer's opinion these characters not
seem to jiistify such an assumption.

Comparing the fertile pinnules of C. koraiensis with those
of the two above mentioned, some differences still remain ; also
it is not wholly impossible, that the form belongs to a genus
entirely different from the latter two.

To sum up, at present it seems advisable to treat the Korean
form, as a new species of Gladophlehis, a genus created for the

1) BjiWARD : Weaklen Flora, I, i). 116.

2) Seward : Fossil Flora of the Cape Colony, p. 12.

3) Fontaine: Potomac Flora, p. 101, pi. XXII., fig. 9.

MESOZOIC PLANTS FKOM KOREA. 37

reception of sterile fronds of ferns whose real relation to the
living forms is unknown.

CLADOPHLEBIS cfr. DUNKERI (Schimper).
PI. IV., fig. 9.

Among numerous examples of fronds of the Cladophleh'is
type, there is one which closely resembles the Wealden species
called G. Dunkeri. Possibly it may turn out to be an extremely
modified form of the preceding species ; but at present it is better
treated as a differrent fossil.

The specimen shows a bipinnate fragment of a frond, with a
stout principal axis from which the pinnae branch off at a wide
angle. The pinnae are opposite and close ; the pinnules are
small, oblong, often slightly falcate, very close, attached to the
rachis with a broad base.

It is almost impossible to draw a distinction between this
fern and C. Dunheri from the Potomac flora (Foutaine, PL XXVI. ,
fig. 3.).

CLADOPHLEBIS sp. indet.
PI. IV., fig. 8.

There are several fragments of isolated pinnae of a Cladophlebls
which though too imperfect for specific determination, may be
briefly described as follows : —

Pinnae lanceolate, rapidly tapering to an acuminate apex ;
pinnules elongated, gradually becoming oval toward the anterior
part of the pinna ; slightly pointed at apex and somewhat decurrent
at base which is auriculate on the posterior side ; margin slightly

38 ART. S. — H. YABE.

undulating. Median vein thick, evanescent near the apex ; lateral
ones obsolete, remote, subopposite, oblique and single-forked.

One of the pinnae measure 3 cm. in length and bears nine pin-
nules on one side of the rachis.

This species is evidently distinct from the foregoing two.

SPHENOPTERIS, Beongn.

SPHENOPTERIS NAKTONGENSIS sp. nov.

PI. IV., figs. 10, II.

Frond tripinnate, deltoid ; principal rachis slender, flexuous ;
primary pinnae subopposite, rather remote, attached to the rachis
at a wide angle, linearly oblong, pinnatifid ; ultimate pinnae
alternate, very short, passing anteriorly into lobed pinnules ;
pinnules oblong or cuneiform, alternate, oblique, decurrent on the
slightly winged rachis, more or less deeply lobed ; lobes denticu-
lated ; venation indistinct, veins very close, flabelately dichtomous.

Among the hitherto described ferns, there is none which is
identical with this Korean fossil. Ruffordia Geopper0 is the
only one which more or less closely approaches it.

SPHENOPTERIS sp.

PI. L, fig. 15.

Frond bipinnate, rachis slender ; pinnae linear, springing
almost at right angles from the rachis, pinnately lobed ; lobes

1) Seward Wealden Flora, I., p. 75.

MESOZOIC PLANTS FROM KOREA. 39

obtuse, acuminate with entire margin ; somewhat falcate ; venation
of a Sphenopteris type, that is to say, with a distinct median vein
sending off an obsolete secondary vein into each lobe. Sori in
two rows, one on each side of the median vein, placed near the
base of each lobe and on the summit of the uppermost branch
of the vein.

At a first glance, this form comes near to DicJcsonia Bin-
drahmensis Feistmantel^^ from the Jurassic of the Rajmahal
Hills, India, and Alsophila polonica Raciborsky^^ from the neigh-
bourhood of Cracow. But, no detail of the fructification being
observable in tlie only specimen at hand, its generic position is
not certain.

ADIANTITES, Goep.
ADIANTITES SEWARDI sp. nov.

PI. I., figs. 1-8.

1888. Rhacopteris ? sp. Matsushima : A Geol. Rep. of the

East. Part of Echizen including the Upper Tetori-

gawa region, in Kaga. (MS.) P. 188, pi. VII.,

fig. 5.

Frond bipinnate, rachis thin and straight ; pinnules alternate

to subopposite, close but rarely imbricated ; rounded at the upper

margin and attenuated below to a narrow base, with a short de-

1) Feistmantel: Jurassic Flora of the Rajmahal group in the Rajmahal Hills. P. 23,
pi. XXXVII., figs. 2, 2a.

2) Raciborsky: Flora Kopalna ognitrawalych glinek Krakowskich. Archaegoniatae,

P. 29, pi. IX., figs. ?., 4.

40 ART. 8. — H. Y A BE.

current petiole. Lateral pinnules generally inequilateral, varying
from subquadrilateral to fan-shaped, with inner margin parallel
and very close to the rachis. Apical pinnules nearly equilateral,
being broader than the others. Upper margin irregularly crenu-
lated. Veins numerous, fine, uniform and distinct, united below
at the base, widely spreading and dichotomously divided upward.

According to the general usage, this elegant fern is referred
to the provisional genus Adicmtltes on account of its undoubted
resemblance to some living sjDCcies of Adiantum or Lindsaya.

Among the recorded species referred to Adimitites or some-
times to Adiantum itself, there are three species more or less
allied to the present species ; Adiantites Tietzi Schenk^^ from the
Khaetic of the Albours-chain, Adiantum Szechenyi Schenk^^ from
the middle Jurassic of China and Adiantites yuasensis Yok.^'^
from the Ryoseki series of Yuasa in Japan. The resemblance
of the Korean frond to the last species has already been pointed
out by Mr. Matsushima.

In the first of these three species, the pinnules are more
densely crowded and deeply imbricated ; the other two species
exhibit a somewhat different venation.

A species which is nearer to the Korean than any of the
above three is Adiantites lomjiqmiis Saporta (Nouvelles Contribu-
tions a la Flore Mesozoique 1894. p. 40, i^l. X., fig. 1.) from the
Neojurassic deposit of Portugal. This species was founded by
Saporta on a few fragmental pinnae with subopposite pinnules ;
however, so far as one can judge from the characters revealed

1) FciiENK : Fossile Pllanzeii aus der Alboiirs-kette, gesammelt von E. Tietze. 1887,
p. ?,, 1.1. 11., fig. 9.

2) Schenk: Fossile Pflanzen (Graf. Szecheny's Reise Vol. III.) P. IßS, pi. I., fig. P>.

3) YoKOYAMA: Mesoz. Plants from Kozuke, Kii, Awa and Tosa. P. 210, pi. XXI.,
flg. 15.

MESOZOIC PLANTS FROM KOREA. 41

by the figures, the species so much resembles the Korean in
form, size and the venation of the pinnules, that these two are
no doubt very closely allied to each other, if not identical. The
only distinction lies in a slight difference in venation.

The Japanese examples of this species have been obtained
by Mr. Matsushima from Kinebashi, Uchinami and Otani, Ono-
gori, prov. Echizen.

FILICES?

SAGENOPTERIS, Presl.

SAGENOPTERIS BILOBATA sp. nov.

PI. III., figs. 16a, b, c.

There are two forms of Sagenopteris hitherto described from
the Jurassic deposists of Yorkshire and others, which are usually
found together. The pinnules in the one are lanceolate in out-
line, while those of the other are cuneate at base ; and these had
been kept for a long time as two distinct species, but at present
the cuneate form is considered to be an imperfect or abnormal
state of the frond of the same species ; thus in the case of the
Yorkshire flora, Seward^^ distinguished these two forms under
the names oî S. Pkillipsii var. cuneata, and var. major.

In the Korean specimens, there are two types ; the one re-
presented by three examples, shows an outline precisely similar
to S. cuneaia or 8. PhilUpsli var. cuneata, while the other three
slightly differ from var. major.

1) Sewaed: Jurassic Flora, I. P. 162.

42 ART. 8. — H. YABE.

The first of them consists of two obcuneate pinnules, in
contact, borne on a stipe ; the pinnules bear no median vein, but
are traversed by spreading anastomosing veins. The second is a
linear pinna, apically cleft into two symmetrical lobes, with a
median vein. Its lateral veins are very fine, anastomosing, leaving
the median vein at an acute angle.

Though at first the writer took the latter type as specifically
identical with S. Phillipsii var. major, the bilobed nature of the
pinnules of the Korean plant led him at last to regard it as a
different species. Such being the case, he believes it best to treat
the two types of the Korean examples as a single species, but
calling them for distinction, var. cuneata and var. major directly
after the example of S. Phillipsii.

Another point of distinction of some value between S. Phillipsii
and S. hilobaia is found in the outline ef the pinnules which are
lanceolate in the former and linear in the latter.

Another allied species which merits mention is S. tasmanica
Feistmantel^^ from the Carbonaceous beds (upper Mesozoic) of
the Jerusalem basin, Tasmania. The species was founded by
Feistmantel on a single example which shows a portion of two
lobes of a three-lobed frond as he understood it. Each lobe is
linear, attenuated above and provided with a median vein, quite
distinct in the greater part of the length though disappearing in
the apical portion, and with lateral, veins passing out from the
median vein at an acute angle and forming a single anastomosis.

A comparison of this species with the Korean is rendered,
however, almost impossible so far as the outline is concerned,

1) Feistmantel: Geological and Palœontological Relations of the Coal- and Plant-
bearing Beds of Palreozoic and Mesozoic age in Eastern Australia and Tasmania, p. 135, pi.
XXIX., fig. G.

MESOZOIC PLANTS FROM KOREA. 43

owing to the want of the basal portion of the frond in the former
and of the apical portion in the latter ; but if Feistmantel's
explanation is correct, then their relationship is only a slight one,
for he understands his specimen to be a probably three-lobed
frond while the Korean species is only two-lobed, the resemblance
being solely due to the mode of preservation.

A similar dichotomization of the frond is also well shown
by Phlebopteris (?) d'lchotoma Shirley^^ from the Denmark Hill,
Ipswich, Queensland.

EQUISETALES.

EQUISETUM, L.

EQUISETUM USHIMARENSE Yok.

PI. III., fig. 10.

1889. Equiselum ushimare7ise Yok. : 1. c. p. 39, pl. XI.,
figs. 1-3.

A few fragments of rhizomes of an equisetaceous plant are
in the writer's collection. Their breadth varies between 2.5 mm.
and 6 mm., with three to four, strong ribs and internodes,
2.0-2.5 cm. apart. Tubercles rarely found are ovate in form,
measuring 1.8 cm. in length and with no surface markings visible
on thier surface.

A comparison between the present examples and the figures
of Prof. Yokoyama's species leaves no doubt about their specific
identity.

1) J. Shikley : Additions to the Fossil Flora of Queensland, p. 24, pl. XV., fig. 2.

44 AET. S. — H. YABE.

The Japanese species is so much Uke E. Burchardi Schenk
from the European Wealden formation, that there is a doubt
that they are really different plants. However, their identity
being not yet fully verified, tlie above name has been here
adopted.

GENERAL REMARKS AND CONCLUSIONS.

Constitution of the fossil flora ' — Though sufficient material
has not yet been collected to warrant an exhaustive discussion of
the fossil flora, yet there are some points of importance with
regard to the geological age of the Naktong series and its floral
constitution, which call for a brief statement in this place. In-
deed, the geological importance of these fossils lies in the cir-
cumstance that they are derived from the base of a very thick
complex which affords no other reliable organic remains for de-
termining its geological age.

As will be seen from the above description, the writer was
enabled to distinguish the following twenty-one species of plants,
of which 3 are new, and 7 of doubtful affinity.

Diciyozamites falcatits (Moeris) common.

Nilssonia orienialU He. abundant.

N. sp. rare.

Diooniies (?) sp. rare.

Clenophyllum (?) sp. rare.

Podozmnites Reinii Geylee rare.

MESOZOIC PLANTS FKOM KOREA.

45

P. lanceolaius (Lindle. & Hutton) rare.

Pinus sp. common.

Pinus sp. common.

Onychiopsis elongata (Geyler) abundant.

Coniopteris Heerianus (Yokoyama) common.

Ö. hymenophylloides (Brongn.) (?) rare.

Cladophlebis cfr. denticulata (Brongn.) rare.

C. koraiensis sp. nov. abundant.

C. cfr. Dunheri (Schimher) rare.

C. sp. rare.

Sphenopteris nalclongensis sp. nov. common.

S. sp. rare.

Adiantltes Sewardi sp. nov. abundant.

Sagenopteris hilohata sp. nov. rare.

Equisetum ushimarense Yok. common.

In the constitution of tlie flora, Filices form of the most
important part, being represented by 6 genera and 11 species.
Among these, three species are found in great abundance. Cy-
cadophyta, though fairly common, are less frequent than the
ferns, being represented by 5 genera and 7 species. Equisetaceae
and Coniferae, though not rare, are all not clearly determinable.
It is very striking that the fern fronds are generally in a fertile
state, just as in the fire-clay of Cracow, described by Raciborsky.
Such are Onychiopsis elongata, two species of Conlopieris, and
one species of Sphenopteris ; likewise the pinnae of Cladophlebis
koraiensis mostly show the fructification.

Geological relation of the fossil flora : — From the above list
it is quite evident that the flora is Jurassic, for neither typically
Khaetic nor Cretaceous forms are found in it ; the form which may
be assigned with hesitation to the Wealden type is but a single

46 ART. 8. H. YABE.

one, that is, Gladophlehls cfr. Dankeri.^^ As usually accepted by
palœophytologists, the vegetative character of the world from the
upper Triassic to the Wealden seems to have been remarkably
uniform and constant in its main features.^^ Hence it is a very
difficult task to make out the exact age of a given plant-bed if
it is ])oor in fossil contents, as in the present case. Though we
have now twenty-one species in all, many are not available for
this purpose.

Even of specifically determinable forms, Onychiopsls elo7igata
seems to be of no value in settling this question, for it is a fern
type of decidedly east Asiatic origin, thence becoming widely
diffused in other lands after the Jurassic age. The fern with
fronds of the Cladophlebis denticidata type is not only exclusively
Jurassic, but is found also in the older as well as in the younger
strata. The two fertile pinnae which the writer has compared
with Coniopteris hymenophylloides, and the detached leaves of
Coniferae also aftbrd us no sure basis for a chronological correla-
tion, because a determination based on such fragments must always
be uncertain. Podozamites lanceolakis and equisetaceous remains
seem equally of little value in this respect.

Excluding the above species and such as are new or doubt-
ful, there remain only five species, viz. Adiantites tSewardi, Conio-
pteris Heerianus, Diciyozamites falcatus, JSilssonia orienialis, and
Podozamitas Reinii, which are available for determining the age
of the strata.

According to Mr. S. Matsushima/^ Adiantites Seivardi
occurs at Kinebashi, Uchinami and Otani, Onogöri, prov. Echizen,

1) Seward: took Equisetum Ushimarcnse as a Wealden type.

2) Sewakd : Floras of the Past : their Composition and Distribution. Pp. 13-22.

3) Matsushima: a Geol. Rep. uf the East. Part, of Echizen etc. (MS.), p. 188.

MESOZOIC PLA.NTS FROM KOREA. 47

associated with Podozamites lanceolatus, Cladophlehis sp. {C. afï!
Dunheri) and C. sp, {C. aff. distans).

Coniopteris Ileerianus and Podozamites Reinii are exclusively
Japanese Korean species ; the latter is especially widely spread in
the Japanese Tetori series ; Prof. Yokoyama^^ described it from
Okamigö, prov. Hida, Tanimura, prov. Echizen, Ozo, Yanagi-
dani and Shimamura, prov. Kaga ; at the last locality it is found
together with Co7iiopteris Ileerianus. The other fossils found in
association with these species are, according to the above author,
at Shimamura, Tliyrsopteris prisca, T. hagensis, Diohsonia gracilis,
D. acutiloha, D. nephrocarpa, Onychiopsis elongata, Adiantites
Kochibeanus, Asplenium Wliiibiense, A. argvtulum, A. distans,
Pecopteris exilis, P. Saportana, Äfacrotaeni opter is cfr. Richthofeni,
Anomozamites sp., Nilssonia nipponensis, Dioonites Kotoei, Zamitis
parvifolius, Podozamites lanceolatus, P. tenuistriatus, Dictyozamites
grossinervis, Cycadospermum jap)onicum, Ginhgodium Nathorsti,
Ginkgo sibirica, Taxites sp., Pinus prodromus, P. Nordenshjoldi,
Palissya sp., and Vallisnerites jurassicus (?) ; at Yanagidani,
Onychiopsis elongata, Asplenium Whitbiense, A. distans, Taenio-
pteris (?), Sagenopteris sp., Nilssonia ozoana, N. (?), Dictyozamites
falcatus, Czehanowshia rig i da, Taxites sp., Vallisnerites jurassicus (?)
and Carpolithes ginhgoides ; from Tanimura Thyrsopteris hagensis,
Onychiopsis elongata, Dioonites Kotoei, Podozamites lanceolatus,
Pinus. cfr. prodromus ; from Okamigö, Thyrsopteris Murrayana,
Dichsonia gracilis, Onychiopsis elongata, Asplenium argutulum,
Equisetum sp., Nilssonia nippone7isis, Podozamites lanceolatus, P.
tenuistriatus, P. sp., Ginhgodium Nathorsti, Ginkgo digitata,
Czekanowskia rigida and Taxites sp.

1) YoKOYAMA : Jurassic Plants from Kaga, Hida and Echizen.

48 ART. 8. — H. YABE.

NiUsonifi orientalis is also known from Japan and Siberia.
According to Prof. Yokoyama, it seems to be abundant in a
plant-bed exposed at Hakogase, Onogöri, pro v. Echizen, where
Dicksonia Glehniana, Onychiojysis elo7igata, Adianiites lancens,
Asplenium argutulum, A. distans., Sphenoplei'is sp., Podozamites
lanceolaius and Glnhgo cfr. lepida are also found. The Jurassic
deposit of Ajakit on the Lena^^ whence the species was first
recorded, contains, besides, Dichsonia gracilis, D. horeaUs, Rhizo-
carpiies singularis, Anomozami tes angulatus, Nilssonia comtula,
Podozamites lanceolaius, P. gramineus, Phoenicopsis ajigustifolia,
Ginkgo Huttoni, G. sibirica, G. integriuscula, Czekanowskia cre-
tacea, C. rigida and Pinus Nordenshjoldi. Recently, Nathorst"^
described an identical or closely allied form from the southern
side of Sassen Bay, Spitzbergen ; the only other fossil associated
is Ptilozamites sp.

According to present knowledge, the distribution of the species
of the genus Dictyozamiics is confined to India, Japan, Korea,
Bornholm and Yorkshire. A full account of the genus has been
very recently wTitten by Seward"^^ who has pointed out the uni-
formity of the general features of the Jurassic floras of these
widely distant places.

From these localities, three or four species are already known.
The Korean and the Japanese fossils make a closer approach to
D. falcatus from India than to those from other countries. In
Japan, the species is found in the plant-bed of Ozö, pro v. Kaga
and Ushimaru, prov. Hida ; the associated fossils at the former
place have already been given above ; those at the latter are
Asplenium distans and Podozamites lanceolaius.

1) Heer : Beitr. z. Juraflora Ostsibiriens u. d. Arnurlandes, p. 18.

2) Nathorst: Znr Mesoz. Flora Spitzbergens, p. 24.
'S) Seward: Occurrence of Dichjozamltes in England.

MESOZOIC PLANTS FROM KOREA. 49

In the Rajmahal group of India, Dictyozamites falcatus is

found associated with the following species : — at Ainrapara, An-

giopteridium Mc. Clellandi, Ptilophyllum cidchense, Williamsonia

sp., Echinosirohus rqjmahalensis ; at Murrero, Angiopteridium

spathulatum, Macrotaeniopteris crassinervis, Pterophyllum Medlicot-

iiamim, Zamites proximus and Zamiost7'obus ; at Godavari, Palissya

eonferta, P. indica, Gheilolepis cfr. Münsteri, Araucarites maci^o-

pteruSf Angiopteridium ensis, A. spathulatum, Ptilophyllimi acuti-

folium, P. cutchense, Pterophyllum Morrislanum, P. Kingianum,

P. distans, P. Carterianum and Williamsonia gigus, Alethopteris

indica and Asplenites macrocarpus. A similar association is also

observed in the Sripermatur group and in the Vema varam beds

of the Madras coast.^^

It has been above pointed out that there are five species

surely common to the Japanese and Korean floras. But there

are, besides, seven less certain or simply allied forms, so that

the total number which admits comparison is twelve. They are

the following : —

Japan. Korea.

Dictyozamites falcatus (Morris) ...Dictyozamites falcatus (Morris)

Nilssonia orientalis Hr Nilssonia orientalis Hr.

. - . XT (Dioonites sp.

Dioonites Kotoei Yok <

yCtenophyllum (?) sp.

Podozamites lanceolatus Lindl Podoza7nites lanceolatus Lindl.

and HuTTON. and Hutton.

P. Reinii Geyler P. Reinii Geyler.

Pinus Nordenskjoldi Hr Pinus sp.

1) Feistmantel : Jurassic Flora of the Eajmalial Group, in the Rajmahal Hills. —
Jurassic Flora of the Rajmahal group of the Golapili. — ^The Fossil Flora of the Upper
Gondwanas Outliers of the Madras Coast.

\ Cladophlebis koraîensis

50 ART. 8. H. YABE.

Pinus cfr. prodromiis FTr Pinus sp.

Coniopteris Heerianus (Yok.) ...Coniopteris Heerianus {Y ok.)

Onychiopsis elongata (Geyler) . . . Onychiopsis elongata (Geyler)

Dicksonia nephrocarpa (Bunb.)...C1 hymenophylloides (Brongn.)

Asplenmm distans Hr.

A. argutulum Hr.

A. argutulum Hr. ] ^ ^ , .. 7 . /-r.

> C. cfr. denticulata (Brongn.)

Pecopteris exilis Phill.J

Adiantite's Sewardi Adiantites Sewardi

Equisetum ushimarense Yok Eq^dsetum ushimarense Yok.

As may be seen from the above, the general features of the
two floras are essentially the same, so that their contemporaneity
is quite evident. ^^

Prof. YoKOYAMA,^^ in his elaborate work on the Jurassic
plants from Kaga, Hida and Echizen, assigned most of the fossili-
ferous beds to the Bathonian age, which view was subsequently
accepted by Nathorst^^ and others. However, Ward^^ and
Seward^^ among others point out the presence in the flora of
many plants of Wealden types, suggesting that the plant-beds are
of the lower Cretaceous age.

In a paper recently written on the Ammonites from Echizen,
Prof. YoKOYAMA*^^ states his conclusion that at least a part of
the Tetori series is younger than the Bathonian and must be of
the Malm age.

1) In this place, it should also be borne in mind lliat tlie Jurassic districts of Kaga,
Hida and Echizen in Japan and of Kyöng-syan,g-Do in Korea are nearly in the same
latitnde.

2) Page 19.

3) Nathobst: Beitr. z. Mesoz. Flora Japans.

4) Ward: Geographical Distribntion of Fossil Plants, p. 789.

5) Seward: Wealden Flora, I., p. ?,2.

6) YoKOYAMA : Jurassic Ammonites from Echizen and Nagato, p. 3.

MESOZOIC PLANTS FROM KOREA. 51

As it appears from previous writings on the geology of the
said region, the plant bed of Kinebashi, Uchinami and Ötani
with Adiantites Sewardi, and that of Hakogase with Nilssonia
orienialis lie immediately upon the Ammonites-bed of propably
the lower Malm age, and Prof. Yokoyama, in his paper, suspects
that the plant bed of Ozö, prov. Kaga, with Dictyozamites fal-
catus, may possibly be somewhat older than the Bathonian ;
taking it as a whole, the Tetori series seems to represent a
phase of the Jurassic in the narrowest sense of the word (i.e.
Malm and Dogger).

The resemblance of the Japanese flora to the Siberian was
pointed out by Geyler ;^^ Heer^> also mentioned that Thyrsopteris
elongata of Geyler is a type of fern very common in the Siberian
Jurassic. Prof Yokoyama 's renewed examination of the rich
material of the Tetori flora revealed that 16 out of 36 species
are found in Siberia.^^

Beside Nilssonia orientalis, there are other forms common to
Siberia and Korea, such as Coniopteris kymenophylloides, Clado-
phlebis denticulata, Finus Nordenskjoldi, P. prodromus and Podo-
zamites lanceolatus. These, however, are cosmopolitan forms, also
found in Japan and China.

From China and Mongolia, Nathorst,''^ Schenk,^^ Zeiller,"^^

1) Geyler: Ueber Fossile Pflanzen aus der Juraformation Japans. — Ueber Einige
Palseoutologische Fragen, insbesondere über die Jurafortnation Nordostasiens.

2) Heer : Nachträge z. Jura-flora Sibiriens, p. 4.

3) Yokoyama : 1. c. p. 16.

4) Nathorst: Om forekomsten af Dictyophyllum Nilssoni Brongn. sp. i. Kinas Kol-
orando bildningen. (By review.)

5) Schenk : Pflanzlich Versteinerungen (Richthofen's China IV.). — Die während der
Reise des Grafen Bela Szechenyi in China gesammelten fossilen Pflanzen.

6) Zeiller : Remarques sur la Flora Fossile de 1' Altai, à propose des dernières de-
couvertes Palseobotanique de M.M. les Drs. Bodenbender et Kurtz dans le république Ar-
gentine.

52 AET. 8. — H. YABE.

Schmalhausen/^ Newberry^^ and Beongniart^^ have described
many Jurassic and Rhaetic plants but there is relatively little
resemblance between these and the Korean flora. The identical
or closely allied forms are restricted to those fern types of a
very wide vertical and horizontal distribution. Among the Meso-
zoic plants, now in Prof. Yokoyama's possession, brought from
various parts of Northern and Southern China, there are also no
particular forms which indicate the presence of a close floral
relationship with the Korean fossils.

The Oroville flora^^ of California is the best known of the
Mesozoic vegetations in the Pacific border of North America.
It is assigned to the Oolitic age. The predominance of Ptero-
phyllurtiy Ctenis and Ctenophyllum consititutes a peculiar feature
more or less like that of the Jurassic flora of India. But between
this North American and the Japanese Tetori floras, there is
only a slight resemblance though they are nearly contemporaneous ;
the distinction between the former and the Korean flora is more
striking, apparently none of the species being either allied or
identical. If what Fontaine calls Pinus Nordenshjoldi is really
identical with the plant so named from Korea, then it is the
only species which shows any relation between the two floras.

In some respects, the Jurassic flora of India, seems to be
more closely related to the Korean. There is in common the
important species, Dictyozamites faleatus, above alluded to. It is

1) ScHMAliHAUSEN : Pflanzen aus der nordwestlichen Mongolei.

2) Newbekry: Notes on Some Fossil Plants from Northern China. Description of
Fossil Plants from the Chinese Coal-bearing Rocks in R. Pumpelly's Geological Researches
in China, Mongolia and Japan.

3) Bkongniaet : Note sur les Plantes fossiles de Tinkiako envoyés en 1875 par M.
l'Abbl. A. David.

4) Wabd : Status of the Mesozoic Floras of the United States. The Older Mesozoic.
P. 340.

MESOZOIC PLANTS FEOM KOREA. 53

very noteworthy, however, that the species is absent in the Oolitic
flora of Kach, though common in the older Eajmahal group.

On the whole, so far as evidence goes, the writer has little
hesitation in announcing the contemporaneity of the Naktong flora
of Korea with that of the Japanese Tetori series, the affinity of
the former to those of the corresponding age in Siberia, China,
India and CaUfornia being apparently more distant.

Nature of the plant-bed of Naktong : — Next the question
arises whether the plant bed is a fresh water deposit or not. In
this connection, the following facts must be taken into considera-
tion.

1) There are very few animal remains ; a very incomplete
bivalve and few Foraminifera with perforated calcareous shells
were found in the same deposit.

2) There is no trace of marine plants.

3) Plant impressions of a more or less large size are generally
arranged in the plane of stratification, often heaped one upon
another.

4) In some parts, however, only crumbled pieces of plants
are accumulated.

From the above facts, it is to be inferred that the deposition
took place at least not far from the sea coast and there is a
probability that the deposit was a beach formation in very shallow
brackish water.

54

AET. 8. — H. YABE.

Glossary of Korean Geographical Names Found
in the Text.'

An-eui.

Chhyön-san.

Chhyüung-chhyong-Do.

Chhyu-phung-nvöng.

Chin-an.

Chin-hai.

Chin-jyii.

Chyan-syöng.

Chyol-la-Do.

Chyöng-jyu.

Fusan (Pu-san).

Hai-nam.

Ham-chhyang.

Hoang-gan.

Im-sil.

In-chhyön.

Keum-chhyön.

Keum-san.

Koang-jyu.

Ku-ryoi.

Kyöng-syang-Do.

Masampho.

Mokpho.

Mu-jyu.

^È» f« 3Ë

m ^

m iij

M

iz )\\
# Mi

^v Oj m

^ m
m ^

Mun-gyöng.

Na-iyu.

Nak-tong.

Nak-tong-gang .

Nam-phyöng.

Nam-uön.

Ok-chhyön.

Ok-ka.

Phyöng-yaiig.

Po-eiin.

Pul-t ang-kokai .

Pu-sang.

Sa-chhyon.

Söul.

Syaujyu.

Syong-iii-san.

Tol-pa.

ïong-pok.

Tu-mau-gang.

Yong-am.

Yong-dam.

Yong-dong.

Yong-il.

Yuk-sim-nyöng.

m m

^ M

n m

m }\\

2l£

1) Koto and Kanazawa: A Catalogue of the Eomanized Geographical Name of
Korea. 1903.

MESOZOIC PLANTS FROM KOREA.

Bibliography.

C. J. F. BUNBURY

W DUNKER :

O. Felstmantel :

E. V. Ahnert : Reisen in der Mandsclirci. Ber. d. k. Russ. Googr.

Ges. Bd. XXXV. (By review.)

A. Brongniart : Histoire des Végéta nx Fossiles. 1838.

„ : Note sur les Plantes fossiles de Tinkiako (Schensi

meridional) envoyés en 1875 par M. l'Abble A.
David. Bull. Geol. Soc. France. Seris 3, vol. III.
On Some Fossil Plants from the Jurassic Strata of
the Yorkshire Coast. Q. J. G. S. Vol. VII. 1851.
Monographie der Norddeutschen Wealdenbildungen.
1846.

C. V. Ettingshausen : Beiträge zur Flora der Wealden Periode. Abh. d.
K. Gcol. Reichsanst. I. Bd., 3 Abth., Xo. 2. 1852.
lieber die Indischen Cycadengattungen Ptilopliyllmv ,
MoRR. u. Dictyozamites Oldh. Palaeontographica,
Supplement III. 1876.

Jurassic (Liassic) Flora of the Rajmahal Group in
the Rajmahal Hills. Mem. Geol. Surv. India, Pal.
Indica, ser. 2, vol. I., pt. 2. 1877.
Fossil Flora of the Gondwana System. Jurassic
(Oolitic) Flora of Kach. 1876. Ditto, vol. IL,
pt. 1. Flora of the Jabalpur Group (upper Gond-
wana) in the Sonnerbad Region. 1877. Ditto, vol.
IL, pt. 2.

Jurassic (Liassic) Flora of the Rajmahal Group from
Golapili. 1877. Ditto, Series 2, vol. L, pt. 3.
Outliers of the Madras Coast. 1879. Ditto, vol.
I., pt. 4.

Tlie Potomac or Younger Mesozoic Flora. U. S.
Geol. Surv. Mon., vol. XV. 1879.
Ueber Einige Palœontologische Fragen, insbesondere
ueber die Juraformation Nordasiens. Bericht ueber
die Senckenbergische Naturforschende Gesellschaft.
1878.

Ueber Fossile Pflanzen aus der Juraformation Japans.
Falseontegraphica. Vol.-^XXIV- 1877.

W. M. Fontaine

T. Geyler :

5ß

AKT. 8. — H. YABE.

gottsche
Heer :

Die Fossil
., pt. 2, Die

ß. KXOCHEXHAUER

B. KoTo

F. Krasser

Geologische Skizze von Korea. 1886.
Flora Fossilis Arctica. Vol. I., part. 2,
Flora der Polarlander, 1868. Vol. III.
Kreideflora der Arctischen Zone, 1875. Vol. IV.,
pt. 1, Beitr. zur Fossilen Flora Spitzbergens, 1877.
Pt. 2. Beitr. z. Jura. Flora Ostsibiriens n. d.
Amurlandes, 1877. Vol. V., pt. 2, Beitr. z. Fos-
silen Flora Sibiriens u. d. Amurlandes, 1878. Vol.
VI., pt. 2, Nachtrage z. Fossilen Flora Sibiriens,
1880. Pt. .3, Nachtr. z. Fossilen Flora Grönlands,
1880.

Korea. Verhandl. d. Abtheil. Berlin-Charlottenberg
d. Deutschen Kolonial Ges. 1901.
An Orographie Sketch of Korea. Jour Sei. Coli.
Tokyo, vol. XIX., no. 1. 1903.
B. KoTö and S. Kanazawa : A Catalogue of the Romanized Geogra-
phical Names of Korea.

Beiträge z. Kenntniss der Fossilien Kreideflora von
Kunstadt in Maehren. Beitr. Palaeont. u. Geol.
Osterr .-Ungarns u. Orients. X. 1896.
Die von W. A. Obroutschew in China u. Cen-
tralasien 1893-94 gesammelten Fossilen Pflanzen.
Denkschrift d. k. Akad. d. Wiss. Wien. 1901.
On the Sandstone and Shales of the Oolites of
Scarborough, with Descriptions of Some New Spe-
cies of Fossil Plants. Q. J. G. S. XX.
Bidrag till Bornholms Fossila Flora, Pteridofyter.
Kongl. Fysiogr. Sallsk. Handl., vol. XIII. 1902.
Gymnosperum.

K. Svenska Vet. Ac. Hndl., vol. XXXVI., No.
6. 1903.

Om Floran i Skanes Kolforande Bildinger. Floran
vid. Bjuf 1878-86. Floran vid. Hoganas och.
Helsingborg. 1878.

Beiträge z. Mesoz. Flora Japans. Denkschrift d.
k. Akad. Wiss. LVII. 1890.

Om forekomsten af Diclyophyllum Nihsonia Brogn.
sp. i Kinas Kolforande Bildninger. Ofversizt af
Kongl. Vet. Akad. Forhandl. 1890.

J. Leckenby

H. Möller

A. G. Nathorst

MESOZOIC PLANTS PKOM KOREA.

57

A. G. Nathorst

J. S. Newberry

T. Oldham and J.

H. POTONIE :

M. Raciborski :

G. Romanowski :

G. DE S APORTA :

A. Schenk :

J. Schmalhausen

Zur Mesoz. Flora Spitzbergens. Kongl. Svonska
Vetenskap Akademieus Handlinger. 1897.
Sur la Flore Fossiles des Regions Antarctiques.
C. R. Ac. Sc. CXXXVIII. 1904.
Notes on Some Fossil Plants from Northern China.
Amer. Journ. Sei. 3rd series, vol. XXVI. 1883.
Description of Fossil Plauts from the Chinese Coal-
bearing Rocks (in R. Pumpelly's Geol. Researches
in China etc.) 1866.
Morris : Fossil Flora of the Rajmahal Series in the
Rajmahal Hills. Pal. Indica. Ser. 2, vol. I., pt.
1. 1863.

Pflanzenreste aus der Juraformation. (Futterer's
Durch Asien III.) 1903.

Flora Kopalna Ognitrwalych Gelinsk Krakowskich.
I. Archaegoniatae. Abh. d. Wiss. Krakow. XVIII.
1894.

Ueber die Osmundaceen u. Schizaceen d. Juraform-
ation. Bot. Jahrb., vol. XIII. 1891.
Geol. Palseont. Uebersicht d. Nordostl. Thian-schan.
1881.

Plantes Jurassiques. Pal. Française. Ser. II, vols.
I-IV. 1873-91.

Flore Fossile de Portugal. 1894.
Die Fossilen Pflanzen d. Wernsdorfer-schichten in
den Nord-Karpathen. Palseontographica XIX.
1863.

Die Fossile Flora der Nordwestdeutchen Wealden-
formation. Ditto. 1871.

Jurassische Pflanzen. (Richthofen's China, vol.
IV.) 1883.

Fossile Pflanzen aus der Albours-kette, gesammelt
von E. TiETZE. 1887.

Fossile Pflanzen (Szechenyi's Reise, vol. III.)
1899.

Beiträge z. Jura Flora Russlands. Mem. Akad.
Imp. Sei. St. Petersb. Ser. VU, Vol. XXVII.,
No. 4. 1879.
Pflanzenpaleontologische Beiträge. (Melanges Bio-

58

ART. 8. — H. YABE.

A. C. Seward

J. Shirley :

J. Velonowsky

L. F. Ward

M. YOKOYAMA :

A. DE ZiGNO

logiques. Bull. d. I'Acad. Imper. d'Sci. d. St.

Petersburg. T. XI.) 188.3.

The Wealden Flora. Pt. I., 1894; pt. II., 1895.

Notes ou Some Jurassic Plauts iu the Mauchester

Museum. 1900.

The Jurassic Flora. Vol. I., 1900.

On the Oecurreuce of Dictyozamltes iu Englaud,

with Remarks on European and Eastern Flora.

Geol. Mag., no. 466. 1903.

Fossil Flora of Cape Colony. Annals of the South

African Museum. Vol. IV., pt. 1. 1903.

Additions to tlie Fossil Flora of Queensland.

of the Dep. of Mines of Queensland No. 7.

Die Fame der Böhmischen Kreideformation.

K. Böhm. Ges. Wiss, VII. folge, vol. II. 1

The Geographical Distribution of Fossil Plauts.

U. S. Geol. Surv., Annual Rep. VIII. 1887-88.

Status of the Mesoz. Floras of the United State.

U. S. Geol. Surv., 20th Ann. Rep. 1898-99.

Jurassic Plants from Kaga, Hida and Echizen.

Journ. Sc. Coll., vol. III., pt. 1.

Mesoz. Plants from Kozuke, Kii, Awa and Tosa.

Ditto, vol. VII. 1895.

On Some Fossil Plants from the Coal-bearinsc Series

of Nagato. Ditto, vol. IV. 1891.

Flora Fossilis Formationis Ooliticae. 1856-85.

Bull.
1898.
Abh.

MESOZOIC PLANTS FEOM KOREA.

59

Contents.

Introduction

Descriptions ,

Diciyommites

I>. fakatus

Nilssonia

N. orientcUis

N. sp

Dioonites

D. sp

Ct&aophyllum (?)

a sp

Podozamites

P. Rhcinii

P. lanceolaius

Firms

P. sp

Onyckiopsis

0. elongata

Coniopteris

C Heerianus

C. hymenophylloides ?

Cladophlebis

C. cfr. dmticulata

C. koraie7isis

C. cfr. Dunkeri

(7. sp

Spheriopteris

S. naktongensis

S. sp

Adianiites

A, Sewardi

Sagenopieris

S. bilobaia

Equisetum

E. ushimarevse

General Remarks and Conclusions

Constitution of the Fossil Flora

Geological Relation of the Fossil Flora

Nature of the Plant-bed of Naktong

Glossary of Korean Geographical Names found in the
Bibliography

Text.

Page.
.. 1

9

.. 9

11
.. 13

13
.. 14

14
.. 14

15
.. 15

16
.. 16

17
.. 18

18
.. 19

22
.. 26

27
... 29

32
.. 32

32
.. 37

37
... 38

39
39
41
41
43
43
44
44
45
53
54
55

H. TABE.

MESOZOIC PLANTS FROM KOREA

PLATE I.

Plate I.

(All^the figures are in natural size.)

Adiantiies Seioardi sp. nov. P. 39.

Figs. 1-7, Specimens from^Korea.

Fig. 8. A better specimen from Kami-uchinami, Onogori, prov.
Echizen in Japan.
Onychiopsis elongata (Geyler). P. 22.

Figs. 9-11. Specimens from Korea ; fig. 9 is a sterile pinna, while

the others are fertile ones.
Figs. 12-14. Specimens from Japan ; fig. 12 shows a specimen
from Kami-uchinami, Onogori, which was referred by
Mr. Matsushima to Sphenopteris Mantelli ; and figs. 13
and 14 are fertile ones from Shimamura, prov. Kaga.
Sphenopteris sp. P. 38.

Fig. 15. A fertile pinna.

Jour. Sei. Co//. Vo/. XX. Art. 8, P/. /.

H. Yäbe: Mesozoic Plants from Korea.

H. TABE.

MESOZOIC PLANTS FROM KOREA.

PLATE IL

Plate II.

(All the figures are in natural size.)

Cladophlehîs horaiensis sp. nov. P. 32.

Fig. 1. A sterile frond.
Dictyozamites fahatus (Morris). P. 11 .

Figs. 2-4. All pinnae sliow tlie reticulated venation, though not
represented on the figures.

Fig. 5. A pinna, figured specially to show the venation.

Figs. 6-7. Two detached pinnae shorter and broader than others.
Goniopteris hymenophylloides Brongn. ? P. 29.

Fig. 8. A fertile pinna.
Pinus sp; P. 18.

Fig. 9. P. cfr. Nordenfikjoldi Hr.

Fig. 10. P. cfr. prodromus Hr.

Jour. Sei. Co//. Vo/. XX. Art.8, P/. II.

H. Yäbe: Mesozoic Plants from Korea.

H. YABE.

MESOZOIC PLANTS FROM KOREA.

PLATE III.

Plate III.

(All the figures, except fig. 14, are in natural size.)

Nihsonia oi'ientalis He. P. 13.

Figs. 1-5.
Nilssonsia sp. P. 14.

Fig. 6.
Dioonîtes (?) sp. P. 14.

Fig. 7.
Goniopteris hymenopliylloides Bkonon. ? P. 21).

Fig. 8.
Coniopteris Heerianus (YoK.) P. 27.

Fig. 9. A fertile pinna.

Fig. 4. A pinnule, magnified two diameters.
Equisetum usliimarense Yok. P. 43.

Fig. 10.
Cladophlehis cfr. denticulata Beongn. P. 32.

Fig. 11.
Cladophlehis koraieoisis sp. nov. P. 32.

Figs. 12-13. Both represent the fertile pinnae ; on fig. 13, tJie
round dots are omitted.
OnycMopsis elongata (Geylee). P. 22.

Fig. 15.
Sagenopferis bilohata sp. nov. P. 41.

Fig. 16. a, var. major; b, c, var. cuneata.

Jour. Sei. Co//. Va/. XX. Art. 8, P/. ///.

Ftg.l.

H. Yabe: Mesozoic Plants from Korea.

H. Y ABK.

MESOZOIC PLANTS FROM KOREA.

PLATE IV.

Plate IV.

(All the figures are in natural size.)

Podozamites lanceolatus (TiixDL. and Huttox). V. 17.

Figs. 1-5.
Fodozamites Rheinii Geyi.er. P. 16.

Fig. 6.
Ctenophyllum (?) sp. P. 17.

Fig. 7.
Oladophlebis sp. P. 37.

Fig. 8.
Cladopldehis cfr. Dunheri (Schimper). P. .37.

Fig. 9.
Splienopieris naktongensis sj). nov. P. 38.

Figs. 10-11.
Gom'opteris sp. P. 26.

Fig. 12. A specimen with fertile pinnae, from Shimamura, prov.
Ivaga in Japan.

Jour. Sei. Co//. Vo/. XX. Art. 8, P/. /V.

Fvg.9.

H. Yähe: Mes ozoic Plants from Korea.

JOUENAL or THE COLLEGE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 9.

Modulus of Elasticity of Rocks'-^

AND

SOME INFERENCES RELATING TO SEISMOLOGY.

By
S. Kusakabe, Rigakushi.

With two plates.

CONTENTS.

I. Introduction.

II. Arrangement and Flexure-apparatus.

III. Process of Observation and Calculation.

IV. Experimental Results.

V. Some Inferences relating to Seismology.

INTRODUCTION.

The present experiments, detailed descriptions of which are
to be found in " The Publications of the Earthquake Investiga-
tion Committee in Foreign Languages " No. 17, Tokyo, 1904,
serve as a complement to the note, recently published by the

* A short abstract is to be found in the " Proceedings of the Tokyo Physico-Mathematical
Society" Vol. IL, No. 11. May, 1904.

£t AKT. 9. S. KUSAKABE.

author, on the modulus of rigidity of rocks.* Some of the speci-
mens were identical with those used in the last investigation,
and the others were prepared in a similar manner. The principal
object of the present investigation is not to determine any accurate
value of the modulus of elasticity, but to determine whether the
modulus is constant within tolerably wide limits or not, and if
it is not constant, how it varies with the amount of stress or
with time and other factors which affect the change. The modulus
is measured by the method of flexure ; but the apparatus is more
complex than the one which is generally employed. It may
therefore not be superfluous to give the following detailed descrip-
tion of it.

ARRANGEMENTS AND FLEXURE-APPARATUS.

In the measurement of flexure, the use of the cathetometre
or of the micrometrescrew are generally dispensed with. The
method with mirror and scale, as modified by A. KöNiG,t is
generally adopted, though that by optical interference is more
accurate. The apparatus as designed in the present experiment
combines the advantages of König's arrangement with other
necessary appliances. The principal improvements are : — (1) the
bending of the specimen cyclically from one side to the other,
with increasing and decreasing force passing through zero conti-
nuously ; (2) the elimination of external disturbances such as any
minute rotation of the specimen or slight displacements of the
scale and telescope.

* Ibid. Vol. I., No. 14, Oct. 1902. Pub. of the E. I. C. in F. L. No. 14, Tokyo, 1903.
The Journal of the College of Science, Imperial University of Tokyo. Vol. XIX., Art. 6.
1903.

+ A. König. Ueber eine neue Methode zur Bestimmung des Elasticitätsmoduls. Wied.
Ann. 28, 1886.

MODULUS OF ELASTICITY OF ROCKS. 3

A rough sketch of the arrangement is shown in Fig. 1. The
specimen is placed horizontally and, when it is bent, its plane
of curvature is also horizontal. There are necessarily one scale,
four fulcrums and four mirrors, of which two mirrors Mj and Mo
are attached to the specimen as in König's method, while the
others M3 and M4 are rigidly fixed to the support. Four different
images of one and the same scale S are to be seen in the field
of the telescope T, Fig. 2. They are all reflected twice by the
following mirrors respectively : —

Right upper image reflected by the mirrors Mj and Mo,

right lower „ ,, „ ,, Mi and M4,

left upper „ „ ,, „ M3 and Mo,

left lower „ ,, „ „ Mg and M4.

The apparatus is shewn in Figs. 3, 4 and 5, in its front-

and side-views as well as in its plan. The two mirrors Mi and

M2 rotate as the specimen is bent, while the other mirrors M3

and M4 are fixed unless the apparatus itself is displaced. The

fulcrums Fi, Fo, F3 and F4 are so adjusted that the edges of any

two of them lie in a vertical plane. A small framework F, which

is shown in Fig. 6, serves to apply bending force to the specimen.

The frame- work consists of two wedges, one fixed (W,) and the

other movable (W2) inside a proper case. After placing a proper

specimen between the two wedges, the movable wedge W.. may

be pushed firmly against the specimen by the fixed screw S.

At the extremities of the strings Si and So, which run over small

pulleys Pi, P2 etc. towards the observer, some weights are hung

which supply the bending force. The support of the fulcrums

is made of soft iron, which is rigidly screwed on a wooden block.

From what has been just described, it may be easily seen

that, when equal weights are hung on both Si and S2, no bending

4 AET. 9. S. KUSAKABE.

force is exerted on the specimea, and that it is only the difference
of the weio'hts attached to the two strino-s which is effective in
bending it. That is to say, if m^ and m-i are the two weights
attached to the strings Si and S2 respectively, then their sum
^1 + ^2 = ^1(1 regists the action of the bending force, tlie last of
which is due to their difference m2—')ni = ^l. For future reference,
Mo and M will be called the resisting 7nass and the effective mass
respectively.

AVhen the effective mass is positive, the specimen is supported
by the fulcrums F3 and F4, and it becomes convex towards the
righthand side. In the other case, it becomes convex towards
the lefthand side, supported by the fulcrums Fj and F2. The
fulcrums standing face to face, i.e. Fj and F3 or Fo and F4, are
clamped so as not to push too tightly against the specimen, as
there is a possibility that the bending of the specimen will be
hindered by friction.

A telescope, provided with a micrometer-screw, is rigidly
clamj^ed on a tripod. The scale, engraved on a ground-glass
plate, 20 em. long and 2 cm. wide is covered with a black board
having a slit, 8 mm. wide, and is illuminated by a row of small
gas jets.

PROCESS OF OBSERVATION AND CALCULATION.

The observation is generally as follows : —

1. To begin with, equal weights, each i Mq, are hung on
the strings S^ and S2.

2. A specimen is put between the fulcrums, passing through
the frame-work F, the last of which is to be clamped on the

MODULUS OF ELASTICITY OF ROCKS. 5

middle part of the specimen. The pLanes passing through the
edges of the fulcrams standing face to ftice shouhl be normal to
the length of the specimen.

3. The four mirrors are so clamped in their proper positions
that the images of the scale reflected by them stand side by side
within the telescope-field. To adjust them properly requires much
practice.

4. The constants of the micrometer-screw for all images are
determined. They are nearly equal to each other but not strictl'y
so. One mm. of the scale division is equal to about 23 divisions
of the micrometer-screw, which is again equivalent to a rotation
of 0-176x10-' rad. i.e. l"-068.

5. Zero-readings are taken for all images in a fixed order ;
i.e. (i) right upper image, (ii) right lower image, (iii) left upper
image, (iv) left lower image.

6. The suspended weights consist of some forty pieces of
equal weight. A definite number of pieces, say i m., are taken
off from one string and added on the other. The bending force
due to this is obviously mg, where y represent the value of gravity.
The time-record corresponding to this transposition of weights is
taken.

7. After a certain time, the readings are noted for all images
in the same order as in the case of the zero-reading.

8. Second transposition of weights ; the time recorded ;
scale-readings noted : and so on till a definite amount of bending
force is reached.

9. The weight is then transposed in the opposite way so
that the force diminishes gradually and ultimately becomes op-
positely directed. In this way, a series of observations is made
to complete the cyclic process several times.

6 AKT. 9. S. KUSAKABE.

10. From the amount of the deviations of the images, the
amount of bending due to each corresponding force is calculated,
by the following method.

In Fig. 7, let the zero-reading be taken when the telescope
is in T while the mirrors Mj and Mg are in the position Mmj
and the mirrors M2 and M4 in the position M'm2. In reality,
the reflections of light by the mirrors take place, as a matter of
course, in the space of three dimensions ; but, for the sake of
simplicity, let us assume that the path of the ray of light lies
wholly on the plane of the paper. Let ab be the position of the
scale, and suppose that <x is a point which gives its images in the
field of the telescope after reflecting at s and s'. Suppose that,
after a certain number of operations, the specimen is bent, it is
rotated and also the telescope is displaced and rotated relatively
to the scale. Let their respective values be given by
a = angle through which the mirror M2

is rotated as the specimen is bent,
— a=anlge through which the mirror Mi

is rotated as the specimen is bent,
^3=angle through which the specimen is rotated,
o=the component of the displacement of

the telescope parallel to the scale.
Note that the other component is negligible relatively
to the distance between the scale and the telescope.
/'=the amount of rotation of the telescope.
Then, if 6 and co denote the angles between the mirrors M2
and M4, Ml and M3 respectively, we have

MODULUS OF ELASTICITY OF ROCKS.

TABLE I.

Specimen convex to :

Righthand side.

Lefthand side.

Rotation of the specimen.

Clockwise.

Cüunter-clock-
w ise.

Clockwise.

Counter-clock-
wise.

d

a-ß

a + /9

-a + ß

-a-ß

to

-a~ß

-a + ß

a + ß

.-ß

That is to say, provided a and ß are taken as algebraic quantities
having proper signs, we have simply

d=a-}-ß
cij=ß — a
Let t' be the last position of the telescope, and put

R.U. =The deviation of the right upper image,
Il.L.= ,, „ ,, ,, right lower image,

L.U.= ,, ,, ,, ,, left upper image,

L.L.= ,, ,, „ ,, left lower image.

Then, from simple geometry, it may be easily proved that
L.L.=o^-f (D-fc + d)}',
L.V. = d+(D + c + d)r-2dw,
'R.h.=^d+(D + G + Q\)r+2{c + d)d,
E.U. = o + (D + c + d)j'-2d.^+2(c + d)^,
where ss' = c , sa=d , ts' = D.

If there is no external disturbance, evidently we have

ß = 0 , /' = 0 , 0 = 0,
so that L.L. = o

L.U. = 2d«,
R.L. = 2(c-fd)a,
R.U.=2(c-fd)« + 2d«,

= 4{d-f|}a,

the last of which is a well known form.

8

ART. 9. S. KU8AKABE.

In all cases, we have between the four values the following
functional relation.

L.L. + K.U.=E.L. + L.U.
Thus, the difference of the two sums indicates an error of observa-
tion : whence it gives the means of rejecting from numerous
observations those which are incorrect. For instance, in the
case of a piece of sandstone w^e had :

TABLE II.

M.

R.U.

R.L.

L.U.

L.L.

R.U. + L.L.

R.L. + L.U.

Error.

900"

1-129

0^529

0-593

-0-004

1^125

1-122

0-003

1200

1-670

0-789

0-877

-0-006

1-664

1-666

-0-002

1500

2-204

1-039

1-159

-0-007

2-197

2-198

-0-001

To calculate the amount of bending, we have four equations
containing four unknown quantities. There is, however, one
functional relation between the four equations. At the same
time, the unknown quantities also may be reduced to three, as
d and Y appear always in one and the same combination.
Put x=L.L.

x + y=L.U.

x + z=E.L.
then x+y + z = K.U.

Taking any three of the four equations, we may solve them. . It
is preferable, however, to use all equations, since none of them
is strictly correct. Applying the method of least squares we
have

x = i[3L.L. + L.U. + R.L.-K.U.],

MODULUS OF ELASTICITY OF EOCKS.

9

y=J[(L.U. + R.U.)-(L.L. + R.L.)],
z = è[(R.L. + R.U.)-(L.L. + L.U.)],
where x = o + (c + d + I))y,

J=-2dco,

z = 2(c + d)6',
and d=a + ß

CO — ß — a.

Eliminating x, y, z, 6 and co from the above equations, we
have

a = 5

1 c + 2d

d(c + d){(R-U-L-L-).(L.U.-R.L.)^}_
In the example above cited, we have

TABLE III.

c = 12"4 cm.

d = 241-5 cm.

c+2d

=2-503x10

C+2J
d(c+dj

= 1-0099x10

M.

R.U.-L.L.

R.L.-L.U.

ß

900
1200
1500

1-133
1-676
2 211

-0-064
-0-088
-0-120

11-46x10 rad.

16-95

22-36

■093X10 rad.

■1-31

■1-76

It is to be noticed that, in the above calculation, the tangent and
arc of an angle are taken to be equal to each other. The greatest
angle to be dealt with is of the order of 10~- radian : whence
the difference between the tangent and the arc is of the order
lO"'^, that is to say, it is of the order of 10~^ of their own
amounts, which is within the error of observation.

10 ART. 9. S. KUSAKABE.

The relation between the modulus of elasticity and the
amount of bending is given by the well known formula.

X1j = ï pi

ab a
where a and b are the breadth and thickness of the specimen,
while I is the distance between the corresponding fulcrums.

EXPERIMENTAL RESULTS.

The last investigation with regard to the modulus of rigidity
proved a great deviation from Hooke's law even in the case of
the least strain. Preliminary experiments showed it to be the
same also in the case of bending. Looking at the curve in Fig.
8, we see that there is a tendency on the part of the rock to
persist in any strained state which it may have acquired, especially
when the variation of the stress changes its sign. The curve is
closed and it is also of simple and regular form, though its path
during the increase of stress differs entirely from that during the
decrease. All rocks, so far as the author has investigated, have
this jDroperty in common, though they differ in the character of
the curves and in other minute details.

It may be suspected that, as the resisting mass increases
with the total mass to be moved, this may have some in-
fluence upon the hysteresis curve. The comparison of the results
of four successive experiments, in which the resisting mass was
1000, 1300, 1600 and 1900 grams respectively, while all other
conditions remained the same throughout, showed that the influence
of the resisting mass might be safely neglected.

There is one important fact which deserves to be mentioned
here. Altliough the hysteresis curve is of a definite form and

MODULUS OF ELASTICITY OF ROCKS. 11

traces one and the same curve when a specimen is bent and
unbent many times cyclically, the direction of the elongation of
the curve does not remain fixed when the amplitude of the cycle
— i.e. the greatest amount of stress applied to the specimen during
the cycle — is varied. As a general rule, the hysteresis curve
becomes more and more vertical when the amplitude of the
cycle is further and further increased. One instance is given in
Fig. 9.

The amount of hysteresis, which is to be measured by the
area enclosed by the curve or by some function of it, is least for
Archaean rocks and increases rapidly for new rocks.

Although the modulus of elasticity is never constant during
a cyclical strain, its variation obeys one and the same law for
both the on- and the ofP-curve, in so far as the centre of the
cycle coinsides with the neutral state of the specimen. It may
be necessary to make a remark on the meaning of the term
" Modulus of Elasticity." As there is a great amount of hysteresis
in the relation of stress to strain, the ordinary conception of the mo-
dulus of elasticity is ultimately vague and uncertain. The actual
resistance to the deformation in any state whatever, be it already
bent or twisted, elastic or plastic in that state, will be taken as
the measure of elasticity in that state, so that in the present
experiment, the modulus is measured, step by step, by the in-
crease of bending per 200 grams increase of the effective mass.

The curve expressing the relation between the modulus of
elasticity and the amount of strain is symmetrical \vith respect to
the axis of ordinate. One instance for Limestone No. 29, is
given in Fig. 10. Each kind of rock seems to have its own
special character. If Hooke's law were to hold good, four branches
of curve, of which the right and left branches correspond to the

12 ART. 9. S. KUSAKABE.

cases ^Ylle^e the specimen is bent convex rightliand or lefthand
side respectively, while the lower branches correspond to the
increasing stress and the upj^er to the decreasing one, would all
shrink to a single horizontal straight line. In the case wdiere
no hysteresis exists, both the upper and the lower branches
would coincide mth each other to make a line not necessarily
straight.

For all cases of rocks here experimented upon, the upper
branch is concave towards the positive axis of the ordinate. As
to its character, however, the variety is very abundant : circular,
hyperbolic, oval and other curves of higher order of complexity.
The curvature of the lower branches is turned sometimes upwards
and at other times downwards. Although it is not easy to de-
termine any law according to which the modulus varies with the
phase of the cycle, we may find, as a first approximation, an
empirical expression for each specimen. For instance, in the
case of sandstone, we have.

for the upper branch, yi =0*243 + 0'92x"

for the lower branch, y2 = 0*243 +0'043x"

where y and x represent Ex 10"^^ and the phase respectively.
As a matter of fact, the constant term of yi is equal to that of
y2, representing the modulus of elasticity at the state where no
external force is acting.

In the following table, the constant term of the expression
for every specimen is given as the modulus of elasticity of several
rocks. It corresponds therefore to the value of the modulus of
elasticity in the state when the bending force became zero, during
which the specimen, whose section was about one centimeter
square and the distance between the fulcrums was 10 centimeters,
was bent cyclically on both sides by a force varying between

MODULUS OF ELASTICITY OF ROCKS.

10
O

those due to M = ±3000 grams-weight. The vakie in any other
state under different conditions must necessarily be different from

those given in the table.

TABLE IV.

No.

Rock.

Locality.

Kind.

Density.

Mod. of
elasticity.

Mean E.

Velocity (if
Loiif;. \V;ive.

ARCH^AÎf ROCKS.

X 10"(c

.g.s.)

Km.

Sec.

31i.

Quartzschist.

Chichibu.

Metaniorphic

2-67

10-48-7-07

8-78

5-73

46i.

)>

Gnmma.

))

2-62

8-41— 8-40

8-41

5-67

81.

Serpentine.

Chichibu.

Eruptive

(altered).

2-72

7-73— 7-21

7-47

5-24

40i.

Micaschist.

Ibaraki.

Metamorphic.

2-54

6-49— 5-92

6-21

4-94

18,.

Chloriteschist.

Chichibu.

>)

2-88

8-63— 5-39

7-01

4-93

72-

Peridotite.

Kuji.

Eruptive

(altered).

2-61

6-73— 5-83

6-28

4-91

26i.

Chloritescliist.

Cliiohibu.

Metamorpliic.

2-82

7-03— 6-29

6-66

4-86

22i.

Gabbro.

))

Eruptive.

2-71

6-21-5.57

5-89

4-66

24i.

Graph itesch ist.

)i

Metamorphic.

2-59

5-12— 4-93

5-03

4-41

23i.

')

»>

))

2-56

3-69— 3-37

3-53

3-71

422.

Micaschist.

Ibaraki.

"

2-63

1.29—1-16

1-23

2-16

PALAEOZOIC ROCKS.

34i.

Adinoleslate.

Gumma.

Sedimentary.

2-64

10-99—10-23

10-61

6-34

122.

Clayslate.

Aumi.

))

2.71

] 0-71— 9-08

9-90

6-04

9i.

Granite.

Mikage.

Eruptive.

2-54

4-31— 3-66

3-99

3-96

2I1.

Limestone.

Chichibu.

Sedimentary

(Jletiuuoriihosed).

2-64

4-14-3-65

3-90

3-84

63.

Marble.

Kuji.

))

2 68

3-51—3.24

3-38

3-55

14i.

Red Schal stein.

Aumi.

Sedimentary.

2-43

3-09-2.39

2-74

3-36

32i.

Pyrosenite.

Gumma.

)>

2.90

2-96— 2-91

2-94

3-18

lOo.

Granite.

Kagawa.

Eruptive.

2-57

2-30—2-10

2-20

2-93

29i.

Limestone.

Gumma.

Metamorphic.

2-66

2-06-1.92

1-99

2-74

14

ART. 9. — S. KUSAKABE.

No.

Rock.

Locality.

Kind.

Density.

Mod. of
elasticity.

Mean E

Velocity of
Long. Wave.

TERTIARY ROCKS. xlO"(c.g.s.) ~^

3Ô1.

50.

23.

5.^.

43.
34.

Sandstone.

Two Pyroxene
Andésite.

Tuff.
Rhyolite.

Sandstone.

Cliichibu.

Mutsu.

Izu.

Yechizen.

Kii.

Chöshi.

Sedimentary.

Eruptive.

Sedimentary.

Eruptive.

Sedimentary.

2-47
2-70
1-90
2-40
2'25
2-21

3-55 -3-51

4-04 2-3S
1-39-136
0-90-0-77
0-6S— 0 57
0-34— 0-20

3-53
3-21

1-38
0-84
0-63
0-27

3-78
3-44

2-69

1-87
1-67
Ml

DILUVIUM ROCKS.

172.

16.

Andésite.

Gumma.

Eruptive.

2-63
2'32

4-36— 4-31

0-68— 0-63

4-34

0-66

4.06
1-69

As the hysteresis curve becomes more and more vertical
when the amplitude of the cycle increases further and further,
the mean elasticity necessarily weakens when the amplitude of
the strain increases. For instance, in a case of sandstone where
the curves conld all be represented by a series of parabolic ex-
pressions, the constant term of them was : —

TABLE V.

Amp. (in gramsweiglit.)

300

600

1200

1800

2400

3000

E. (c. g. s. X 10").

0-65

0-46

0-33

0-27

0-21

0-15

Thus it is important to notice how the modulus of elasticity di-
minishes when the amplitude of the strain increases.

Here it will suffice to remark that, as in the case of the
modulus of rigidity, the modulus of elasticity also is comparatively
greater in a strained than in the neutral state.

MODULUS OF ELASTICITY OF ROCKS. 15

The phenomenon of yielding, though it is not so enormous as
in the case of torsion, is still sufficiently great to be dealt with.
For a piece of sandstone, e.g., wliich was loaded with Mo = 3oOO
and M =3000 gramsweight, the amount of bending increased, in
the course of two and a half days, to, at least, more than three
times its initial value. It is, indeed, questionable whether there
is any limit to the yielding.

Also the amount of residual surviving the bending force does
not I'emain constant, but recovers gradually and uninterruptedly.
The amount of recovery, in the case of the above specimen just
referred to, increased, in the course of about four days, by more
than twice its initial value.

The yielding of specimens under a constant force having
become comparatively small after a few days, the temperature-
variation of the flexure may be clearly observed. The relation
between temperature and bending for a piece of sandstone is given
in Fig. 11. The curve, as a whole, expresses the simple propor-
tionality between the two elements. We find, however, the amount
of flexure has a minimum value in the neighbourhood of about
9°C. In the case of the rigidity-modulus, we had a result strictly
analogous of this effect. It may be, however, the effect of moisture.
To determine any general relation between elasticity and tempera-
ture requires further investigation by a special arrangement.

SOME INFERENCES RELATING TO SEISMOLOGY.

In the author's publications above cited, it was experimentally
as well as theoretically explained that, in the case of distorsional
waves, the velocity of propagation is a function of the amplitude

16 ART. 9. — S. IvUSAKABE.

of the wave, as there exists more or less yielding in the rocks
through which the waves propagate, and also that, in view of
this inference, we do not see the necessity of assuming the path
of the tremors to be different from that of the principal shocks.
The present experiment relating to other modulus give, it seems to
me, still stronger foundations to support the above view. We must
not however forget that, it is unfortunately the common rule ra-
ther than the exception that a theory, however perfect it may
be, does not explain all the facts connected with it and also that
almost every phenomenon has more than a single cause, and this
is particularly true in the case of earthquakes.

As the elastic constant varies during one cycle of bending
and all values at different phases of the cycle equally play their
parts in causing the vibratory motion, the apparent value of the
elastic constant during one complete vibration must be the mean
value of all the values at different phases. Now the mean elas-
ticity for one complete cycle being distinctly greater than what
is commonly adopted, the actual velocity of propagation for seis-
mic waves must be correspondingly greater than those given in
the above table, which are calculated by taking the square root
of the elasticity-density ratio. In the case, e.g., of a piece of
sandstone, the result of the experiment shows that the mean
value is 3'67 times greater than the constant term. Whence we
may infer that the actual velocity, in this case, would be probably
twice the value given in the table.

Again, the velocity must necessarily diminish with an increase
of the amplitude of the wave, since the elasticity diminishes in
that case as explained above. From the example given there,
we may deduce the following to show how the velocity changes
with the amplitude.

MODULUS OF ELASTICITY OF ROCKS.

17

TABLE VI.

Ratio of Amplitudes.

1

2

4

6

8

10

Ratio of Velocities.

2-08

1-75

1-48

l-;54

1-18

1.00

Though the variation of the elastic constant due to temperature-
rise is comparatively small, it can never be neglected in so far
as the velocity of seismic waves is concerned, since the under-
ground temperature rapidly rises with the increase of depth.
Although the elastic constants increase from Cainozoic ito Archaean
rocks in a greater ratio than the density does, to attain the main
stratum of Palaeozoic rocks we must go deep down some ten
kilometres, at least, and for a stratum of Archaean rocks, at least,
thirty kilometres, where the underground temperature must be
tolerably high. Any conclusive deductions should, however, be
postponed until the more accurate observations on the change of
the elastic constants due to temperature-rise, which are in course
of preparation, shall have fully elucidated the relation between
the elastic constants and the temperature.

As a matter of fact, there are scattered everywhere within the
earth's crust veins and dikes of different kinds of old rocks, uplifted
by geological disturbances, some of which run over many hundreds
or thousands of kilometres. The velocity along such a vein or
dike must necessarily be greater than that through any of the
surrounding strata, so that the seismic waves mainly propagate
through that region. As a consequence of the above result, if
an observing station be situated near such a vein, not only will
the number of earthquakes observed at the station be greater
than those observed at any place in the vicinity, but the direction
of the motion will not necessarily indicate the position of the

18 ART. 9. — S. KUÖAKABE.

seismic centre. It is a matter of daily experience with us who
live in an earthquake country located in the ' girdle of fire of
the Pacific, that observers in some districts feel all shocks as if
coming from one particular direction even when the seismic centre
lies in an entirely different direction.

As another consequence, there may exist seismic shadows ;
or, in other words, seismic waves may be partially shielded by a
vein or dike of old rocks. Earthquakes originating in one region
may always be well observed in the station while those originating
in another region may fail to be observed in the station. In
Prof. F. Omoei's paper* we find a most interesting example to
support the above consideration. Of the earthquakes which hap-
pened between Sept. 1889 and July 1886 in Central Japan, those
whose origins were situated within certain boundaries were never
felt in Tokyo, though the weaker ones of more distant origins
were clearly felt there.

The frequency of earthquakes as related to the geological
distribution of rocks will be fully discussed in a following paper
under a special title.

In conclusion, I wish to express my great indebtedness to
Mr. FuKUCHi for valuable information concerning the geological
characters of the specimens. My best thanks are due to Professor
H. Nagaoka, under whose kind guidance I have carried out this
experiment.

* F. Omori. The Pub. of the E. I. C. in F. L. No. 11. 1902.

Jour. Sei. Coll. Vol XX. Art. IX PI. f.

Ftg.l.

L^ïhandaide.

JUffhtfijandaide.

Fi^.2.

Thejielfi of telescope-.

INI llllllllll
llllllll

II

6

11

hm 1
iiliiii

IM

lllllllll

*

1

IUI iiiiliiii 1
llllllll

II
6

II

1 7

lull 1
lllllll

III

7

lllllllll

*

1

Ftg.6.

- — ^; »»5

<?/>«>. I

FÙ/. .V.

Front view.

Fig. 5.

Plan, tif tfie J7eücure apparwbivs.

li'—

Fi&Jl.

Flg. 10.

m.29,.

Limestone.

\

Y

16.

•lO'

(n

diai

^J

\

/

/

X

\

s

/

y^

\

\,

y*

/

X

y"

"s

NJ

y

/^

^

^

.

y'

{y

S

\

^

^

»-

^

S^~

>

-r

H»-

—

~K

\

^

^

xto

^v>1 '

1 —

tf

b° iF 13^ TP IT 13

Jour. Sei. Coll. Vol. XX. Art IX PI. II.

Fig. 7.

(r*26)c

(r+2l)-2w)ci

"^

!

~

1
1

\

1

1

1

—

1

^

;

.-/

«t.

/

^ 1

/

/

i

F Iff. 8.

/

/

1

\Fvg.9.

/

/'

—

/

/

M. 16.

/

/

/

A/tdesite:

/

1

t

1

/

4

/

/

J-^^

/

/

/

1

i^iri

/

/

i / i/

j

'

/

//^

/

V

r

/

1

/

/

/

\

/

Foi

p».

-30

-a

0

-10

1.

10

^

0

80

/

fj

\

/

/

1/

/

/

/

/

V

h

(

'

^

/

f

1

4

1

1

1

k

/

1

/

/

f

J

/

1

1

/

/

.

/

/

/

/

/

/

/

/

/

/■

/

/

/

/

/

/

y

/

/,

/

/

^

l

y

1

1

JOURNAL OF THE COLLEGE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 10.

Kinetic Measurements ol the Modulus of Elasticity

FOR

158 SPECIMENS OF ROCKS:

AND

A NOTE ON THE RELATION BETWEEN THE STATIC AND THE KINETIC VALUES

OF THE SAME.*

S. Kusakabe, Rigakushi.

With one 'plate.

CONTENTS.

I, Introduction.
II. Method of Measurement and Apparatus.

III. Test of the Apparatus with a Tuning-fork and a Prism
of Soft Iron.

IV. Preliminary Experiments, and the Process of Calcula-
tion.

V. Experimental Eesults.
VI. Effect of Moisture.
VIL A Note on the Relation between the Static and the
Kinetic Values of the Modulus.

* The essence of this report was read in the meetings of the Tokyo Physico-Mathema-
tical Society for Sep. 1904 and April 1905 ; and was published in Nos. 15 and 22, Vol.
II, Proc. of the Society.

ART. 10. — H. KUSAICABE.

I. INTRODUCTION.

As to the elastic constants of several rocks wliicli compose the
so-called outer-crust of our planet, we have already a valuable
paper joublished by Professor H. Nagaoka,* in which the elastic
constants for one hundred specimens are given, with some notes
relating to seismology. The author also has reported some ex-
perimental researches relating to the hysteresis and the variation
of the constants under different conditions.^ A little considera-
tion, however, will make it evident that the values of elastic
constants determined by a statical method may be far from what
ought to be used in the discussion of seismic waves.

The importance of measuring them again by a kinetic
method may thus be readily recognized. Not only are the rapid
a,lterations of state concerned in the propagation of a wave attended
with a thermal effect, which goes to change the elastic constants
beyond their value statically obtained ; but also in the case of
rocks, the phenomenon of yielding may have great influence in
making them deviate further from one another.

In the author's papers above cited it has been shown that
the modulus of elasticity varies, during one cycle of strain, ac-
cording to a definite law. When a piece of rock is bent by a
force and unbent by virtue of its own elasticity, it is not, evidently,
the modulus of elasticity at any particular state, which determines
the vibratory motion of the rock. Afoduli of elasticity at all
the different phases of the vibratory motion equally take part in

* H. Nagaoka.. The Pub. of the E.I.C. in RL. No. 4, 1900, and Phil. ]\Iag.
1900.

t S. KusAKABE. This Journal. Vol. XIX., Art. 6. and Vol. XX., Art. 9. The Pub.
of the E.I.C. in F.L. Nos. 14 and 17.

KTNETrC MEASUREMENT OF THE ELASTICITY OF ROCKS. 3

causing the motion. Hence for the apparent modukis of elasticity
during one complete vibration, we must take the mean value of
the moduli at all different phases.

For a j)iece of sandstone, e.g., when the maximum bending
force during the cycle was equal to that due to J/=^3000 grams-
weight, we had a mean value 3.67 times greater than that taken
at the state of no bending. In other words, the mean kinetic
modulus for this specimen, provided the vibration be assumed to
take place at as slow a rate as in this case, is nearly three and
a half times greater than the static modulus.

Possibly sound-experiments furnish the best means of ascer-
taining the kinetic modulus of elasticity. The results deduced
from such infinitely small strains as occur in sound vibration is
no doubt of great significance as regards the elastic property of
rocks. The fact that it seems strange to speak of the vibration
of such loose rocks as sandstone, shows at once that the measure-
ment of the kinetic modulus of elasticity in them is much more
defficult than in metallic substances.

The method to be adopted here is a new application of
Melde's experiment*, combined with the principle of resonance.
The number of specimens examined amounts to one hundred and
fifty eight, collected from various lacalities in the main islands of
Japan, containing 23 Archœan, ß5 Palaeozoic, 12 Mesozoic and 58
Cainozoic rocks. This whole series of experiments is but a begin-
ning in this field of inquir}^ and may be thrown aside as a wreck
in the path of progress ; yet the author ho^^es it may prove a help
to later investigators.

* Melde. Phil. Mag. Vol. 47, 1874. Lowery. Ibidem. Vol. 48, 1874.

ART. 10, — S. KUSAKABE.

II. METHOD OF MEASUREMENT AND APPARATUS.

The essence of Melde's experiment is that, on one hand, a
massive fork excited by a bow or sustained in permanent opera-
tion by an electro-magnet, j)roduces its vibrations in approximate
independence of the reactions of any light body, which may be
connected with it ; and, on the other hand, the period of the
forced vibration of the light body is determined solely by the
period of the force which is supposed to act on the system from
without. The principle of resonance is merely that the kinetic
energy or the amplitude of any forced vibration is the greatest
possible, when the period of the external force is that in which
the system would vibrate freely under the influence of its own
elasticity.

The present method is simply as follows : —

a. A specimen of rock, one end of which is tightly clamped
by a massive vice, is maintained in free vibration.

b. A fine wire of known linear density w is connected with
the free end of the specimen and stretched with a known

' tension.
c. The length I of the above wire, in which the wire vibrates
with maximum amplitude is measured.
The annexed figures show the plan and elevation of the
whole arrangement, which is laid directly on stone floor of the
laboratory.

KIXETIC MEASUREMENT OF THE ELASTICITY OF ROCKS.

0

Elevation.

ß = bench; jS,..=bridge ; J5^ = battery ; 7?=electro-magiiet ;
IZ'=hamnier ; il/=mirror; il/c- = microscope ; P=pully;
6i> = specimen ; »S'-. = stniig i.e. copper wire; <S'^=stan(l;
F=vice ; 11'= weight.

A specimen Sp. clamped in the vice V is tapped by a hammer,
H, which is supported by a stand St. and maintained in constant
vibration electro-magnetically. A fine copper wire Sr. of ca.
0.05 mm. in radius is connected by means of bee's wax to the
upper end of the specimen and stretclied over a bridge Br. by a
tension due to the suspended weight, TK The copper wire, wliose
breaking tension is ca. 140 grams, is strained by a tension of
100 grams during some tens of minutes. Both the Ijritlge and
the microscope 3Ic. may sHde along an iron bench B of ca. 118
cm. in length, in which a scale is graduated for each mm.

The distinction of forced and free vibrations is very impor-
tant, and must be clearly explained. If a vibration is the
response of the system to a force imposed upon it from without

6 ART. 10. S. KUSAKABE.

and is maintained by the continued operation of that force, it
is obviously a forced vibration. It must, however, be remembered
that any free vibration which we shall have in a laboratory ex-
periment takes its origin necessarily from a force acting upon it
from without. At first, there is a forced vibration not less im-
portant than its rival, but when the force is removed, though
there is no discontinuity in velocity or displacement, yet the
period of the force is at once exchanged for that natural to
the system and the forced vibration is converted into a free
vibration.

In the present case, the frequency of the hammer-blow,
which is easily adjustable weithin wade limits by varying its
moment of inertia, is about ten per second ; while that of the
vibration of the specimen lies between some three hundreds and
a thousand per second. In other words, between two consecutive
tappings, the number of vibrations of the specimen amounts to
some thirty or one hundred, so that the mode of vibration is
necessarily of a free nature.

in. TEST OF THE APPARATUS WITH A TUNING-FORK
AND A PRISM OF SOFT IRON.

So far as the principle is concerned, the above statements
are sufficient and nothing more is required. In laboratory work,
however, there are several experimental difficulties to be overcome ;
especially, maintenance of the vibration and the fulfilment of
the condition of a clamped end. The fixedness of the clamped
end is absolutely important, since the frequency varies inversely
as square of the length of the vibrating portion.

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS.

To test the apparatus, the experiment was made with a
tuning-fork of known frequency, giving the following result.

Tuning-fork. Solo. 384 VS. June 8, 1904.

Tr=40-ö95 grs. w=4-77 X lO--" grs./cm, gf = 979-8 in the laboratory.

Measured to

1st Ohs.

2nd Obs.

3rd Obs.

4tli Obs.

]\Iean.

F

calculated.

I node.

24-0

24-1

24- î

24.0

240

24-00

24-0

TI node.

47-7

48-4

48-0

47-6

47-9

23-95

47-8

Ill node.

71-0

71-ß

71-4

71-4

71-5

23-83

71 -f;

IV node.

95-1

95-8

95-2

9Ô-6

95-4

23-85

95-4

Here it is necessary to remark that, although the vibration
of the string is due to the periodic force imparted from the
specimen, the point of application of the force never corresjDonds
to a loop, but, on the contrary, there is a node in that vicinity.
Assuming the impressed force given by the vibrating rock, which
is connected with the string at a point whose distance from the
bridge is x=b, to vary as F cos pt, the amplitude of the motion
between the bridge and the rock is approximately given by

IÖ1U r

I a

r<

4 a'

cos

px
a

•2A2

ism"^^ i- "^ 9 cos""-^— I

2 P^ ^ ft>

a

4a'

pb
a

where r is the amplitude of the rock measured at the point where
the string is connected, and / is the coefficient of friction of the
string. From this expression it is evident that the amplitude of
the forced vibration attains its maximum value when the point
of application of the impressed force is a node.

In the above example, it is reasonable to assume that a
nodal point lies somewhere at a point near the origin of the

\

8 ART. 10. — S. KUSAKABE.

scale-measurement. Let this distance be denoted by a, then tbe

(I
above values of I' correspond to 1+ —^ where N is the number

of loops contained in the observed segment of the string. Thus
a and I being two unknown quantities, they may be easily cal-
culated by the method of least squares. The result is : —

I =23"^79.

The result of the second experiment with another string and
smaller tension is

Tf=20-908

I =16-32.
The number of free vibrations of gravest mode for a perfectly
flexible string is given by

1 / W.g

where W is the suspended weight to which the tension is due.
But if we consider the string as not being infinitely thin, its
stiffness must be taken into account. In the case where the extre-
mity of the wire is constrained to be a node by stretching it over
a bridge but no couple acts to fix its direction, the correction
for a circular wire is given by

_ T.imJE!

where p' and E' are the specific density and the modulus of
elasticity of the wire respectively.

Thus, the number of vibrations n of the specimen, which
is identical with that of the connected wire, is given by

7:wE 1

n=ni + 7io =ni +

32/>'' n,l'

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS. 9

Another term of correction clue to the effect of the wire
being connected at the free end may be easily ascertained as we
see that the effect of the small load d3I is the same as a
lengthening of the sj^ecimen, whose weight is M, in the ratio

M: M+cUL
Now, calculating the number of vibrations for the tuning-
fork from the data above given, we have

91 = 191 '9 from the first experiment,
7i=191'6 ,, ,, second ,,

The difference between each of these and the registered value
192 is within the error of observation.

As to the verification for the fulfilment of the condition of
a clamped end, the case of a tuning-fork with two prongs is
wholly out of place, so that it may not be superfluous to cite
here the following observations with a prism of soft iron. Any
continuation of the specimen beyond the clamped section would
be without effect, as it acquires no motion ; but as the first clamp
is relaxed, the pitch rapidly falls, in consequence of the increase
of the length. Hence, in tapping the specimen care must be
taken to give no impact to the clamped section, i.e. the specimen
should be tapped at a position corresponding to the centre of
percussion with respect to the clamped section.

The result of the first observation is as follows : — (c.g.s. units).

f> =7-779
b =0-763
Z = 13-4
Tr=30-718
«;=.5-19xl()-'
^' = 1-221x10^'
// =8-667

10 ART. 10. — S. KUSAKABE.

g =979-8

/ =10-87

whence 7ii = oô0"o

7/2=0-2
n =350-5
The relation between the number of vibrations n and the
modulus of elasticity E of the specimen is given by the well
known formula

k .IE

where k is the radius of gyration and /■' is the specific density of
the specimen whose length is L ; while m is a constant satisfying
the equation

cos m cosh m+ 1=0.
The smallest root, 1*875, corresponds to the gravest mode of
vibration. The first over-tone is ca. 2*6 octaves higher than the
gravest tone, so that in an actual case the succeeding roots of
the equation have no importance.

Now, calculating the modulus of elasticity for the specimen
of soft iron from the above data, we have

^= 2-029 xlO^' c.g.s. unit.

In the above experiment, the direction of the motion of the
specimen was perpendicular to the elongation of the string. It
is well knowai that, if the direction of motion is parallel to the
string, the period of the vibration of the string becomes double
that of the specimen. That is to say, the frequency n of the
specimen is equal to twice the frequency {nx + n-ij of the string.
The second observation with another string and smaller tension
was made under the last mode of vibration, giving the following
result : —

KINETIC MEASUREMENT OF THE ELASTICITY OF EOCKS. 11

Tr=ll-0B5
w =5-67x10-'
/ =12-47
whence 7ii=175'o

??2=--0-2
n =351-0
^ = 2-034x10'- c.g.s. unit.

IV. PRELIMINARY EXPERIMENTS AND THE PROCESS
OF CALCULATION.

Although the above testings gave satisfactory result, preli-
minary experiments with sandstone aud tuff were so very ambiguous
as to wholly confound the observer. At first sight, there seems
to be no definite length with which the string may vibrate witli
maximum amplitude. In the case of sandstone, e.g., lengths
corresponding to maximum amplitude were as follows : —

(1^ 9-28

(1) 9-93

'1)10-64

(1)11-42

(1)12-34

(1)13-40

(1)14-63

(1)16-23

(2'17-39

(2a8-41

12)19-69

(2)21-04

(2)22-61

(2)24-49

(3)26-06

(2)26-70

(3)27-58

(1)29-43*

(3)31-53

(2)32-53

(3)33-89

(-')34-62

(1)30-76*

etc. etc.

The small numbers in brackets are the number of loops contained
in the segment. Those marked with ^r correspond to a peculiar
mode of vibration.

It may be argued prima facie that the elasticity of such
loose materials as rocks which compose the earthcrust is not
unique and therefore, as the facts prove, that the velocity of the
propagation of seismic waves is diverted between wide limits. If
we assume that all these lengths equally correspond to the proper
vibrations of the specimen, then the velocity of longitudinal waves

12 ART. 10.— S. KUSAKABE.

should be proportionally many-valued. In the case of sandstone
No. 84, e.g. we have,

I = 13-48 15-33 16-36 17-20 19-25 21-68 24-67 28-72 34-82 42-89
whence V 00 12-9 11-3 10-6 lO'l 9-0 8-0 7.0 6-0 5-0 4-0

In Professor F. Omori's papers relating to seismometory we fre-
quently find what correspond to the above, calculated as the
velocities of seismic waves in their successive phases.

Repeated experiments, however, showed that this confusion
was an effect of the tapping by the hammer, so that varying the
period of the impressed force we might obtain another series of
maximum values. Although the vibration is really of a free
nature, it is rendered intermittent by the periodic interposition
of an obstacle, so that a very dillerent result is arrived at. In
this case, a vibration of a frequency n varies in its amplitude
with a frequency m, which last is the frequency of the hammer.
The amplitude increases very suddenly and it is always positive
so that the motion may be assumed, though by a very rough
approximation, to be represented by the expression,

•j=a

Y=Ao cos 27r 7it+ 1' Av [cos 2;: {n+2o m) + cos 2;r (n~2u m)].

It is obvious that, in such a case as the above, the amplitude
takes its maximum value when the length of the strings corresponds
to any oiie of the numerous component vibrations. The relative
magnitudes of the several maximum amplitudes differ very much
from one another and in such a way that the greatest maximum
corresponds to the vibration in the natural period of the specimen,
and the smaller the amplitude the more it is affected by the
impressed force.

When m is not too small to be compared with n, each
maximum may be distinctly observed ; but their consécutives
more and more approach each other as the ratio m/n becomes

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS. 13

smaller and smaller. In the case Avliere the ratio is one-fiftieth
or smaller than that, the series of maxima become approximately
continuous and practically constitute one maximum with a small
gradient, as in the case of soft iron.

Again, varying the period of the impressed force from m to
m', all the lengths corresponding to maximum amplitudes, except
that which corresponds to the uatili-al period of the specimen,
are changed. When these two facts are taken into account, it is
not a difficult matter to determine the natural period of vibration
for any specimen at hand.

Let U and /p be the lengths of string, vibrating with one
loop, corresponding to the number of vibrations {71 + '2am) and
{7i + 2ßm) respectively, then for a certain length /, which is a
common multiple of Ix and l[-i, the string vibrates in a peculiar
manner apparently with one loop, nodes of one mode of vibra-
tion being over-lapped by loops of another mode and vice
versa.

For example, in the case of sandstone, the result of one
experiment with a brass hammer showed that the string might
vibrate with a maximum amj)litude corresponding to any one of
the number of vibrations

• 410-0±25'7 u
where u is an integer, so that we have

^1=410-0

2m = 20-7.
Another experiment with a lead hammer, in which the moment
of inertia was increased, gave a different result, as follows : —

7i= 409-3

2m' =15-3.

14 ART. 10. S. KUSAKABE.

These two values for the frequency in free vibration are equal to
€ach other within the error of observation.

In the above observations, the tension of the string remained
constant and equal to

TF= 30-718 grams.

To test whether the error of observation is affected by the varia-
tion of the constant tension or not, two other observations were
made on the same specimen, giving the result ;

TF=30-718, n = i09'7

TF=20-908, 71=411-4

F=ll-065, n =409-1

Now, taking the mean of the above values, the data required
to calculate the modulus of elasticity of the sandstone are

L=9-9
^>=1-16
P =2-25
n =410-1
whence -ÈV=10-86x 10^'^ c.g.s. unit.

It may be here noted that tlie value determined by the
static method is nearly half of the above, i.e.

^s =5-7 X 10'*^ c.g.s. unit.
So far as the principle is concerned, the present method is
very simple, and it may be easily understood at a glance. The
mode of observation and the process of calculation, however, are
so tedious and complicated that the author considers it in no
case superfluous to recapitulate them.

With a known tension ]F= TFi and a linear density tv = iVi
of the string and a certain frequency of the hammer m = nii, we
proceed as follows : —

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS. 15

XNode.

X,,^^

1st

•Un\

,.ll,

,,tli

Obs. \^

1st

Ml

hi

lu

—

h'j

2ml

/l>i

l-2->

kr

—

l'2rj

o"l

/31

hi

h.

—

hrj

Mean.

l\

I'o

1'r

—

If
' 'J

I'r/r

h

h

Ir

—

'7

Now, a and I, being two unknown quantities, we have the relation

7 Ci 7

L+ — =1-

where r = l,2, q.

By the method of least squares, 4 may be easily found and
from it the frequency of the string ria may be calculated by the
formula given in the third chapter.

Wi 712 fi^ — n-j — 71, _i til,

which are connected by the relation

?^ + 2??^l'J = /^u where ') = 1,2, t.

From these values, the frequency of the free vibration of the
specimen may be calculated by the method of least squares, n
and m^ being two unknown quantities. Thus we have the first
value for the frequency, i.e.

n = n.
Now, varying the tension, linear density of string and period
of the hammer, i.e.

we proceed as in the former case and obtain a second value

n=n .
A third series of observations, with TF= TT^, w = Wo and m=m^„
gives a third value

16 ART, 10. S. KUSAKABE.

These three values of the frequency, i.e. n ^ n" and n" are
nearly equal to one another provided there be no blunder or
mistake in the observation or in the redaction of the result.
Taking the mean of these three values as the frequency of the
free vibration of the specimen, with a small correction due to
the stiffness of the string, the modulus of elasticity is calculated
by the formula given in the third chapter.

V. EXPERIMENTAL RESULTS.

For the complete discussion of the elastic nature of rocks,
as many different elastic constants as the number of symmetry
planes, which can be drawn in the rock, must be determined.
As we have, however, no simple means of examining these sym-
metry planes, a single modulus of elasticity was determined
relating to two mutually perpendicuhir directions, on the supposi-
tion that the material was isotropic.

The above enormously complicated method was applied to
one hundred and fifty eight specimens of different kinds of
several ages, and cost the author immense labor during one
complete year. For all this labor and trouble expended to obtain
but a poor result, he is consoled with Boyle's thought that,
" men are oftentimes obliged to suffer as much wet and cold
and dive as deep to fetch up sponges as to fetch up pearls."

The table at the end of the text contains the results arranged
in the order of geological age ; for the same geological age, those
with the larger modulus come before those with the smaller.
The velocity for longitudinal waves, calculated by the formula

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS. 17

is also given in the table. The actual velocities of longitudinal
waves in various rocks may differ from those given ; the table,
however, will probably furnish a rough estimate which may be
of some use in seismometry.

Expressing the elastic constant of the rocks, classified accord-
ing to the age of formation, by means of " the height from a fixed
base line," Fig. I, we find a distinct gradation as we pass from the
rocks of the Archaean age to those of the Cainozoic. Some of
recent age may, of course, have a greater modulus of elasticity
than those belonging to the older periods. As a whole, however,
Archfean rocks come in first of all, while Cainozoic rocks come
in last. The greatest and the least of the Archaean group are
greater than those of the Palaeozoic group respectively, and so on
in turn for other periods. In the mean, the greater part of
Archœan rocks have a greater modulus of elasticity than the
greater part of Pakçozoic rocks and so on.

The modulus of elasticity for a given rock may vary within
wide limits as the density and other physical properties differ
for each specimen. In the case of granite, e.g., No. 59i has
the greatest value ^=5'93xlO" while No. 71"^ the least value
E=l'25xlO^\ c.g.s. unit. For engineers, it would be well to
remember that in the case of any rock with unusually great
density it does not necessarily follow that the modulus of elasticity
is correspondingly great. For example, a piece of granite No. 63i
has a density twelve percent greater than that of the other piece
of granite No. 9i, while the latter has its modulus of elasticity,
on the contrary, forty-eight percent greater than the former. The
possession of a greater modulus of elasticity and yet less density
is what makes a material the more valuable.

As a general rule, however, rocks of recent formation have

18 AKT. 10. — S. KUSAKABE.

a smaller modulus of elasticity and, at the same time, less den-
sity than those of older periods. The modulus of elasticity of
old rocks increases very rapidly, more rapidly, indeed, than is
proportional, as the density increases slowly.

The velocity of propagation for longitudinal waves as shown
in the table, also increases with the age of formation of the rock
in question. It may here be noted that, in so far as the present
experiments go, the curve expressing the relation between the
density and the velocity is somewhat concave towards the posi-
tive part of the axis along which the velocity is measured. That
is to say, the increase of velocity is more rapid than the corres-
ponding increase of density as the age of the formation becomes
older.

VI EFFECT OF MOISTURE.

In the present case, it being impossible to give specimens
any desired amount of moisture, it is only intended to test
whether the modulus of elasticity is or is not largely affected by
the amount of moisture which might be 2:)resent within the
specimen.

To begin with, a specimen of sandstone was clamped in the
usual way, and the wire was stretched with a known tension.
When the specimen was in the ordinary dry state, it was found
that the wire vibrated most violently when the half wave length
was

/= 10-39 cm.
Then the specimen was wrapped in a wet cloth in its clamped
state and fed with a constant supply of water dropped upon it

KINETIC MEASUREMENT OF THE ELASTICITY OF KOCKS. 19

for forty-two hours to moisten it throughout. It was then found
that the corresponding half wave length had increased to about
double, i.e.

1=20-25 cm.
which indicated that the modulus of elasticity had decreased to
about one-fourth of its original value ! Not without some doubt
as to the result the author waited for one complete day till the
specimen had apparently become dry, when it was found that
it had nearly returned to its original state of elasticity ; i.e.

1=10-65 cm.
On heating the specimen by hot air to drive out all the moisture,
and then rapidly cooling it to the ordinary temperature, its
elasticity increased slightly ; i.e.

/=9'89 cm.
After three hours, it having returned to its ordinary state
mth regard to temperature and moisture, the elasticity became
weakened to

l=10-ß2 cm.
The result of such cyclical observations shows clearly that
the enormous diminution of the modulus of elasticity is actually
caused by the effect of moisture. The difference between the
initial /=10'o9 and the final l=10'62 may be due to some varia-
tions of surrounding conditions, — probably a little relaxation of
the clamp. As the first clamp is relaxed it results in an increase
of the effective length of the specimen. The last difference
corresponds to an increase of one mm. (or a little more) of the
effective length.

The following result of the experiment may serve to give a
rough notion as to the effect of moisture.

20

ART. 10. — S. KUSAKABE.

Eock.

Density.

Elasticity X 11 -10

Dry.

Wet.

Dry.

Wet.

Sandstone.

Sedimentary, Mesozoic.

2-230

2-3.51

10-15

3-17

Mica sell ist.

Metaraorphic, Palaeozoic.

2-647

2-669

17-54

9-21

Serpentine.

Eruptive, Archœan.

2-708

2-711

74-55

6286

A glance at the above table will show what an important effect
on the elastic nature of the earth crust is produced by the drops
of water, which permeate beneath the earth's surface.

To study more closely the effect of moisture, and especially
its effect as combined with high temperatures, a special arrange-
ment is now in preparation, the author hopes to be able to give
some observations on that special subject in the near future.

VII. RELATION BETWEEN THE KINETIC AND
STATIC VALUES.

Of the one hundred and fifty eight specimens given in the
table, a greater part have their elastic constants determined accord-
ing to the statical method by either Professor H. Nagaoka or
by the author himself. Now it is of no small interest to compare
the kinetic and the static values for one and the same specimen
of rocks.

A short remark may here be added with respect to a new
system of co-ordinates specially fitted for percentage representation.

KINETIC MEASUREMENT OF THE ELASTICITY OF EOCKS. 21

Take a rightangled-isoscelestriaiigle ABC with its vertex C below
the horizontal base AB. Divide each of the three sides into one
hundred equal segments, and also in the produced part BD mark
off any number of segments equal to one of the last. Join the
vertex A with each of the points of section on the side BG and
its produced part.

Let a and ß be the corresponding values of any physical
quantity in two different conditions 21 and ^. From A measure
a length Ax equal to a, the length of one segment being
taken as the unit of measurement, on the side AC or its pro-
duced part. Then, from the end point x measure again a length
xy equal to ß in the direction parallel to the other side CB.
If the last point y falls on the n^^ of the previously drawn lines
numbered from the base AB, then the physical quantity is said
to have increased or decreased n percent wdiile the condition
varied from 21 to 33, according as the line is above or below the
base.

These premised, we have a new^ system of co-ordinates, in
which the radius vector and the percentage are two independent

22 ART. 10. — S. KUSAKABE.

variables. It resembles a system of polar co-ordinates, but one
variable ' percentage ' is not proportional to an angle measured
from any fixed line. If a point p (f>i, n^ is given, then describe
a rightangled-isosceles triangle a2')h, with the given point as the
vertex and its base ab coinciding with AB. The two segments
Aa and Ab represent the corresponding values in the two dif-
ferent conditions referred to, provided one of the last segments is
taken as the unit of measurement.

In Fig. 2, the modulus of elasticity is represented in this
co-ordinate system, its static and kinetic values corresponding to
a and ß respectively. It will be seen, at a glance, that when
the radius vector is relatively small almost every point lies on the
positive side, while a majority of those in which the radius vector
is somewhat greater lie on the negative side. Generally speaking,
so far as is shown by the present experiment, the percentage is
enormously large for a small radius vector, but diminishes as the
radius vector increases and ultimately it becomes even negative.

That the percentage diminishes as the radius vector increases
is a matter of course, in so far as the phenomenon of yielding is
the principal cause which makes the two values differ from each
other. A negative value, however, can never be expected unless
some other cause or causes exist beside the phenomenon of
yielding.

Examining more closely, we see that the percentage rapidly
diminishes and becomes even negative especially for those rocks
which have a foliated structure as in the case of schists and slates.
In schistose rocks, the percentage is generally negative as they
have large moduli of elasticity and, at the same time, a distinctly
foliated structure. Any eruptive rock, of whatever age it may be,
has a positive percentage. Of sedimentary rocks, those in which

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS. 23

the moduli of elasticity are comparatively small Inve also a posi-
tive percentage, but some of them having large moduli may have
negative percentages. The following table vvill show the facts
more clearly.

Eock.

Total no. of
specimens.

Mean value of
tlie peicentage.

Mean value of
ihe modulus
of elasticity.

Sandstone.

G

+,1S

l-r)7xlO"

Tuff.

7

+2;',

1-92

Andésite.

i;:;

+ 20

2-12

Granite.

5

+ 11

;M4

Slate.

17

- G

4-()9

Peridotite-serpentine.

G

+ 4

G-29

Schists.

10

-10

G-;i9

Here it must be noted that I do not mean to say that the
percentage would be always negative for any old rock having a
distinctly foliated structure. That the percentage diminishes as
the radius vector increases would in all probability be true ; its
becoming negative, however, may be due to some other causes.
The two series of experiments have been made under very different
surrounding conditions. The effect of moisture, for instance, is
never negligible as some observations were made during the wet
summer season ; while the others, during the dry cold of winter.

Any complete discussion of the relation between the static
and the kinetic values must be postponed, at least, until the
effects of temperature and moisture on the elastic constants have
been clearlv investi 2;ated. The numerical values hitherto given
for the moduli of elasticity of several rocks must be taken only
as rouo'li estimates.

24 ART. 10. — S. KüSAKABE.

The only closing remark that can be properly added is that
the result of these experiments as a whole has only brought the
author, who at first sight thought he had found a quarry of
research that would soon be exhausted, to the threshold of a
labyrinth, where many paths invites him to proceed further.

In conclusion, I wish to express my great indebtedness to
Dr. N. FuKUCHi of the Geological Laboratory, for valuable in-
formation concerning the geological characters of the specimens.
My best thanks are due to Professor H. Nagaoka, under whose
kind guidance I have carried out the experiments.
June 190Ô.

Physical Laboratory, College of Science,

Tokyo Imperial University.

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS. 25

Modulus of Elasticity for 158 Specimens of Rocks.

.5 o

Rock.

Locality.

Kind.

ft

^[odulus of
Elasticity.

l'À

Bi

E

o a >
■2. 2. 5

ARCH^^N ROCKS.

c.g.s.XlOii

50.*
9.*

8i.
46i.

;î1i.

ISi.
16.*
22i.

2G2.

Dil.
4Ô1.

2Gi.

28.

41a.*

24i.

49i.

4U.*

(jOi.

IT.*

47i.
2?n.

;50.

Chlorite schist.

Peridotite ")
(Serpentinized).j
Serpentine.
Quartz schist.

Quartz schist.

Ciilorite schist.

Peridotite )

serpentine.)

Gabbro.

Chlorite schist.

Chlorite schist.

Graphite schist.
Peridotite )
serpentine. \

Graphite schist.

Chlorite schist.
Peridotite )
serpentine.)

Gneiss.
Peridotite )
serpentine.)

Quartz schist.

Graphite schist.
Chlorite schist)
(Decomposed). J

Chichibu.

Kuji, Ibaraki.

Chichibu.
Kashiwagi, Gumma

Onishi, Gumma.
Nogami, Chichibu.

Ibaraki,

Nogami, Chichibu,
Kunikami,»
Chichibu. j"

Chichibu.

Nogami, Chichibu.

Kunikami, (

Chichibu. )

Onishi, Gumma.

Kuji, Ibaraki.

Kunikami,)

Chichibu. )

Nogami, Cliichibu.

Kuji, Ibaraki.

Shinshiro, Mikawa.

Kuji, Ibaraki.

Onishi, Gumma.
Kunikami, )
Chichibu. )

Onishi, Gumma.

Metamorphic.

Eruptive )
(Altered).)

»
Metamorphic.

2-96

2-98

2-61

2-72
2-(54

9-79
10-69

7-72

9 05
599

12-39
10-88

9-74

6-98
8-91

11-09
10-78

8-73

8-02
7-45

Metamorphic.

2-f>7
2-88

7-66
6-59

7-19
7-19

7-43
6-89

Eruptive.

2.83

6-55

7-07

6-81

i>

2-71

5-95

7-26

661

Metamorphic.

2-87

5-21

6-35

5-78

Metamorphic.

2-88
2-77

5-78
5-46

5-37
5-54

5-58
5-50

»

2-82

5-55

5-35t

5-45

»»

2-59

5-39

5-07

5-23

Eruptive.

2-78

5-13

4-9G

5-04

Metamorphic.

2-59

4-88

5-15

5-02

)>

2-77

4-84

5-00

4-92

Eruptive.

2-79

4-52

4-86

4-69

Metamorphic.

2-59

4-02

4-98

4-50

Eruptive.

2-57

4-21

4-67

4-44

Metamorphic.

2-G3

4-82

3-34

4-08

»

2-5G

4-05

4-02

4-04

»

2-64

3-9G

4-11

4-03

Km.

Sec.

6-12
6-02

5-78

5-43
5-31

5-28
4-89

4-91

4-94

4-49

4-41
4-46

4-39

4-50

4-27

4-41

4-22
4-10
4-17
4-16

3-94
3-98
3-91

PLAiEOZOIC ROCKS.

67i.

Schalsteine.

Asakuma-dake, Ise.

Sedimentary.

2-98

9-35

9-90

9-62

5-68

122.

Clayslate.

Miyanomaye, Aumi.

I)

2-71

9 26

8-84

9-05

5-78

69i.

Pyroxenite.

Isosuzu-gawa, Ise.

Metamorphic.

3-05

8-85

8-87

8-86

5-40

78,,.*

Schalsteine.

China.

Sedimentary.

2-77

8-01

8-60

8-30

5-48

78„.*

jj

»

)>

2-77

8-34

7-87

811

5-41

26

ART. 10. — S. KUSAKABE.

{Continued.)

o

>-.

Modulus

of

i^-ti

•.§0

œ

Rock.

Locality.

Kind.

a

Elasticity.

•« *tc aj

Fa

El

E

G5i.

Schalsteine.

Kamuro, Kii.

Sedimentary.

2-84

8-32

7-80

8-06

5-33

■71i.

Limestone.

Akasaka, Gifu.

)>

2-71

7-82

7-54

7-68

5-33

32i.

Pyroxenite.

Hominoyama, 1
Gumma. J

»

2-90

7-81

7-52

7-67

5-15

2i.*

Clayslate.

V

,,

2-69

7-61

7-47

7-54

5-30

34i.

Adinolslate.

Yonowo, Gumma.

)j

2-64

7-54

7-45

7-50

5-33

13*

Linaestone.

Kanagawa.

Sedimentary.

2-05

7-43

7-52

7-48

5-31

2„.*

Clayslate.

?

,,

2-67

7-77

7-13

7*45

5-28

29.*

Limestone.

Mikawa.

3,

2-68

6-68

7-n

6-89

5-07

79>

Schalsteine.

RikuchO.

2-65

6-75

6-70

6-72

5-04

74,.

Limestone.

Akasaka, Gifu.

J)

2-69

6-60

6-65

6-65

4-97

55*

Limestone.

Mnsashi.

Sedimentary.

2-63

6-55

6-62

6-58

5-01

7l2.

))

Akasaka, Gifu.

jj

2-69

6-67

6-30

6-48

491

3U.*

Marble.

Ibaraki.

}>

2-63

6-26

6-14t

6-20

4-86

73*

Sandyslate.

Rikuchü.

jî

2-64

6-13

6-05

6-09

4-81

SO.*

Schalsteine.

»

»

2-82

5-77

6-23

6-00

4-62

72i.

Limestone.

Akasaka, Gifu.

Sedimentary.

2-69

5-87

6-13

6-00

4-72

59,.

Granite.

Okazaki, Mikawa.

Eruptive.

2-63

5-92

594

5-93

4-75

74.*

Clayslate.

îvikkô.

Sedimentary.

2-15

5-69

5-72

5-70

5-15

la.*

Weathered )
clayslate. j

Umebatake.

>>

2-31

5-13

5-93

5 "53

4-89

87i.

Ophicalcite.

Arakawa, Chichibu.

Eruptive.

2-65

4-91

5-30

5-11

4-39

21,.

Ivimestone.

Nogami, Chichibu.

Sedimentary.

2-64

5-31

4-76

5-04

4-37

3 6.*

Weathered 1
cliyslale. l

Umebatake.

5>

2-30

4-97

4-58

4-77

4-.J5

9i.

Granite.

Mikage, Settsu.

Eruptive.

2-54

4-88

4-61

4-75

4*;)2

76,.

Slate.

Atago, Yamashiro.

Sedimentary.

2-24

4-61

4-83

4-72

4-59

40,.

Micaschist.

Fudö-tüge, Ibaraki.

Matemorphic.

2-54

4.25

4-81

4.53

4.22

69*.

Granite.

Kagawa.

Eruptive.

2-57

4-53

4-42

4-47

4-17

81i.

»

Tamba.

2-62

4-46

4-46

4-46

4-13

(34,.

»

Nishiura, Mikawa.

5)

2-61

4-53

430

4-42

4-1 1

m-z.

>»

Shirakawa, Kyoto.

2-62

4-28

4-01

4-14

3-98

79i.

Slate.

Karutaki, Kyüto.

Sedimentary.

2-45

3-90

4-26

4-08

4-09

75i.

Limestone.

Akasaka, Gifu.

Sedimentary.

2-72

3-55

4-o9t

4-07

3-87

66i.

Granite.

Shirakawa, Kyoto.

Eruptive.

2-62

4-03

—

—

3-93

«3.

Marble.

Maiyama, Ibaraki.

Seiliiiieiitaryl

(Metaiiior- \

lilioscd.) )

2-68

3-48

3-89

3-69

3i.*

Clayslate.

Tamba.

,,

2-39

3-63

3-90

60i.*

Weathered)
clayslate. J

Yamashiro.

»

2-31

365

3-49

3-57

3-94

KINETIC MEASUREMENT OF THE ELASTICITY OF ROCKS.

27

{Continued.)

c
Jo

Keck.

Locality.

Kind.

1

Modulus of
Elasticity.

Hi

El

I'h

E

44].
602.

12a.*
IO2.

63i.

Granite.

^larble.
Grauite.

Hyögo, Sett su.
NislHura, Mikawa.

Ibaraki.
Kitaki, Kagawa.
Nishiura, Mikawa.

Eruptive.

Sedimentary.
Eruptive.

2-.59

2-84
2-65
2-.57

2-84

3-49

3-.35

.•'.-35t

3-24

.••.•40

3-33
.3-37

3-02t

3-41
3-36

3-21

3-6.3
3-44
3-56
3-.55
3-36

60„.*

Weathered)
clayslate. J

Yamashiro.

Sedimentary.

2-32

3-66

2-72t

3-19

3-72

.-la.*

782.

78i.

Clayslate.

Slate.

Marble.

Slate.

Tamba.
Narutaki, Kyoto.

Ibaraki.
Narutaki, Kyoto.

!)
J»
»

2-.37
2-42
2-6.5
2-44

.3-17
3-13

.•V68t
3-08

.3-16

2-60t

.3-14
314

3-65
3-59
3-44
3-55

8O1.

Slate.

Atago, Kyoto.

Sedimentary.

2-3G

2-99

300

3-00

3-56

27i.

Micaschist.

Motoizumi,)
Chichibu. J

Metamorphic.

2-63

2-71

315

2-93

3-35

29,.

64a.*

52i.

Limestone.

Clayslate.

Contact slate.

Onislii, Gumma.
Tocliigi.
Tamba.

Sedimentary.

3>

2-66
2-46
2-33

2-40
2-63
2-75

2-85
2-46
2-23

2-63
2-54
2-49

3-15
3-21
3-27

68.*
52*

Granite.

Kitaki, Kagawa.
Ibaraki.

Eruptive.

2-5.5
2-50

2-19
l-80t

l-67t

1-98

2-79

2-68

7a*

Weathered)
clavslate. )

Î)

Sedimentary.

2-50

1-53

1-.58

1-56

2-50

<3.5a.*

7i.*

3)

2-49
2-50

1-54
1-51

1-53

1-44

1-.53
1-47

2-48
2-43

6.5i.*

Weathered)
clayslate. J

Ibaraki.

Sedimentary.

2-50

1-38

1-37

1-37

2-.34

56.*
71.*
42,.
61.

Granite.

Mica schist.
Pegmatite.

))

Tsukioka, Ibaraki.

Hojô, Ibaraki.

Eruptive.

)>

Metamorphic.

Eruptive.

2-53
2-59
2-64
2-57

1-36
1-28
1-08
1-16

1-20
1-23
1-29
1-10

1-28
1-25
1-19
1-13

2-25
2-20
2-12
2-10

MESOZOIC ROCKS.

53.*

Clayslate.

Rikuzen.

Sedimentary.

2-70

7-80

7-.39

7-59

5-.30

77.*

Schalsteine.

Akamagaseki.

2-78

6-46

6-52

6-49

4-85

72.*

Clavslate.

Rikuzen.

2-71

6-.54

6-38

6-46

4-89

76.*

9

2-71

6-42

6-38

6-40

4-86

62a.*

»

Tsushima.

ji

2-68

4-39

4-25

4-32

4-02

G8,.

Granite.

Yoshima, Kagawa.

Eruptive.

2-61

3-42

3-40

3-41

3-62

5.*

Sandstone.

Kii.

Sedimentary.

2-22

1-32

1-50

1-41

2-52

85i.

Kvushu.

2-34

1-29

1-30

1--30

2-36

6i.*

Kii.

2-22

1-23

1-00

1-11

2-23

83,.

J'

Kyushu.

))

2-09

1-03

1-07

1-05

2-24

28

ART. 10. — S. KUSAKABE.

(Continued.)

.= O

Rock.

Locality.

Kind.

'ai

Modulus of
p]lasticity.

ocity
ongit.
ave.

Ev \ 32

E

'^ o

84,.

Sandstone.

Kyüshü.
Kii.

Sedimentary.

2-24
2-24

1-02
1-02

1-05
0-95t

1-03
0-98

2-15
2-09

CAINOZOIC ROCKS.

51i.

172.

li.
54.*

Andésite.

Shinano.

Haruna, Gumma.

Nebukawa, Sagami.

Eruptive.
»

2-61
2-63
2-59
2-56

4-37
4-17
.".•96
4-12

4^35
3^94

3^53

4^36
4^06

3^83

4-09
'3-94
3^91

3^87

35i.

Sandstone.

Mitagawa,)
Chichibu. }

Sedimentary.

2-47

3-65

3-41

3-53

3-78

l2.

15*.
52i.
57i.
51.*

Andésite.

>)

Rhyolite.

Porphylite.

Rhyolite.

Nebukawa, Sagami.

Komatsu, Sagami.

Hôraiji, Mikawa.

Arumi, Mikawa.

Honbun, Izu.

Eruptive.

))
))

»

2-59
2-20
2-20
2-29
2-32

3^45
3^32
3^20
3-31
3-12

3^59
3-10
3-08
2^94
3-10

3-52
3^21
3-14
.3^12
3^11

3-69
3-82
3-80
3-70
3-66

50.

Andésite.

Nakatsugaru, )
Mutsu. J

Eruptive.

2-69

2^79

2^90

2-85

3-26

14.*

28.*
30.*

8a.*

Tuff.

Andésite.

Tuff.

Rhyolite tuff.

Xanasawa, Sagami.

Izu.

Kiga, Izu.

lyo.

Sedimentary.

Eruptive.

Sedimentary.

J»

2-22
2-17
2-17
2-35

2-91
2^49
2^51
2-48

2^66
2^87
2-64
2-30

2^79
2-68
2^57
2^42

3-55
3-52
3^45
3-21

.^Î9.*
10.*
70.*
58i.
84.*

Andésite.

Tuff.

Andésite.

Sandstone.

Rhyolite tuff.

Yokone, Izu.

Yoshida, Izu.

Komatsu, Sagami.

Yebi, Mikawa.

lyo.

Eruptive.
Sedimentary.

Eruptive.
Sedimentary.

2-40
2-28
2-46
2-21
2-32

2-39
2-29
2-40
2-33
2-22

2-35
2-39
2^23
2-27
238

2-37
2-34
2-31
2-30
2^30

.3^14
3^21
3-07
3-23
3^15

19i.*
56i.

Tufaceus )
sandstone.)

Rhyolite tuff.

Tomioka, Gumma.

Yebi, Mikawa.
Nakura, Mikawa.

Sedimentary.

2-.32

2-12
2-15

2^15

2-17
2-15

2-31

2^12
2-11

2^23

2^1ö
2^13

3-10

3-19
3^14

19«.*
59i.*

Tufaceus 1

sandstone.)

Rhyolite.

Tomioka, Gumma.
Gumma.

Eruptive.

2-31
2-4.3

2^08
1-94

2-01

l-93t

2-04
1-94

2^97
3-82

5:5i.

•ta.

59„.*

r.7„.*

57.*

Rhyolite tuff)
(Weathered)./

Tuff

Rhyolite.

Andésite tuff.

Tuff

Hôraiji, Mikawa.

S.awada, Izu.

Gumma.

Midera, Yechizen.

Nawazi, Izu.

Sedimentary.

»

Eruptive.

Sedimentary.

2-16

1-84
2-47
2-44
2-04

1-78

1-78
1-88
1-74
1-G7

1-87

r84t

1-64
1-73

r70t

1-83

1-81
1-76
1-73
1^69

2-91

3-14
2-67
2^66

2^88

KINETIC MEASUREMENT OF THE ELASTICITY OF KOCKS.

29

{Continued.)

0/

s^

Modulus (

if

i^-^

.So

02

Rock.

Locality.

Kind.

S

E

asticit\

^ o -

Ex

Ei

E

676*

Andésite tuft'.

Midera, Yeehizen.

Sedimentary.

2-40

1-03

1-72

1-68

2-05

54,.

Tufaceus |
sandstone, j

Xebukawa, Sagami.

>)

2-07

1-.55

1-03

1-59

2-77

40,.*

Andésite.

Manazuru, Sagami.

Eruptive.

2-55

1-56

1-55

1-55

2-47

2,.

Tuff.

Kawatsu, Izu.

Sedimentary.

1-91

—

1-50

—

2-80

40.*

Andésite.

Sagami.

Eruptive.

2-30

1-54

1-45

1-49

2-54

OU.*

Rhyolite tuff.

Iwashiro.

Sedimentary.

2-20

1-46

1-48

1-47

2-59

;jo.*

Tuff.

Izu.

ÎÏ

1-8Ô

1-05

1-27

1-40

2-81

49*.*

Andésite.

)'

Eruptive.

2-21

1-40

1-39

1-42

2-54

4ô.*

Tuff:

Kawatsu, Izu.

Sedimentary.

1-82

1-45

1-39

1-42

2-79

55i.

Tufaceus )
sandstone.)

Hôraiji, Mikawa.

)>

2-09

1-44

1'35

1-39

2-58

47.*

Andésite.

Yenoura, Shiruga.

Eruptive.

2-43

1-30

1-35

1-.35

2-30

01,,.*

Rhyolite tuff.

Iwashiro.

Sedimentary.

2-23

1-37

1-30

1-34

2-45

22.*

^ Tuff

Teishi, Izu.

}}

2-01

1-30

1-35

1-32

2-50

58.*

Sandstone.

Ciiöshi, Chiba.

ÏÏ

2-35

1-35

1-20

1-30

2-35

03.

Rliyolite.

Midera, Yecliizen.

Eruptive.

2-40

1-09

1-28

1-18

2-22

10.

Andésite.

Haruna, Gumma.

Eruptive.

2-32

1-21

1-14

1-18

2-26

;!2.*

Rliyolite tuff.

iMitaka, Izu.

Sedimentary.

1-89

1-21

1-02

rii

2-42

87*

Quartz sandstone

Hizen.

))

2-2;5

1-02

1-21

1-11

2-23

38.*

Andésite.

Izu.

Eruptive.

1-94

105

1-17

1-11

2-.39

43.

Sandstone.

Shinjô, Kii.

Sedimentary.

2-25

1-04

1-13

1-08

2-19

23.*

Andésite tuff:

Y'ema, Izu.

Sedimentary.

1-83

1-00

0-98

0-99

2-33

34.*

Andésite.

Manazuru, Sagami.

Eruptive.

2-U2

0-90

0-95

0-95

2-17

00.*

Rhvolite tuff.

Amaknsa, Hizen.

Sedimentary.

2-26

0-96t

0-02t

0-79

1-87

31.*

■ Tuff.

Teishi, Izu.

}f

1-92

0-78

0-77

0-78

2-02

48.*

Andésite

Izu.

Eruptive.

2-10

003

—

1-73

n>.

Sandstone.

Chöshi, Chiba.

Sedimentary.

2-20

0-50

0-62

0-59

ro3

21.*

Andésite tuff.

Izu.

M

1-.50

0-42

0-50

0-46

1-70

35.*

Tuff

)!

))

1-29

0-20t

—

—

1-43

N.B. The specimens marked with * are those whose elastic constants were statically deter-
mined by Professor H. Naoaoka and pnblislied in "The Pub. of the E.I.C. in F.L."
No. 4. 1900. and Phil. Mag. 1900.

The specimens marked with f aie tliose which were found to have been crushed or
broken at tiie clamped section so that the value given in tiie table may be erroneous.

Jour. Sei. Coll. Vol. XX. Art. X. Platt

MODULUS OF ELASTICITY.

K -12X10 Çc-S-^O

ABBffEVIAT/OJVS.

Ad.Sl. Adùnolslala.

Peu.

Pericloôit^.

An. Andésite-.

Fe.

Pegmatile'.

Ch.-S. CJilorite scAiet.

Po.

Porpfijylite.

Cl.al. ClaysliUe..

Py-

Pi/riuee/uie .

C0.SI. Coniiiet skcte.

Q.S.

Qiutrfz sc/iist.

Ufa. Gahbro.

Rh.

HhipUt<: .

iir. Granite'.

Sœ.

Siuulstorie .

Gp. Grapkile.

.Sci.SL.

ôwi^/f/ slcUe.

Gp.S. Gn^kiteiTMit.

Seh.

S'c/iaistein^.

Grv. Gneiss.

Se:

S'erperttine.

Lnv. Limestone^.

SI.

SlaU^.

Mb. Marble,.

Tw.

Tuff.

Mi.6'. Mica, schist.

Tu.Sa.

TilfaeeiM ga;i//»tone.

Op. Op/iicalci/^.

W.Cl.Sl

Weathered a. SI.

Fié. 2.

JODBNAL OF THE COLLEÖE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 11.

On Two New Species of Chimsera.

By
Shigeho Tatiaka, Rigahishi.

With hvo p/afes- and one woodcut.

There are eight sjoecies of Chwiœra known to bo extant,
viz : —

C. monstrosa Ltnn^us. GIjnther, Catalogue of fishes, Vol.

VIII., 1870, pp. 349-350.
C. affinis Capello. Günther, Catalogue of fishes, Vol.

VIII., 1870, pp. 350-351.
G. colliei Lay and Bennet. Günther, Catalogue of fishes.

Vol. VIII., 1870, p. 350.
C. ogilbyi Waite. Mem. Australian Mus., IV., Pt. I.,

1899, p. 48, PI. VI.
C. negleda Ogilby. Werner, Zool. Jahrbücher, Bd. 21,

1905, p. 270.
C. Phantasma Jordan à Snyder. Proc. Nat. Mus., Vol.

XXIII., 1903, pp. 338-339.
C. mitsuhurii Dean. Journal College Science, Vol. XIX.,

Art. 3, 1904, pp. 6-9, PI. I., figs. 1-2.
C. purpiirascens Gilbert Ms. Jordan & Snyder, Smith-
sonian Miscellaneous Collections, Vol. 45, 1904, p.

235.

2 ART. 11. — S. TANAKA.

Last winter, from the last part of January to the middle of
February, there were brought into Mr. Owston's office in Yoko-
hama twenty-three specimens of Chimœra, all taken in the Sagami
Sea, off the village of Inatori, Idzu. He generously turned them
over to me for examination. There also came about the same
time to the Zoological Institute of the Science College two speci-
mens from the Tokyo market, said to have come from the same
locality off Idzu. A careful examination of all these specimens
has convinced me that they do not belong to any of the hitherto
known eight species and that they must be referred to two new
species, for which the specific names jordani and owstoni are
proposed.

It is necessary for me to add here that I am under very
deep obligations to Prof. Dr. Mitsukuei for superintending my
study and for his careful corrections of the manuscripts in the
preparation of this article. I wish also to thank Mr. Owston
for his generosity in placing his specimens at my disposal for
examination.

CHIMiERA JORDANI n. sp. PL I., fig. 1.

The points of the species which strike us at a glance are
(1) that the lateral canal of the sensory canal system runs straight
on the side of the body almost without any sinuation, (2) that a
deep notch lies between the anal fin and the lower lobe of the
caudal, and (3) that the head and body are of a uniformly
brown color.

The following is a detailed description of the species.

Head measured from tip of snout to first gill-opening con-
tained 4f times, and greatest height of body in front of the posterior

ON TWO NEW SPECIES OF CHIMERA. 3

end of the insertion of first dorsal 51 times, in the total length of
body (excluding the parts from the end of the second dorsal to the
end of caudal filament) ; snout about 2, eye 3è, interorbital 4i-5 in
head ; height of head in front of the insertion of pectoral a little
lower than the greatest height of body ; the height in front of eye
li-li in the height of body ; spine of the first dorsal curved back-
wards, a little longer than the length of head and higher than the
first ray of the first dorsal, triangular in cross section, smooth with
a median keel anteriorly, grooved posteriorly, the distal third of its
length with recurved spinules ; this spinous roughness is weaker
than in phantasma and stronger than in mitstikurii ; first dorsal
subcontinuous with second dorsal by a fold of rayless skin ; height
of the anteriormost ray of first dorsal exclusive of its fleshy base
li in head; free margin of the first dorsal almost straight; ex-
ternal margin of the second dorsal even and slightly concave, its
posterior margin rounded ; height of the fin, measured perpen-
dicular to the upper margin of body at a point two-sevenths of
its length from the anterior end, 41 in head. Upper lobe of caudal
slightly higher than the lower ; height of the former measured
perpendicular to the upper margin of body 7 in head ; second
dorsal and upper lobe of caudal continuous with a deep notch.
Anal very low, falcate, continuous with the lower lobe of caudal,
but separated by a deep notch opposite the dorsal notch and
similar to but deeper than that of pha7itasma. Both caudal lobes
nearly as high as in mitsuhurii ; as in that species as well as
in phantasma the lower lobe extends more posteriorly than the
upper lobe. Distance from the dorsal notch to the end of caudal
filament 2 in body exclusive of head and the parts behind dorsal
notch ; in this proportion the species agrees with phantasma,
but not with mitsukurii in which the distance from the dorsal

4 ART. 11. — S. TANAKAi

notch to the end of caudal filament is equal to the length of
body.

Pectoral pointed and slightly falcate, with a deep notch at
its lower insertion in the fleshy base ; its tip reaches far behind
the insertion of ventral when depressed. Insertion of the ventral
midway between tip of head and notch of second dorsal, its tip
subtruncate, its inner margin rounded. Distance between origin of
second dorsal and insertion of pectoral a little shorter than the
distance between the latter and the insertion of ventral.

There exists no difference in external features between the male
and the female except in the sexual characters. Cephalic organ
in the male directed forwards and downwards, situated on the
snout in front of the anterior margin of the eye ; its length 2
in eye ; on its lower side about sixty or more recurved spinous
denticles, very much like those of phantasma or mitsukurii.
Clasper tripartite ; its tip covered with fine shagreen as usual ;
its length from the insertion 1 i in head ; its division at a point
two-thirds of its length, as in purpurascens and owstoni (PI. I.,
fig. 3), but unlike phantasma in which the point of division is
at one-third of the length (PI. I., fig. 4), while in mitsukurii the
clasper is rather short, bipartite, and the point of division situated
halfway of its length. In a young specimen, which measures
42.5 centimetres in total length exclusive of the parts behind
the dorsal notch and in which the cephalic organ is still em-
bedded in the skin, the claspers measure 1.6 centimetres in
total length and 1 centimetre in the length from the insertion
to the point of division ; they are already tripartite. In a
young phantasma 35 centimetres long and with the cephalic
organ still embedded in the skin, the claspers are still bipartite.
Ill all the Japanese species I have found the anterior ventral

ON TWO NEW SPECIES OF CHIM^KA. 5

clasping organ armed with five spinous denticles on the inner
side.

Teeth confluent ; five or six rods in each anterior lamina of
the upper jaw ; the margin of the lamina sinuate and slightly
convex ; lateral rods behind the anterior lamina low, oblique,
their tips directed towards the median part. Mandibular laminae
included within the upper laminse ; rods in the laminae rather
indistinct, margin of lamina with two concavities and more or less
indistinctly sinuate. Peritoneum lining the body cavity unpigment-
ed ; wall of digestive canal of a dusky color.

Lateral canal of the sensory canal system straight almost

Lateral view of the head of a C'Idmœm, explaining nomenclature of the sensory
canal system adopted. (After Mr. Collinge's nomenclature.) Ju., Jugular branch ;
L. C, Lateral canal; Mx. Mn., Maxillo-mandibular branch; Mx., Maxillary division of
the maxillo-mandibular; Mn. Mandibular division of the maxillo-maudibular; S. or.,
Suborbital branch.

without any sinuation. It shows an interesting feature in the sub-
orbital part. From the suborbital branch of the canal system

b AJIT. 11. — S. TANAKA.

below the posterior margin of the eye, there branch off a maxillo-
mandibular and a jugular branch sometimes at the same point (PL
II., fig. 15), at other times the latter from the former (PL II., fig.
12), and at still other times the two separately from the suborbital
(PL II., figs. 13 and 14). This variation may go so far that
in one and the same individual the two branches, the jugular and
the maxillo-mandibular, may arise at the same point on one side,
while on the other side, the origin of the jugular may be shifted
a greater or less distance along either the suborbital or the
maxillo-mandibular branch as mentioned above. In phantasma
and mitsukurii such variations seem never to occur. The origin
of the jugular seems to be constant for each of the species. In
Phantasma it arises always from the suborbital (PL II., fig. 6),
while in mitsukurii it is given off invariably from the maxillo-
mandibular (PL II., fig. 16). So far as I was able to examine
specimens of jordani, I came to the conclusion that in this species
the points of origin of the maxillo-mandibular and of the jugular,
whether the latter arises from the former or directly from the
suborbital, are considerably nearer each other than in phantasma
or in mitsukurii (cf. PL II., figs. 6, 12, 13, 14, and 16). In-
dividual ^differences with respect to this part of the sensory
canal system I have observed in purpurascens also (PL II., figs.
I and 5).

Color in formalin uniformly dark brown ; as Mr. Owston
tells me, a female when brought in by a fish-monger was beauti-
fully iridescent ; one indistinct pale line above the lateral canal
and three indistinct pale lines below it run parallel to one an-
other in the posterior part of the body ; posterior margin of
the pectoral and ventral fins blackish ; free margins of all other
three fins with a more or less broad blackish band ; color of second

ON TWO NEW SPECIES OF CHIMERA. 7

dorsal fading out abruptly towards its base ; numerous transverse
series of dark dots above the lateral canal on the side of body as
shown in the figure of the next species.

This species and the next are intermediate in many respects
between phantasma and mitsukurii. According to our collector,
KuMA, phantasma is taken with long line at a depth of two
hundred and fifty to three hundred hiro*^ in the fishing grounds
near Misaki, Province of Sagami, and mitsukurii at a depth of
three hundred to four hundred and fifty hiro in the same grounds.
The ground off Inatori where jordani and owstoni have been
taken is some five hundred or more hiro deep. From the ground
of capture as well as from their dusky coloration and large eyes,
the two species seem to be residents of deeper regions than either
phantasma or mitsukurii. The male specimen here figured and
a female are deposited as types in the museum of the Zoological
Institute, Tokyo Imperial University.

To sum up, the species is characterized by the following
points : — uniformly dark coloration ; large eye ; first dorsal spine
showing rather distinct roughness, the same spine projecting out
beyond the end of the anteriormost ray of first dorsal ; large
pointed pectoral ; distinct anal ; caudal filament as long as that of
phantasma ; height of caudal much greater than in phantasma ;
almost straight lateral canal.

The species is named after Dr. David Stakk Jordan, of the
Leland Stanford Junior University, in recognition of his extensive
work on fishes of Japan.

The measurements of the species are as follows :

I) One hiro is a little less than five feet.

8

ART. 11. S. TANAKA.

MEASUIi EMENTS

MALE.

Total length (exclusive of parts behind dorsal)
notch) j

Length from dorsal notch to end of filament

Length of head (to first gill opening)

Height of head behind eye

Height of head in front of eye

Greatest height of body (in front of the posterior)
end of the insertion of first dorsal) j

Long, diameter of eye

Diameter of pnpil

Long, diameter of iris

Length of snout

Interorbital width

Length of dorsal spine

Height of anteriorniost ray of first dorsal (ex-)
elusive of its fleshy base) )

Height of hijjhest part of second dorsal (measured)
perpendicular to upper margin of body) j

Height of lowest part of second dorsal

Upper margin of pectoral (excl. of its fleshy base).

Breadth of pectoral at insertion of base

External margin of ventral (exclusive of its fleshy)

base) J

Breadth of ventral at insertion of base

Height of upper lobe of caudal

Height of lower lobe of caudal

Length of cephalic organ

Length of clasper from its insertion

Length of clasper from insertion to the point)
of division J

Distance between insertion of pectoral and that)
of ventral 3

Distance between insertion of the dorsal spine)
and that of pectoral 3

Distance between base of ventral and anal notch.

G2.0

28.0
? 10.7
8.9
8.15

10.7
3.6
1.65
3.1
4.1
.3.1
? 11.6

11.4
2.3
1.5

19.0
5.0
9.0
2.6
1.25
1.25
2.05
7.0
4.65

20.0
9.1

30.0

60.0

23.5

10.6

9.0

7.5

9.2

3.3

1.7

3.0

5.7

2.7

12.1

10.5

2.3

1.85

16.0

5.0

8.1

2.2

1.1

1.0

1.7

6.0

4.1

18.0

9.0

31.5

55.5

23.5
? 9.5
8.8
7.15

10.5
3.7
1.6
2.9
4.4
2.8

13.0

10.1
2.2
1.55

17.2
4.0
7.6
2.0
1.2
1.2
1.7
6.0
3.9

18.5
8.4

26.5

42.5

25.0
9.2
6.5
5.5
7.5
2.8
1.2
2.2
4.4
2.1
8.7
6.6
2.0
1.7

11.3
3.0
5.8
1.5
1.1
1.05
?

1.6
1.0

12.0
5.8

21.0

All the measurements are

ON TWO NEW SPECIES OF CHIMERA.

OF THE SPECIES,

FEMALE.

63.5

60.0

58.5

57.1

63.0

39.0

50.5

56.5

69.0

62.5

51.0

64.0

23.0

27.0

2.55

23.0

?26.0

22.5

23.5

28.5

?20.0

?

29.0

25.0

12.8

12.4

?11.9

12.0

12.6

8.1

10.4

11.8

?12.0

?12.9

11.1

?11.5

9.3

8.85

9.0

9.9

10.3

6.0

7.7

9.0

8.8

10.35

8.3

9.3

8.4

8.35

7-7

7.85

8.4

5.35

6.6

7.5

8.3

9.15

7.25

8.45

11.3

10.8

9.9

11.4

11.3

6.6

9.1

11.0

11.0

10.7

8.5

11.2

3.7

.3.6

3.3

3.5

3.8

2.6

3.0

3.5

3.8

4.0

,3.2

4.1

1.5

1.6

1.6

1.65

1.7

1.4

1.7

1.75

1.65

1.8

1.5

1.9

2.7

2.9

2.8

2.7

3.0

1.9

2.8

3.0

2.95

3.5

2.5

3.1

6.2

5.8

5.0

6.3

5.6

4.1

5.0

4.9

5.2

5.3

6.0

5.2

2.3

2.8

2.8

2.7

2.3

1.5

2.6

2.6

2.8

2.9

2.2

2.6

12.9

?10.8

12.5

12.7

? 9.3

6.6

10.1

? 9.8

12.2

12.0

9.4

12.6

9.9

10.5

10.5

10.9

10.0

5.5

7.7

9.6

10.0

9.4

7.1

10.5

2.7

2.5

2.0

2.6

2.7

1.8

2.5

2.8

2.6

2.2

2.3

2.3

2.2

1.95

?

1.8

1.75

1.3

1.8

1.9

2.0

1.7

2.0

1.8

16.6

16.2

15.7

16.1

17.2

10.3

13.1

15.8

16.6

16.5

1.3.5

16.9

5.1

4.3

4.3

4.S

5.0

2.3

4.0

4.0

4.1

4.8

3.2

4.8

8.9

8.9

S.O

8.(i

9.2

5.4

6.7

8.3

8.6

8.5

6.2

9.5

2.8

2.6

2.3

2.3

2.6

1.4

1.8

2-2,

2.5

2.5

1.9

2.6

1.45

1.65

1.65

1.65

1.8

0.97

1.45

1.55

1.7

?

?

1.3

1.3

1.5

1.45

1.45

1.55

1.0

2.7

1.25

1.5

?

?

1.3

21.0

20.0

19.0

18.5

18.8

10.5

15.6

18.5

20.5

20.0

16.0

21 .5

8.9

9.1

8.8

8.6

9.5

5.5

7.3

8.6

8.0

9.2-

7.5

9.3

31.0

30.0

29.0

26.0

30.0

19.5

25.0

28.5

,32.0

29.0

25.0

.30.0

in centimetres.

10 AKT. 11.— S, TANAKA.

CHIMiERA OWSTONI n. sp. PI. I., figs. 2 and 3.

Of this species I have seen only two specimens, a male and
a female, in which rather a robust body, minutely lateral canal
undulating, distinct and falcate anal fin, very deep anal notch,
and brown color with more or less indistinct pale spots or blotches
are characteristic points. Description in detail is as follows :

Head measured from tip of snout to first gill-opening (snout
being rather shortened after preservation) equal to greatest height
of body in front of the posterior end of the insertion of first
dorsal ; both contained 4§ in total length exclusive of the parts
behind the dorsal notch ; height of head in front of insertion
of pectoral a little less than height of body ; the height in
front of eye If in height of body ; snout 2, eye 3|, inter-
orbital 4? in head. Spine of first dorsal curved backwards,
its length equal to that of head, longer than the anteriormost
ray of first dorsal, its anterior surface smooth, keeled in the
median line, its posterior surface for a little less than half its
length with recurved spinules as strong as those of phantasma ;
anteriormost ray of first dorsal exclusive of its fleshy base 1|
in head ; free margin of the fin almost straight, subcontinuous
to second dorsal by a fold of rayless skin ; the distance between
origin of second dorsal and insertion of pectoral a little less
than the distance between the latter and insertion of ventral ;
free margin of second dorsal may be either gently convex or
concave as shown in the figure ; its posterior margin rounded,
continuous to caudal lobe with a deep notch. Anal low, falcate,
its notch opposite the dorsal notch and very deep ; lower lobe
of caudal higher than the upper, the former extending further

ON TWO NEW SPECIES OF CHIMERA. 11

posteriorly than the latter ; each lobe of caudal nearly as high
as in mitsukurii ; no tail filament as in purpurascens ; pectoral
pointed and slightly falcate, its tip reaching far behind insertion
of ventral when depressed ; ventral truncate at the free margin,
its inner margin rounded, its insertion midway between tip of
head and dorsal notch.

Lateral canal with fine sinuation which is not so apparent
as in Phantasma ; from the insertion of ventral backwards for
some distance, the sinuation grows somewhat more pronounced ;
maxillo-mandibular and jugular branches of the canal system arise
at the same point from the suborbital branch (PL II., figs. 8
and 10), or the latter directly from the suborbital at a point a
short distance behind the maxillo-mandibular (PI. II., fig. 2).
Six rods in the anterior lamina of upper jaw ; margin of the
lamina sinuate and slightly convex ; lateral rods behind the lamina
low, oblique, their tips directed medianly. Each lamina of lower
jaw with tW'O concavities more or less sinuate in the margin.
Peritoneum white ; wall of the digestive canal whitish. The
species has more a robust body than jordani.

There exists no difference in external features between the
male and the female except in the sexual characters. Cephalic
organ on the snout in front of eye, its inner surface armed with
about seventy spinous denticles, its length 2i in eye length.
Clasper tripartite, its length from insertion 1| in head, its divi-
sion at a point about two thirds its length, its tip covered with
fine shagreen skin ; anterior ventral clasper with five spiny den-
ticles on the median side.

Color in formalin dark brown, with lighter dot-like and
elongate spots, often indistinctly vermiculate by the spots fusing
together. This marking extends to head and to bases of pectoral

12

AßT. 11. S. TANAKA.

and ventral fins. On the side of body behind the ventral, three
pale broad lines below the lateral canal and one above it. Dorsals,
anal, caudal, and free margins of pectoral and ventral blackish ;
proximal parts of dorsals marked similarly as on the side of body ;
numerous transverse series of dark dots above the lateral canal
on the side of body.

The male specimen here figured as the type is deposited in
the museum of the Zoological Institute.

To sum up, the species has the following characters : — dusky
coloration with lighter spots ; great eye ; rather robust body ; spine
of first dorsal with spinules as strong as those of phantasma, and
projecting beyond the end of the anteriormost ray of first dorsal ;
large, pointed pectoral ; no caudal filament ; anal larger than
that of jorda7ii ; lateral canal finely sinuate in the anterior part
of body.

The species is named after Mr. Owston, the well-known
naturalist of Yokohama.

MEASUBEMENTS OF THE S F EC I ES.

Total length (exclusive of parts behind dorsal notch)

Length from dorsal notch to end of caudal

Length of head (to first gill opening)

Height of head behind eye

Height of head in front of eye

Greatest height of body (in front of the posterior end of tlie "(
insertion of first dorsal) /

Long. diam. of eye

Diameter of pupil

Long. diam. of iris

Male.

Female.

71.0

69.0

? 22.5

14.5

14.2

? 15.0

12.6

13.1

11.1

11.9

14.7

14.5

4.25

3.9

1.8

1.8

3.6

3.0

ON TWO NEW SPECIES OF CHIM^ERA.

13

Male.

Female.

Length of snout ...
Interorbital width

Length of dorsal spine

Height of anteriormost ray of first dorsal (exclusive of itsl
fleshy base) i

Height of highest part of second dorsal (measured perpen-1
dicular to upper margin of body) /

Upper margin of pectoral (excl. of its fleshy base)

Breadth of pectoral at insertipn of base

External margin of ventral (exclusive of its fleshy base)

Breadth of ventral at insertion of base

Height of upper lobe of caudal

Height of lower lobe of caudal

Length of cephalic organ

Length of clasper from its insertion

Length of clasper from insertion to the point of division...

Distance between insertion of pectoral and that of ventral...

Distance between insertion of the dorsal spine and that of\
pectoral j

Distance between base of ventral and anal notch

7.0 1

7.2

3.4

.3.5

15.8

? 12.6

11.2

10.9

2.8

,3.4

20.ri

19.2

5.5

5.1

1L9

10.9

2.9

2.8

1.4

1.35

1.8

1.75

2.0

—

8.5

—

5.2

—

24.0

24.0

11.6

10.3

.33.5

31.5

All the measurements are in centimetres.

Synopsis of Species of Chiniœra in the Waters
of Japan,

Anal notch present. (With two dorsal fins. Claspers long,
tripartite.)
a) Caudal filament present, moderate in length.

1) Distance between insertion of clasper and its point of
division almost one half that between the latter and
tip of clasper ; spinous roughness of dorsal spine dis-
tinct ; color silverly, with two longitudinal bands near

14 AKT. 11. — S. TANAKA.

back, and one less distinct band along the lateral
canal phantasma.

2) Distance between insertion of clasper and its point of
division almost twice that between the latter and tip
of clasper ; spinous roughness of dorsal spine less
distinct ; anal notch deeper ; color uniformly dark
brown, with four indistinct bands in the posterior
part of body jordani.

b) No caudal filament.

3) Distance between insertion of clasper and its point of
division nearly twice that between the latter and tip
of clasper ; sjiinous roughness of dorsal spine as dis-
tinct as in phantasma ; anal notch deeper than in
jordani ; color brown, distinctly marbled with small
pale spots, with four indistinct lighter bands in the
posterior part of body. owstoni.

II. No anal notch. (Two dorsal fins.)

a) Claspers rather short, bipartite. Caudal filament very
long.

4) Point of division of clasper in the middle of its length ;
spinous roughness of dorsal spine rather indistinct ;

color whitish with dusky parts here and there

mitsukurii.

b) Claspers long, tripartite. No caudal filament.

5) Distance between insertion of clasper and its point of
division nearly twice that between the latter and tip
of clasper ; spinous roughness of dorsal spine very
indistinct; color uniformly deep purple... ^^^rp^^rasce?^s.

üi. TANAKA.

TWO NEW SPECIES OF CHIMJIRA.

PLATE I.

Explanation of Plate I.

Fig. 1. Lateral view of Chimœra jordani N. sp. About f nat. size.
Fig. 2. Lateral view of Chimœra oiostoni n. sp. About | nat. size.
Fig. 3. Ventral view of the claspers of Chimœra oiostoni. About § nat.

size.
Fig. 4. Ventral view of the claspers of Chimœra phantasma .Tordax *

Snyder. About § nat. size.

Tanaka, New Species of Chimsera.

Jour. Sei. Coll., Vol. XX , Art 11, PI. I.

s. TA>AUA.

ON TWO NEW SPECIES OF CHIMJIRA.

PLATE II.

Explanation of Plate II.

Course of tho sensory canal system in the Chimœra of Japan. All the
lines have been traced on paper placed on the specimens themselves.
Figs. 1-4. Several parts of the canal system taken successively from

anterior parts of the left side of Chhnœra purpurascens. Fig. 3.

Lateral canal above insertion of ventral fin.
Fig. 5. Anterior part of the canal system on the left side of another

specimen of purpurascem.
Figs. G-7. Anterior parts of the canal system un the left side of Clil-

inœra phautasiua .
Figs. 8-9. Anterior parts of tlic canal system on the left side of Chiinœra

oii'donl.
Figs. 10-11. Anterior parts of tlie canal system on the right side of the

same individual as above.
Figs. 12-15. Anterior parts of the canal system in lour individuals of

Chimœra jordani.
Fig. IG. Anterior parts of the canal system on the left side of Chimœra

iniisuharii.

Tanaka. Chimrrra.

Jour. Sei. Coll. Vol. XX., Art. II, PI. II.

JOUENAL OF THE COLLEGE OF SCIENCE, IMPERIAL UNIVERSITY,
TOKYO, JAPAN.

VOL. XX., ARTICLE 12.

A Revised List of Corallinse.

By
K. Yendo, Rigaliushi.

SYNOPTICAL KEY TO THE GENERA.

A, Mother cells of the propagating cells generate! in the cortex : genicula
unizonal or nuiltizoiial : articuli cylindrical, compressed or sagittate:
ramification dichotomous, pinnate or irregular. Conceptacles verrucose,

hemispherical or conical, sessile oen. j. Amphiroa.

I. Medullary portion of articuli constructed with several zones of articoli
interposed between zones of otricoli.

a. Articuli cylindrical, irregularly branching. Genicula multizonal rarely
unizonal, lineœform Sect. I. Euamphiroa.

b. Articuli compressed or subcompressed, mostly dichotoniously branch-
ing. Genicula multizonal, mostly fenestraform in younger portion,
bandform in older portion Seet . //. Eury tion.

II. Medullary portion of articuli constructed ^Yith zones of articoli.

a. Articuli compressed sagittate, reniform, or truncate, pinnately
branched ; genicula unizonal, lineseform. Conceptacles on the sur-
face of the compressed articuli Seei. III. Arthrocardia.

b. Articuli cylindrical or compressed sagittate, or linear, pinnately or
irregularly branched. Conceptacles conical or hemispherical, pro-
minent on the margins of the compressed articuli, or on the surface
of cylindrical articuli spct IV. Marginisporum.

ß. Mother cells of the propagating cells generated in the medulla ; genicula
unizonal or multizonal ; articuli cylindrical, compressed, sagittate or trun-
cate ; ramification pinnate, dichotomous, verticillate or irregular. Con-
ceptacles verruculose, globular or pear-shaped.
I. Genicula multizonal ; ramuli starting from the genicula.

a. Articuli cylindrical, ramuli verticillate

<ie„. ji. Metagoniolithon.

b, Articuli flat, di-trichotoinously branching oon. iii. Litharthron.

AKT. 12. K. YEN-DO.

II. Genicula not specially differentiated ; main hranches dicliotomously
divided, with compressed articuli ; ramuli pinnate, with cylindrical

articuli Gen. ir. Lithothrix.

III. Genicula unizonal, rarauli starting from articuli.

a. Conceptacles sessile immersed in articuli or pinnulets

Ge-«. r. Cheilosporum.

i. Fertile articuli sagittate with pointed wings, conceptacles immersed

in the upper margins of the sagittate articuli ..

Sect. I. Eucheilosporum.

ii. Fertile articuli sagittate or reniform with round or truncate
wings, conce^Dtacles immersed in the outer margins of the articuli,
or in the surface of the articuli Sect. II. Alatocladia.

iii. Articuli compressed hexagonal, conceptacles wartlik ■, on the sur-
face of the articuli or immersed in the pinnulets

Sect. III. Serraticardia.

b. Conceptacles stalked, mostly taking the place of a segment.

i. Branches pinnated Gen. rr. Corallina.

a. Pinnules simple or less decompound Sect. I. Officinales.

ß. Pinnules densely decompound Sect. II. Halyption.

ii. Branches dichotomous Gen. rif. Jania.

SPECIAL PART.

Ge.,. I. AIVIPHIROA LAMX.
Sect. I. EUAMPHIiiOA Dcnk Memoir, p. 111.

Amphiroa anastomosans AV. v. B. Coral. Sib. Exp. p. 91. PL IG. fig.

3-4.
A. valonioides Yendo. Coral, ver. Jap. p. 5. PL 1. fig. 1-3 ; PL IV.

fig. 1.
A. fragilissima f. fragilissima W. v. B. Coral. Sib. Exp. p. 00.

= Amp. fragilissima Lamx. Exp. Method, p. 26. t. 21. fig. d.

= Cor. frngilissima Ellis et Sol. Coral, p. 123. Tab. 21. fig. 9.

= Amp. debilis Kütz. Spec. p. 700. no. 3.

= Cor. rigens Pall. Elench. Zooph. p. 429. no. 11.

A REVISED LIST OF CORALLINE. 6

A. fragilissima f. cuspidata W. v. B. Coral. Sib. Exp. p. 90.

= Amp. cnspidiita Lamx. Exp. Method, p. 26. t. 21. fig. f.

= Cor. cuspidata Ellis et Sol. Coral, p. 124. tab. 21.

= Amp. fragilissima KÜTZ. (nee Lamx.) Spec. p. 700. no. 2.
A. fragilissima f. cyathifera W. v. B. Coral. Sib. Exp, p. 90.

= Amp. cyathifera Lamx. Freyc. Voy. Zool. p. 627.
A. cryptarthrodia Zanarb. Coral, p. 21.
A. involuta Kl rz. Tab. Phyc. VIII. p. 23. Tab. 48. %. 2.
A. rigida Lamx. Histoir. p. 297. t. 11. fijr. 1.

= Amp. cladoniœformis Menegpi. in Kütz. Spec. p. 700.

= Amp. spina, Kütz. Spec. p. 701.

= Amp. irregularis Kütz. 1. c.

= Amp. amethijstina Zanard. Coral, p. 211.

= Amp. spina ß amethystin a Kütz. p. 700.
A. rigida var. inordinata Zanard. Iconogr. III. p. 265. Tab. 99. B.

%. 5.

=-• Amp. inordinata Zanard. Coral, j). 21.
A. tribulus Lamx. Histoir. p. 302.

= Cored, tribulus Ellis et Sol. Coral, p. 124. Tab. 21.

Ampliiroa fragilissima has its articuli much swollen at their
distal ends. That this character is also found in Metagoniolithon
{Amphiroa) Charoide and 3Ietag. stelligeriim has been remarked
by Areschoug.^^ Madam Weber van Bosse^^ has clearl}'- illus-
trated the structure in her recent work. The intumescence is
evidently due to the secondary growth of the circumgenicular
cortex as was formerly proven by me in the case of Melag. stelli-
gerumF' I had an opportunity of examining only of the specimen
of Am'p. fragilissima Lamx. kept in the herbarium of the Higher
Normal School of Tokyo. The specimen is in a much distorted
condition with most of the articuli separated one from another.

According to Madam Weber van Bosse the plant has in
its internal structure no less a tendency to approach to Meta-

1) Sp. Alg. II. p. 531.

2) Coral, of Siboga Exp. PI. XV T. fig. 2, 5.
8) Stii(!y of Genic'ula. p. 29.

4 ART, 12. — K. YENDO.

goniolithon and Lltharthron. Bat we can not find good reason
to separate the species in question from the present genus and
bring it into either of these two gener;i.

The characters which distinguish the present section from the
next are rather ambiguous. Some of the members of the Eurytion
gradually approach this section in shape of articuli and in rami-
fication. Yet it is often convenient to reserve the name to re-
present the cylindrical A7)iphiroa.

Sect. II. EURYTION DcNE. Memoii-. p. 113.

A. nodulosa Kütz. Tal), Phyc. VIII. }>. 19. Tab. 41, %. 1.
A. Yerruculosa KiJTz. Phyc, GeD. p. 387, Tab. 79, fig, 3.

= Amp. cpijptarthrodia ß verruculosa Hauck, Meeresalg, p, 276.
A. BeauYoisii Lamx. Histoir, p. 299.

= Amp. exilis Harv. Ner. Austr, p, 95.
A. BeauYoisii ß crassiuscula (Harv.) f

= A}np. exilis ß crassiuscula Harv, Ner, Austr. p. 95.
A. zonata Yendo. Cor. ver. Jap. p. 10. PI, 1, fig, 11-14 PL IV, fig. 9.
A. parthenopea Zanard, Icoiiogr. III. p. 268. Tab, 100. B.
A. echigoensis Yendo, Cor, ver, Jap. p. 11. PI. 1. fig. 15-16: PL IV

fig, 10,
A. ephedrsea Done. Mémoi]-. p, 112,
« fusoides (Aresch,) t

= Amp. fusoides Lamx, Histoir, p. 298, t. 11. fig, 2,

= Amp. Poppighii Exdl, et Dies. Alg. NataL p, 290,

= Amp. ephedrœa Harv, Ner. Austr, p. 95. t, 39.

= Amp. uentricosa Lamx. (sec. Dcne. Memoii- p. 112.)
ß Gaillonii (Aresch.) f

= Amp. Gaillonii Lamx, Histoir. p. 298. t. 11. fig. 3.

= Amp. Guenzii Harv. Ner. Austr. p. 95.

= Cor. ephedrœa Lamk. Mem. Mus. II p. 238.

= Cor. Chara y Lamk. (sec. Done. Memoir, p. 112.)
A. anceps Done. Memoir, p, 113.

= Cor. anceps Lamk. Mem, Mus. I. p. 238.

= Cor. dilatata Lamx. Histoir. p. 299.

= Anp. dilatata Aresch. iu J. Ag, Spec. IL. p, 536,

= Amp. galaxauroides Hoxd. PLiut, Preiss, VoL II. p, 188.

A REVISED LIST OF CORALLINE. 5

= Amp. Cwningii Rvph. (sec. Weber van Bossr. Coral. Sib. Exp. p. 93.)
-= Amp. Karstalskii Rui>r. ditto.

= Amp. nobilis Kütz, .Spec. p. 708. (sec. Weber van Bosse. 1. c.)
= Amp. ephedrœa Harv. New South Wales Algto. no. 458. (sec.
Weber van Bosse. 1. c.)
A. Bowerbankii Harv. Ner. Austr. p. 97. t. 37.

= Amp. dilatata Krauss. Flora des Cap. u. Natal, p. 206. (sec.
Aresch.)
A. crassa Lamx. Freyc. Voy. Zool. p. 627.

f. Godeffroyi W. v. B. Coral. Sib. Exp. p. 98.

:=^ Amp. Goddefroyi Grun. Ale;, d. Fidschi.
f. minuta W. v. B. Coral. Sib. Exp. p. 98.
A. foliacea Lamx. Freyc. Voy. p. 628. t. 93. fig. 2-3.

= Amp. squarrosa Grun. in Bot. Herb. Hamburg, (sec. Weber van
Bosse.)
f. procumbens W. v. B. Coral. Sib. Exp. p. 92.
f. erecta W. v. B. 1. c.

= Amp. nobilis Hauck. (nee. Kütz.) Ueber einige von Hildebr.
Gesam. Alg. (sec. W. v. B.)
A. pusilia Yendo. Cor. ver. Jap. p. 13. PI. I. fig. 22-23 : PI. V. fig. 11-13.
A. misakiensis Yendo. 1. c. p. 14. PI. I. fig. 24-25 : PI. VI. fig. 1.
A. canliculata Mart. Preus. Exj). Tange, p. 29. PI. VI.

SPECIES DOUB TFUL.

Amphiroa linearis Kütz. Tab. Phyc. VIII. p. 22. Tab. 46. fig. 2.

A. cdgeriensis Kütz. Tab. Phyc. VIII. p. 21. Tab. 44. fig. 2.

A. Boviesii Kütz. 1. c. p. 21. Tab. 43. fig. 4.

A. pustidata Mart, in Reg. Flor. no. 31. (sec. Aresch. in J. Ag. Sp. II.

p. 535.)
A. brasiliana Done. Memoir, p. 113.
A. Gomplanata Kütz. Phyc. Gen. p. 388.
A. canalieidata Yendo (nee Mart.) Cor. ver. Jap. p. 14. PI. I. fig. 26 :

PL IV. fig. 14-15.
A. Cwningii Mont. Zellenpl. auf den Philipi)in. p. 66U.

The essential characters of this section in distinction from those
of tlie others lie in the regularity of the dichotomous ramification
and the flatness of the articuli in the middle and upper portions.
In Amphiroa canaliculala Mart, and some others the dichotomy
is more or less disturbed. The articuli of the species in this

6 AEÏ. 12. — K. YENDO.

section are mostly linear, but subterete in some species ; in this
respect gradually approaching the preceding section. Amp. 7iodu-
losa has nearly cylindrical articuli, slightly compressed in the
ultimate articuli. Amp. canaliculata Mart, has essentially broad
and ancipitous articuli at the upper portions of the fronds. Both
margins, however, are inflexed very much to one side forming a
pair of canals along the margins. This is found to be the case
in some other species, though in a slight degree, such as Amp.
foliacea and Amp. misakie7isis.

Some of the members, such as Amp. BovjerhankU, Amp.
anceps, Amp. canaliculata Mart,, Amp. foliacea, etc., are habit-
ually decumbent. This causes the differentiation of dorsiventrality
in the upper articuli. The canaliculatad species have the canals
at the ventral side only, or in other words, on the shaded side :
and the conceptacles are to be found on the dorsal side only. In
the uncanaliculated and decumbent members the conceptacles are
restricted to the shaded surface with a few conceptacles occasionally
found on the opposite side. In the erect forms they are found
on both surfaces.

The decumbent habit may have been caused by the fact that
the genicula of the basal portion in these species develope further
and finally form large horny flexible joints. These joints, i. e.,
the bandform genicula, allow the ujoper portions of the fronds to
bend downwards. In the preceding section the lower genicula
are very short and form the constrictions of the moniliforni stems
at the basal portions of the fronds. The short linear genicula
cause the fronds to stand upright.

Amp. anceps Done., according to Harvey's description and
illustrations,^^ is identical with Amp. dilatata Aresch. Areschoug

1) Ker. Austr. p. 'J8. t. 37.

A REVISED LIST OF CORALLINE. 7

chose to mention tliem separately. Madam Weber van Bosse
is of tlie same ojiinion as Harvey ; and this view I think pre-
ferable. Amp. Cumingii Rupr. and Amp. Karstalskii Rupr. have
been challenged by Areschoucj, who thought that they have
close affinity with Amp. exilis Harv. Madam Weber van
Bosse has kindly Avritten me after examination of authentic
specimens that Ruprecht's plants are identical with Amp. anceps.
This disposal of Ruprecht's species is entirely on her own
authority.

Areschoug doubtingiy referred Amp. Beauvoisii, Amp.
pustulata and Amp. complanata to Amp. exi/is Harv. Madam
Weber made it clear that Harvey's plant was similar to Amp.
Beauvoisii Lamx. The other doubtful species described by Kütz-
I^'G are so insufficiently defined that I can not find their exact
relation to other plants ; nor are they sharply enough defined to
be reckoned as independent species. Madam Weber remarks
that she saw in Kutzing's herbarium a plant called Amp. linearis
KüTZ. which she thinks a stunted form of Amp. ephedrœa Lamx.
She does not mention anything about the connection between the
specimen and the illustration in Kutzing's Tab. Phyc. VIII.
Tab. 46. Amp. caiialiculata Yendo is a doubtful plant, as has
been already mentioned. ^^ It is probably a broad form of Amp.
anceps. The scarcity of the specimen precludes me from further
discussion.

Se:t. HI. ARTHROCARDÏA Dcne. p. p. Memoir, p. 112.

= Arthroeardia Aeesch. p. p. in J. Ag. Spec. Alg. II. p. 547.
= Pseudarthrocardia Yendo. Eimmer. Cor. Alg. p. 191.
Amphiroa corymbosa Dcne. Memoir, p. 112.

1) Enumeration of Coral. Alg. p. 190.

8 ART. 12. K. YEN DO.

= Arthrocardia eorymbpsa Aresch. in J. Ag. Spec. Alg. II. p. 550.

= CoraUina magnifica Leach, (sec. Harv. Ner. Aiistr. p. 99.)
a. Aresch. in J. Ag. Spec. Alg. II. p. 551.

= CoraUina eorijmbosa Lamk. Mem. Mns. II. p. 234.
ß. Aresch, in J. Ag. Spec. Alg. II. p. 551.

== Ampiroa eorijmbosa Harv. Ner. Austr. p. 99. t. 38.
A. aberrans Yendo. Cor. ver. Jap. p. 16. PL I. fig. 1-5 : PI. V. fig. 1-3.
Ä. Wardii Harv. Ner. Austr. p. 99. t. 38.

= Arthrocardia Wardii Aresch. in J. Ag. Sjiec. Alg. II. p. 551.
A. Mallardise Harv. Ner. Austr. p. 99.

= Arthrocardia MallarcUœ Aresch. in J. Ag. Spec. Ag. II. p. 552.

SPECIES DO l B TFUL.

CoraUina rigida Kütz. Spec. p. 708.

Amphiroa constricta Kütz. Tab. Phyc. VIII. p. 24. Tab. 49. fig. 3.

CoraUina gomphonemacea Kïtz. 1. c. p. 30. Tab. 63. fig. 1.

CoraUina frond escens Kütz. (nee Post et Rupr.) 1. c. p. 29. Tab. 59.

fig. 2.
CoraUina arbuscida Kütz. (nee Post, et Rupr.) 1. c. fig. 3.

Arthrocardia was first established by Decaisne as the second
section of the genus A^nphiroa. Areschoug elevated it to a distinct
genus. Both writers enumerated two differing groups under a
sinofle section or genus. It is reo;rettable tliat Schmitz, when he
united Arthrocardia Akesch. with Chilosporu7n, did not recognize
this confusion. Madam Weber re-established Arthrocardia
Aresch. She^seems to have overlooked the peculiarities in the
position of the coneeptacles of those species such as Cheilosporum
frondescens, C%eiL yessœnse, etc., which are enumerated under the
section Alatocladia in the present paper. In distiguishing Arthro-
cardia from Cheilosporum she states : " The position of the con-
ceptacula, however, marks a difference between the genera : in
Arthrocardia we find conceptacula on the joint as in Amphiroa ;
in Cheilosporum we find them immersed in the margin of the
broadened, wing-like joint " This distinction holds good

A REVISED LIST OF COLALLINJE. 9

only for separating the above listed species from the section
Eucheilosjjorum .

I am not inclined to discuss the question whether Arthro-
cardia is worthy of generic rank or not. Were the characters
of Arthrocardia important enough to elevate it to a genus, the
Marginisporum, the Alatocladia and the Serraticardia should be
equally mentioned as genera. But this is a matter of personal
preference.

I was no less in doubt whether the two species, Amji. Wardii
and Amp. Mallardiœ, described by Harvey, should be legitimately
reckoned under the present section or not. We have no informa-
tion about the conceptacles of these species ; and I w^as not so
fortunate as to study authentic specimens. Judging, however,
from the original descriptions and Areschoug's notes, I ventured
to arrange them as above. They may have close affinity with
the Serraticardia.

Arthrocardia palmata Aresch. seems to be unsatisfactorily
defined. Areschoug considers Cor. filicula Lamk. to be its
forma. But the latter belongs without doubt to the Alatocladia,
while the former has several characters allied to Amp. aberrans.
Grunow^^ remarks that Cor. flabellata KtJTZ. seems closely related
to Arth. palmata Aresch. As it is impossible at present to
study original specimens of these species, I must satisfy myself
with the references of these noted algologists.

The genicula of Amp. aberrans are formed of single zones
of the periclinal cells. In explaining the fig. 5. pi. II. of " Cor.
ver. Jap.," I remarked that probably the geniculum was built
up with several zones. This was disproved after repeated closer
examination.

1) Weltreise der Novara p. 79.

10 ART. 12. — K. YENDO.

Scet. IV. MARGINISPOB.UM Yendo. Enumer. Cor. Jap. p. 191.

Amphiroa declinata Yendo. Cor. ver. Jap. p. 15. PI. I. %. 29 : PI. VI.

ûg. 4.
A. crassissima Yendo. 1. c. p. 16. PI. I. %. 27-28 : PI. V. %. 5-6.
A. cretacea Ende. Mantissa. Suppl. III. p. 49.

= CoraUina cretacea Post, et Rupr. Illustr. p. 20. t. 40. f. 104.
= Arthrocardia cretacea W. v. B. Coral. Sib. Exp. p. 105.
f. rosariformis Yendo. Cor. ver. Jap. p. 7. PI. IV. fig. 3.
f. capensis f

= Amphiroa capensis Aresch. in J. Ag. Spec. Alg. II. p. 533.
= Amp. dubia Kütz. Tab. Phyc. VIII. p. 24. Tab. 49. fig. 1.
Yar. tasmanica Kütz. Tab. Phyc. VIII. p. 23. Tab. 47. fig. 2.
= Amp. tasmanica Sond. Plant. Müll. p. 686.
f. breviarticulata. f
= Ajup. breviarticulata Aresch. in J. Ag. Spec. Alg. IL p. 532.

SPECIES DOUBTFUL.

Amphif^oa lucida Lamx. Histoir. p. 297.

Amp. nodularia Dcne. Memoir, p. 111.

= CoraUina nodularia Lamx. (sec. Dcne. Memoir, p. 111.)

= Cymopolia rosarium Lamx. (do.)

= CoraUina rosarium Lamk. Mém. Mus. I. p. 234.

The first two species which are assigned to the Japan coast
only, have a peculiarity in the position of the conceptacles, which
are always found on the margins of the articuli, growing in a
row, pointing obliquely toward the ventral side of the frond. In
other respects, these species are closely related to the preceding
section.

Madam Weber transferred Atnp. cretacea Endl. to Arthro-
cardia Aresch. The principal reason for doing so seems to be
grounded on the internal structure of the articuli. In the ex-
ternal appearance of the articuli the species suggests a position
near to Amp. rigicla Lamx. or C/ieilosporum iuberculosum. From
the former it is distinguished by the unizonal genicula and the

A REVISED LIST OF CORALLINiE. 11

superficial position of the conceptacles ; and from the latter by
the cortical origin of the mother cells of the spores.

I am no less in doubt in classifying Aynp. cretacea Endl. in
the same section with Amp. decUnata. The position of the con-
ceptacles is very irregular in the former species, while in the latter
the conceptacles are arranged in a definite regularity. Amp. cre-
tacea and its forms are widely distributed in the colder seas of
both the northern and the southern hemispheres. It may claim
an independent position. The above disposition is a provisional
one until we find a more special character in Aiiip. cretacea.

Amp. breviarticidata Aeesch. and Amp. capejisis Aresch.
seem to be nearly allied to Amp), cretacea Endl. According to
the descriptions given by Areschoug^^ we can hardly distinguish
the former from var. tasmanica Kutz. This has been already
noted by Sonder."^ Amp. capensis Aresch. is also unseparable
from Amp. cretacea., except that the basal genicula are compara-
tively longer. Amp. dubia Kütz., judging from the figures of
Tab. Phyc. VIII. Tab. 49. fig. 1., is referrible to Amp. capensis
Aresch. These species have hitherto been distinguished by the
relative size and forms of the articuli ; but these points can hardly
be relied upon to warrant a specific character.

Madam Weber doubts Amp. hreviarticulata Aresch. She
found an Artkrocadia called by that name in Suringar's her-
barium, and tried in vain to see the type of Areschoug's species.
Judging from the original descriptions, I believe the specimen
found by her is different from what AreschouG meant.

Amp. cretacea has been defined as often having recurved or
flexuous articuli. Areschoug put too much stress on this character

1) Spec. Alg. IT. p. 532.

2) Linnea. XXV. p. 687.

12 ART. 12. — K. YENDO.

in distinguishing his two species. In the specimens which I have
referred to Äm,p. cretacea Endl., found commonly in north Japan,
the articuli with such characters occur frequently but not invari-
ably. Neither Areschoug nor Kützing gives anything about
the conceptacles of their plants, and I have not seen any au-
thentic specimens of these species. But the distinguishing charac-
ters mentioned by these authors are by no means fixed ones.
Their species, if worth mentioning separately, may be disposed as
above.

Amp. nodularia Done., according to Areschoug, has close
affinity with Amp. hrevim^ticulata. It should be united with
Amp. cretacea or its forma.

Gen. II. IVIETAGONIOLITHON W. v. B.

Metagoniolithon Charoides W. v. B. Coral. Sib. Exjo. p. 84.

= Amphiroa Charoides Lamx. Histoir. p. 301.

= Amp. Chara Done. Classif. des Alg. p. 63. PL 17. fig. 9.

= Corallina Galioides Lamk. Mem, Mus. II. p. 239.

= Amp. steUulata Kütz. Spec. p. 702.
M. stelligerum A¥. v. B. Coral. Sib. Exp. p. 84.

= Amp. stelligera Aresch. in J. Ag. Spec. Alg. II. p. 540.

= Corallina stelligera Lamk. Mem. Mus. II. p. 239.

= Amp. jubata Lamx. Histoir. p. 301. t. 11. fig. 6.

= Amp. elegans Sond. Plant. Preiss. p. 55. no. 53.

= Amp. verrucosa Lamx. Histoir. p. 300. PI. XI. fig. 4.

= Amp. stelligera ß. interrupta Kütz. Tab. Phyc. VIII. p. 26, Taf.
52. fig. h.

= Amp. interrupta Lamx, Histoir. p. 301. PI. XL fig. 5. A.
M. gracile f

= Amp. gracile Harv. Phyc. Austr. PI. CCXXXI.

= Amp. intermedia Harv. Syn. Catal. of Austr. and Tasm. Alg.
p. 30.

SFECIES DOUBTFULL.

Amphiroa granifera Harv. Syn. Catal, of Austr, and Tasm. Alg. p. 30.
= Metagoniolithon graniferum W. v. B. Coral. Sib. Exp. p. 84.

A REVISED LIST OF CORALLINE. 13

= Amp. setacea Kutz. Spec. p. 700.

= Amp. similis Sond. (sec. Weber 1. c. p. 84.)
Corallina Chara Lamx. Mem. Mus. I. p. 240.
Cor. radiata Lamx. Mem. Mus. I. p. 240.

This genus is easily distinguished from the others by having
verticillate ramuli which start from genicula and not from articuli.
The articuli are exclusively cylindrical, varying in their length
according to the position in a frond and to the species. In some
species, both extremities of the articuli bulge out, owing to the
secondary development of the circumgenicular cortex ; and in
some species the verticillate ramuli are slightly curved upward
and inward.

The conceptacles are, as far as I could observe from the
accessible material, generated from the medullary cells. The
diameter of the conceptacles is very large when compared with
that of the ramuli ; in Metagoniolithon stelligerum the former is
subequal to the latter. This circumstance, no dobut, forces con-
ceptacles to stai>t from the deep places in the medulla.

The species under this genus are characterised by much
elongated genicula. In Metag. stelligerum the length of a matured
geniculum often attains more than twenty times its breadth, and
the greater portion of the racliis is formed by the genicula.
These genicula, quite strange in comparison with those of plants
belonging to other genera, can send out the verticillate ramuli
around a certain region. No ramuli is found starting directly
from an articulus. This peculiarity, together with the medullary
origin of the propagating cells, is enough to detach the present
group from the genus Amphiroa and to place it under a new
one. In my former paper I made a preliminary announcement
of this matter. A few months later, Madam Weber, evidently

14 ART. 12. K. YENDO.

not knowing what I had suggested, established a new genus for
this peculiar species.

Harvey remarks that Amp. gracilis Harv. has some affinity
with Metag. stellig er urn, and still closer affinity with A7np. inter-
media Harv. Cf. Phyc. Austr. t. 231. I consider that both of
Harvey's plants belong to the same species. Madam Weber
refers Amp. granifera to the genus Iletagoniolithon reducing Amp.
intermedia, Amp. setacea, Amp. similis and Amp. stellata to the
position of its spnonymes. I can not decide at present whether
Amp. granifera Harv. is the same as Amp. gracilis or not. If
it be the same, the name Amp. gracilis Harv. should be kept
on account of priority. Amp. similis Sond. was united by Madam
Weber to Metag. graniferum. But I could not find the descrip-
tion in Bot. Zeitung 1845, which she mentioned as the " quelle "
of A7np. similis Sond.

Gen. III. LITHOTHRIX J. GRAY.

Lithothrix aspergillum J. Gray. Journ. Bot, Vol. V, p. 33.
= Amphiroa aspergillum Anderson, Zoe, II. p. 225.
= Amp. nodulosa Farl. (nee Kütz.) Report U. S. Fish, Comm,

1875. p. 715.
= Amp. nodulosa Coll., Hold, et Setch, Phyc. Bor.-Amer. no. 498.
f. nana f

= Amp. aspergillum. f. nana Setch. et Gard. Alg, N. W, A, p, 359.

The monotypic genus has been established by J. Gray, who
does not, however, give clear reason for its establishment. But
it is not difficult to understand from his remarks that the plant
has some relation to Amphiroa by having wartlike conceptacles
and at the same time to Corallina in its general habit. The
generic diagnosis given by him is not sufficient to separate it from
the other genera. The plant was referred to Amphiroa by An-

A REVISED LIST OF CORALLINE. 15

DERSON^^ witbout any comment, but probably strengthened by
Prof. Farlow's remark on this species.^^*

It has, however, unique characters not mentioned by its dis-
coverer or others, as far as my research has extended. The
peculiarty lies in the genicular structure and in the branching
mode of the fronds. It has no geuiculum, in the exact sense,
similar to any of the other Corallince. The geniculum is formed
by the lower half portions of the filamentous periclinal cells,
the upper half of which practically corresponding to the medulla
of other species. The details are given in my former paper. ^^
The axial stems ramify dichotomously while the ramuli start from
the margins of the articuli of the stem. An articulus which sends
forth dichotomous branches from its top has a longitudinal furrow
along its median. Hence the articulus apparently seems to be
two compressed articuli fused together by their periclinal sides
with a geniculum on the top of each. The furrow or the canaliza-
tion is also visible in the succeeding lower articuli, gently dis-
appearing as we trace downwards. The two genicula on the top
of the branchino; articulus are not in a strai2;ht line but form
a geometrical supplimentary angle to the axile divergence of the
branches. The angulation of the genicula is also distinct in the
succeeding articuli, but gradually approaches a straight line as
canalization vanishes.

The canalization and the mode of pinnation of ramuli are
unique characters among Corallince. The solitary conceptacles on

1) Zoe. Vol. II. p. 225.

2) Proc. of Amer. Acad. Vol. XII. p. 239.

* These two papers were not accessible to me. But findin;j; them indispensable in
preparing the present chapter, I asked Mr. F. S. Collins for references. Mr. Colliks
was so kind as to send me a transcript of all the lines relating to the subject. I desire to
express to him my heartiest thanks.

3) Study of Genicula. p. 16.

16 ART. 12. — K. YENDO.

the articuli of branchlets are also an unusual phase. These
characters are ample enough to restore the genus, with slight
modification in the original diagnosis. This was also mentioned
in my former paper and noticed by Madam Weber.

Gen. IV. LITHARTHRON W. v. B.

? Rhodopeltis Schmitz, p. p. in Eng. Pflanzenfamil. Alg. p. 530.
Litharthron australe W. v. B. Coral. Sib. Exp. p. 104.
= Ampliiroa australis Sond. Plant. Preiss. p. 55.
? = Rhodopeltis australis Schmitz (nee Harv.) p. p. in Eng. Pflanzen-
famil. Alg. p. 530.

Conceptacles have not yet been found in this species, which
has, however some relation to MetagonioiUthon in the structure
of genicula and the mode of ramification. It was regarded by
some old writers as having close affinity with Eurytion. But
this is merely a superficial likeness in the shape of the articuli.
A study of the internal structure proves that it is an independent
genus.^^ KijTzmG^^ tended toward this view but did not actually
appropriate it.

Harvey^^ described a nemathecium-like plant, calling it
Rhodopeltis australis, epiphytic on Litharthron australe. Schmitz^^
seems to have regarded the epiphyte to be the propagating organ
of the host, which he classified among Rhizophyllidacece. Madam
Weber has written me that Prof. De Toni called her attention
to this matter. It is also mentioned by De Toni^' in the last

1) After the publication of "Study of Genicula" Mr. Reinbold was so kind as to
send me a piece of the original specimens of Am'p. australis Sond. The specimen was of
great value in preparing the present paper. I am very glad to express my gratitude to him.

2) Tab. Phyc. VIII. p. 25.
■ 3) Phyc. Austr. PI. 264.

4) ENGii. u. Pkantl : Pflanzenfamilien. Algte. p. 530.

5) Serie. XV. p. 175.

A REVISED LIST OF CORALLINvE. 17

October number of La Noiiva Notarisia. Madam Weber doul^ts
ScHMiTz's view. So do I. The problem is reserved for future
until an authentic specimen of Rhodopeltis australis or a fertile
exemplar of Litharthron australe has been examined.

Gen. V. CHEILOSPORUm ZANAKD. emend.

The genus Cheiloèporum was first instituted by Zanardini
to include most of the members of Decatsne's fourth division of
Amphiroa. Schmitz^^ extended the generic sense so as to cover
the genus Arthroeardia Aresch. The definition of Arthrocardia
given by Areschoug is somewhat different from that of the sec-
tion Arthrocardia Done. Many of the members of Arthrocardia
Aresch. may more legitimately be referred to Amphiroa than to
Gheilosporwn. On the contrary, the greater number of species
comprised in the section Arthrocardia Dcne. evidently belong to
Chellosporum. Hence the genus Gheilosporwn, as conceived in
the present paper, comprises both the sections Arthrocardia Done.
p.p. and Cheilosporum Dcne. The scope given by me to the
genus thus practically agrees with that given by Schmitz, except
that some species of Arthrocardia Aresch. have been removed
from it.

The members enumerated under the present genus show
certain characters by which they may be easily distinguished
from those of the other genera. The most important characteristic
lies in the fact that the spores are situated deep in the medullary
portion.

The genus is divided into three sections, viz., Eucheilosporum,
Alatocladia and Serraticardia. The Eucheilosporum is a group

1) System. Uebersicht. p. 455.

18 ATE. 12. — K. YENDO.

sharply defined from all others, and its members belong exclusively
to the warmer seas. The extreme forms of the Eucheilosporum
and of the Alatocladia show indeed some differences with regard
to their habit and the external appearance. But I consider the
differences not to be of sufiicient tenor to justify their separation
into different genera. In the Alatocladia the position of the
conceptacles is rather indeterminate, so that we can not delineate
a sharp boundary to separate it from the Eucheilosporum.

The Serraticardia has the external form closely resembling
that of the Arthrocardia. By the position of the conceptacles
and the phi nation of the branchlets, these two sections may easily
be distinguished one from another. Only two species are assigned
to the former section. They have two sorts of conceptacles, which
may occur in an individual or in one branch at the same time.
One is imbedded at the top of the branchlet, or articulated lobe,
as it were. Another is situated upon the surface of articulus.
The former character suggests Corallina and the latter, no doubt,
is the peculiarity found in the members of the Alatocladia.

Sect. I. EUCHEILOSPORUM Yendo. Enumer. Cor. p. 193.

Cheilosporum elegans Aresch. in J. Ag. Spec. Alg. II. p. 546.

= Amphiroa elegans Hook, et Harv. Ner. Austr. p. 101. PL
XXXVIII.
C. sagittatum Aresch. in J. Ag. Spec. Alg. II. p. 545.

= Corallina sagittata Lamx. Freyciu. Voy. p. 625. t. 05. fig, 11-12.

= Amp. sagittata Done. Memoir, j). 113.
C. jungermannioides Rupr. in J. Ag. Spec. Alg. II. p. 546.
C. spectabile Harv. Friendly Isl. Alg. no. 31.
C. cultratum Aresch. in J. Ag. Spec. Alg. II. p. 545.

= Amp. cuUrata Harv. Ner. Austr. p. 102. PI. XXXIX.
ß debilis f

= Amp. cultrata ß debilis KützJ Tab. Phyc. VIII. p. 27. Tab. 54. fig. k. 1.

= Amp. cultrata y globulifera Ktixz. 1. c. Tab. 55. fig. 1.

A REVISED LIST OF CORALLINE. 19

d pectinata f

= Amp. cultrata d pectinata Kütz. 1. c. Tab. 55. fig. 2.
f. multifida f

= Amp. multifida Kütz. 1. c. Tab. 56. fig. 1.

SPECIES DOUBTFUL.

Amphiroa aeutiloba Done. Memoir, p. 113.
Amp. Lamourouxiana Dcne. 1. c. p. 113.

= Corallina Lamourouxiana Leach, iu Herb. Lamx. (sec. Done. 1. c.)
Amp. fastigiata Dcne. Memoir, p. 113.

Amp. heterodadia Kltz. Tab. Phyc. VIII. p. 27. Tab. 55. fig. 1.
Cheilosporum pukhrum Harv. (sec. Sonder, in Fragmenta, p. 20.)

The doubtful species above mentioned are too briefly defined
to make it safe to count them as independent species. A^np.
heterodadia KiJTZ. is referrible to either Oieil. Stangeri or Cheil.
flabellatum. The linear or cylindrical articuli in some part of
the frond are often found in an abnormal form of this section,
and their presence is never a constant character. A)i)p. cultrata
Harv. ï globulifera Kutz. is the name given to a plant having
the conceptacles of Ghoreonema.

Except these questionable plants, most of the members are
clearlv defined and acknowledged by many eminent systematists.

Sect. II. ALATOCLADIA f

= Arthrocardia Dcne. p. p. Memoir, p. 112.
= Arthrocardia Yendo. Enumer. Coral, p. 192.
= Arthrocardia Aresch. p. p. in J. Ac4. Spec. Alg. II. p. 548.
Cheilosporum californicum Yendo. Cor. ver. Port Renfrew, p. 715. PL
LIV. fig. 2: PI. LVI. fig. 3.
= Amphiroa californica Dcne, Memoir, p. 112.
= Amp. tuberculosa f. californica Setch. et Gard. Alg. N. W. Amer,
p. 361.
C. frondescens Yendo. Cor. ver. Port Renfrew, p. 715.

f. typica Yendo. 1. c. p. 715. PI. LIL fig. 1. : PI. LVI. fig. 4, 5 ami 8.

20 AKT. 12. — K. YENDO.

= Corallina frondescens Post, et Rupr. Illustr. p. 20. t. 40. fig. 103.
= Arthroeardia ? frondescens Aresch. in J. Ag. Spec. Alg. II. p.

549.
= Amphiroa Hoolieri Harv. (sec. Harv. Ner. Bor. Amer. p. 86).
= Amp. tuberculosa f. frondescens Setch. et Gard. Alg. N. W.
Amer. p. 362.
f. maxima Yendo. Cor. ver. Port Renfrew, p. 716.
f. intermedia Yendo. 1. c.
f. polymorpha Yendo. 1. c.

C. filiculum t

= Cheilosporum palmatum ß filiculum Yendo. Euumer. Cor. p. 192.

= Arthroeardia palmata, et ß Aresch. in J. Ag. Spec. Alg. II. p.
550.

= Corallina filicula Lamk. Mem. Mus. II. p. 237.
var. planiusculum f

= Cheilosporum planiusculum Yendo. Cor. ver. Port Renfrew, p. 717.
PI. LIII. fig. 1-3 : PL LVI. fig. 9-10.

= Corallina planiuscula, et ff. Kütz. Tab. Phyc. VIII. p. 31. Tab.
63. fig. 3.

= Amp. tuberculosa f. planiuscula Setch. et Gard. Alg. N. W.
Amer. p. 363.
C. Stangeri Aresch. in J. Ag. Spec. Alg. II. p. 544.

= Amphiroa Stangeri Harv. Ner. Aiistr. p. 101. PI. XXXIX.
C. flabellatum Aresch. in J. Ag. Spec. Alg. II. p. 544.

= Amp. flabellata Harv. Ner. Austr. p. 101. PI. XXXIX.
C. Darwini f

= Amp. Darwini Harv. Ner. Austr. p. 100.

= Arthroeardia Darwini W. v. B. Coral. Sib. Exp. p. 106.
C. anceps Yendo. Cor. ver. Jap. p. 18. PI. II. fig. 6-S : PI. VI. fig. 2.

= Corallina anceps Kütz. Pliyc. Gen. p. 388.
Yar. modesta Yendo. Cor. ver. Jap. p. 19. PL II. fig. 9 : PL VI.

fig. 3.
C. Orbignianum f

= Amp. Orbigniana Done. Memoir, p. 112.
C. latissimum Yendo. Cor. ver. Jap. p. 21. PL II. fig. 16-17 : PL VI.

fig. 7.
C. yessoense Yendo. 1. c. p. 19. PL II. fig. 12-13 : PL VI. fig. 5.

f. angusta Yendo. 1. c. p. 19. PL II. fig. 14-15 : PL VI. fig. 6.
C. tuberculosum t

= Corallina tuberculosa Post, et Riipr. Illustr, p. 20 t. 40. fig. 100.

= Amp. tuberculosa Endl. Mantissa. SuppL III. p. 49.

A REVISED LIST OF CORALLINiE. 21

= Arthroeardia ? tuberculosa Dcne. Memoir, p. 110.

= Arthro. tuberculosa W. v. B. Coral. Sib, Fxp. p. 110.

= Amp. tuberculosa f. tijpiea Setch. et Gard. A1<^-. N. W. Amer.

p. 361.
= Amp. epiphlegmoides J. Ag. in Harv, Alg. from N. W. Amer. p.

169.
= Arthro. epiphlegmoides W. v. B. Coral. Sib. Exp. p. 106.

SPECIES DOUBTFUL.

Amphiroa firma Kütz. Spec. p. 704.

Corallina Filieula Kütz. Tab. Phyc. VIII. p. 30. Tab. 61. fig. 1.

Cor. carinata Kütz. Tab. Phyc. VIII. p. 30. Tab. 61. %. 2.

Amp. chilensis Dcne. Memoir, p. 118.

Amp. vertebralis Dcne. 1. c. p. 112.

= Arthroeardia vertebralis W. v. B. Coral. Sib. Exp. p. 106.
Amp. proliféra Dcne. Memoir, p. 113.

= Cor. proliféra Lamx. Histoir. p. 291. Tab. 10. %. 5.

The section Arthroeardia, as before stated, was first distin-
guished by Decaisne^^ as the second division of Amphiroa ; and
comprised many species of the present section and a few of
Arthroeardia in the sense in which the present writer uses that
term. It was elevated to a genus by Areschoug with modifica-
tion however in the sense, on account of the characteristic posi-
tion of conceptacles. Schmitz and Hauptfleisch^^ combined it
with the Eucheilosporu7n, including both under the genus Ghei-
losporum. Setchell and Gardner^^ in their joint work remarked
that they could not discover why the present writer''^ referred
some of the Canadian forms, such as Cheilosporum f7'ondesGens,
Gkeil. filiculum f. planiusculum { = Gheil. planiusculu77i), etc., to
the genus Gheilosporum. To remove their doubt it will be

1) Memoir, p. 112.

2) Engler. u. Pbantl: Pflunzenfamilien. p. 543.

3) Notes on X. W. Amer. Alg. p. 3G9.

4) Cor. ver. of Port Renfrew, p. 715.

22 ART. 12. K. YENDO.

sufficient simply to remark that Corallina frondescens Post, et
RüPR. was referred to Arthrocardia by Areschoug,^^ though with
some doubt ; and that Arthrocardia was reduced to Ckeilosporum
by ScHMiTZ.^^ But I referred the Canadian algse to Cheilosporum
not merely because I was strengthened by Areschoug's opinion.
Some further remarks will be added below.

The principal difference between this genus and Amphiroa
lies in the fact that in the former the mother cells of the propagat-
ing cells originate in the medullary portion of the fronds, while
in Amphiroa they are in the cortical. The scars of the dropped
conceptacles prove the difference. This is easily recognized by the
practiced eye. But to get the exact view, one microtome section
of the fertile articuli through a conceptacle is sufficient. This is
quite clear if the reader refers to the figures which I have re-
peatedly delineated in my former papers. ^^

The distinguisliing point between the typical forms of the
EucheilospoTum and the Alatocladia lies in the position of the
conceptacles. In the formei' the normal position of the con-
ceptacles is the upper margin of the sagittate articuli ; as a rule,
only one conceptacle is found in a wing of an articulus. In the
latter, one or more conceptacles are on the flat surface or the
outer margins of the wings of the reniform or sagittate articuli,
Hence the fertile branches of the Alatocladia have articuli with
a pair or more of conceptacles. Cf. PI. LVI. figg. 5, 7 and 18
in '' Cor. ver. Port Renfrew : " and PL II. figg. 12, 14 and IG
in " Cor. ver Japon."

1) J. AG.: Spec. Alg. II. p. 549.

2) Syst. Uebersicht. p. 455.

3) Coral, ver. Japon. PI. II. figg. G, 8, 14 and 16: Cor. ver. Port. Renfrew. PI. LVI.
lîgg. ö and 8.

A REVISED LIST OF CORALLINiE. 23

There are, however, numerous species which have the normal
conceptacles at the external margins of the articuli, displaying,
in a manner, the link between the Eucheilospormn and the
Alatocladia. Cheilosporum yessoense and C%eil. latissimum are
the best examples. Cf. PI. II. figg. 12 and IG, "Cor. ver.
Japon." We should find some difficulty in reckoning the former
species under the Alatocladia, had not some conceptacles been
found on the flat surface of the articuli, before examining a
microtome section of a conceptacle.

The sharp boundary between the two sections above men-
tioned is by no means easily delineated. In practice, however,
we can distinguish them with the utmost safety by the size of
the frond, the shape of the articuli and the locality of the plants.
Almost all of the Eucheilospoi'um are inhabitants of warmer
climates : the articuli are sagittate with sharp pointed wings ; and
the height of the frond hardly exceeds a few inches. These
characters are of-course arbitrary and may not be clearly ascer-
tained in some intermediate forms. Still they are often useful in
determining a sterile plant.

Setchell and Gardner, ^^ after their long study, hold the
opinion that Clieil. frondescens, etc., must be united to the formée
of Amp. tuberculosa Endl. The present writer wishes to be
allowed to add a few words concerning their view.

Cheilosporum (Amphiroa) tuberculosum is a peculiar plant.
It has the characters of both the Arthrocardia and the Alato-
cladia in the external as well as in the internal points. Decaisne'^
has already noticed that it belongs to the section Arthrocardia

1) Alg. N. W. Am. p. 360.

2) Memoir, p. 110.

24 ART. 12. — K. YENDO.

DcNE., and Areschoüg/^ without hesitation, assigned the species
to the genus Amphiroa. 1"^ have enumerated it under the genus
Amphiroa but with much uncertainty. The fronds have both
sorts of articuli, the homogeneous cylindrical ones and the com-
pressed reniform. These forms occur in different branches separately
but frequently in the same branch. Cf. PI. LI. " Cor. ver. Port
Renfrew." The conceptacles are always found upon the flattened
articuli, in the same manner as is characteristic of the Alatocladia.
This character suggests Cheilosporum much more than it does
Amphiroa. The microtomic section of the fertile articuli shows
some conceptacles originate deep in the medulla and some ap-
parently in the cortex. This fact greatly perplexed me in dis-
tinguishing the generic position of the plant, but I provisionally
followed the opinion of Endlicher,^* and reckoned it under
Amphiroa. The reader may notice that the diagramatic figure^^
of the cross section of the articulus with four conceptacles is the
only one figure in my former papers which does not give the
boundaries in the fine dotted line between the medulla and the
cortex. The origin of the spores in this species, as above men-
tioned, is uncertain in some degree. But the presence of the
conceptacles exclusively on the compressed articuli is a character
which suggests disposition as above. It might well be considered
as a transitional form linking the Alatocladia to the Arthrocardia.
The reduction of a species to a forma or variety of a distinct
species depends upon the view of the author. A character may
be taken as specific or as formal. No one can judge which should

1) J. Ag. : Spec. Alg. II. p. 538.

2) Cor. ver. Port Renfrew, p. 714.

3) Mantissa.iSuppl.lIII. p. 49.

4).. Cor. ver.^Port Renfrew. PI. LVI. fig. 1.

A REVISED LIST OF CORALLINE. 25

be absolutely legitimate. But the leduction of the Canadian
species to the formas of Cheilospori(m, tuberculosum, as has been
done by Setchell and Gardner, is by no means acceptable.
The latter species undoubtedly has variable characters, apparently
fluctuating from one to the other ; while the others have, not
whithout reason, been described by many systematists for a long
time as distinct species.

In my former paper on Corallmce of Port Renfrew I had
doubt regarding Amp. epipJilegmoides J. Ag. But as I had not
been fortunate enough to see the original specimen, I placed it
provisionally under Gheil. {Amphiroa) tuberculosum. Madam
Weber,^^ after examining the specimen in the British Museum,
thinks it very likely that it is only a form of latter species, as
I had supposed. Strengthened by her remark, I here reduce the
species in question as synonymous with Clieil. tuber cidosum.

Corallina palmata Ellis et Sol. was transfered to Arthro-
cardia by Areschoug. But I had no less doubt regarding his
opinion in doing so when I compared the definition of Arthro-
cardia pahnata Aresch. with the description of Corallina palmala
Ellis et Sol. given by various authors. Lately I was fortunate
enough through the kindness of Madam Weber van Bosse to
study the specimens, labelled " Arthrocardia px^^lmata Aresch."
and " Corallina pahnata Ellis et Sol.," which were kept in the
Herbarium of Suringar. After careful examination of the speci-
mens, and referring to almost all of the literature concerning
both species, I have come to the conclusion that the species
Arthrocardia palmata Aresch. is quite different from Corallina
palmala Ellis et Sol. Suringar's specimen of Arth. palmata
is sterile. But it accords very well with the description of Arth.

1) Coral. Siboga Exp. p. 106.

26 ART. 12. — K. YENDO.

palmata f. ß of Areschoug. It stands closely to Oheil. j^^ccnivs-
culum Yendo, and the latter may well be taken as its variety.
The difference lies in the shape of articuli ; the axial articuli of
the former having much more projected lobes, and those in the
latter being nearly triangular. Hence the specific name Corallina
Filicula should be reserved by priority. The illustration of Cor.
Filicula delineated by Kutzing in his Tab. Phyc. VIII. Taf. 60.
fig. 1. seems to represent a different species.

Not a few of the present members have been unsatisfactorily
defined. Some of Decaisne's species which he described from
American plants should probably be reduced as synonymous with
others, or some described by other authors afterwards may be
identical with his. Decaisne's original definitions are often in-
complete and good for more than two distinct forms. Actual
examination of the authentic specimens is necessary to the solu-
tion of this problem.

Sect. III. SEB-RATICARDIA Yendo. Enuraer. Cor. Alg. p. 193.

Cheilosporum maximum Yendo. Cor. ver. Jap. p. 21. PI. II. fig. 18-19 :

PL VI. fig. 9.
C. McMillani Yendo. Cor. ver. Port Renfrew, p. 718. PI. LII. fig. 4-5 :

PI. LVI. fig. 11-14.

SPECIES DOUBTFUL.

Corallina officinalis L. /. robusta Setch. et Gard, (nee Kjellm.) Alg. N. W.
Amer. p. 365.

The present section is characterized by the peculiar position
of the conceptacles. Those on the hexagonal articuli, mostly in
pairs, show the relationship to the Alatocladia ; those imbedded in
the apices of the pinnated rarauli suggest a transition to Corallina.
It differs from the latter genus by having the compressed con-

A REVISED LIST OF CORALLINyE. 27

ceptacles as in the JEucheilosponmi. The conceptaculiferous pinnules
may be considered as the apices of the sagittate articuli enor-
mously prolonged and finally jointed. I have formerly mentioned
such a circumstance in a frond of the Eucheilosporum}^

Setchell and Gakdner^^ have doubted my species and think
them the robust forms of Corallina. They hesitatingly referred
C/ieil. Mc31illani to a coarse form of Californiau Corallina : and
further stated that they could not find any conceptacles on the
surface of the articuli of their Corallina except some which
seemed that of Choreonema. A single glance at a section of the
conceptacles under the microscope will determine whether it be-
longs to the plant proper or to the parasite. PI. II. fig. 19. of
*' Cor. ver. Japon " shows the cross section of the conceptacles of
Cheil. maxi7num. No one can confuse the comparatively con-
spicuous conceptacles of Corallinœ with the small sized ones of
Choreonema. I am quite cunvinced by Areschoug when he
writes in J. Ag's. Spec. Alg. II. p. 564 as follows : "■ Quœ sub
hoc nomine {Corallina officinali L.) inde (e Capite bonœ spei)
accepimus, sunt cum Arthrocardiœ palmatœ, tum Arthrocardice
Gorymbosœ formae juniores, quae pro Corallina officinali ab in-
cauto facile haberi possunt."

Gen. ri. CORALLINA LAMX.

= TitanephiUum Nardo. (sec. Endlicher).

Although we can not find a detailed account, it is not hard
to understand that it was on the following ground that Solms-

1) Cor. ver. Japon, p. 12. Cf. also Kütz. : Tab. Phyc. Vlil. Tab. 54.

2) Alg. N. W. Amer. p. 3(55.

28 ART. 12. K. YENDO.

Laubach^^ united Jania Lamx. with Corallina Lamx. He noticed
tlie pinnated ramuli in Jania corniculaia and referred it to a
variety of Jania ruhens. He also, at the same time, remarked
that the pinnation is a character which has hitherto played the
important part in separating Corallina from Jania, which was
defined as branching dichotomously.

I can not acquiese in the opinion that the pinnation of Jania
corniculata Lamx. is identical in its nature with that of Corallina.
The members of Jania seem to have the ability of issuing pin-
nate or subpinnate ramuli at the lower portion of a frond. Jania
micrarthrodia f. antennina (Kütz.) is the actual example, besides
the above mentioned species. Cf. Tab. Phyc. VIII. Taf. 84. fig. 1.
Yet all the members of Jania are sharply defined as having the
dichotomous branches at the apical portions of the fronds. The
conceptacles of Jania, even in the pinnated fronds, are always
found at the dichotomous points of the upper ramuli, or at the
apical points. In Corallina, as a rule, they are at the ends of
pinnae or the pinnules. I cannot discover any necessity of uniting
both genera simply for the reason that there is a species in a
genus which shows, in a manner, a similar character to one found
in the other genus ; and more so, if this common character is a
vegetative one. This circumstance suggests to us that both genera
are related to one another and not that they must be united
into one.

I held this opinion long ago and mentioned Jania, as a sub-
genus in the list of Japanese Corallinaceœ.^^ But I found it
better to restore the genus in Lamouroux's sense. The regularity
of the dichotomous ramification, which is rarely disturbed, seems

1) Corallina. p. 6. foot note.

2) Enumeration of Cor. Alg. p. 19o.

A RENISED LIST OF CORALLINiE. 29

to have special signification in the phylogeny. The other char-
acters limited to the group indicate generic rank rather than
subgeneric.

Decaisne and his followers ascribe the establishment of the
genus CoralUna to Tournefort, while Areschoug chooses La-
MOUROUX as its founder. The latter author remarks that the
genus Gorallina in Tournefort's sense comprised Älgce and
Zoophyte together. I am inclined to think that the recent de-
finition of Gorallina is largely drawn from the elaborate work of
Lamouroux.

Sed. I. OFFICINALES t

Gorallina officinalis L. f. typica Kjellm. Alg. Arct. p. 86.

= CoralUna officinalis, a. Aresch. in J. Ag. Spec. Alg. II. p. 562.
= Cor. officinalis vulgaris Kutz. Tab. Phyc. VIII. p. 32. Tab. 66.

lig. 2.
= Cor. officinalis L. Fauna 8uec. u. 2234. (sec. Aresch. 1. c).
= Cor. articulata Ellis. Coral, p. 60. t. 24. fig. a. A.
= Cor. nana Zanard. Coral, p. 20. (sec. Zanard. Iconogr.)
= Cor. laxa Lamk. Mem. Mus. II. p. 231.
= Cor. spathulifera Kutz. Spec. p. 709.
= Cor. longieauUs Lamk. Mem. Mus. II. p. 232.
= Cor. officinalis y Yekdo. Cor. ver. Jap. p. 29. PI. III. üg. 13 :

PI. VII. fig. 12.
= Cor. officinalis f. typica Setch. et Gard. Alg. N. W. Amer. p.
364.
f. paltonophora Kutz. Tab. Phyc. VIII. p. 33. Tab. 68. fig. 2.
f. flexilis, Kjellm. Alg. Arct. p. 86.
f. robusta, Kjellm. 1. c.

f. mediterranea Kutz. Tab. Phyc. A^III. p. 32. Tab. 66. fig. 3.
f. Farnoesis Kutz. 1. c. p. 33. Tab. 68. fig. 1.

b. Aresch. in J. Ag. Spec. Alg. II. p. 563.

= Cor. angelica Ellis et Sol. Coral, p. 63. t. 24. fig. 3.
= Cor. elongata Ellis et Sol. 1. c. p. 119.

c. Aresch. in J. Ag. Si)ec. Alg. II. p. 563.

= Cor. nodularia Pall. Elench. Zooph. p. 421.

30 ART. 12. — K. YENDO.

= Cor. lorieata Ellis et Sol. Coral, p. 17.

a Yendo. Cor. ver. Jap. p. 28. PL III. fig. 11 : PL VII fig. 10.
? = Corallina officinalis fastigiata Kütz. Tab. Ph3^c. VIII. p. 32.
Taf. 67. fig. 1.
ß Yendo. Cor. ver. Jap. p. 28. PL III. fig. 12 : PL VII. fig. 11.
? = Cor. officinalis elator Kütz. Tab. Pliyc. VIII. p. 32. Taf. 67.
fig. 2.
3 mediterranea Hauck. Meeresalgen, p. 281.

= Cor. mediterranea Aresch. in J. Ag. Spec. Alg. II. p. 568.
= Cor. granifera Kütz. (nee Ellis et Sol., uec Sond.) Spec.

p. 709.
= Cor. densa Kütz. Spec. p. 705.
Yar. chilensis Kütz. Tab. Phyc. VIII. p. 32. Taf. ß6. fig. 1.

= Cor. chilensis Dcne. in Herb. Paris, (sec. Harv. Ner. Austr. p.

103).
= Cor. officinalis o Yendo. Cor. v. Jap. PL VII. fig. 13.
Yar. profunda Farl. Mar. Alg. New Eng. p. 179.
Corallina squamata Ellis et Sol. Coral, p. 117.
= Cor. cupressina Lamk. Mem. Mus. II. p. 233.
= Cor. abietina Lamk. 1. c.

= Amphiroa heterarthra Trev. Flora, no. 27. p. 416. (sec. Trev.).
Cor. Yirgata Zanard. Kütz : Phyc. Gen. p. 397.

= Cor. attenuata Kütz. Tab. VIII. p. 37. Tab. 77. fig. 1.
= Cor. granifera Ellis et Sol. Coral, p. 120. t. 21. fig. c. C.
= Cor. Bertiana De Not. (sec. Freda. Primo Contr. p. 80).
= Cor. gihbosa Kütz. Tab. Phyc. VIII. p. 39. Tab. 82. fig. 2.
Cor. ceratoides Kütz. Tab. Phyc. VIII. p. 36. Tab. 75. fig. 2.
Cor. elegans Lenorm. in J. Ag. Spec. Alg. II. p. 570.
Cor. pectinata Lamk. Mem Mus. II. p. 233.
Cor. pilulifera Post, et Rupr. Illustr. p. 20. t.40. fig. 101.

= Cor. officinalis L. /, pilulifera Setch. et Gard. Alg. N. W. Amer,
p. 366.
f. flabellata Rupr. Tange, p. 344.
f. filiformis Rupr. 1. c.
f. Sororia Rupr. 1. c.
f. intermedia Yendo. Cor. ver. Jap. p. 30. PL III. fig. 16 : PL VII.

fig. 16.
f. arbuscula t

= Cor. arbuscula Post, et Rupr. Illustr. p. 20. Tab. 40. fig. 102.
Cor. capensis Leacii. in Decaisne: Memoir, p. 107.

= Arthrocardia capensis Aresch. in J. Ag. Spec. Alg. II. p. 552.

A REVISED LIST OF CORALLINiE. 31

Cor. sessilis Yendo. Cor. ver. Jap. p. 32. PL III. tig. 18 : PI. VII. fig. 18.

Cor. Berterii Mont. Flor. Chil. (sec. Harv.: Ner. Au.str. p. 103),

Cor. kaifuensis Yendo. Cor. ver. Jap. p. 33. PL III. fig. 19: PL \U.

fig. 19.
Cor. confusa Yendo. L c. p. 34. PL III. fig. 20 : PL VII. fig. 20.
Cor. armata Hook, et Harv. Ner. Austr. p. 103. PL XL.
Cor. Yancouveriensis Yendo. Cor. ver. Port Renfrew, p. 719. PL LIV.
fig. 3 : PL LV. fig. 1-2 : PL LVI fig. 16-17.

= Cor. officinalis f. multiramosa Setch. et Gard. Alg. N. W. Amer,
p. 366.
f. typica Yendo. Cor. ver. Port Renfrew, p. 719. PL LIV. fig. 3 : PL
LVI. fig. 16.
= Cor. officinalis f. multiramosa siibf. laxa Setch. et Gard. Alg.
N. W. Amer. p. 367.
f. densa Yendo. Cor. ver. Port Renfrew, p. 719. PL LV. fig. 1 : PL LVI.
fig. 17.
= Cor. officinalis f.multiramosa siibf. densa Setch. et Gard. Alg.
N. W. Amer. p. 367.
Cor. aculeata Yendo. Cor. ver. Port Renfrew, p. 720. PL LV. fig. 3 : PI.
LVI. fig. 18-19.

= Cor. officinalis f. aculeata Setch. et Gard. Alg. N. W. Amer,
p. 367.

SPECIES DOUBTFUL.

Amphiroa variabilis Harv. Ner. Austr. p. 98.

= Arthrocardia variabilis W. v. B. Coral. Sib. Sxp. p. 106.
Corallina bifurca Kütz. Tab. Phyc. VIII. p. 41. Tab. 86. fig. 3.
Cor. lobata Lamx. Histoir. p. 285.
Cor. cubensis Kütz. Tab. Phyc. VIII. p. 37. Tab. 77. fig. 2.

= Jania cubensis Mont. (sec. Kütz : Spec. p. ' 708).
Cor. Deshaysii Mont, fide Harv. Ner. Austr. p. 103.
Cor. muscoides Kütz. Tab. Phyc. VIII. p. 42. Tab. 84. fig. 5
Cor. polijchotoma Lamx. Histoir. p. 285.
Cor. simplex Lamx. L c. p. 290. t. 10. fig. 4.
Cor. pinnata Ellis et Sol. Coral, p. 117.
Cor. racemosa Kütz. Tab. Phyc. VIII. p. 41. Tab. 85. fig. 1.
Cor. hemisphœrica Fosl. Nye Havsalg. Tromö Mus. Aarsh. X.
Cor. Calvadosii Lamx. Histoir. p. 290.

= Cor. officinalis f. d. Aresch. in J. Ag. Spec. Alg. II. p. 563.
Cor. palmata Ellis et Sol. Coral, p. 118. Tab. 21. fig. a. A.

= Cor. flabellata Kütz. Tab. Phyc. VIII. Tab. 60. fig. 2.

32 ART. 12. — K. YENDO.

The representative member of the present section, Corallina
officinalis L., is universally distributed. The plant undergoes much
morphological change according to the condition of the place
where it grows. Numerous formte have been described and many-
forms were mentioned as distinct species or varieties. The best
authorities, however, agree in the opinion that many of the formie
are local variations and not worth mentioning separately. If a
widely differing form be reduced to the variety or forma, because
of some similarity to Corallina officinalis L., only a few species
among those listed above may retain their specific position. There
is no doubt, as I formerly mentioned, that the forms hitherto
mentioned as Corallina officinalis may have been confounded with
various distinct plants ; or, on the contrary, some of the plants
described as independent species may be reduced to formal or
often to synonymous position. At present, however, the revision
of the specimens of Corallina officinalis, reported from innumer-
able localities, is beyond my power. I must confine myself here
to mentioning the formœ hitherto described with more or less
exact definitions ; and at the same time trying to make the nearest
possible references thereof.

Setchell and Gaedner^^ combined Coriallina pilulifera Post.
et RuPR., Cor. Vancouver iensis mihi, and Cor. aculeata mihi, as
the formte of Cor. officinalis L., and mentioned a number of sub-
formse under them. It is a matter of personal preference whether
the plants with such affinity should be taken as form« or as
distinct species. The plants, however, which have characters
more or less constant and easily distinguishable from the typical
forms of Cor. officinalis L. may be considered as distinct species.

1) Alg. N. W. Amer. p. 366.

A REVISED LIST OF CORALLINE. 33

Coralliiut sessilis is a peculiar plant. It was first described
by the present writer in " Cor. ver. Japon." ]i. o2. Lately I
also found it among a collection of Gorallina from Australia, sent
by Dr. Okamura for determination.^^ The position and the form
of the conceptacles show the characters of Corallina ; and at the
same time are referrible in some respects to both the Serraiicardia
and the Alatocladia. Cf. " Cor. ver. Japon." 1. c. Judging from
the description given by Harvey in Ner. Austr. 1. c, Amphiroa
variabilis Harv., which was described from a sterile specimen,
is probably related to this species.

Areschoug doubted the generic position of Cor. eapensis
Leach., but he provisionally included it under Arthrocardia
Aresch. He did not know the propagating organ of the species.
There are two fertile specimens of "Arthrocardia capensis Aresch."
in Suringar's Herbarium, collected at Cape Agulhas, South
Africa. They differ slightly in minor points but evidently accord
with the description given by Areschoug in J. Ag.'s Spec. Alg.
II. p. 552, and one of them especially with /. ß of Areschoug.
The conceptacles have a peculiar character resembling those of
Gorallina sessilis in general. As there has nothing been reported
concerning the matter, it will not be superfluous to give some
points in detail.

The articuli in the middle and upper portions are cuneate,
more or less compressed above. Each upper corner of an articulus
is mostly extended to a short spinous process. Very often we
find one more process at the external side of the former situated
very close and parallel. Both are never jointed, and are alike
one to another. They elongate further, keeping their relative

1) Bot. Mag. Tokyo. Vol. XVIII. p.

92.

34 ART. 12. K. YENDO.

positions as before, but the inner one becoming gradually thicker
and thicker, till finally it results in a pyriform, short-stiped con-
ceptacle. The external process is now found at the lateral margin
of the conceptacle, fused to the wall of the latter. The apex of
the former is often found much more pronounced than the ter-
minal end of the conceptacle, or, not infrequently, it takes place
just upon the latter and thus the conceptacle seems to have
apiculated.

The apparent difficulty in reckoning Cor. sessilis and Cor.
capensis in the genus Corallina lies in the point that the con-
ceptacles are not jointed. This character may suggest an extreme
form of Alatocladia in some way. We must, however, remember
that there are many conceptacles found in the Officinales without
any joint to separate them from the axial articuli. Cf. " Cor.
ver. Port Renfrew." PL LVI. fig. 16. I believe it will be safest
to group these species under the present section, as I could not
discover any conceptacle upon the surface of the hexagonal articuli
of the axial stems.

Corallina palmata Ellis et Sol. is a doubtful plant, as has
been stated before (p. 26). Grunow mentions Arthrocardia pal-
mata Aresch. in the list of plauts collected during the Novara
Expedition (p. 79) : and remark that the plant is similar to
Cor. flabellata KÜtz. illustrated in Tab. Phyc. VIII. Tab. 60.
The specimen bearing the name of " Corallina palmata Ellis et
8oL.," kept in Suringar's Herbarium, accords very well with
the description of Cor. |9a/ma^a Ellis et Sol. and at the same
time with the figures of Cor. flabellata KÜtz. 1. c. It is far
from doubtful that this species should be better placed near Amp.
corymbosa under the section Arthrocardia.

Most of the other doubtful speciee in the above list are

A REVISED LIST OF CORALLINiE. 35

unsatisfactorily described, or referred to sterile specimens. Cor.
hemisphœrica Fosl. seems an unfixed local from of Cor. offici-
nalis L.

Sect. II. HALIPTYLON Done. (Mut. strict.) Memoir, p. 111.

Corallina Cuvieri Lamx. a Ciwieri Aresch. in J. Ag. Spec. Alg. IL p.
572.

= Corallina Cuvieri Lamx. Expos. Method, p. 24. t. 69. fig, 1-3-14.
= Jania Cuvieri Done. Memoir, p. 111.
= Jania granifera Sond. Plant. Preiss. 11. p. 187.
? = Jania granifera Done. Memoir, p. 111.
ß. crispata Aresch. in J. Ag. Spec. Alg. II. p, 572,
= Cor. erispata Lamx. Histoir. p. 289. t. 10. fig. 3.
= Jania crispata Done. Memoir, p. 111..

= Jania subulata ß crispata Harv. (sec. Sond. Plant. Müll. Cont. p.
522.)
0. subulata Aresch, emend, in J. Ag. Spec. Alg. II. p, 572.
z= Cor. subulata Ellis et Sol. Coral, p. 120. t. 21. fig. B,
= Jania subulata Sonü. Plaut. Preiss. II. p. 180.
= Cor. Cuvieri ß subulata Kütz, Tab. Phyc. VIll. p. 33, Tab, 70,
%. IL
f. denudata Sond, Plant. Müll. Cont. p. 521.

= Cor. denudata Kütz. Tab. Phyc. VIII. p. 34, Tab, 73,
f. densa f

= Cor. gracilis f. densa Collins. Phyc. Bor. Amer. uo. 650,
Yar. calliptera Grun, ISTovara Exp. p. 78.
= Cor. calliptera Kütz, Spec, p. 705,

= Cor. gracilis Collins, Hold, et Setch, Phyc. Bor. Amer. no.
399,
Cor. rosea Lamx. Mém. Mus, 1. p. 235.

=z Jania rosea Done. Memoir, p. Ill,
Cor. pilifera Lamx. Histoir, p. 290,

Cor. trichocarpa Kürz. Tab, Phyc. VIII. p. 35. Tab. 74. fig, 1,
Cor. clavigera Kütz, Tab, Phyc. VIII. p. 36. Tab, 75. fig. 1.
Cor. pistillaris Aresch. J, Ag. Spec. II. 574,

= Jania pistillaris Mont. Pôle Sud. Bot. p. 147.
Cor. Hombronii Kütz. Spec. p. 708.

= Jania Hombronii Mont. Pôle Sud. Bot. p. 146.

36 ART. 12. K. YENDO.

SPECIES DOUBTFUL.

Corallina graeilis Lamx. Histoir. p. 288. t. 10. fig. 1. a. B,

= Jania graeilis Mont. Pôle Sud. Bot, p. 147.
Jania elegans Done. Memoir, p. 111.
Cor. Turneri Lamx. Histoir. p. 289. t. 10. f. 2. a. B.

= Cor. Cuvieri j Turneri Kütz. Tab. Phyc. VIII. p. 34. Tah. 70.
fig. 2. h.
Cor. panieulata Lamx. Freyc. Voy. p. 626.

= Jania panieulata Dcne. Memoir, p. 111.
Cor. plumifera Kütz. Spec. p. 705.
Cor. rosea Lamx. ß crispa Kütz. Spec. p. 708.

This division is characterized by the dense ramuli aggregated
on botli sides of the main branches. Morphologically speaking,
the segments corresponding to the pinnae as well as to the pin-
nules of the Officinales are stunted in the present group, forming
a sort of " kurztrieb," The simpler form suggests close affinity
with the preceding section and there can be no sharp line drawn
between the two. Corallina Cuvieri var. calUftera, Cor. rosea,
etc., of this section are closely related to Cor. Berterii of the
preceding section. Some vexatious and erroneous references, as a
consequence, have been made in the genus. For example, Jania
granifera Sond. was taken by Areschoug^^ as synonymous with
Cor. Cuvieri Lamx. ; Kutzing'^ transferred the former to the
genus Corallina keeping the original specific name : finally Coral-
lina granifera KiJTZ. was referred by Hauck'^' to Cor. officinalis
L. Consequently, if these referrences have all been correctly
made, Cor. Cuvieri Lamx. must be united with Cor. officinalis L.,
or at least there must be a close similarity between both species,
which can not be readily admitted.

J) J. Ag. Spec. Alg. II. p. 572.

2) Spec. Alg. p. 708.

3) Meeresalg. p. 281.

A REISVED LIST OF CORALLINE. 37

Decaisne assigned some of the members of the present section
to the genus Jania. Lamouroux's original deHnition of Jania,
which was followed by many distinguished systematists, can by
no means include the present section. Areschoug wisely reckoned
them under Corallina instead of Jania.

Corallina gracilis Lamx. has been held in doubt by Are-
schoug."^ This species has not been fully defined and is hardly
referrible to other plants without authentic specimens. In Phyc.
Bor. -Amer. '^^ this name is assigned to one of the plants. The
specimen is more applicable to Go7\ Gavieri var. caUvptera. Grun.
than the imperfectly described species of Lamouroux. Collins
mentioned, in no. 650. of the same exsiccate, a plant bearing
the name of Corallina gracilis Lamx. var. densa Collins. But
the plant labelled no. 650 b. in the copy kept in my herbarium
is quite distinct from the type species with regard to the characters
of the articuli of the axial stems. It is related in several respects
to a form of Cor. Cuvleri. It probably belongs to an undescribed
species and must wait further study with a large stock of speci-
mens. The specimen labelled 650. a. contains many fragments of
distinct forms, all of which it is improper to refer to no. 650.

Almost all of the members of this section are inhabitants of
the Australian seas.

Gen. ril.

JANIA LAMX.

Jania adhaerens Lamx. Histoir. p. 270.

= Corallina adhœrens Kütz. Spec. p. 710.
J. decussato-dichotoma f

1) Memoir, p. 111.

2) J. Ao. Spec. Alg. II. p. 572.

3) Collins, Holdkn and Setchell. No. 395).

38 AET. 12. — K. YENDO.

= Cor. deeussato-dichotoma Yendo. Cor. v. Jap. p. 25. PL III. fig.

1-3 : PI. VII. fig. 3-4.
= Cor. adhcerens Kütz. p. p. Tab. Phyc. VIII. p. 40. t. 83.
J. micrarthrodia Lamx. a tenuissima Aresch. in J. Ag. Spec. Alg. II.
p. 555.

= Cor. tenuissima Kütz. Tab. Phyc. VIII. p. 40. Tab, 84. fig. 3.
= Jania tenuissima Sond. Plant. Preiss. p. 39.
f. antennina f

= J. micrarthrodia ß Aresch. in J. Ag. Spec. Alg. II. p. 555.
= J. micrarthrodia Lamx. Histoir. p. 271. t. 9. fig. 5. a. B.
= J. antennina Kütz. Phyc. Gen. p. 389.

= Cor. antennina Kütz. Tab. Phyc. VIII. p. 40. Tab. 84. fig. 1.
J. rubens Lamx. Histoir. p. 272.

= Cor. rubens L. Syst. Natur, ed. 12. Vol. I. p. 1304.
= Cor. dichotoma Ellis et Sol. Coral, p. 65. t. 24. fig. f. F.
= J. cristata Endl. Mantissa. Suppl. III. p. 49.
= J. rubens C cristata Lamx, Histoir. p. 272.
= Cor. cristata Ellis. Coral, t. 27. n. 7.

= Cor. rubens cristata Kütz. Tab. Phyc. VIII. p. 38. Tab. 80. fig. 2.
E concatenata Lamx. Histoir. p. 273.

= Cor. rubens ß concatenata Kütz. Tab. Phyc. VIII. p. 40. Tab.

84. fig. 4.
= Cor. intermedia Kütz. Tab. Phyc. VIII. p. 37. Tab. 79. fig. 1 :

p. 42. Tab. 86. fig. 4.

var. corniculata t

= Cor. corniculata L. Syst. Natur, ed. 12. Vol. I. p. 1305.

= J. corniculata Lamx. Histoir. p.

= Cor. corniculata Ellis et Sol. Coral, p. 121. t. 24. fig. d. D.

= Cor. alba Ellis. Coral, p. 65. t. 24. fig. d. D.

= J. pulumula Zanard. Coral, p. 21.

= Cor. plumula Kütz. Tab. Phyc, VIII. p. 41. Tab. 86. fig. 1.

= Cor. rubens var. corniculata Hauck. Meersalg. p. 279.
J. longifurca Zanard, Coral, p. 21.

= Cor. longifurca Zanard. Iconogr. II. p. 63. Tab. 56.
J. capillacea, Harv. Ner. Bor. Amer. H. p. 84.
J. ungulata f

= Cor. ungulata Yendo. Cor. ver. Jaj), p. 26. PI. III. fig. 7-8 : PI,
VII, fig, 8,
f. brevior f

= Cor. ungulata f. brevior Yendo. 1. c. p. 27. PI. III. fig. 9 : PI.
VII. fig. 9,

A KEVISED LIST OF CORALLINE. o9

J. purpurata Blainv. in Kütz. Spec Alg. p. 710.

= Cor. jmrpurata Lamk. Mem. Mus. II. p. 237.
J. crassa Lamx. Expos. Method, p. 23. t. 69. f. 9-10.

= J. micrarthrodia Lamx. d. Aresch. J. Ag. Spec. Alg. II. p. 555.

= Cor. crassa Coll. Phyc. Bor. Amer, no. 500.
J. nipponica f

^ Cor. nipponica Yendo. Cor. ver. Jap. p. 23. PL II. fig. 20 : PI.
VII. fig. 1.
J. pacifica Aresch. in J. Ag. Spec. Alg. II. p. 556.
J. fastigiata Harv. Ner. Austr. p. 107.
J. natalensis Harv. 1. c.

= Cor. natalensis Kütz. Tab. Phyc. VIII. p. 38. Tab. 79. fig. 2.
J. yenoshimensis f

= Cor. yenoshimensis Yendo. Cor. ver. Jap. p. 23. PI. II. fig. 21-24 :
PI. VII. fig. 2.
J. Novae Zelandige Harv. Fl. N. Z. II. p. 237. (sec. Hooker : Handbook.

p. 2.)
J. affinis Harv. (sec. Sonder : Alg. Austr. p. 21.)
J. spermaphros Kütz. Spec. p. 708.

= J. rubens var. D. spermaphros Lamx. Histoir. p. 272.

= J. rubens var. B. pyrifera Lamx. 1. c.

= Cor. spermaphros Ellis. Coral, p. 122. Tab. 24. fig. g.
J. arborescens f

= Cor, arirescens Yendo. Cor. ver. Jap. p. 25. PI. III. fig. 5 : PI.
VII. fig. 5.
J. tenella Grun. Novara Exp. p. 78.

= Cor. tenella Kütz. Tab. Phyc. VIII. p. 41. Tab. 85. fig. 2.
J. radiata f

Cor. radiata Yendo. Cor. ver. Jap. p. 26. PL III. fig. 6 . PL VII. fig. 7.
J. pusilla t

= Cor. pusilla Sond. Alg. Austr. p. 21.

= Cor. nana Lenorm. (sec. 1. c.)

= Cor. Lenormandiana Grun. (?) ?

SPECIES DOUBTFUL.

J. pygmœa Lamx. Histoir. p. 269. t. 9. fig. 1.

= Cor. pygmœa Kütz. Tab. Phyc. VIII. p. 37. Tab. 78. fig. 3.
J. pumila Lamx. Histoir. p. 269. t. 9. fig. 2.

= Cor. pumila Kütz. Tab. Phyc. VIII. p. 39. Tab. 83. fig. 1.
J. pedunculata Lamx. Histoir. p. 269. t. 9. fig. 3. a. ß.

= Cor. pedunculata Kütz. Tab. Phyc. VIII. p. 37. Tab. 78. fig. 2.

40 ART. 12. K. YENDO.

J. verrucosa Lamx. Histoir. p. 269. t. 9. fig. 4. «. ß.

= Cor. verrucosa Kütz. Tab. Phyc. VIII. p. 38. Tab. 80. fig. 3.
J. parvula Zanard. Coral p. 21.
J. gibbosa Lamx. Histoir. p. 269.

J. compressa Lamx. Freyc. Voy. p. 624. t. 90. fig. 8-10.
Cor. constricta Kütz. Tab. Phyc. VIII. p. 40. Taf. 84. fig. 2.
Cor. tridens Kütz. Tab. Phyc. VIII. p. 41. Taf. 85. fig. 3.

This genus is sharply distinguished from the others by the
great delicacy of its fronds and the regular ramification. The
greatest number of the species ramify dichotomously in a plane.
J. decussato-dichotoma branches in an aberrant manner as was
described in my former paper.^^ And it is highly probable that
this species has been hitherto reckoned among J. adhcerens Lamx.
There are many species closely related to J. adhcerens Lamx.,
found in the warmer parts of the Atlantic and the Pacific Ocean,
such as J. capiUacea, J. micrarthrodia f. antennina, J. micrarthro-
dia f. tennissima, J. tenella, J. novœ- Zelandce and J. ungulata.
The essential distinguishing character lies in the relative size
of the articuli. It is a question whether or not it is worthy to
serve as a specific character. /. adhœrens Lamx. seems to undergo
some change in the minor character. This species stands in the
present genus in the same relation as Corallina officinalis L. in the
Officinales.

The articuli are, in the majority of the genus, cylindrical, a
few only have compressed ancipitous, and some, slightly com-
planated articuli. The pinnate ramules, as has been already
remarked, are frequently found at the basal and middle portions of
J. rubens var. corniculata and J. micrathrodia f. antennina. At
the base of certain species a sort of a delicate ramulet is found in a
form of the rhizoidal processes. These ramulets are proliferated

1) Study of Genicula. p. 8.

A REVISED LIST OF CORALLINE. 41

from the lateral surface of the basal articuli and seem to be of the
same nature as those found in some Corallina.

J. adhœrens, J. pusilla, J. radiata, etc., are always found
epiphytic upon other algse. The first one and some of its allies are
mostly upon Digenia smplex Ag. and the latter two upon Sar gas-
sum, Gysto'phyllum, Cystophora or other brown alga*.

I have several specimens of J. pusilla Sond. kindly sent by
Major Reinbold, Prof. Miyabe and others. They seem to have
been distributed from the herbarium of Dr. F. von Müller, and
are named " Corallina Lenormandiana = Gor. nana Lenorm. nee
Zanard." The author of the species was not mentioned with any
of the specimens. I am not certain where Lenormand mentioned
the plant. In asking Mr. Eeinbold about the matter a few years
ago, he answered that he thought the author of Cor. Lenorynandiana
is probably Grunov^^ but not quite sure. So far as I could
ascertain, the specific name has never been published.

Tokyo, May, 1905.

42

ART. 12. K. YENDO.

SPECIES INDEX.

Page.

Amphiroa aberrans Yendo 8

acutiloba DcNE 19

algeriensis Kütz 5

(unethystina Zanard 3

anastomasans W. v. B 2

anceps DcNE 4

aspergillum Anders 14

/. nana Setch. et Gard.... 14

aiistralis Sond 16

Beauvoisü Lamx 4

ß crassiuscula (Harv.) f ... 4

Boviesii Kütz 5

Bowerbankii Harv 5

brasiliana Dcne. 5

bnviarticulata AviESCH 10

californien Dcxe 19

caiialiculata Mart 5

canaliculata Yendo 5

capensis Dcne 10

Cham Dcne 12

Charoides Lamx 12

chilensis Dcne 21

dadoniœformis Menegh 3

complanata KÜTZ 5

contracta Kütz 8

corymhosa Harv 8

corymbosa Dcne 7

a. Aresch .'. - 8

ß. Aresch 8

crassa Lamx 5

f. GodeöVoyi W. v. B 5

f. minuta W. v. B 5

crassissima Yendo 10

cretacea Endl 10

f. capensis Yendo 10

f. losariformis Yendo 10

var. tasmanica Kürz 10

f. breviarticulatat 8

cryptarthrodia Zanard 3

ß verrucnlüsa Hauck 4

Page.

Amphiroa cultrata Harv 18

ß dehilis Kütz 18

Y gJobulifera Kütz 18

5 pectiuata Kq 19

Ciuningii Rüpr ö

Cumingii Mont o

euspidata Lam 3

cyathifera Lamx 3

Darwinii Harv. ..'.......... 20

dehilis Kütz 2

declinata Yendo 10

dilatata Aresch. 4

dilatata Krauss 5

dubia Kütz 10

echigoensis Yendo 4

elegans Sond 12

elegans Hook, et Harv 18

ephedrsea Dcne 4

a fusoides (Aresch.) t 4

ß Gailionii (Aresch.) f ... 4

epheirœa Harv .^.. 4,5

epiphlegmoides J. Aq 21

exilis Harv 4

ß crassiuscula Harv 4

fastigiata Dcne 19

firma Kütz 21

flabellata Harv 20

foliacea Lamx 5

f. procmnbens W. v. B. ... 5

f. erecta W. v. B o

fragilissima Kütz 3

fragilissinia Lamx 2

f. fragilissima W. v. B 2

f. euspidata W. v. B 3

f. cyathifera W. v. B 3

soldes Lamx 4

Gailionii Krauss 4

GaiUonii Lamx 4

galaxnuroides Sond 4

Godeffroyi Grün ö

A REVISED LIST OF CORALLINJE.

idt

", • Page.

Amphiroa gracilis IIarv 12

; granifera IIarv 12

< Guenzii Hahv 4

heterarthra Trev 30

Iieterocladi:i KÜTZ 19

Hookvri IIarv 20

inordinnta Zanard 3

intermedia Harv 12

interriipta Lamx 12

jnvoluta KÜTZ 3

irregularis Kütz 3

juhata Lamx 12

Karstarski Rupr. 5

Lamourouxlana Done 19

linearis Kütz 5

lucida Lamx 10

Mallardiie IIarv 8

misakiensis Yenpo 5

multißda Kütz 19

nonlis Kütz 5

nohilis Hauck 5

noduluria DcNE 10

nodalosa Kütz 4

nodidosa Farl 14

nodulosa Collins, Holdex et

Setch 14

Orbigniana Dcne 20

partiienopea Zaxard 4

Poppigii Endl. et Dies 4

proliféra Dcne 21

))usilla Yendo 5

pustulata Mart o

rigida Lamx 3

var. inordinata Zanard. ... 3

sagittata Dcne 18

setacea Kütz 13

si?uilis SoND 13

spina Kütz 3

ß amethystina Kütz 3

squarrosa Grün o

Stangerii IIarv 20

stelligera Aresch 12

ß interriipta Kürz 12

stelliilata Kürz 12

tasmanica Sond 10

tribulus Lamx 3

tuberculosa Endl. 20

/. calif arnica Setgh. et Gard. 19

Page.
Ampliiroa tuberculosa f. frondßscens

Setch. et (jard. , 20

/. planiuscula Setch. et

Gard 20

/. typica Setch. et Gard. 21

valonioides Yendo 2

variabilis Harv 31

ventricosa Lamx 4

verrucosa Lamx 12

verruculosa Kürz 4

vertebralis Dcne 21

Wardii Harv 8

zonata Yendo 4

Arthrocardia capensis Akesch 30

corymhosa Aresoh 8

cretacea W. v. B 10

Darwini \V. v. B 20

frondescens Aresch 20

Mallardiœ Aresch 8

palmata Aresch 20

palmata ß Aresch 20

tuberculosa W. v. B 21

tuberculosa Dcne 21

variabilis W. v. B 31

vertebralis W. v. B. 21

Wardii Aresch 8

Cheilosporum anceps Yendo 20

var. modesta Yendo. 20

col i fern i cum Yendo 19

cultratura Aresch 18

ß debilist .., 18

S pectinata t 19

f. niultifidaf 19

Darwini Y''endo 20

elegans Aresch 18

filiculnmt ., 20

var. planiusculumf ••• 20

flabellatuiii Aresch 20

frondescens f. typica Yendo. 19

f. lUîixiina Y'"endo 20

f. intermedia Y'endo 20

• f. polymorpha Y'eni^o 20

jungermannioides RuPR 18

latissimum Y'endo. 20

McMillani Yenpo. 26

maximum Y''endo 26

Orbignianumf -'f 20

palmatum f. filicidum Yendo. 20

44

AKT. 12. — K. YENDO.

Page.

Cheilosporum planiuseulum Yendo 20

pulchrum Harv 19

sagittatura Abesch 18

spectabile Harv 18

ötaiigeri Aresch 20

tuberculosum f 20

yessoense Yendo 20

f. angusta Yendo 20

Corallina dbietina Lamk 30

aculeata Yendo 31

adhœrens K-VTZ 37

alba Ei.Lis 38

aneeps Lamk 4

aneeps Kütz 20

angelica Fllis et Sol 29

ante nni na KiJTZ 38

arboreseens Yendo 39

arbuM'ula Post, et Rupr 30

arbuscula Kütz 8

armata lIooK et Harv 31

ort iCHZa^a Ellis 29

attenuata Kütz 30

Berterii Mont 31

Bertiana De Not 30

bifurca KÜTZ 31

calliptera Kütz 35

Calvadosii Lamx 31

capensis Leach 30

carinata KÜTZ 21

ceratoides KÜTZ 30

Chara Lamx 13

Chara Lamx. y 4

chilensis Dcne 30

clavigera Kütz 35

confusa Yendo 31

constricta Kütz 40

cornieulata L 38

eorniculcita Elt.is et Soi,. ... 38

eorymhosa Lamk 8

eras^a Co LLiNH 39

eretaeea Post, et Küpr 10

crispata Lamx 35

cristata Ellis 3>8

cubensis Kütz 31

cupressina Kütz 30

cuspidata Ellis et Sol 3

Cuvieri Lamx 35

f. densa t 35

Page.

Corallina Cuvieri Lamx. f. denudata Sond. 35

« Cuvieri Aresch 35

ß crispata Ahesch 35

ß subulata Kütz 35

S subulata Aresch 35

Y Turneri Kütz 36

var. calliptera Grün 35

deeussato-dichotoma Yendo.... 38

densa Kütz 30

denudata Kütz 35

Deshayii Mont 31

dichotoma Ellis et Sol 38

dilatata Lamx 4

elegans Lenorm 30

elongata Ellis et Sol 29

ephedrœa Lamk 4

Filicula Lamk 20

Filicula Kütz 21

f. raraiaossima KÜTZ 8

flahellata Kütz. et var 31

fragilissima Ellis et Sol. ... 2

frondeseens Post, et Eupr. ... 19

frondescens Kütz 8

Galioides Lamk 12

gibbosa Kütz 30

gomphoneniacea Kütz 8

gracilis Collins, Hold, et

Setch 35

gracilis Lamx 36

/. densa Collins 35

granifera Kütz 30

granifera Ellis et Sol 30

heniispliperica Î^OSL 31

Huir.broni Kütz 35

intermedia Kürz 38

kaifuensis Yendo 31

Lamouj'ouxiana Leach 19

laxa Lamx 29

Lenoruiandiana (!) 39

lobata Lamx 31

longicaulis Lamk 29

longifurea Zanard 38

loricata Ellis et Sol 30

magnifica Leach 8

inediterrania Aresch 30

muscoides Kütz 31

nana Zanard 29

nana Lkxorm 39

A REVISED LIST OF CORALLINE.

45

Page. I
Corallina natalensis Kütz 39

nipponiea Yendo o9

nodularia Pall '29

nodidavia Lamx 10

officinalis L 29

a Aeesch 29

a Yendo 30

b Aresch 29

ß Yendo 30

ß mediterranea Hauck. ... 30

c Aresch 29

Y Yendo 30

d Aresch 31

0 Yendo 31

/. aculeata Setch. et Gard. 31

/. elator Kütz 30

f. faroensis Kütz 29

/. fastigiata Kütz 30

f. flexilis Kjellm 29

f. mediterranea KÜTZ 29

/. muUiramosa Setch. et

Gard 31

S!(ö/. ?axa Setch. et Gard. 31

suhf. densa Setch. et

Gabd 31

f. paltonophora Kütz 29

/. püulifera Setch. et Gard. 30

f. robusta Setch. et Gard. 26

f. robusta Kjellm 29

f. typica Kjellm 29

/. typiea Setch. et Gard. 29

/. vulgaris Kütz 29

var. chilensis Kütz 30

var. profunda Fai^l 30

palmata Ellis et Sol 31

paniculata Lamx 3(i

pectinata Lamk 30

pedunculata Kütz 39

pilifera Lamx •">'>

pilulifera Post, et Rupk. ... -'îO

f. arbuscula Y''endo 30

f. filiformis RüPR 30

f. flabellata RuPR 30

f. intermedia Y'endo 30

f. Sororia Rufe 30

pinnata Ellis et Sol 31

pisitillaris Aresch 35

planiuscula Kütz. et ff. 20

Pago.

Corallina plumifera Kütz 36

plumula KÜTZ ;'>H

polyclioloma Lamx 31

proliféra Lamx 21

pumila KÜTZ 39

purpurata Lamk 39

pusilla SoND 39

pjjgmœa Kütz 39

racemosa Kütz 31

radiata Lamk 13

radiata Yendo :'>9

rigens Pali 2

rigida Kütz S

rosarium Lamk 10

rosea Lamx 35

ß crispa Kütz 36

riibens L 38

ß eoneatenata Kütz 38

var. cristata Kütz 38

var. eorniculata Hauck IW

sagittata Lamx 18

sessilis Y'endo 31

simplex Lamx 31

spathulifera Kütz 29

spermaphros Ellis 39

squamata Ellis et SoL 30

stelligera Lamk 12

subulata Ellis et Sol 35

tenella Kütz 39

tenuissiina Kütz 38

tribulus Ellis et Sol 3

trichocarpa Kütz 35

tridens Kütz 40

tuberculosa Post, et Rupk. ... 20

Turneri Lamx 36

ungulata Yendo 38

/. brevior Yendo 38

vancouveriensis Y'endo 31

f. typica Yekdo 31

f. densa Yendo 31

verrucosa Kütz 40

virgata Zanard 30

yenoshimensis Yendo 39

Cymopolia rosarium Lamx 10

Jania adhserens Lamx 37

affinis Harv. ^9

antennina Kütz 38

arborescensf '■^'^

46

rARÏ. 12.— -K. YEKDO.

Page.

Jania capillacea Haev. 38

compressa Lamx 40

cornieulata Lamx 38

crassa Lamx 39

crispata Dcne 35

eristata Endl 38

eiibensis Mont 31

Cuvieri Done 35

deciissato-dichotoina t 37

elegans Dcne 36

fostigiata Harv 39

gibbosa Lamx 40

gracilis Mont 36

granifem Sond 35

Hombronii Mont 35

longifurca Zanaed 38

mierarthrodia Lamx 38

a tenuissima Aresch 38

ß Aresch 38

8 Aresch 39

f. anteanina t 38

natalensis Harv 39

nipponica Yendo 39

Novie Zelandiae Harv 39

pacifica Aresch 39

jmniculata Dcne 36

parvula Zanard 40

pedunculata Lamx 39

pistillaris Mont 35

plumula Zanard 38

Page.

Jania puniila Lamx...... 39

purpnrata Blainv 39

pusilla t 39

P3'gu5a3a Lamx 39

radiataf 39

rosea Dcne 35

rubens Lamx 38

B. pyrifera Lamx 39

C. eristata Lamx 38

D. spermaphros Lamx 39

E. concatenata Lamx 38

var. cornieulata L 38

spermaphros KÜTZ 39

subulata Sond 35

ß crispata Harv 35

tenella Geun 39

tenuissima Sond 38

ungulataf 38

f. brevior Yendo 38

verrucosa Lamx 40

yenoshimensis t 39

Litharthron australe W. v. B 16

Lithothrix Aspergillum J. Gray 14

f. nanaf 14

Metagoniolillion Charoides W. v. B 12

gracile t 1^

graniferum W. v. B 12

stelligerum W. v. B 12

Rhodopeltis australis Schmitz 16

riï.~-ai.a'Sj~

iTjrtit.vrA.

p. 4. Insert next to " a I'usoides (Abescii.) "

= Amp. Gaillouii Krauss. Beitr. Flor. Caput. p. 206.
1'. 5. for " = Amp. Gotldefroyi Grun." read " = Amp. Godeffroyi Grux."
P. 8. for " ^- Ampiroa corymbosa Harv." read " = Amphiroa corymbosa Hakv.
,, I'or "Amphiroa eoustricta Küxz." read "Ampbiioa contracta Kütz."
,, insert next to " Corallina froudescens Kütz." etc.

Coraliina Filicula f. ramosissima Kütz. Sjiec. p. 707.
1". 2'J. for " f. Farnoensis Kütz." etc., read " f. faroensis Kütz." etc.
P. 3S. for " -- J. pulumula Zakard." read "^ J. plumula Zanard."