CONTENTS.

Pars 1.

(Published Aug. 28, 1901.)

PAGE

Negative Phototaxis and Other Properties of Littorina as Factors

in Determining its Habitat.—K. MITSUKURT. ..................... i

On the Lampreys of Japan together with Notes on a Specimen from

Sera ey CELA AR, O Ri al.

Gonyaulax and the Discolored water in the Bay of Agu.—

RESTA A tee > Selena. oR MR I 31.

Reminiscence of Holoblastic Cleavage in the Egg of the Shark,

Heterodontus (Cestracion) Japonicus Macreay. Tab. I.—

La ANI e AO TRA RE 35.

On the Relation of the Metameric Segmentation of Mesoblast in

Petromyzon to that in Amphioxus and the Higher Craniota.—

DREAM. a Jah ANAL REEL RANCORE 43,

Pars II.

Published March 22, 1902.

Uber das Blutgefässsystem der Hirudineen.—A. OKA................... 49.

On the Habits of the Japanese Lingula.—N. YATSU. .................. 61.

The Salmon and Trout of Japan.—D. S. JORDAN. ..................... 69.

On a new Enteropneust from Misaki, Balanoglossus Misakiensis

MENO Ce. RM iii Tale

ii CONTENTS.

Pars III

Published Aug. 12, 1902.

Notes on a Specimen of Amphitretus obtained in the Sagami Sea.

Tab. DIE I toms et S. TKEDA eee nn... 85.

Pars IV?

Published Dec. 30, 1902.

Preliminary Notes on Cœloplana.—J. F. ABBOTT. ..................... 103.

On Two New Species of the Family Maldanidæ from the Sagami

Bay. Tab, III:—A. IZUKA 22.0... RE 109.

On the Occurrence of Phoronis australis Haswell near Misaki.

Tr IREDA! 1.2.2.2 SONORE ee moe

Note on Walteria leuekartı Is. SING eee eee 119

Pars V.

Published Sept. 25, 1903.

Bosminopsis in Japan. Tab. IV.—E. KLOGKE 123.

On a New Polygordius from Misaki (P, Ijimai n. sp.) A. Izuka... 137.

On the Development of the Sexual Organs and of Their Products in

Phoronis. Tab. V.—I.-IKED&. ‘...:.. See 141:

ESTAS:

= 2 BA B

Fr À 6 IU $

FR ©

=+BAEM+=%M

5 H

ANNOTATIONES

ZOOLOGICA, JAPONENSES

Vol. LV; Part.

PUBLISHED

The Tokyo Zoological Society

TOKYO

August 1901.

11:

SEPE,

”

IV.

= =

Factors in i its Habitat.

By K. Minsorunt. a ra

On the SU of Japan together win Notes

men from Siberia i et

CFO CMA CE A e Or

A By. Sam. era

| Gonyalax a and the Discolored Water in the eBay Age | 2

Reminiscence of: Holoblastie —

OCT 24 1901

Negative Phototaxis and other Properties of

Littorina as Factors in Determining its

Habitat.

K. Mitsukuri.

Professor of Zoölogy, Imperial University.

During my stay at the Marine Biological Station, Misaki in the

summer of 1900, I spent a part of my time in making some experiments

on Littorina, and the results embodied in the following pages are pub-

lished with the hope that they may prove of some interest to those who

are engaged in similar studies. I wish to express my thanks to Mr. N.

Yatsu avd Mr. T. Tsuchida for various assistance rendered during the

progress of the experiments.

The species of Littorina that is commonest about the Station is

L. exigua DUNKER. There is a second species also very abundant, L.

sitchana PHIL. of which the var. brevicula is very conspicuous. Both

species live from the highest tide-level down for distances equal to a

vertical height of about two to three feet. They cease to be found long

before the lowest tide-mark is reached which is perhaps four or five feet

still further down. The experiments recorded were made mostly on L.

erigua, as this is an easily diagnosed species and as it was desirable that

the experiments should be performed on one and the same species. I

shall not attempt to describe all the experiments performed but only such

as are Important in bringing out the salient points.

After the fact that Littorina shows a strong negative phototaxis was

established, I performed the following experiments :—

Experiment 2 *—The following arrangements were fixed up. A

* 'l'he number is that used in my note-book.

2 K. MITSUKURI.

glass-plate A (Fig. 1) which happened to be semi-circular in shape with

the diameter of about one foot was placed on a table. This was made

slightly wet so that the molluse could easily crawl over it. On this, a

glass-vessel B, about 10 cm. in diameter and 3 cm. in height was placed

and filled with water to the depth of about 1 cm. Various articles C, D.,

F, and G among which a small dissecting-dish C inverted and tilted

up by a small box D was very conspicuous were placed in the quarter

toward which the animals were expected to move.

100 individuals* out of about 200 which had previously been used

N in a similar experiment were pick-

ed up at random and placed in

the glass-vessel B. Light was

shining from the two windows N

and E placed at a right angle to

i each other. This was at 2 P.M.

Va À (Aug 6). Almost immediately in-

dividuals started in the direction

ofthe arrow BH which represent-

ed the direction of the least light,

climbed over the edge of the

glass-vessel B and crept over the

glass-plate A in the same direc-

tion. Before long the vessel B

was nearly empty. At 5 P.M.

three and half hours after the

commencement of the expari-

ment, the individuals were disposed as follows :—

Remaining in the dish B.

At the spot a

At the spot d

At the spot c

D + + ©

* It is possible that in earlier experiments there was some mixture of the individuals

of LL. sitchana.

NEGATIVE PHOTOTAXIS ETC. OF Littorina. 3

At the spot d ON PERS is ie u

PR BCDC Ne n . ... dl

Atthe spot f ... "... e... I

Under the inverted er vesting: Aish €. Se ET ie ana, CA MTS

RG tal RP Re RE. 7 en n aa (0) |

The whole arrangement was left intact during the night, but when

observed next morning (Aug. 7), no change was visible, In the course

of the same morning, one more got out of the glass-vessel B, and placed

itself at the spot g.

Nearly all the individuals that got out of the glass-vessel B, did so

between the lines marked x and y, i. e. at the arc turned toward the

darkest corner of the room. Only three or four got out at the other

points (e and f).

It is clear from the above table that by far the largest majority (90%)

of the individuals started toward the part that appeared darkest, and a

large portion of this (78 individuals) hid themselves under a shaded

cover.

Experiment 3 (Aug 6):—A number of individuals was placed in

a glass-vessel (similar to that marked B in Exp. 2). One side of this was

then tilted up by a number of glass-slides (Fig 2) so that the water in the

vessel was deepest toward the darker side of the room. The result was

that most of the individuals got over the edge at the darker arc in spite

of deeper water on that side.

After making several more experiments of a more or less similar

nature and purpose as the above two, it occurred to me that this property

* One of these got out of the glase-vessel B at the north-west point, crawled about on

the glass-plate toward the north and then finally came here.

4 K. MITSUKURI.

of negative phototaxis shown by these molluscs must necessarily lead

them to move landward instead of sea-ward in their native habitat, and

must lead them to act as if they had a notion as to which way lay the

land and which way the water. Accordingly I walked along the shore-

line, carefully observing how light struck my eyes. There could be no

question as to which gave out more light: wherever 1 might put myself

along the shore, (except on detached rocks) there was always more light

facing the sea than when facing landward. In order to test this point

further, I tried the following

Experiment 5 (Aug. 7):— An oblong flat glass-dish was made by

erecting narrow strips of glass along the

edges of a window pane about 35x30 cm.

and by holding them there by putting a

mass of putty outside them. This I took to

Naközaki, a place near the Station. The

a a) E È

i dish was placed on the rocky and sandy

| beach with its longer axis as à normal to

| | the beach-line and as nearly horizontally as

w ee S possible (Fig. 3). The glass-dish was made

just a little wet by having a thin layer of

sea-water. 100 individuals were picked off

| at random from the rocks near about and

Fig. 3. placed in the center of the dish. The time

was 2:48 P. M. (Aug. 7) and the midsummer sun was pouring down

its rays from overhead equally all over the plate. At 4:30 P.M. the

individuals were distributed as follows :—

The quarter with theicormer a. "ME | a

The quarter with the:comer db è Ve eee

The quarter with the corner © LOS 66

The quarter withthe corner d |. ARM AZZ

7 90= 90

Missing ali où. ic OS re

Total... "OMR

NEGATIVE PHOTOTAXIS ETC. OF Littorina. 5

Thus the half of the plate turned toward the land which appeared

naturally darker on account of bushes, grasses etc, had nearly three

times as many individuals as the half turned toward the sea. When it

is recollected that the sun was at the time shining from the zenith, the

reaction must be considered as very decided. It was evident that

whether these molluscs have other means or not, the property of

negative phototaxis alone must induce them to crawl landward and

to make them behave as if they knew the differences between the land

and the sea.

There was another series of experiments which I shall detail next.

Experiment 7 (Aug. 7):— In a square aquarium about a foot

on each side (Fig. 4), two

glass-plates A and B* were

placed, forming an in-

clined plane as shown in

the annexed woodcut of

the side-view. A third

glass-plate C was put on

the upper edges of the

aquarium against the in-

clined plane so as stop

any mollusc that may

come up the incline. The

whole was placed so that

light was falling on the

inclined plane in the direc-

tion of the arrow in the Fig. 4.

The upper figure is the horizontal view ; the lower

figure is the side-view.

lower figure.

A large number of shells was placed at the spot f on the bottom of

* There was no reason why two plates should be used instead of one. This was

rather a drawback than otherwise. Moreover one of the plates, B. was not a whole one.

Only these were the best arrangement I could make at the time. I must apologize for

the erudity of arrangements in my experiments.

6 K. MITSUKURI.

the aquarium. It was intended that water poured drop by drop from

above should gradually submerge them and their action under the condi-

tion was to be observed. But the aquarium and the glass-plates

being somewhat wet, the shells all began to march up the inclined plane,

as soon as placed there, toward the dark side and was only stopped

by the third glass-plate C.

(In another experiment (Exp. 9) with exactly the same arrangements

as the above, every thing was thoroughly dried and the shells were also

dried in the sun. 200 individuals were placed at 11: 80 A. M. (Aug. 8)

at the point f and water was poured to a slight depth. At 4 P. M.

distribution was as follows :—

3) DIS 3 a (These were der

remaining at f... … .…. … … … 12 er: half dead ones.

AG the corner den NN 3

At the corner e...

13 Several of these had =

At the vertical back-wall Cc. \bably slid off the plate B.

Crept up the incline and stopped at ab ... 167

Total RO Ea eee 2 D

This shows that an overwhelming majority goes up the incline toward

the dark side.)

On the next morning (Aug. 8) the position of the aquarium was

reversed, bright light being made to strike on the abe side. This induced

some that were at ab to move down the incline and to creep across the

depth of water and to gain the de side. It was noticed that in coming

down, as soon as they struck water, most of them hesitated and crawled

horizontally along the water’s edge. They showed reluctance to enter

deep water, and seemed only driven by negative phototaxis which

appeared to be a stronger instinct with them. Some indeed crawled

straight across to the de side. Some others, while crawling horizontally

at the water’s edge got unfortunately on the under side of the inclined

glass-plates. Most of the individuals, however, did not stir aed held on

to the abc side. Even the bright sun-rays induced only few to move down.

NEGATIVE PHOTOTAXIS ETC. OF Littorina.

The foregoing experiments showed that it was easy enough to drive

these molluscs up an incline, i. e. from the sea toward the land but nothing

indicated what induced them ever to come down again in state of nature,

for the change ot the light and dark sides, such as was brought about in

Experiment 7, can never take place in the native habitat of these

molluses. After these experiments, therefore, two questions confronted

me :—

(1) What makes these mollnses scatter themselves somewhat

uniformly over rocks in state of nature, instead of being

amassed together at the highest tide-mark as at the line ad

in the Experiment 7 ?

(2) What induces these molluses to come down toward the sea

after they are once driven up toward the land ?

The first question was comparatively easy of solution, while the second

puzzled me considerably. ‘The following experiments will elucidate

these points.

xperiment 10 :— In Experiment 9, there happened to be two

glass-slides stuck on the glass-plate B, and it was noticed that these

obstacles stopped for more or less lengths of time some of the moliuses

that were going up the incline, and one actually settled there. Acting

on this bint I made the following changes in the arrangement. I took the

smooth giass-plates A and B (replacing B with a whole one) and poured

on them plaster of Paris mixed with coarse sand making an irregular

uneven rough surface as nearly resembling the rocks on which these

molluscs live as I could make them. These were then placed as before

inclined against one side of the square aquarium. About 150 individuals

were placed at the spot f as before, and a little water was poured until

the bases of the glass-plates were just touched. ‘The result was very

striking. The molluscs started up the incline as before, but instead of

marching straight up to the line ab and of being stopped only by the

glass-plate C, they now scattered themselves all over the glass-plates A

and B. "They availed themselves of the nooks and corners produced by

the rough surface and settled down in them. Fig. 5, A shows roughly the

5 K. MITSUKURT.

Fio. 5, A; — The dotted line marks the level of water, and the black dots show the

ST positions in which the molluses settled down.

unevennesses on the glass-plates A and B, and how the moliuscs settled

themselves on this first day. It will be noticed that they are most

abundant in valleys and were more numerous in the lower half.

On the next day no spontaneous changes were visible : so the level

of the water was raised to that indicated by the dotted line in Fig. 5, B.

All the molluscs that were threatened with submergence marched upward

again ; a few were finally left at the water’s edge but none below it.

On the third day, the level of water was raised to that given in Fig. 5, C.

The molluses shifted their positions upward for the third time. A

NEGATIVE PHOTOTAXIS ETC. OF Littorina 9

Fig. 5, B.

colin diminumon observed in the number of individuals is due to two

causes: one was that a wet roll of paper was kept at the junction of the

glass-plates A and B with the glass-plate C and many reaching there hid

themselves in the folds of the roll and could not be observed. Secondly,

many slid off the glass-plates at .the sides and went to the vertical back-

wall c.

This experiment seemed to answer the first of the two questions given

above. It was evident that the reason why the mollsues scatter them-

selves uniformly over rocks is because the rocks are very uneven and

rough, and there were nooks and corners in which the animals could hold

0 K. MITSUKURI.

Fis 5 C:—The break in the dotted line is due to a slight shifting of the glass-plates. The

PUP ONE e loop indicates that the molluse on that elevation was above the water-level.

and conceal themselves.

The conditions given in Fig. 5, C was left intact for two days.

During the interval, not a single mollusc stirred. On the fifth day (from

the beginning of the experiment), water was again lowered to the level

of Fig. 5, B. with the hope that there might be some downward move-

ments of the molluscs. But none stirred, and although the whole was

left untouched for four days more, no movement was visible.

Under what conditions, then, the molluses moved downwards, re-

mained as much of a mystery as ever. I was greatly puzzled. At one

NEGATIVE PHOTOTAXIS ETC. OF Littorina. INI

time, I thought, these molluscs left by the receding tide and dried by the

sun were mechanically blown down by wind or washed down by rain.

But this idea could not be entertained for any length of time, as the

following facts show. I marked out with paint areas on rocks around

certain counted numbers of individuals, which were also marked with

paint. On visiting these areas day after day, it was noticed that some

individuals had clearly moved downwards out of the area. Here is an

example of such an observation :—

Station XI. Established Aug. 30 A. M. 20 individuals within the

circumscribed area.

Aug. 31, A. M. No change.

Sept. 1, A. M. Tide has touched the area. Five individuals

have wandered out of the area but none more than 8 cm.

3 of these have moved downwards below the area.

Explanation must then be sought elsewhere. Meanwhile I had

accidentally discovered the fact that if a stream of water is dropped

rather rapidly from above on these molluses, som» of them begin to move

down, after the stream is stopped.

Here is a case. :—

xperiment 11 (Aug. 29) :—Arrangements as described in Experi-

ment 10. In fact they had stood over from that experiment and the in-

dividuals were scattered over the artificial plaster-of-Paris rock as in

the last phases of that experiment. I now rigged up a siphon arrange-

ment by which water could be poured on the molluses. Dropping of

water was begun at 8 : 30 A. M. and continued for 20 minutes. ‘There

were about 56 individuals subjected to the action of the stream. When

the stream was stopped, 15 individuals moved distinctly downwards.

Most of these 15 descended to the water’s edge, some rather sinuously, and

then wandered horizontally along that edge. 2 made a loop and went up

again. 6 unfortunately slid off the glass-plates and went to the vertical

back-wall c. About 12 of the 56 individuals moved upwards.

The stream, therefore, induced some to move down toward the light-

12 K. MITSUKURI.

er side. The same stimulus, however, caused some others to go further

upward toward the dark.

Another case :—

xperiment 12 (Aug. 30) :—The same arrangement as before. Water

sprayed on for 30 minutes on a number of individuals. After 10 minutes

for another 3 minutes. After 40 minutes the reaction nearly ceased.

Then water sprayed another 3 minutes.

During that time, 2 individuals were observed to go upwards.

Several individuals were seen moving down.

One of these down-moving ones was picked up and as quickly as

possible placed in the center of the flat dish described in Experiment 5.

It crawled toward light, i. e, was positively phototaxic. The dish was

turned around 180 degrees. The individual turned around also and

resumed its positive course. It made no difference whether the light side

of the flat dish was lifted and made higher than the darker side.

A second individual which was also moving down the incline was

quickly put in the center of the same dish. This became negatively

phototaxic. The process of moving this one had probably disturbed the

condition, whatever that may be, under which it was moving down the

incline in its positive course. ‘The first individual and this one were

moving in two opposite directions on the same dish, at the same time.

A third individual also moving down was similarly transferred to

the flat dish. It became negatively phototaxic. On being placed back on

the incline it marched upward.

A fourth also going down was similarly transferred. This was

positively phototaxic. Put back on the incline it went down.

A fifth individual, going down, removed on the flat dish was posi-

tive.

A sixth (L. sifchana var. brevicula), going down, removed on the

flat dish, negative.

A seventh, going down, removed on the flat dish, positive.

NEGATIVE PHOTOTAXIS ETC. OF Littorina. 13

An eighth, going down, removed on the flat dish, negative.

A ninth, (L. sitchana, var. brevicula), going down, removed on the

flat dish, at first turned about as if it could not make up its mind, and

then became negative.

The results of these experiments were confirmed by another.

Aug. 30, 8 A. M. Went to Nakozaki. A group of Littorina

lying attached to a nearly vertical face of rock, just above

the water level was chosen, and water was splashed on with

hands for about five minutes. On splashing being stopped,

many began downward movements and reached water’s

edge in about half an hour. One went upwards.

Two other lots of these shells (A) and (B) near highwater mark

were selected. First, some individuals from the both

groups were tested in water and found to be negatively

phototaxic. After this, water was splashed on them, on

one (A) 50 bucketfuls. In one group (A), no movements

followed. The shells could not be roused, so to speak, from

their quiescence. In the other group, (B), two were dis-

tinctly seen to go downwards, while four went upwards.

Drying of rooks, however, stopped the motions of all the

individuals quickly. First splashing made a few roll down.

It seemed now tolerably certain to me that the molluscs in being

driven upward by the advancing tide must necessarily be subjected for

hours to the splashing of little waves, and therefore when the receding

tide leaves them comparatively quiet, they must become positively

phototaxic and move down toward the sea. 1 therefore went to Nakozaki

and made the following observations.

Station XII. Aug. 30, 8:30 A. M. Tide receding. An area

circumscribed on a vertical face of rock around 20 indivi-

duals which were also marked.

14 K. MITSUKURI.

5 P. M. Tide was just advancing on the lowest limit of the

somewhat squarish circumscribed area. Individuals had

not stirred from their positions in the morning. As tide

arose, the individuals were gradually driven upwards.

Movements of the animals took place, when splashes of

waves receded and left them momentarily quiet. 3 indivi-

duals remained below submerged, but others crawled up

higher and higher and were driven out of the circumscribed

area, until the individuals were about 5 cm. above the

highest point of the circumscribed area. This was at 6:

20 P. M. when I had to leave. The tide could not have

risen much higher and receded during the night.

g. 31:—When I went to the spot next morning at 5 A. M.,

tide was rising again but had not yet reached the circums-

cribed area. I was able to recognize 17 out of the 20

individuals which had originally been in the circumscribed

area, with one other doubtful one, distributed as follows:—

1 was in a crevice about 25 cm. to the left (looking from

seaward) of the original area, a trifle below the level

as when I left last night. All the rest had moved

down.

4 were in the original circumscribed area.

5 were on the same level as the original area but outside

it, farthest one being abont 10 cm. to the right.

1 doubtful, at the same level.

7 were below the level of the original area, the lowest

having gone down about 15 cm., below the lowest

side of the area, at the then water’s edge.

It was evident that after the tide receded, some of the molluses

went down seaward. I have further notes on the same area as follows : —

Aug. 31,8:45 A. M. Nearly full tide. The tide did not rise

as high as the preceding evening. It affected only the

lowest individuals which moved up about 15 cm.

NEGATIVE PHOTOTAXIS ETC. oF Littorina. | 15

Sept. 1, 8: 30 A.m. Another high and low tide during the

interval. Iudividuals scattered but still more or less near

the original area. One abont two feet from it.

I will conclude with another experiment :—

There is a large rock lying near the landing place of our laboratory.

When the tide was at the level of the dotted line (See Fig. 6.), I placed some

individuals of L. erigua which were known to be negatively phototaxic

at the spot A of this rock a little below the water line. The spot was

about ten feet from the high bank near the landing place. The question

was whether these individuals would go in the direction of the high bank

or toward the higher part of the same rock which was turned away from

the general mass of land. There was no ambiguity about the answer.

As soon as they were placed there, they startel in the direction of the

arrow toward the higher part of the rock A.

I now took some of these negative ones to a spot on a ledge of rock

vr,

“wen

amp a

Sm vs E rations e

ESA RS SS

c >

Sl _ = ; i a do ES » SESSI B

SR A es o

“a -—. Si _— = Ne

=— [>

—m— In

erg ar =—=

16 K. MITSUKURI.

D. Here the depth of water was about 15 cm., but considerably deeper

at the very base of the bank at y, while it was shallower in the direction

of x : —

1 went up in the direction of x.

3 started toward the land in spite of a greater depth of water. 1 of

these crossed the deepest part to the side y. The second went

as far as the deepest part. The third to a somewhat shallower

place.

2 did not stir at all.

Several individuals that were moving upwards at À were placed at

B and C several inches deep in water. There was a deep cut abut 2 feet

deep in the rock between B and C. B was 2 feet from the bank. In-

dividuals placed at B did not stir at all. Of those placed at C, one went

in the direction of the bank. The others did not stir at all.

I take it that in this experiment, A was out of the influence of the

shade of the high bank, at least from the molluscan standpoint. Hence

they went up in the direction that must have appeared darkest to them

i. e. in the direction of the highest point of that rock, although this was

no more than a few inches in height. B, C, and especially D are

in the shade of the bank : hence most of those that stirred at all went

landward, even if they had to cross water considerably deeper than that in

which they were placed. In B, the darkest part was toward the high

bank but water being very deep must have influenced themolluses in some

way so that they did not stir at all.

Once a lot of Littorina sitchana var. brevicula kept in a glass-bottle

for about ten days showed positive phototaxis for several days conti-

nuously. This I can not account for in any way, as this species also is

ordinarily negatively phototaxic. Nor could I induce other lots to act

in the same way by subjecting them to a similar treatment.

CONCLUSIONS.

1. Littorina exigua shows a strong negative phototaxis under ordinary

circumstances. That it is not negative hydrotaxis is shown

by Exp. 3.

NEGATIVE PHOTOTAXIS ETC. OF Littorina. 17

Note. This property enables the mollusc to creep up from

the sea towards the higher level which corresponds in

most cases with land, but in cases of detached rocks may

be away from the general mass of land.

2. Littorina exigua shows a disinclination to be submerged. This is

proved by the fact that a rise of water drives the molluse up-

ward, as well as by the fact that if the mollusc touches the

water’s edge in coming down, it hesitates to plunge in, and in

many cases travels horizontally along the edge as if in search

for a better or more comfortable route. In Experiment 11,

two actually turned back and went upwards again.

Note. When negative phototaxis and negative hydrotaxis

act together, the mollusc is driven upward without hesita-

tion. When these two properties are acting against each

other as in the case of Experiments 3 and 7, or at D, Fig.

6, negative phototaxis seems to be stronger and to over-

come negative hydrotaxis, at least in small depths. In

greater depths, the result is not entirely clear, as at B.

Fig. 6, where the mollusc did not stir.

3. Littorina exigua are in nature scattered over rocks, because there

are unevennesses in rocks which provide them with holes and

crevices to settle down in. On smooth glass-plates, they move

on until they are stopped by some obstacle which they can not

overcome,

4. When Littorina exigua is splashed on by water, for some length of

time, as by waves in rising tide or artificially by a jet of water,

and then is left quiet, it may become positively phototaxic and

then it goes down from land soward the sea. That this is pho-

totaxis and not hydrotaxis, is shown by the hesitating manner

which these exhibit on touching water's edge, in one case at

least, such dislike seeming to send the animal upward again (Exp.

11). This property is not as unfailing and strong as negative

18 K. MITSUKUKI.

phototaxis, and seems to disappear soon and on slight distur-

bances.

5. . While disliking deep water, Littorina erigua can not live on dry sur-

faces. It must have a cartain amount of moisture to crawl and

probably to feed.

Note. This is the reason why the mollusc left dry at the

highest tide-level sleeps on, so to speak, unmindful of the

boiling heat of the midsummer sun, till the next spring-

tide comes round again in a fortnight and wets the region

about.

6. Individuals of Littorina exigua do not probably wander about over

any large extent of space.

Note. One individual which I marked, probably in August

or at the latest, early in September, I found, had not stir-

red from the identical spot nearly four months later, on

Dec. 27, 1900.

What is said here of Littorina exigua is probably true to a large

extent of the second species, L. sitchana, var. brevicula and of other

sp2cies found elsewhere.

In thinking over the facts thus brought out, it seems to me that the

facts under the second and fifth headings are probably the primary ones.

That is, the animal lived by preference on slightly moist surfaces of

rocks. Whenever the water became too deep for its comfort, it would

naturally try to escape toward shallower places more adapted to its life.

This they found, they could do by guiding themselves by their eyes toward

the quarter that appeared darkest to them. We know that this corresponds

with the direction of land, but it is not likely that the mollusc can have

any conception of the things presented to our mind by the words, land

aud water. To the mollusc the dark quarter represents the comfortable

quarter, and nothing more. In course of ages, by the process of natural

selection, the mollusc would acquire the property of negative phototaxis

which has in actuality become so strong that even when the conditions

which originally made this property useful are artificially reversed, that

NEGATIVE PHOTOTAXIS ETC. OF Littorina. 19

is, when the dark quarter is made to correspond with the deep water, it

plunges in, unconscious of any disadvantage that might follw. It knows,

also, that after it is beaten by waves for hours—known to us as due to

rising tide—and then is left quiet, it ought to move down, if it wishes to

feed perhaps, and if it does not desire to be left dry—as, we know, would

surely follow on account of receding tide. Thus, even when judged from

our standpoint, the actions of the mollusc are rational and easily com-

prehensible. But we must not of course for one moment suppose that to

the mind—or whatever there is that represents our mind—of the mollusc

there is present such conceptions as land, water, tide &c. We must be

careful not to read our thoughts into the actions of animals in studying

their psychology. The mollusc probably knows iustintively only to go

toward dark, when water begins to splash it, and to move toward light

when splashing has stopped.

While it is undoubtedly true that the environment in which Litto-

rina has lived, has developed the instincts or properties enumerated above,

the contrary statement is true at the present day that, given these

properties it can not live in any other habitat. They limit the horizon of

the animal's activity. To that extent the animal has become specialized.

Imperial University, Tokyo.

May, 1901.

j è (260, after

nOi: i ‘deu TAN

di | JO In A DEE ath

4 Ve proce È WAR

ii torah) A

Kay

3 a:

! ’ x ) i eee a

. 1 en! È alin an ori Ù

Ayers:

ane En iti AT

‘ pi Ta 7

7 ; j 7" d SU Et EU. ut

pre: EPL 4 LE ‘a 4

re «TAR ART tal IA

nt)

. ST te

« u 2

tata Ma

DS cre

On the Lampreys of Japan together with Notes

on a Specimen of Lamprey from Siberia.

S. Hatta.

For some years past, I have been engaged in collecting materials

for the study of the species of Petromyzon which occur in Japan, and of

their distribution. The present notes embody the results so far obtained,

and while not yet complete, will, it is hoped, prove of some interest. In

addition to this, I append at the end a brief notice on a specimen from

Siberia.

The materials employed in the present investigation are (a) the

specimens in the Zoological Institute of the Science College, (b) the

specimens in the Imperial Museum, !) and (c) those in my own possession

partly collected by myself at several localities in the year 1891-1899,

and partly collected and sent to me by the undernamed gentlemen in

various parts of the country, 2) to whom my best thanks are due :—

Prof. A. Oka, Mr. S. Honda, Mr. S. Fujishima, Mr. J. Nakanishi,

Mr. T. Yubuki, Mr. T. Yasuye, Mr. K. Ando, Mr. N. Takahashi,

Mr. K. Okamura, Mr, S. Morohashi, Mr. T. Kitahara, Mr. T. Awoyama,

Prof. J. Hara, Mr. T. Nozawa, Mr. T. Yamanaka, Mr. M. Yamaoka,

Mr. Y. Yanagisawa, &c

1) For the free access of these collections I am indebted to Pror. MıTsukurı, the

Director of the Institute, and Pror. IsHrgAwA, the Director of the Natural History

Department of the Museum.

2) I am also indebted to the undermentioned gentlemen, from whom I received

the important informations on Lampreys :—

MR. S. OnIsHI, Mr. K. Forusawa, MR. S. Honpa, MR, Y. NAKAMURA, MR. Y. Nawa, MR.

K. Nrwa, Mr. K. Kato, Mr. H. Karo. Mr. 'l. Hara, MR. T. NozAkı, Mr. T. OGASAWARA,

Mr. K. Oaura, Mr. T. Iro. MR. Y. WATANABE, Mr. T. Taxesuita, Mr. S. Harrorr, MR.

S. Kasrya, Mr. K. IsHIiKAWA, Mr. K. Fostwara, MR. M. SanEvosHI, MR. N. HAYASHIDA,

MR. T. YAMANOUTCHI, &c.

D2 ON THE LAMPREYS OF JAPAN AND SIBERIA.

A. On the Species of the Japanese Lampreys.

A collection of the Japanese Lampreys falls naturally into two

groups ; (1) a larger form measuring 39.0—50.7 em in length and (2) a

smaller one having length of 8.0—15.6 cm. The larger form is undoubt-

edly the one described by v. MARTENS under the name of Petromyzon

japonicus. Unfortunately I have been unable to see the original descrip-

tion. GUNTHER!) makes the following remark on it: “ Petromyzon

japonicus, Martens, Wiegm. Arch. XXXIV, p. 3, does not appear to be

specially distinct from P. fluviatilis. The dentition in both is extremely

similar, only the Japanese example has an additional transverse series of

small teeth hehind the mandibulary tooth. The figure accompanying the

description of P. japonicus is incorrect and I am indebted to Pro-

fessor Peters for re-examination of the typical specimen ” (p. 504)

PROF. JORDAN has called it, in his ‘ Preliminary Check List of

the Fishes 2) of Japan,” Petromyron japonicus following MARTENS’

nomenclature.

As GÜNTHER states, this form of the “ Japanese Lamprey closely

resembles P. fluviatilis, that is, the position of the fins, the coloration of

the body, the branching mode of the marginal tentacles of the suctorial

disk, the maxillary tooth, the lingual tooth, the teeth of the suctorial

disk, the structure of the brain (see below), &c. all agree with those de-

scribed in P. fluviatilis. But there are found two points of constant

differences :—

(1) The first point has already been noted by GÜNTHER. He

notes “an additional transverse series of small teeth behind

the mandibulary tooth.” I would replace the word “behind”

with the word “ inside,” because this series of the teeth is

found immediately inside the mandibulary tooth-plate. It

should be noted that these teeth which are 3—4 in number

are, so far as I am aware, not yet enough hardened to deserve

1) GuNTHER. Cat. of Fishes, vol. VIII.

2) This Journal. vol. III, pp. 31—159, 1891.

S. HATTA. 23

the name of teeth, but are the merely hardened horny pro-

jections of the skin.

(2) The number of the cusps of the mandibulary tooth-plate is

described in P. fluviatilis as being seven or about seven. In

the Japanese species, this tooth-plate has constantly six cusps

of which the cusp at each lateral extremity of the plate is

bifurcated.

I propose, therefore, to regard the Japanese form as a distinct spe-

cies and to call it Lampetra 1) japonica.

This Lamprey does not occur in all the parts of the empire, but is

limited to certain localities (see below).

There has not been published any description of the smaller Lam-

prey 2); it differs from the larger form just stated not only in size but

also in the following essential characters :—

(a) The suctorial disk is protruded more than in the larger one ;

(b) The tooth-cusps are less prominent and more obtuse than in the

larger form, and the series of the teeth outside the mandibulary tooth-

plate consists of a lesser number than in the larger forın ; the cusps at

lateral extremities of the mandibulary tooth-plate are not bifurcated ;

dorsal, but there is only a notch befween them; the anal fin in the

female attains in the spawning season a considerable height; this is

called by S. H. GAGE the fin-like appendage ;

(d) The labial tentacles are mostly palmate ;

(e) The skin is dark brown and shows faint irregular spots. This

species is thus in the external characters very close to P. planer? (BI.).

In addition to these, two species show a great similarity in the external

configurations of brain (see below). But (1) the smaller size and

(2) the smaller number of the mandibulary tooth-cusps are the constant

1) For the reason brought out further on, I have adopted the generic name Lampetia

for this species.

2) V. Martens only touches it in his “ Die Reise nach Ostasien.”

24 ON THE LAMPREYS OF JAPAN AND SIBERIA.

differences which distinguish it from the latter. It must, therefore, be

regarded as a distinct species; and I have great pleasure in calling it

Lampetra mitsukurii in the honour of PROF. MrrsukuRt.

The size varies within certain limits ; it is not infrequent that the

individuals from some localities are double those from others.

The American Brook Lamprey, Lampetra wilderi (GAGE) 1,2) is very

close to this species ; it is, as it seems to me, probably its nearest ally 2).

I may perhaps be permitted to add a few words in regard to the brain

of two Japanese species of Lamprey. I find that the adult individuals

of the both species are remarkably alike in the external configurations of

their brain, which are again very close to those of P. fluviatilis of

Europe. (see AHLBORN’S4) figs 4—6). Therefore, if AHLBORN'S statements

and figures (figs. I—3) on P. planeri be trusted which maintain some

difference between its brain and that of P. fluviatilis, they might be taken

to be opposed to the idea of a near affinity between P. planeri and L.

mitsukurii mentioned above. But there is a strong reason doubting the

accuracy of AHLBORN'S results. It appears to me that the brain of P.

planeri as given in his fig. 1 is that of a young individual, for the brain

obtained from the oldest Ammocoetes stage of the smaller Japanese

species shows very much the same configurations. This doubt is further

confirmed by the fact that the figures (figs. 5 & 6) of the brain of P. planeri

given by WIEDERSHEIM 5) agree closely with the brain of L. mitsukurii

of Japan.

I accept mainly on account of the brain configurations just stated,

the. views of recent writers in maintaining that the genus Petromyzon

should be split into two: retaining the old name Petromyzon for P.

Dies è

1) Gage, The Lake and Brook Lampreys of New York: WıLprr’s Quarter-century

Book, 1893.

2) JoRDAN AND EVERMANN, Fishes of North and Middle America, Pt. I, 1899.

| 3) P. marinus has not been hitherto met with in Japan.

4) AHLBORN, Untersuch. ii. d. Gehirn der Petromyzonten ; Zeitschrift f. w. Zool., Bd.

XXXIX, 1883.

5) WIEDERSHEIM, Das Gehirn von Ammocoetes und Petromyzon planeri ; Jen. Zeit-

schrift, Bd. XIV, 1880.

S. HATTA. 25

marinus and assigning GRAY'S name Lampetra to the portion represented

by fluviatilis, planeri, mitsukurii, wilderi, &e.

B. Distribution of the Lampreys in Japan.

The smaller Lamprey, Lampetra mitsukurii, is found in almost all

localities of the empire, while the larger form, L. japonica, is limited to

the rivers and ponds on the slope facing the Japan Sea on the main island

(the Hondo); the latter is especially abundant in the Shinano River, the

Mogami River, the Omono River, the Jintsu River, the Lake Kawakita,

the Kozan Pond, &c. It has been maintained by some persons that the

larger Lamprey is occasionally met with in the marshy regions of the

Kwanto-Plain and in-the mouth of the Kiso River and of the Kitakami

River and their adjacent waters. As these localities all belong to the

Pacific side, I have been at great pains to obtain the specimen from

them, but all my efforts were unfortunately so far proved fruitless.

Let us now try to explain this peculiarity in the distribution of the

Lampreys on the main island.

As is well known, the long chain of the islands, of which our empire

is composed, extends from the parallel of about 52° north to the parallel

of 22° with an oblique north-east or south-west direction. As regards

the main island, the long range of high mountains running lengthwise

in the central part constitutes the water-shed which divides rivers into

two sets of systems: the rivers on the slope facing the Asiatic Continent

empty their water into the Japan Sea, and the streams on the opposite

slope flow into the Pacific Ocean. The rivers on the Pacific slope have

in general a very steep incline, and their waters rush down in torrents

after heavy rains, and in most of the rivers, almost drying up or dwind-

ling into mere threads of water at other times. ‘Thus their bed consists,

in most parts, of sand and gravel. On the contrary, the rivers on the

opposite side are less steep, and their bed is muddy. ‘hese topograph-

ical conditions exert, it seems to me, a great influence up u the distribu-

tion of Lampreys in Japan, as the following considerations will show.

26 ON THE LAMPREYS OF JAPAN AND SIBERIA.

Sluggish creatures like the Lampreys can not naturally flourish in

rapid currents even when other conditions are favorable. Thus the

larger Lamprey which inhabits preferably the muddy bed of rivers, lakes,

ponds, and river-mouths ; and in the spawning season alone ascends rivers

to deposite eggs on the sandy bed of their tributaries, would not find

such conditions in the rapid streams of the Pacific slope, The poor

creatures could not be able to hold themselves. This, it seems to me,

sufficiently accounts for their absence from the river systems of that

side.

On the other hand, the smaller Lamprey does not live in rivers, but

generally in small streamlets between the fields, in springs, or in small

canals near towns. The animal being thus independent of rivers their

habitation has been able to extend itself all over the main island wher-

ever other natural conditions allow them.

There will be nothing to clash with the views put forward here, if

the larger Lamprey should be found hereafter, as is very likely, in

marshy regions of the Kwanto-Plain as well as in estuaries with the

muddy bed as the mouth of the Kiso River and of the Kitakami River,

although these localities belong to the Pacific side. On the contrary,

such a discovery could go a long way in proving the correctness of the

above hypothesis.

Let us now turn to the examination of other islands: Shikoku,

Kyüsbü, and the Hokkaido, &e.

Geologists tell us that the inland sea (the Setouchi-Umi) between

Shikoku, Kyüshü, and the main island was brought about by faults

and constitutes a geological moat. The distribution of the Lampreys

affords the excellent evidence for this view. ‘The two great southern

islands, Shikoku and Kyüshü, show, as regards the distribution of the

Lampreys, the same condition as the Pacific side of the main island; on

these islands the smaller species alone is found ; they are to be regarded,

in this respect, as belonging to the Pacific slope.

In the Hokkaido, only the slope facing the Japan Sea has been

explored ; in this part, the larger as well as smaller forms occur, just as

S. HATTA. 27

in the corresponding slope of the main island. The former species is

found in the Ishikari River and the Teshiwo River, whilst the latter form

is met with abundantly in streamlets in that part of the island.

Sado, a small island in the Japan Sea off the Peninsula of Noto, is

very interesting ; it is the same, as to the Petromyzon distribution, as the

localities on the opposite shores of the main island, viz: Noto, Toyama,

Niigata, Sakata, &c.

In the Ryükyü (Loochoo) islands, the Lampreys have not yet hith-

erto been found; these seem to be destitute of Lampreys.

I have not yet explored the Lamprey fauna of Formosa ; this I hope

to deal with on a future occasion together with that of the opposite coast

on the Asiatic Continent.

If the above assumption be correct, it is clear that the distribution

of the Lampreys in our empire is dependent on the topographical features.

The influences of other natural conditions, if such truly act apon the life

of the animal, are very little as regards the distribution.

C. On a Specimen of the Lamprey from Siberia.

I have recently received a specemen of the Lamprey from Eastern

Siberia, which was caught in a brook callad Pervaya Rechka, a tributary

of the Amur River, not far from the Seaport of Vladivostock. It was

taken last summer and sent to me by my friend Mr. ABeno who has

resided there for about seven years. He writes to me that the Lamprey

is very rare in the neighbourhood of the city, and that he has seen only

two individuals during that time of his residence.

The specimen is just transformed individual and is preserved in good

condition in alcohol. It measures 15 cm ; therefere, it is smaller than

the larger specimen of L. mitsukurii and larger than smaller ones

among it. The second dorsal fin is continuous with the caudal ; the first

and second dorsals are separated by a deep notch. The first dorsal is

3 ınm high, while the second is 5 mm in height. The distance from the

snout to the anterior limit of the first dorsal is 7 cm. The head is about

as long as the branchial region or the thorax. The supraoral lamina is

28 ON THE LAMPREYS OF JAPAN AND SIBERIA,

provided with 2 well separated obtuse cusps ; the infraoral lamina posses-

ses 6 obtuse cusps. 3 bicuspid teeth are found on each side of the gullet.

Other buccal teeth are simple and describe a circle immediately inside of

the fringe of the iabial tentacles. Each labial tentacle is palmate, termi-

nating in a blunt point. The specimen is female, but wants the anal

lobe (the fin-like appendage), probably owing to being caught out of the

spawning season. Coloration is quite like the Japanese specimen of L.

mitsukurii.

The specimen is thus very close to L. mitsukurii of Japan; in fact,

both are, I think, of the same species. It is, therefore, also very close to

L. wilderi (GAGE) of North America.

Let as now examine other species of the Lamprey of East Siberia.

I have not seen the description of the Lamprey of Kamtschaka by

PALLAS;!) but by the account of Dr. JoRDAN,23) I know that two

species of the Lamprey, Eutosphenus camtschaticus (TiLESIUS) and

Lampetra camtschatica (PALLAS), are found in Kamtschaka. On the

former species he remarks: ‘ A larval lamprey obtained by STEINEGER

in the Paratunka River, near Petropaulski, Kamtschaka, is apparently of

some species of Entosphenus. It can not be distinguished from the larva

of Entosphenus tridentatus, though the adult may show peculiar charac-

LE]

ters ” (p. 434). Entosphenus tridentatus is, according to JORDAN, common

southward along the coast of Unalaska. The auther puts the second

species (Lampetora camtschatica) synonymous with Lampetra aurea

(BEAN), which is found also in the Yukon River and other streams of

Alaska. Thus the two species, Hutosphenus tridentatus and Lampetra

aurea, occur common on the two continents, Asia and North America,

separated by the Behring Strait.

Lampetra aurea is, according to JORDAN, very close to Petromyzon

fluviatilis of Europe, and therefore, also to the larger Lamprey (L. japon-

ica) of Japan.

1). PALLAS, Zoographia Rosso-asiatica, 1767.

2). JoRDAN, Report of Fur-Seal Investigations, Part. 3, 1899.

3). JORDAN & EvEeRMANN, Fishes of North and Middle America, Pt. I., 1899.

S. HATTA, 29

My friend Mr. Uyepa in Nikulaevsk informed that there are found

Lampreys abundantly which much resemble the larger Lamprey of

Japan ; these may be of the same species as Lampetra camtschatica

(PALLAS) or L. aurea (BEAN).

Thus together with Lampetra camtschatica which is the same or at

least very close to Lampetra japonica of Japan, the present species is

represented on the Japanese Island. I do not hesitate in concluding,

therefere, that the coasts round the Okhotsk Sea and the Japan Sea

show, with regards the Lamprey fauna, a uniform character with excep-

tion of Entosphenus tridentatus which is wanting in Japan.

From the accounts given above, it is also true that the Lamprey

fauna of Japan is connected through Kamtschaka, on one hand, with

North America and, on the other, with Asiatic Continent.

I wish to express my deepest thanks to my teacher, Pror. Mitsukuri,

who kindly looked through the manuscripts.

Biological Laboratry, The Gakushuin, Tokyo.

Nov., 1900.

| O i nd nee

nf ey dunia A

“ays Vis ee ok À

fr >

5 Ses ia

4 t \ lie Di |

h | lo

fut fs

De nea

road M aan na

nhituber; è AR tennis

Gonyaulax and the Discolored Water

in the Bay of Agu.

T. Nishikawa.

During the latter part of September, the streaks and patches of

brownish-yellow water emitting an unpleasant odor were observed by

fishermen in the Bay of Agu, situated in the southern part of the

Province of Shima, on the Pacific side of the Hondo. As this “ red-tide”

is said to have been in former times highly destructive to the beds of the

pearl-oysters which form one of the principal productions of the Bay, a

great alarm was felt for the safety of the Molluscs. Fortunately,

however, on the 28th of the month, a heavy storm arose and cleared

the waters of the Bay, so that very little damage was actually done.

By the order of His Excellency, the Minister of Agriculture and

Commerce, I visited the Bay to determine, if possible, the cause of the

unusual coloration of the sea-water.

According to the observations of Mr. Mikimoto, a pearl-oyster

culturist of Tadoko Island in the Bay, the surface waters on the southern

coast of the Island and along the shore of Hasako, were on the 23 rd of

the month densely discolored. But, when I arrived at the Island on the

25th, this discoloration of the water had almost disappeared. I was told

by a fisherman that patches of discolored water were to be seen on

he northern shore of Masaki Island, but unfortunately it was already -

twilight when I arrived there by a boat on the evening of the 25th

Consequently it was impossible to distinguish clearly any coloration

but by an unpleasant odor I was able to recognize the presence

of the discolored water. The odor of the discolored water was

similar to that which is emiited by the Algae when collected in a great

quantity. On next day, I was fortunate enough to observe well such a

32 T. NISHIKAWA.

patch, at about half mile west of Masaki Island. It was deep brownish-

yellow in color, and was 3 or 4 feet beneath the surface, forming ir-

regular cloud-like masses, interspersed with the water of the ordinary

color. The thickness of the discolored patch was 2—4 feet.

On taking a little of it into a glass-cup, it was observed that an

enormous number of minute organisms swarmed in the water which was

as clear as usual. There could be no question, that the discoloration

of the water was due to the presence of numerous minute organisms.

On examination with a microscope, it was found that the organism

belonged to the Peridiniaceae. It has a polygonal shape, with an

areolated shell. The anterior pole is pointed, and there are two proces-

ses, one larger than the other, on the posterior pole. The longitudinal

groove is very broad in its posterior part having a conspicuous ‘ Geis-

sel-spalte”, from which a long flagellum extends backward. Anteriorly

the groove becomes very narrow aud scarecely communicates with the

polar opening. The left end of transverse groove in the ventral side is

below the right. The chromatophores are small and numerous, brownish-

yellow in color. The length of the body is 0.046 mm. and the larger

process is 0. 0062 mm. long. The organism moves very actively by

rolling itself on the longitudinal axis of the body. Sometimes two

individuals were connected in a chain like Ceratium tripos.

The organism occurred in such a great abundance that it gave to

the water very noticeable color and peculiar odor, 800 or even nearly

3,000 in the densest part, having been estimated to be present in a drop

of the discolored water. George Murray Esq. of the British Museum,

to whom the specimens were sent, has kindly identified the organism.

It is Gonyaulax polygramma Stein. According to his statements, it is

widely distributed ; he has found it frequently in the tow-nettings from the

Atlantic, but he has never before seen it in such an abundance.

When disturbed, the active Gonyaular gives a very beautiful

phospholescent light in the dark. By keeping the discolored water

in a glass bottle, the minute organisms gather near the surface of the

water. After a short time, some of them sink to the bottom and the

GONYAULAX. 33

anterior half of the shell breaks longitudinally, the contents come ont

from the shell, and gradually change to the spherical shape. If the

discolored water is kept without any preserving reagent, the water

emitts, after about a week, a pungent odor of hydrogen sulphide and

the shell of Gonyaulax swells which is finally dissolved. The proto-

plasmic contents which have come out of the shell retain the spherical

shape, with the chromatophores gathered in the centre. This fact is

interesting, as a sample of similarly discolored water which was observed

in September of the previous year (1899) in the Bay of Toba and which

caused a great mortality among fishes, presented exactly the same ap-

pearance when it came to my hand from Mr. Mikimoto who had bottled

it ten days before without adding any preserving reagent. This makes

it highly probable that the cause of discoloration on that occasion was also

Gonyaulax polygramma.

Besides this species of Peridiniales, the discolored water contained

a species of Ceratium and Prorocentrum micans, Ceratium tripos,

Chaetoceras, and some nauplius larvae were also rarely found. In the

discolored water which I observed on the northern shore of Masaki

Island, Prorocentrum and Ceratium were very abundant. At the same

time, in the ordinary clear water of the Bay, the plankton is rich in

Bacillariaceae (such as Bacteriastrum, Rhizosolenia, Chaetoceras) Cope-

poda, Noctiluca, Appendicularia, Sagitta, Nauplius and the larvae of

Polychaeta and Gasteropoda. 'Thus, there was a great difference between

the living organisms of the discolored water and of the ordinary sea-water

close by. This difference seems to show that the discolored water

is unfitted for the existence of Copepoda, Noctiluca, Bacillariaceae &c.

while it seems specially suitable for the propagation of the unicellular

organism, Gonyaulax. The chemical properties of the discolored water

are probably different from those of the. ordinary sea-water, but I have

been unable to make investigations on the matter.

The occurrences in an excessive abundance of minute unicellular

organisms, as the cause of unusual coloration of sea-water have been

described before. Dr. Carter described Peridinium sanguineum as

34 T. NISHIKAWA.

causing the red coloration of the sea on the shore of Bombay. According

to Dr. Whitelegge the cause of discolored water of Australia in 1891 was

Glenodinium rubrum. lu 1898, Dr. Mead discovered a species of

Gymnodenium in the “red-water” of Narragansett Bay, Mass, U.S. A.

Usually the appearance of discolored water is accompanied by a great

mortality among fishes, mollusks and shrimps. According to the

observation of a pearl-oyster culturist, in the later part of August, 1899,

large streaks and patches of yellowish-red water floated abont with the tide

in the Bay of Toba. Fishes, which were kept in baskets floating on the

surface of the sea, were damaged by them. Fishermen easily caught

the littoral fishes by spearing, for the fishes had became very sluggish

in the discolored water. Even Haliotis or Ear-shell seemed to suffer.

But whether the presence of minute organisms, the Peridiniales, per

se is the immediate cause of this mortality or not is uncertain. I put

some pearl-oysters, Meleagrina martensii, in the most highly colored

water kept in a porcelain jar. After 24 hours, I found that they

suffered no injury, although the water was still swarmed with

Gonyaulax. A simillar effect was obtained by Dr. Mead with

Gymnodenium. I observed that the stomach of the pearl-oysters

was filled with Gonyaulax and many empty shells were found in

the intestine, which shows that Gonyaulax forms suitable food for the

pearl-oysters, although these usually feed upon the Bacillariaceae. Thus

the presence of Peridiniales, it appears to me, can not be the immediate

cause of the destruction of fishes. The water which is fitted for the

sropagation of Peridiniales and unsuitable for the existence of the usual

plankton, is probably also unsuitable for the fish life, or the dead bodies

of an enormous numbers of peridiniales sinking to the bottom decompos-

ing and putrifying there may eventually become injurious to other

organisms. This poiut however needs further examination.

Imperial Fisherus Bureau,

Tokyo.

Reminiscence of Holoblastic Cleavage in the Egg

of the Shark, Heterodontus (Cestracion)

japonicus_Jlaceay.

Bashford Dean.

In studies on the yolk characters of vertebrates the eggs of Elas-

mobranchs have always been described as presenting typical meroblastic

cleavage. This character has here indeed been deemed so axiom-like

that we find that the meroblastic egg is usually, if not always, made the

point of departure in various works dealing with the problem of the gain

or loss of yolk in eggs of the gnathostomes. Thus it has long been

discussed whether the eggs of the higher classes have been capable of

losing and reacquiring food yolk, or whether the eggs of the highest forms

have merely lost the ancestral supply. This, however, illustrates but

one line of discussion in which the embryologist has treated the eggs of

elasmobranchs as irrevocably meroblastic. It is upon strictly theoretical

grounds (reducing the matter to its simplest terms, for as a matter

of fact segmentation has been observed in but very few forms) that

the early characters of the eggs of elasmobranchs have been compared

with those of other ichthyopsids. But in spite of this no one, I fancy,

would have been bold enough to have prophesied that the wide difference

between the typically meroblastic egg of the shark and the holoblastic

egg of such a teleostome as a sturgeon might come to be bridged over

within the limits, not of fossil sharks, but of recent sharks themselves.

Indeed it seemed all that embryoogists could hope for, that meroblastic

features should be discovered in such ancient forms of teleostomes

as Lepidosteus and Amia. It was naturally, therefore, a great surprise

to me to find in my studies on the development of Cestracion that there

occurred in this ancient shark features which are best interpreted as

36 BASHFORD DEAN.

well marked reminiscence of holoblastic cleavage. And I summarize

the present evidence for such a conclusion in the following notes :

In all early stages of Cestracion examined (I have notes upon up-

ward of a hundred examples) the surface of the egg, notably the surface

which keeps uppermost, and which we may call the animal pole, is tra-

versed by a series of definite and sharply marked lines. These are disposed

so conspicuously* that the observer is given the impression that he is

examining an egg of Lepidosteus, magnified some 20 diameters. He thus

sees at the upper pole of the egg a number of sharply defined areas,

resembling the blastomeres in a late cleavage stage of Lepidosteus. He

also observes a series of meridional lines passing down the sides of the

egg, and diverging from one another radial-wise. These lines are next

seen to become groove-like below, and then fade away toward the vegetal

pole. There is also similarity in color: the region of the animal pole is

of a pale straw color, the vegetal region is greenish yellow, and there is

an intermediate equatorial zone, in which there is an orange or pale

brownish cast. Thus far it will be seen that the resemblance to the

ganoidean egg is a striking one: a more critical examination, however,

brings out the facts that in the egg of Cestracion a red-colored germinal

disc is present in addition to the above blastomere-like areas, and that

it is situated, not in the centre of the apparent animal pole, but further

down on the side of the egg. The homology of the furrows traversing

the egg of this shark is accordingly not as self-evident as at first appears.

Is it possible to regard them as not due to cleavages, in spite of their

extraordinary resemblance ? Or, on the other hand, is the excentric posi-

tion of the germinal disc to be interpreted as secondary, due, for example,

to some physiological cause? In attempting to answer these questions

I was led to tabulate the reasons favorable and unfavorable for regarding

the markings of the egg of Cestracion as due to cleavage characters, and

this table may be quoted as follows :—

* Their conspicuousness may be judged from the fact that they can be seen at

a distance of from six to seven feet by persons of normal eyesight. 'l'hey were dis-

tinguished at a distance of nine feet by a sharp sighted laboratory attendant.

HOLOBLASTIC REMINISCENCE IN CESTRACION. 37

Reasons for regarding the lines on the egg of Cestracion as remin-

iscent of holoblastic cleavage.

Unfavorable evidence.

Such lines have never

been known to occur in

the egg of any other

shark.

They may be surface

wrinkles merely.

They have not been

traced back to the ear-

liest cleavage lines.

Favorable evidence.

Their occurrence, therefore, is all the more

significant in Cestracion, on account of the

close relationship of this form with palaeo-

zoic genera. Cf. also the breaking up of the

egg of Chimaera into lobes or blastomeres.

In this event they should be superficial,

more or less inconstant, and subject to change

by artificial means. They are, however, not

only present in all early stages examined, but

they suggest distinctly the outlines of blasto-

meres. ‘They show no change for as long a

time as I have been able to keep alive the

opened egg—a matter of 30 hours. They

can be distinguished after the egg has been

broken, and at such a time it is seen tbat they

are not shallow surface markings, but that

they pass deeply into the yolk; they are

deepest (1. 5 mm.) at the upper pole, and

become shallower as they pass downward.

The most serious objection, for as yet I have

failed to secure the earliest stages of segmen-

tation. The egg of Cestracion does not ex-

hibit these lines shortly before it separates

from the ovary. They are present when

the egg is deposited (late segmentation) and

they remain conspicuons for as long a time

(over a month) as the yolk is uninclosed.

In the earlier stages, finally, they are generally

less numerous than in later ones.

They may have no

constant relation to

nuclear structuves.

The germ disc is

not at the animal pole,

where it would be apt to

be if the lines are due to

cleavage.

Why are not these

blastomere-like struct-

ures due to merocytes,

like, for example, the

small irregular emine-

nces surrounding the

germ disc of Torpedo

or Pristiurus ?

BASHFORD DEAN.

I have not as yet been able to prepare sec-

| tious of the surface of the egg. The presence

| of many nuclei in each segment, however,

| would evidently have little bearing upon this

| point, since polyspermy undoubtedly occurs.

| There is evidence that the present position

of the germ disc is a secondary one, for in eggs

just deposited, (1) it is nearer the animal

pole than in later stages: (2) there is a kind of

| track, whitish in color, extending from the

| direction of the upper pole of the egg, sugges-

ting therefore that the disc has shifted its

position, leaving a wake behind it.

There is, however, no evidence that such dis-

tinct structures are ever produced by mero-

cytes. Moreover, there are grounds for believ-

ing that in the older sharks these characters

were less highly developed than in the more

recent ones. ‘Thus it has been shown by

Rückert that the tendency for merocytes to

emigrate is less marked in a shark like Pris-

| tiurus, than in the more specialized Torpedo :

that in the shark the entrance of the nuclei

into the yolk is less precise, and that the nuclei

while in the germ disc are in general more

superficial in position. We can justly infer

that in so ancient a shark as Cestracion the

part played by merocytes during segmentation

would, if anytbing, be less marked than ın a

later form like Pristiurus.

From the foregoing analysis we can, I believe, conclude reasonably

that the egg of Cestracion retains rudiments, at least, of its ancestral

cleavage. But we must also admit that absolute certainty in this con-

HOLOBLASTIC REMINISCENCE IN CESTRACION. 39

clusion will be lacking until the origin of the above described furrows can

be traced back directly into the earlier cleavages. In the mean while,

however, the case is so strong a one that I think we can justly refer

to the lines as expressing cleavage, and to the sharply outlined areas as

blastomeres.

In the figures, Pl. I, Figs. A-I, are shown a series of stages illus-

trating characters of segmentation. ‘The drawings* were made from the

living egg. It should be noted that the egg capsule, when opened from

either end, shows the egg with the animal pole uppermost. ‘The egg itself

measures between 4 and 5 cm. in diameter: it is of semi-fluid consisteney,

and is surrounded by a dense glairy albumen, which in turn in its outer-

most layer is attached to the egg capsule. The albumen is tenacious: it

will flow over the rim of the opened capsule, and if the escaping mass is

not promptly separated, it is apt to draw with it the albumen immedi-

ately surrounding the egg, and will thus destroy the spherical contour of

the yolk. Mention might also be made of a delicate whitish membrane

which surrounds the egg within the mass of albumen, reminding one,

for example, of the vitellina of Necturus. The rupture of this delicate

inner membrane is apt to disturb the contour of the egg. Removal

of the end of the egg capsule together with the attached albumen,

causes a flattening of the egg at the animal pole, more or less pronounced

depending upon the amount of the albumen removed. It is exceedingly

difficult to remove the egg from the capsule without rupturing it,

although in several cases I have succeeded in doing so: it sinks

flatly, very much as the yolk of a hen’s egg. Whenever successfully

removed it has shortly ruptured of its own weight. In examining the

cleavage lines it has accordingly been found best to allow the egg to

remain in the opened capsule. A view of the side and vegetal pole

can then be had by rotating the egg by means of curved needles

thrust into the adjacent albumen.

* I am indebted to Messrs. N. Yatsu and I. Kuabara for their kindness in preparing

them.

40 BASHFORD DEAN,

In Fig. A., is shown an early stage. The egg was collected at the

time of deposition. 'There are few blastomeres and the radial pattern of

the marginal furrows is well marked. The germinal disc is seen at the

top of the figure. A somewhat similar stage is shown, Fig. B., from the

region near the vegetal pole. The germ disc is situated on the side away

from the observer: its position, however, is indicated by dots, as if shown

through the egg. The present specimen is an interesting one since it

shows one of the few cases in which I was able to satisfy myself that the

furrows could sometimes be traced well down toward the vegetal pole, in

fact that one or two of them actually traverse it, This stage well

illustrates the flattening of the groove-like ends of the cleavage lines. A

later stage, Fig. C., shows a greater number of meridional and marginal

cleavages. The germinal disc appears at the lower left hand side of the

figure. In Fig. D., a lateral view is given of an egg somewhat similar

to the foregoing. The spscimen is noteworthy in as much as the radial

(meridional) lines pass quite uniformly as far as the region of the equator

of the egg. More irregular in this regard is the egg shown in Fig. E.

In this several of the lines pass further towards the vegetal pole than

their fellows. In the egg shown in Fig. F., the meridional furrows pass

still further toward the vegetal pole.

In Fig. G. is pictured the animal pole of an egg somewhat finely

subdivided by cleavage lines. In Fig. H., a stage of gastrulation, the

cleavage lines are interestingly shown through the transparent edge of

the blastoderm. From the radial arrangement of these furrows it can be

inferred that the embryo is travelling tail foremost over the animal pole

of the egg. Andina still later gastrula, Fig.I., it will be noted that a

number of large blastomeres are seen conspicuously through the con-

stricting blastopore. From the radiating lines which thus appear, we can

reasonably infer that the embryo has travelled still further backward.

In order to give a more convincing idea of the appearance of

the above described cleavage lines, I have inserted, Figs. J-O, a few

photographs of the living egg. These show the lines clearly enough,

although, technically speaking, the pictures are faulty, for they were

HOLOBLASTIC REMINISCENCE IN CESTRACION. 41

necessarily taken through an irregularly refracting mass of albumen.

In the first of these photographs, J, an early stage, the radial

arrangement of the marginal furrows is clearly marked. In K, the egg

has been rotated so that the flattened groove-like ends of the furrows

can be seen from the region of the vegetal pole. The remaining figures

picture the egg in the region of the animal pole, and indicate various

degrees of its subdivision into blastomeres.

If we accept the foregoing observations as evidence of holoblastic

reminiscence, the egg of Cestracion is evidently of considerable value in

comparison with the cleavage characters of other ichthyopsids. We can

thus conclude that the great size of the eggs of other sharks was attained

before total cleavage became lost; and that accordingly the yolk region

of such eggs is directly, not indirectly, or partially, homologous with the

lower pole cells in other ichthyopsids. Cestracion also indicates that the

change in the position of the germ disc occurred before holoblastic cleav-

age was given up, and we have with it the suggestion that it was from

some new or modified physiological cause that a distinction came to arise

between the germ disc region and the region of the upper pole. In such

a functional change may have arisen an efficient cause for the disappear-

ance of the holoblastic type of cleavage, which up to that time had

continued to develop on ancient lines, pari passu with the differen-

tiation of the lighter and heavier deutoplasmic elements of the egg.

Marine Laboratory of the Imperial University of Tokyo,

Misaki, June 8. 1901.

Postscriptum. Misaki, July 1.

I have recently taken several eggs (early blastula) from the oviduct

of Cestracion. And there can now be little doubt that the lines rep-

resent cleavages. In one specimen the entire region of the living germ

disc was successfully removed and viewed as a transparent object, and one

could then detect cellular outlines bridging the space between the germ

disc and the yolk furrows.

BMD

On the Relation ofthe Metameric Segmentation

of Mesoblast in Petromyzon to that in

Amphioxus and the Higher Craniota.

(Preliminary)

S. Hatta.

It has been very well known for a long time that the mesoblast

in all Vertebrata undergoes the metameric segmentation only in its

dorsal part adjacent to the chorda and neural canal, while the ventral

part remains unsegmented. Further, efforts have been made to bring

the fact in harmony with the corresponding process observable in Am-

phiorus. Setting aside innumerable literature on the subject, I will cite

here O. HeRTwIG'S” formulation of the fact, which runs as follows:

“ Ähnliche Vorgänge (as in Amphiorus) vollziehen sich in etwas abgeänd-

erter Weise bei den übrigen Wirbelthieren ” (p. 139). How far the mode

is modified, we see in the following lines given by the same author”:

“Während nun beim Amphioxus der Process der Segmentirung sich

auf das gesammte mittlere Keimblatt ausgedehnt, ergreift er -bei den

Amphibien und ebenso bei allen übrigen Wirbelthieren nur den Theil,

welcher an Chorda und Nervenrhor angrenzt, lässt dagegen die Seiten-

platten unberührt ” (p. 140). It is easily conceivable that there is a gap

between the case in the Craniota and that in Amphioxus, although there

would be no difficalty in accepting that the process in both cases is

analogous, and the parts thus formed are homologous with each other.

But the attempt to derive one type from the other is, so far I am aware,

has not been supported with any positive evidence for it

1) Lehrbuch der Entwicklungsgeschichte der Menschen und der Wirbelthiere,

3. Auflige, 1890.

2) Loc. cit.

44 MESOBLAST IN PETROMYZON.

The obscurity of the interrelation between the two cases is doubtless

caused by the modified mode of formation of the mesoblast in the higher

Craniota. If such be the case, there is some excuse in describing the

precise mode of the metameric segmentation of mesoblast in Petromyzon,

which forms the subject of present communication.

As discovered by v. KUPFFER 1) and confirmed by myself 2) as far back

as ten years ago, the mesoblast in Petromyzon is brought about, in the

anterior region, by a simple folding of the entoblast, just as in Amphiorus,

The question above referred to must be, therefore, decided by a careful

observation of the metameric segmentation in this part of the mesoblast.

The harmony between the two cases, if such truly exist, ought to be

seen here.

In the course of differentiation of the mesoblast in the head fold of

Petromyzon, I have found great harmony between the case in Potromyzon

and that in Amphioxus ; I will enumerate it in the following lines :—

1) Before the metameric segmentation, the mesoblast is a sack of

the cylindrical epithelium, which is, on a cross-section, of a triangular

shape and fills up the space on either side of the chorda and neural cord,

bordered upon laterally by the epiblast and ventrally by the roof of the

enteric canal.

2) The metameric segmentation of the mesoblast is not confined to

the proximal part of the layer adjacent to tne axial organs, but goes

through the whole substance of it.

3) In a mesoblastic somite thus formed, the proximal (dorsal) and

distal (ventral) portions are distinguished: they constitute respectively

the scleromyotome and a part which loses later the segmental character

and, by the subsequent growth of which the non-segmented lateral plate

is brought about.

4) The scleromyotome is converted, having been cut off from the

rest of the mesoblast, into a polyhedral case which is interposed between

1) V. Kuprrer, Die Entwickelung von Petromyzon Planeri: Arch. f. mikrosk. Anat.,

Bd. XXXV, p. 501, 1890.

2) S. HaTTA, On the Formation of the Germinal Layers in Petromyzon: Journ. Coll.

Sc.Imp. Univ., Tokyo, vol. v. 1891.

S. HATTA. 45

the epiblast, the chorda, the neural cord, the distal part of the mesoblat,

and the roof of the enteric canal.

5) By stages, the shape of the scleromyotome is again converted

into an elongated triangle. The greater part of its inner long limb

which touches the axial organs, is transformed into the body-muscle,

while the outer long limb along the epiblast is flattend to form the cutis

layer. Finally the base of the triangle, which rests on the distal part of

the mesoblast and the roof of the enteric canal, is likewise thinned out

and evaginates, pushing its way between the muscle-layer and the axial

organs ; this is the sclerotome.

In the dorsal region, the process of segmentation is essentially the

same as in the head-fold. ‘The process goes on through the whole

breadth of the layer before the separation of the scleromyotome from

the rest of the mesoblast. Here the circumstance is more or less altered

by the differentiation of the Anlage of the pronephros which follows

immediately the segmentation. But this does not disturbe the parallel,

because the morphological equivalent of tbis section of the somite is

found in the head-fold, which does not develop into the pronephros, but

after having lost its segmental character, takes part in the formation of

the non-segmented lateral plates.

The facts mentioned above agree with those in Amphiorus obtained

by HATSCHEK 1) so far as concerns the stage up to the end of the meta-

meric segmentation and the fate of the scleromyotome. "The parallel of

the further growth of the ventral half of the somite (the Seitenplatte of

HATScHEK), which takes place immediately after this stage, has not

hitherto been demonstrated in Amphioxus. But there is little room for

doubt that the Seitenplatte grows ventrally (distally) after the oblit-

eration of the coelomic discepments and before the foundation of the

subintestinal vein ; for in Petromyzon, this vein appears in much later

stages.

1) B. HarscHEK, Studien über Entwickelung des Amphioxus: Arb. Zool. Instit,

Wien, vol. iv, 1881.

46 MESOBLAST IN PETROMYZON.

As I have already mentioned), it is a fixed fact that the line

of the separation of the scleromyotome (the Urwirbel of HATSCHEK) from

the rest of the mesoblast agrees with each other in both cases : it divides

the segmented part of the mesoblast into the dorsal (Urwirbel) and the

ventral (Seitenplatten) halves. The Urwirbel of Amphiorus undergoes

exactly the same fate as the scleromyotome of Petromyzon. In Pet-

romyzon, the ventral half forms, as above stated, the Anlage of pronephros

and the coelomic projection ; this point has not yet been made clear in

Amphioxus. But the reasonable suggestion of BovERL2) makes it intel-

ligible that this section of the mesoblast should give rise to the formation

of the Nierencanälchen, that is, the excretory organ of segmental

character.

The results arrived at in Petromyzon differ from those in the higher

Craniota concerning the extent, to which the process of the segmentation

takes place. In the Craniota the segmentation is confined to a small

proximal part, while in Petromyzon the whole breadth of tbe layer

undergoes the process. But this difference is, I think, easily explained

by a more precise examination of the stage, in which the segmentation

takes place. In the higher Craniota, this process comes to an end only

in later stages when the mesoblast, i. e. lateral plate, has grown consid-

erably. While the segmentation is going on in the proximal part, the

lateral plate itself grows distally. And thus a phenemenon which leads

one to think of the process confined exclusively to a small proximal part

alone comes into view. The segmented and non-segmented parts uudergo

essentially the same fate as in Petromyzon.

The differentiation of the mesoblast in Petromyzon holds, in this

respect, very important position in the morphology of the Chordata. In

earlier stages, it is exactly parallel with that in Amphiorus, and in later

stages, it shows the feature seen in the higher Craniota. We can thus

observe in one and the same animal, Petromyzon, the stages, through

which the primitive state of the mesoblast in Amphioxus has developed

1) S. Harra, Contrib. Morphol. Cyelost. ii.: Journ. Coll. Sc. Imp. Univ., T'okyo, vol.

xiii. 1900.

2) T. Boveri, Die Nierencanälchen des Amphioxus: Zool. Jahrb. vol. v. 1892.

S. HATTA. 47

into the condition in the higher Craniota. These facts can have no

other significance than as a connecting link between Amphioxus and the

the Craniata, to which reference has already been made at the beginning

of the article.

My obligations are due to PROFESSOR MITSUKURI and PROFESSOR

WATASE for their kindness looking through the manuscripts.

Biological Laboratory, the Gakushuin, Tokyo.

May 29th, 1901.

= Remittances from ee countries | should Le “ride by postal

iii

SITI

LES

OR HE ES Al ht

43h myo

Ki!

FR # iti

i *

Kane

yb FU

BR 4h

H =a

FF ua

vor FA

x #7 u

BB BR,

ANNOTATION ES

ZOOLOGICA, JAPONENSES

Vol. IV, Part II.

PUBLISHED

The Tokyo Zoological Society.

TOKYO

March, 1902.

CONTENTS:

N . 11

i AE PS

sì +

m. On hs Habits of the Ji apanese se Lingula. tir

By Naonrdé YATSU mn

x IV. Onanew Enteropneust from 1 Misti Balanoglossus

: kiensis n. sp. a À

| BI Faso

pes i a

re ñ AN 1,797

Über das Blutgefässsystem der Hirudineen.

Vorläufige Mitteilung

Von

Asajiro Oka, Tokyo.

Im Laufe einer ausführlichen Untersuchung über die Anatomie der

Hirudineen, die schon vor einigen Jahren begonnen wurde, bin ich

allmählig zur Überzeugung gekommen, dass die Auffassungen in Betreff

des Blutgefässsystems dieser Tiere, wie sie gewöhnlich in Lehrbuchern

angegeben sind, dem wirklichen Sachverhalte durchaus nicht entsprechen.

Ich erlaube mir im Folgenden einige Ergebnisse meines Studiums,

welche mich zu dem obigen Schluss geführt haben, schon jetzt kurz

mitzuteilen, da mir dieselben für das richtige Verständniss der Organisa-

tion der betreffenden Tiergruppe von fundamentaler Bedeutung zu sein

scheinen,

Die vorliegenden Angaben über die Bluträume der Hirudineen

verdanken wir hauptsächlich Leydig 1), de Quatrefages 2), Whitman 3),

und Bourne 4). Nach diesen Forschern bestehen die blutführenden

Räume bei diesen Würmern überall aus zwei ganz verschiedenen

Systemen, d.h. einem eigentlichen mit einer deutlichen Wandung

versehenen Blutgefasssystem und einen jeder Wandung entbehrenden

Lakunensystem ; die beiden Systeme kommen jedoch in den verschiedenen

Abteilungen der Hirudineen so verschiedentlich zur Ausbildung, dass es

unmöglich ist, davon ein für die ganze Gruppe geltendes Schema zu

konstruiren. So besitzen z. B. die Gnathobdelliden zwei Seitengefässe,

1) Leydig, Zur Anatomie von Piscicola geometrica mit theilweiser Vergleichung

anderer einheimischer Hirudineen. Zeitschr. f. wiss. Zool. Bd. I. 1849.

2) Quatrefages, Histoire naturelle des Annelés marins et d’eau douce. 1865.

3) Whitman, the Embryology of Clepsine. Quart. Journ. of mier. Se. XVIII. 1878.

4) Bourne, Contributions to the Anatomy of the Hirudinea. Quart. Journ. of mier.

Se. XXIV. 1884.

50 A. OKA

welcher die Glossiphoniden entbehren ; dagegen sind die letzterer mit

einem Rücken- und einem Bauchgefäss versehen, die den ersteren

abgehen. Bourne, welcher das Blutgefässsystem aller Hirudineen von

einem gemeinsamen Grundtypus abzuleiten versuchte, hielt die Organisa-

tionsverhältnisse, wie wir sie bei Pontobdella finden, für die ursprüng-

lichen, indem er glanbte, dass bei dieser Gattung simmtliche Liingsstiimme

sowohl des Blutgefässsystems wie des Lakunensystems, die überhaupt

bei Hirudineen vorkommen, gleichzeitig existirten. Hier soll nämlich

nach diesem Autor, das Blutgefässsystem aus vier Längsstämmen,

einem dorsalen, einem ventralen und zwei lateralen, bestehen, welche je

in einem entsprechenden Liingsstamm des Lakunensystems einge-

schlossen seien. Von diesen vier Längsgefässen entwickelten sich bei den

Gnathobdelliden nur die seitlichen, bei den Glossiphoniden dagegen nur

die medianen. Ähnlicherweise wären bei den Glossiphoniden alle vier

Längsstimme des Lakunensystems wohl entwickelt, während bei den

Gnathobdeiliden nur die medianen, bei der Gattung Herpobdella sogar

der einzige ventrale, zurückblieben. Bei allen diesen Formen sollen

übrigens beide Systeme von Bluträumen in offener Kommunication

stehen, so dass in beiden ein und dieselbe Flüssigkeit zirkulirt. Was die

morphologische Bedeutung der Gefässe und Lakunen betrifft, so sagt

Bourne „as De Quatrefages pointed out, the former (Gefässsystem) seems

to represent the closed vascular system of Chaetopoda, while the latter

(Lakunensystem) is probably coelom, whether in the state of new forma-

tion, or gradually becoming occuluded, i.e. in a state of degeneration.”

Wie aus diesem Überblick hervorgeht, bieten die Hirudineen nach

unserer gegenwärtigen Auffassungsweise im Bau des Blutgefässsystems

eine ganze Reihe von Eigenthümlichkeiten, welche uns zu rechtfertigen

scheinen, diese Würmer als eine höchst sonderbare. Abteilung des

Annulatenstammes zu betrachten; denn ein Gefässsystem, von dessen

vier Längsstämmen bald nur die medianen, bald nur die lateralen sich

entwickeln, weiches ferner mit einer stark reduzirten, gefässartig

aussehenden Leibeshohle in offener Verbindung steht, findet man sonst

wohl nirgends im Thierreich. Wie auffallend diese Organisations-

BLUTGEFÄSSSYSTEM DER HIRUDINEEN. Sil

verhältnisse auch zu sein scheinen, so begegnen wir doch in meisten

neueren Lehrbüchern lauter Angaben, die im Allgemeinen mit der obigen

übereinstimmen. So giebtz. B. Hertwig 1) an, “dass ihre Leibeshöhle

rückgebildet ist, indem sie durch Parenchymwucherung eingeengt und

zu Längscanälen, die mit dem Blutgefiisssystem in Verbindung traten,

umgewandelt wurde.” In ähnlicher Weise sagt Lang 2) “Beide Systeme

von Hohlräumen (Blutgefisse und Leibeshohle) stehen mit einander

derart in Verbindung, dass es schwer fällt, sie von einander zu trennen.”

Selenka giebt in seinem vorzüglichen Taschenbuch an, “Die reduzirte

Leibeshöhle kommunizirt mit dem wohlentwickelten Blutgefisssystem.”

Dass das Blutgefässsystem der Hirudineen allgemein mit der

Leibeshöhle in Kommunikation stehe, erschien mir um so sonderbarer,

als durch die Untersuchungen von mir3) und von Johausson 4) der

‘Nachweis geliefert wurde, dass bei einzelnen Gattungen das Blutgefäss-

system vollkommen geschlossen, also mit der Leibeshöhle in keiner

Weise verbunden ist. Ich habe nämlich gezeigt, dass bei den Gattungen

Glossiphonia, Helobdella, und Hemiclepsis das Blutgefässsystem aus

einem dorsalen und einem ventralen Längsstamm besteht, die an beiden

Enden durch eine bestimmte Anzahl von Schlinggefässen mit einander

kommuniziren, jedoch an keiner Stelle direkt in die Leibeshöhle

ausmünden ; und dasselbe bewies Johansson für die Ichthyobdelliden

Schwedens. Indem wir beide aber unsere Studien je an eine kleine

Abteilung der Hirudineen beschränkten, waren wir damals nicht im

Stande über das Blutgefässsystem der ganzen Gruppe Etwas bestimmtes

vorzubringen. Jetzt vermag ich, nach einer sorgfältigen vergleichend-

anatomischen Untersuchung aller Hauptabteilungen der Hirudineen, den

Gesammtbild der Bluträume dieser Tiere zu entwerfen, welcher eine

Homologisirung verschiedener Abschnitte derselben zu ermöglichen

scheint.

1) Hertwig, Lehrbuch der Zoologie.

2) Lang, Lehrbuch der Vergleichenden Anatomie.

3) Oka, Beiträge zur Anatomie der Clepsine Zeitschr. f. wiss. Zool. Bd. LVIII.

1894.

4) Johansson, Bidrag till kinnedomen om Sveriges Ichthyobdellider Akad. Afh. 1896

52 A. OKA.

Meine Beobachtungen beziehen sich auf die folgenden Genera :

Glossiphonidae, —Glossiphonia, Helobdella, Hemiclepsis.

Ichthyobdellidae,—Ozobranchus, Pontobdella, Callobdella, Pis

cicola,

Gnathobdellidae.—Hirudo, Haemadipsa.

Herpobdellidae.—Herpobdella, Orobdella, Mimobdella.

Von diesen zwölf Gattungen weisen diejenige, die zu derselben Familie

gehören, in Betreff des Verhaltens der Bluträume grosse Ubereinstim-

mung auf, so dass sie zusammen behandet werden können.

Glossiphonidae. Was das Blutgefisssystem von Glossiphonia,

Helobdella, und Hemiclepsis betrifft, so habe ich es in einer früheren

Abhandlung !) ausführlich beschrieben, so dass ich hier darüber nur

ganz kuzz mitzuteilen brauche. Bei diesen Gattungen besteht das

Blutgefässsystem aus einem dorsalen und einem ventralen Längsgefäss,

die an beiden Körperenden je durch einige Gefässschlingen mit einander

in Verbindung stehen. Die wohl entwickelte Leibeshöhle ist in drei

Längsstimme, zwei Seiten- und eine Medianlakune, zertheilt, welche

letztere wieder durch den Darmkanal mit seinen Ausbuchtungen in die

dorsale, die ventrale, und die Zwischenlakune gesondert werden können.

Die Seitenlacune trennt sich von der Medianlakune hauptsächlich durch

die Reihe von dorso-ventralen Muskeln, welche jedoch segmental durch

die Kommunikationslakune unterbrochen ist. Wie oben bemerkt steht

das Blutgefässsystem mit der Leibeshéhie nirgends in Kommunikation.

Die Fig. 1 stellt einen Querschmitt durch den mittleren Körperteil

einer Hemiclepsis (tesselleta) dar. .Die grosse Medianlakune ist

hier durch den Darm (D), den Hoden (T), und die dorsoventralen Muskeln

in mehrere Abschnitte (dl, vl, zl, zl) geteilt. Die Seitenlakune (sl) kom-

munizirt mit der medianen durch einen kurzen Querkanal (cl). Die

Hypodermallakunen, welche aus den Seitenlakunen entstammen und

sich dicht unterhalb der Epidermis ausbreiten, sind in der Figur nicht

1); 2. 2.0,,8.108.

BLUTGEFÄSSSYSTEM DER HIRUDINEEN. 53

repràsentirt. Vom

| Blutgefiisssystem be-

gegnen wir das Dor-

sal-(dg) und das Ven-

tralgefäss (vg); beide ( TUE

liegen frei in der ;

Uy S

dorsalen (dl), resp. ET TK, 7

ventralen Abteilung

(vl) der Medianla-

> 0. ; v

LY f Dein Vy

kune. EN

Ichthyobdellidae. el if ee ime

Die blutfuhrenden a i i À

1 Lo AR

Räume der Ichthy-

obdelliden bieten in Bigs).

sofern grosse Interesse, dass sie gewissermassen ein Übergangs

stadium zwischen Glossiphoniden und Gmathobdelliden darstellen

Bei allen vier Gattungen, die ich untersucht habe, verhalten sich die

eigentlichen Blutgefiisse genau wie bei den Glossiphoniden. Hier wie

dort finden wir ein Dorsal- und ein Ventralgefäss, welche an beiden

Körperextremitäten durch seitliche Aeste mit einander in Verbindung

stehen. Wie Johansson für die Ichthyobdelliden Schwedens ausdrücklich

betont hat, !) ist das Biutgefässsystem vollkommen geschlossen ; eine

_ offene Kommunikation zwischen den Blutgefässen und der Leibeshöhle

findet nirgends statt. So weit stimmen die Ichthyobdelliden mit

den Glossiphoniden überein. Ein merkwürdiger Unterschied aber

findet sich zwischen diesen zwei Familien in der Beschaffenheit

der Leibeshöhle ; er besteht nämlich darin, dass bei den Ichthyobdelliden

ein Teil des Lakunensystems, und zwar die Seitenlakunen, mit

1) a. a. O. 8. 87 “Lika skarpt vill jaz ocksa° beträffande de af mig undersögta

Ichthyobdellidsläktena betona, att blodkirlsystement ir fullkomligt slutet och ingenstädes

kommunicerar med kroppshiilesystemet,............... sd

54 A. OKA.

muskulöser Wandung versehen und dadurch kontraktil und täuschend

gefässartig geworden ist. Es eind dies die sogen. Seitengefässe der

Ichthyobdelliden, die man bisher mit Unrecht zum Gefässsystem

gehören liess. Auch ich habe sie anfänglich als wahre Blutgefässe

aufgefasst, aber eine genaue Untersuchung über ihren Verlauf und ihre

Verhältnis mit den übrigen Abteilungen des Lakunensystems überzeugte

mich bald von ihrer Lakunennatur.

Fig. 2 repräsentirt einen Querschnitt der Kiemenregion eines

Ozobranchus (branchiatus). Die wohl entwickelte Medianlakune ist

durch den Darm (D), den Hoden (T), und die dorsoventralen Muskeln in

verschiedene Abschnitte (dl, vl, zl) geteilt, gerade wie es bei Hemiclepsis

(Fig. 1) der Fail war. Die mit muskulöser Wandung ausgestattete,

gefässartige Seitenlakune (sl) steht mit der Medianlakune durch Vermittel-

ung der feinen Kanäle des Kiemenapparates nur indirekt in Verbindung.

Dass die letzteren keine Blutgefüsse, sondern nur Teile der Leibeshöhle

sind, habe ich schon an anderem Orte hervorgehoben.) Von dem

eigentlichen Blutgefässsystem sind in der Figur nur die beiden Längs-

gefässstämme (dg, vg) zu sehen, welche an der dorsalen, resp. ventralen

1) Oka, Description d’une espèce d’Ozobranchus, Zool. Mag. VII. 1895. 8.5.

BLUTGEFASSSYSTEM DER HIRUDINEEN

Wand des Darmes angehefted erscheinen.

Die obige Beschreibung passt für die drei anderen Gattungen ebenso

gut wie für Osobranchus. Nur hat man dabei statt der feinen Canale

des Kiemenapparates ein kontraktiles Seitenbläschen zu setzen. Die

paarig angebrachten Seitenbläschen von Pontobdella, Callobdella, und

Piscicola kommunizieren durch Verbindungskanäle einerseits mit der

gefässartigen Seitenlakune, anderseits mit der weiten Medianlakune

und stellen somit bloss einen Teil der Leibeshöhle dar.

Der Umstand, dass ein Teil der Leibeshöhle gefässartig und

kontraktil geworden ist, ist ohne Zweifel der Grund, warum die Bluträume

der Hirudineen, trotz zahlreichen Untersuchungen, bis jetzt in so

unrichtiger Weise aufgefasst worden sind. Bourne, welcher die Anwesen-

heit einer besonderen muskulösen Wandung als ein Kennzeichen der

Gefässnatur eines Hohlraums in Anspruch nahm, lässt die Seitenlakunen

von Pontobdella natürlich zum Blutgefässsystem gehören, und so

kam zu dem Gedanke, dass das Blutgefässsystem der Hirudineen

ursprünglich aus vier Längsstämmen sich zusamensetzten. Hätte

er den Bau von diesem Fischegel statt an vereinzelte Schnitte,

sondern an vollkommene Schnittserien untersucht, wie ich es

gethan habe, so könnte er unmöglich die Seitenkanäle als wahre

Gefässe auffassen. Sie laufen beiderseitig dem Körperrand entlang,

vereinigen sich am vorderen und hinteren Enden mit einander,

am Kopfende auch mit der kanalartigen Fortsetzung der Median-

lakune. In mittlerer Region des Körpers stehen sie mit der

Medianiakune durch die paarige Querlakunen in Verbindung, welche bei

den mit besonderen Athmungseinrichtungen ausgestatteten Formen

einen Umweg durch die letzteren machen. Sie stehen, ferner, mit dem

eigentlichen Blutgefässsystem an keiner Stelle in Kommunication.

Kurz, die Seitenkanäle der Ichthyobdelliden stimmen mit den Seitenla-

kunen der Glossiphoniden in allen wesentlichen Beziehungen, bis auf die

muskulöse Wandung, so vollkommen überein, dass man die beiden

Gebilde unbedingt als identisch betrachen muss.

56 A. OKA.

Die Behauptung Bourne’s, !) dass bei Pontobdella die kontraktilen

Seitenkanäle von einem “ lateral sinus” ähnlich umschlossen sind, wie

das Dorsal- und Ventralgefiss vom dorsalen resp. ventralen Lakune,

scheint nicht auf direkte Beobachtung, sondern vielmehr auf seine the-

oretische Betrachtung zu beruhen : denn er giebt selbst zu, dass “ this

is never really developed as a sinus around the lateral vessel (Seiten-

lakune), but there is looseness about the connective tissue outside the

muscular wall of the vessel.” Da sein “lateral vessel” nichts anders

ist, als die Seitenlakune, so kann ein besonderes “ lateral sinus ” aus-

serhaib dieses nicht existiren. Die schematische Figur vom Querschnitt

von Pontobdella, welche Parker and Haswell, Perrier, und Sedgwick

aus Bourne’s Arbeit entlehnen, repräsentirt demgemäss einen Zustand,

welcher in Wirklichkeit nie vorkommt.

Gnathobdellidae. Bei Hirudo und Haemadipsa bilden die sämmt-

lichen blutführenden Räume des Körpers ein einziges, zasammenhängen-

des System, durch welches ein und dieselbe Flüssigkeit zirkulirt. Dieses

System besteht hauptsächlich aus vier Längstämmen, zwei sogen. Sei-

tengefässen und zwei medianen Längskanälen, die oft auch Dorsal- und

Ventralgefäss genannt werden, jedoch offenbar nichts anders darstellen,

als durch Parenchymwucherung eingeengte Leibeshöhle.e. Wie sich

diese vier Längsstämme mit einander verbinden, ist schon durch zahl-

reiche, theils sehr minutidse Untersuchungen 2) zur Genüge bekannt.

Segdwick 3) vermuthet, dass auch bei den Gnathobdelliden ein gesch-

lossenes Blutgefässsystem nebst der reduzirten Leibesh6hle vorhanden

sei, aber diese Annahme ist neulich durch die genaue Untersuchung von

Goodrich 4) endgultig beseitigt worden. Von dem wahren Blutgefäss-

system, wie wir es bei den Glossiphoniden und Ichthyobdelliden vor uns

haben, findet man bei diesen Gattungen keine Spur.

2) z.B. Vogt und Yung, Praktische Vergleichende Anatomie.

3) Sedgwick, Students T'ext-Book of Zoology.

4) Goodrich, On the Communication between the Coelom and the Vascular system in

the Leech. Quart. Journ. mier. se. Vol. 42. 1899.

BLUTGEFÄSSSYSTEM DER HIRUDINEEN. 5T

Ein Blick auf die Fig. 3, welche einen Querschnitt durch den mitt-

leren Körperteil eines japanischen medieinischen Blutegels (Hirudo

nipponia) darstellt, wird al

schon genügen die oben

angeführten Verhältnisse Ve

klarzustellen.. Das enge

Dorsallakune (dl) befindet Vga 1

sich an der Rückenseite stl

des Darmes (D); an der 4

Bauchseite desselben be- oo

obachtet man die etwas 2

weitere Ventrallakune (vl)

mit darin eingeschlossenen

2 ANT c & LR A,

Bauchmark (N). Die sogen ev 3 27777, Yj

È 2 # SH, G Y

Seitengefiisse (sl) kommu- a TELL

ak . È I \

nizieren mit den medianen ul | \

4. T l À

Lakunenstämmen nur |

durch feine Kanäle. Weder Fig. 3.

das Dorsalnoch das Ventralgefass ist hier zu beobachten.

Herpobdellidae. Auch die Egel dieser Familie sind in Hinsicht auf

die Bluträume schon genügend untersucht worden; wir haben in der

Literatur Abbildungen, !) die den Verlauf und die Anastomose der

Hauptkanäle sehr schön wiedergeben, so dass ich es für unnötig halte,

sie hier zu beschreiben In allgemeinen Zügen näheren sich die Gat-

tungen Herpobdella, Orobdella, und Mimobdella, in der Beschaffenheit

dieser Räume, sehr an Hirudo.

Fig. 4, einen Querschnitt von Herpobdella ‘octoculata) darstellend,

zeigt, wie stark die Lieibeshöhle auch bei dieser Gattung reduzirt ist.

Sogar die Dorsallakune, die wir bei Hirudo finden, ist hier verschwun-

den, so dass im Ganzen nur drei Längskanäle (sl. vl) vorhanden sind.

Bei Orbdella und Mimobdella existirt noch eine schmale Dorsallakune.

1) Jaquet, Recherches sur le systeme vasculaire des Annelides. Mitth. a. d. Zool.

Stat. zu Neapel. VI. 1886.

58 A. ORA

Wenn man nun die Fig. 3 und 4 mit der Fig. 2 vergleicht, so wird

man sofort wahrnehmen, dass die sogen Seitengefässe von Hirudo und

Herpobdella gevau

entsprechende Stelle

einnimmt, wie die von

Ichthyobdelliden ; ihre

Verbindungsweise mit

anderen Bluträumen ist

in beiden Fällen auch

fast dieselbe. Sie sind

in der That identische

Gebilde. Wie ich das

Seitengefiiss von Ich-

thyobdelliden als Ho-

mologon der Seitenla-

kune von Glossiphoniden betrachte, so kann ich bei den Gnathobdelliden

und Herpobdelliden auch nicht umhin, sie als solches anzusehen. Nach

meiner Meinung bilden die sämmtlichen Bluträume des Gnathobdelli-

denkörpers eine stark reduzirte Leibeshöhle; von dem, was dem

geschlossenen Blutgefässsystem der Chaetopoden entsprechen sollte,

besitzen die Egel dieser Familie nichts. Was man bis jetzt als wahre

Blutgefässe andeutete, stellt nichts anders dar, als gefässartig umge-

wandelte Teile der Leibeshöhle.

Dass man die Leibeshöhle gewisser Hirudineen mit dem wirklichen

Blutgefäss verwechselt hat, rührt hauptsächlich daher, dass man beim

Untersuchen des Vascularsystems vielfach den medicinischen Blutegel

als Typus nahm, und durch Injektionen den Verlauf einzelner Kanäle

zu verfolgen sich bemühte. Durch solches Verfahren gewinnt man

freilich è priori den Eindruck, als wären die Kanäle Blutgefässe.

Aus den obigen Krörterungen wird man leicht ersehen, dass bei den

Hirudineen, unter dem Namen Blutgefäss zwei Dinge von ganz ver-

BLUTGEFÄSSSYSTEM DER HIRUDINEEN. 59

schiedener Natur umfasst worden sind; einmal, das eigentliche Blutge-

fäss von Glossiphoniden und Ichthyobdelliden, und dann, die gefäss-

artige Seitenlakune von Ichthyobdelliden, Gnathobdelliden, und Her-

pobdelliden. Sogar die Dorsal- und die Ventrallakune wurden manchmal

als Blutgefässe aufgefasst. Es ist einleuchtend, dass so lange solche

Verwirrung herrscht, von einer richtigen Verständnis der Organisation

der Hirudineen keine Rede sein kann.

Kurz zusammengefasst lässt sich meine diesbezügliche Meinung,

wie folgt, ausdrücken :

1. Eine Leibeshöhle kommt bei allen Hirudineen vor; sie tritt

bei den Glossiphoniden und Ichthyobdelliden viel deutlicher hervor, wie

bei den Gnathobdelliden und Herpobdelliden. Bei diesen ist sie durch

Wucherung des Bindegewebes etc. sehr eingegengt.

2. Ein eigentliches Blutgefässsystem besitzen nur die Glossipho-

niden und Ichthyobdelliden. Dasselbe ist vollkommen geschlossen und

ist im Allgemeinen wie das Blutgefässsystem der Chaetopoden gebaut.

Was man bei Gnathobdelliden und Herpobdelliden Blutgefässe nannte,

ist bloss gefässartige Teile der Leibeshöhle.

3. Die Ichthyobdelliden stellen gewissermassen ein Uebergangssta-

dium zwischen den Glossiphoniden und den Gnathobdelliden dar, indem

bei ihnen sowohl das wirkliche Blutgefässsystem, als die gefässartige Sei-

tenkanäle vorhanden sind.

Am Schluss möchte ich noch einige Worte über die Verwand-

schaftsbeziehungen von Hirudineen mit anderen Tieren hinzufügen.

Wie aus den oben angeführten Gedanken ersichtlich, fasse ich die Hiru-

dineen als coelomate Gliederwürmer auf, die den Chaetopoden, und zwar

den Oligochaeten, mit welchen sie auch in der Zwitternatur der Ge-

schlechtsorgane übereinstimmen, am nächsten stehen. Unter den Hirudi-

neen stehen wieder die Glossiphoniden den Regenwürmern am nächsten ;

dabei stellen Branchiobdella und Acanthobdella die Uebergangsformen

zwisehen den beiden Gruppen dar. Die nebenstehende, aus Kowale-

vesky’s Arbeit entlehnte Figur, welche einen Querschnitt durch Acan-

thobdella repräsentirt, zeigt zur Genüge, wie durch diese Tierform die

60 A. OKA

Glossiphoniden sich an Oligochaeten knüpfen lassen. Auf der anderen

Seite, nun, bilden die

Ichthyobdelliden eine

Zwischenform, welche

die Glossiphoniden mit

den Guathobdelliden ver-

bindet ; und so haben wir

hier gleichsam eine Kette

von Tierformen, die mit

den Regenwürmern an-

fängt, und mit dem me-

dieinischen Blutegel und

Herpobdella endigt. An-

gesichts solcher That-

Fig. 5. sachen leuchtet es von

selbst ein, dass man Unrecht hat, die Hirudineen als eine besondere

Unterclasse der Anneliden, oder gar als gegliederte Platt würmer zu

betrachten.

Tokyo, d. 15. Oktober, 1901.

On the Habits of the Japanese Lingula.*

Naohide Ya tsu : Rigakusht.

The present short notes are published as a supplement to the

cecological observations of the previous writers, such as MORSE, SEMPER,

and FRANÇOIS. But before entering into the habits I shall touch upon

the distribution and the specific characters of the Japanese Lingula.

From Japanese waters five species of Lingula have been described.

They have however, been declared by DAvipso to be varieties of one

and the same species (LiscHKE '69 p. 115). Whether this view is a

correct one I cannot at present determine, since as yet I have not had the

opportunity of examining materials from various localities in Japan. In

the museum of the Zoological Institute there are specimens from

Misaki, Hyakkan-zeki (Kyushü), and the islands of Bizen. Of these

localities the second yields very large specimens and those of slightly

different features from those collected at any other localities. At any

rate all the specimens I have examined tally with Lingula anatına

Bruce.

BLOCHMANN (00. p. 94-95) enumerates several distinguishing

characteristics of Lingula anatina BRUG from Lingula murphiana

Kıng. Of these only the mode of branching of the pallial sinus seems

to be a reliable criterion, all others being mere individual differences. In

the specimens examined by me, the peduncle, for exmple, in some in-

dividuals terminates in an ampulla, while in others from one and

the same locality it ends in solid attenuated cuticle. Some gape a little

in the posterior part of the shell. while in others the shells are closed

firmly all around the margins. Proportions of length, breadth and

* This formed a part of my graduating thesis, but, as it has less connection with my

embryological studies, I have thought it best to make it a separate paper.

62 HABITS OF LINGULA.

thickness also vary a great deal; in some specimens the ratios agree

with the figures of L. anatina given by BLOCHMANN, while in others

with that of L. murphiana. In all cases the ventral valve is a little

thicker than the dorsal. The largest specimen I have examined is 45

mm. in length, 20 mm. in breadth and 10 mm. in thickness. Of the

specimens from Misaki the largest measures 36.5 mm. in length and

16.5 mm. in breadth.

As far as I have been able to ascertain, Lingula occurs only along

the coast of southern Japan, but in that part one may expect to find it in

every suitable mud-flat. It is not found in deep water at all. In some

localities it occurs ina great abundance. At Yanagawa, a village by

the Ariake bay, Kyushü, Lingula is so plentiful that it is sold by the

peck and forms an article of diet. In this region the whole animal is

boiled and eaten. At Akura, Bizen, it is said that the peduncle only is

used for food. In the vicinity of Misaki Lingula was formerly so

abundant that we were frequently able to secure hundred or more

specimens in a few hours’ collecting, but of late years the number

seems to have dwindled, owing no doubt to too great a demand made by

naturalists on the supply. Still, in the neighborhood of the Station every

mud flat yields specimens of Lingula. Such flats are exposed at low

water and are covered at high water to the depth of 3-4 feet ; of these

the most favorable for Lingula are composed of sandy mud, blackened

and ill-smelling from the decomposition of organic matter. Over such

black mad a layer of somewhat dirty brown mud 2-3 mm. thick is

found which is probably formed by the excrements of Annelids, Lingula,

etc. The flats of clean sand, do not seem fit for the animal. Also

the flats* in which Lingula lives are covered neither by Phanerogams

nor by Algae, different from the habitat of Lingula at Numea, New

Caledonia, as described by François (91) which is covered by marine

Phanerogams.

* Only one flat, where a few individuals live is covered, as an exception, by Zostera

nana,

N. YATSU. 63

A visit to the collecting ground which is exposed at the time of

low water shows at once that no trace of Lingula can be detected

from the surface. ‘The reason is probably as follows : the retreating

tide disturbs mud, so that the entrance of burrow is stopped up and

at this the animals retire deep in their burrows. This fact stands in

opposition to the observations of FRANÇoIs and those of my friend Mr.

Namive who informs me that he was able at Bizen to find out the

specimens by three holes on the surface of mud. In the localities examin

ed by these observers, I think, the mud must be so hard or heavy that it

is not easily disturbed by the tide. On digging, however, specimens

are soon met with at a depth of about one foot. They show no sign of

attachment to either rock or stone but it is so firmly fixed to the hard

sandy mud that, when suddenly drawn out, the entire peduncle or its tip

is apt to be torn off and remain in the mud. On pulling out the

specimens we find that a bit of mud is usually encrusted at the tip of

the peduncle, and in some cases the latter is attached to a fragment of

bivalves. This stands in opposition to the description of DALL, which

runs as follows: “the true Lingulas are almost always attached to a

fixed rock or stone, while Glottidia attaches itself, if at all, only when

adult, and usually to a very small pebble or bit of shell” (BEYER '86

p. 228). The tube-forming around the peduncle seldom takes place and is

not so usual a phenomenon as described by Morse. Nearer the shore

are usually found younger individuals. It seems that the free-

swimming life terminates in the shallows. The burrow is not

always vertical; sometimes it is oblique or nearly horizontal, or

sometimes it is bent in the middle, the lower limb being vertical. There

are some cases in which two individuals are found in the same burrow,

but this must be regarded as accidental; for in the first place their

power of locomotion is very limited, and secondly as fertilization takes

place in the opeu water it is of little advantage for them to dwell in

pairs.

For the purpose of studying the cecology of Lingula I kept in the

autumn of last year, a number of individuals of various sizes in

64 HABITS OF LINGULA.

cylindrical specimen jars, in which the mud was so placed as to imitate as

far as possible the natural conditions. I buried a specimen deep in the

mud, touching the inner wall of the jar in order to observe how it makes

its burrow. At first the water is forcibly gushed out of the central

funnel formed of the mantle edge and the setae. By this the mud above

the animal was gradually removed. By the above action coupled with

the sliding lateral motion of the shell, the burrow was soon formed,

whose walls were made smoth owing to the secretion of the gland-ridge

(Drüsenwall) and to the up-and-down gliding of the valves. Seen from

above the burrow gives a rhombic aspect a little larger than the section

of the animal. Lingula can extrude itself above half of its shell length

from the surface of the mud; when alarmed it would retire 5-30 cm.

below the mud surface according to the size. At that time we can

perceive how great the contractibility of the peduncle is. When the

animals are dug out of its burrow the peduncle is of the minimum length.

The peduncle, it should be noted, of our Lingula is comparatively short.

In Glottidia according to MoRsE, it reaches nine times the shell

length. This difference in length of the peduncle can readily be noticed

so early as the stage of 7-9 pairs of cirri.

Some Brachiopods can protrude their arm-apparatus out of the

shell as observed by Morse (79). According to him, such is the case

in Lingula, (p. 257). On the contrary SEMPER (64) states of the

animal “dass Arme niemals zur Schale hervor gestreckt werden und

sich niemals enrollen’ (p. 424). To settle

this point I placed a number of individuals

in a glass vessel and looked at them carefully.

At last I found that only the comb-like row

of cirri of the largest whorl of the arm

can be projected out of the shell as in the

cut (A and B) but the tip of the arın-

apparatus is always kept within the mantle

cavity. The protrasion of the cirri, that is

fore-ward and backward movement of the

N. YATSU. 65

arm-apparatus is best observed in younger individuals.

Among the specimens from Hyakkanzeki I found some shells,

both valves of which are partially covered by some tubes of polychaetous

Annelids, lying parallel to the longitudinal axis of the shell.

The peduncle has a great regenerative power. We often meet with

large individuals with a very small and slender peduncle. As for the

mode of regeneration I shall put it off till I study this structure more

carefully.

As for the spawning habits of Lingula the reader is referred to my

embryological paper.*

It is not an easy task to determine how iong the life of Lingula is.

To do this one must observe definite specimens in aquaria as well as in

their natural habitat for many years. MorsE (73) states “ they

(Groruww) were all of the same size and their life does not exceed

one year” (p. 46) Somewhat later FrANgoIS (91) has declared

that Lingula lives “au moins plus dune année” (p. 232). The

small individuals I was able to secure in the autumn at Misaki were

5 mm. in shell length on an average. Assuming that the growth of the

shell goes on constantly and also that there is only one spawning season

in the year, the fact that smaller individuals than the above size do not

occur, proves that it takes one year for the larvae to grow up to this

size, To attain the maximum length (35 mm.) of the shell at Misaki

at the same rate would require 7 years. Accordingly, the conclusion

is not unreasonable, I believe, that at Misaki Lingula lives for five

years or more on an average.

That Lingula could have lived and survived throughout the whole

geological periods is due, one would naturally suppose, to their great

power of withstanding unfavorable conditions. This is proved by the

fact that Lingula lives for a considerable lengthe of time in aquaria

even after the water has become spoiled and foetid. A remarkable

case justifying the above supposition has been brought to my notice

XVII. (in press.).

45 HABITS Ol LINGULA.

by Mr. Hara, Professor of Biology in the Peer's School. Near

Kumamoto there is an inlet called Matsubasé which forms a part of the

Bay of Ariaké. ‘The vast area of the inlet is covered by a sandy

flat (Kitadasu) in which Lingula, Tapes, Cytherea, etc. are found

in a great abundance. At every low tide thousands of the poor

people crowd on the flat to dig the edible Lamellibranchs. In his

collecting trip to that district in July 1887 Mr. Harra observed the

following striking facts. It was after a season of heavy rain which

lasted twenty days or more that he visited that bay. ‘The sandy

flat was found entirely covered by a torelably thick layer of muddg

sediment which had been brought down by a brook (Asakawa) ruuniny

into the creek. All the Lamellibranchs had been completely destroyed

under these unfavorable circumstances and were found already

putrified emitting a horrible smell, and poor villagers were very much

vexed at having thus lost their living. To his great surprise Mr. HATTA

found that Lingula alone had continued to live in as excellent a condition

as it had ever been.

It is a very striking phenomenon that Lingula has continued to

live since the Cambrian period and much more so that its form has

undergone but little change during such a vast length of time. There

is an interesting incident which I have recently learned from

Pror. DEAN. It is as follows. MoRSE succeeded in carrying home

to America with him living specimens of Japanese Lingula, and he had

satisfaction of placing living specimens upon a ledge of Cambrian limestone

among the primaeval but hardly different shells of their ancestors !

Lingula had already acquired as long ago as the Cambrian period

an organisation most favorable for facing all the ambient conditions,

physical as well as chemical, that have taken place since that time, and

there seems to have been no necessity for improving their adaptations

to the environment.

Zoological Laboratory,

Imperial University, Tokyo.

October, 1901.

N. YATSU. 67

REFERRENCE PAPERS.

BEYER (H.-O.) (86). A study of the structure of Lingula (Glottidia)

pyramidata, Stim. (Darr). (Stud. Biol. J. Hopkins Univ. IIT.

p- 227-265.)

BLocHMANN (’00). Untersuchungen über den Bau der Brachiopoden.

2ter Theil.

Frangoıs (‘91)—Choses de Naumea. Observations biologiques sur les

Lingula. (Arch, Zool. Exp., 2 Ser., IX p. 231-239.)

LiscHKE (69). Japanische Meeres-Conchylien. p, 115.

Morse (’73). ‘The Systematic Position of the Brachiopoda. (Proc of

the Boston Society of Nat, Hist. XV. p. 315-373.)

Morse (79). Notes on the Extrusion of the coiled arms in Rhyncho-

nella (Amr. Jour: Sc. and Art. Vol. XVII, 1879 p. 257.)

SEMPER ('64). Reisebericht. Zeit. Wiss. Zool. XVI Band. p. 424.

The Salmon and Trout of Japan.

David Starr Jordan.

Of all the families of fishes none offer greater ditticulties to the

student of classification than the Salmonide. This is due to the ex-

tremely close relation existing among the species themselves and to the

great range of variation offered by the individuals. No fishes respond

more quickly to the influences of environment, none vary so greatly with

the conditions of food supply, and none show greater changes with the

different conditions of seasonal and sexual development.

Having carefully examined all the many specimens of salmon and

trout obtained by Mr. Snyder and myself in the waters of Japan, in the

summer of 1900, I am able to point out the distinctive characters of the

species represented in this collection, which includes all the species thus

far ascribed to the Japanese fauna.

1. The Sake or Dog Salmon ; Oncorhynchus keta (Walbaum).

The genus Oncorhynchus Suckley includes the salmon of the North

Pacific, six known species in all. It is characterized by the increased

number of anal rays, branchiostegals and gill rakers. Its species are

resident in the sea, ascend the rivers to spawn, do not feed in the

spawning season, and hence except when land-locked do not feed in fresh

waters. The males become hook-jawed and greatly distorted in the

spawning season, and after depositing the eggs and milt all of both sexes

die.

The largest of the Japanese salmon is the Sake, called Dog Salmon

on the American coast and Haiko in Siberia, Oncorhynchus keta

(Walbaum). It reaches a weight of 10 to 15 pounds and runs in the fall

in the rivers of Hokkaido and northern Hondo. In the character of its

70 DAVID STARR JORDAN.

flesh it is inferior to all the other salmon. For canning it is entirely

worthless, the flesh being pale and becoming soft and pulpy when boiled.

As a salted fish its rank is higher and many hundred thousands of pounds

are sold each year in Japan.

The question is sometimes raised, why cannot salmon be put in tin

cans with profit in Japan as well as in the United States? Why do the

Japanese rivers produce inferior salmon? Simply because they are

occupied by an inferior species. ‘The Sake is found also in Puget Sound,

Canada and Alaska, but it is everywhere rejected by the Salmon-canners

and used only as salted by the Indians.

It is a matter of great economic importance for the government ot

Japan to substitute a better salmon for the Sake. The most valuable of

all salmon, largest in size, finest in flesh, and most prolific is the

American Quinnat or King Salmon Oncorhynchus tschawytscha. This

species is easily transplanted, reaches an average weight of 22 pounds,

and for every purpose is superior to any other salmon. I urge the

planting of this species in the Kitakami and the Ishikari Rivers as a

matter of vital importance to the fishery interests of Japan.

The Sake may be known from other Japanese salmon by its larger

size, its mottled color, its broad tail, and especially by the medium size

of its scales, there being 160 to 170 in a lengthwise series from head to

tail. The fishes called Oncorhynchus haberi by Dr. Hilgendorf belong to

Oncorhynchus keta. We have the Sake from Ishikari River, from

Aomori and Hakodate.

2. The Yezomasu or Hokkaido Salmon ; Oncorhynchus yessoensis

Hilgendorf.

All the smaller salmon as well as the large trout are called in Japan

Masu, and neither fishermen nor literary men seem to make any distinc-

tion among the three or four species confused under that name.

The present species is known at once from all the others in Japan

by its very small scales, there being about 195 in the lateral line. The

only other salmon-like fish having scales so small are the fresh water

THE SALMON AND TROUT OF JAPAN. 74

gray-spotted trout, Iwana, Amemasu, and the Ayu.

This species is found only in Japan. It reaches weight ot about

eight pounds, and it is caught and salted with the Sake, to which it is

probably not much superior. Our specimens are from the Ishikari

River and from Aomori, besides a number of small ones said to be from

the Daiya River at Nikko.

3. The Gimmasu or Silver-Salmon ; Oncorhynchus kisutch Walbaum.

This is the smallest of the true salmon, rarely exceeding six pounds

in weight. It is common in the rivers of Northern Japan, as also in

Siberia, Alaska and British Columbia. On the American coast it 1s

known as Silver Salmon, and to the Russians as Kisich. Like the two

preceding it ascends the rivers in the fall, and it ascends small streams as

well as large ones. Asa food fish it is regarded in America as superior

to the Dog Salmon (Sake), but much inferior to the Red Salmon

(Benimasu).

From the salmon thus far mentioned it is known by the large size

of its scales, which number about 125 in a lengthwise series. It lacks

the large number of gill rakers found in the Benimasu. It has more

anal rays than the Yamabe, and it lacks the black spots on the base of

the dorsal characteristic of that species. The dorsal fin in Oncorhynchus

kisutch is always tipped with inky black at all ages and stages, and by

this character the species can usually be known. As it is called Silver

Salmon in America, it will not be improper to use tbe equivalent name

Gimmasu in Japanese. Our specimens are from Otaru, Aomori,

Osatsube and Ura River. This is the species wrongly called by

Hilgendorf Oncorhynchus perryi.

4. The Benimasn or Red Salmon ; Oncorhynchus nerka (Walbaum).

This species, the handsomest of the salmon, may be known by the

large number of its gill-rakers, about 36. It runs in the spring, weighing

from four to eight pounds, and has a great liking for cold lakes. In such

localities it often becomes land-locked, never descending to the sea, and

when so placed it reaches a very small size. In running from the sea,

12 DAVID STARR JORDAN.

its color is bright blue and silvery, without spots. In the fall it becomes

very red, when it is known as the Red-fish, or in Japan as the Benimasu.

In spring it is called Blue Back. The young specimens are faintly

spotted, but those land-locked are often profusely marked by black spots.

The flesh of the Benimasu is rich and delicate, much superior to

that of the other Japanese salmon and inferior only to the American

Quinnat Salmon, Oncorhynchus tschawytscha. The range of the species

is northern. In America it is not fouud South of the Columbia. In

Alaska and Siberia it far out numbers all other species and furnishes the

chief product of the canneries. In Japan it is said to be known only in

Kushiro in Northern Hokkaido, where dwarf examples enter Lake Akan.

trom Lake Akan our specimens were received through the courtesy of

Mr. Nozawa of the Museum of Sapporo.

5. The Yamabe or Mountain Trout ; Salmo perryr Brevoort.

In all clear streams through Hondo and Hokkaido, trout are found

in abundance. Small ones are found in small streams and large ones in

larger rivers, while those weighing ten or twelve pounds are often taken

in the sea, especially in the north. To the small ones, the name Yamabe,

with its variants Yamabai, Yamami and Yamomi, is given. The large

ones are called Masw by the fishermen, who do not discriminate between

them and the salmon. |

They belong, however, to a different genus, Salmo, characterized by

the presence of not more than 13 anal rays, not more than 13 bran-

chiostegals, and not more than 20 gill-rakers. The sexual changes are

much less than in Oncorhynchus. The species feed chiefly in fresh

water, and as a rule the individuals do not die after spawning.

In all its stages of growth, the Yamabe may be known by the

presence of black spots along its back and dorsal fin, 3 to 5 on the base

of the dorsal being especially conspicuous. The scales are from 130 to

HO.

It is always a good food fish, its flesh being firm and red, much like

that of the Atlantic Salmon, Salmo salar. It is an excellent game-fish,

THE SALMON AND TROUT OF JAPAN. Ts

active and vigorous, taking the hook readily in the mountain streams.

The fish called Salmo macrostomus by Günther is the same as S.

perryi. There may be other species of true Trout or Salmo in Japan,

but all that we have seen certainly belong to one species.

Our specimens are from Lake Chuzenji, Kinu River at Utsunomiya,

Daiya River at Nikko, Otani River near Nikko, Maebara and Karasaki

on Lake Biwa, Kawajırı in Rikuchu, Kitakami River, Aomori, Hakodate,

and the market at Tokio. It is the only trout or salmon found by us in

Lake Biwa. In Lake Biwa, the fish is called Amenouwo.

Besides the native Yamabe two other species of Salmo, the American

Land-locked Salmon (Salmo salar sebago) and the European Brook

Tront or Forelle, (Salmo fario), bave been introduced into Japan. The

latter species, which much resembles the Yamabe, I found in Chuzenji

Lake.

6. The Ito-Uwo or Blackiston’s Trout ; Hucho blackistoni

(Hılgendorf).

Most interesting of the Japanese trout is the long, slender pike-like

species, called for its slimness the Ito-Uwo. This belongs to the genus

Hucho, characterized by its flattened head, large mouth, and by the

absence of teeth on the shaft of the vomer, which is formed as in

Salvelinus. ‘The scales are silvery, finely speckled with black, and in

this and other respects, the genus is intermediate between Salmo and

Salvelinus. The anal fin is short as in Salvelinus, and the gill rakers

are few. The scales are about 113. Two species of Hucho are known,

one, Hucho hucho, from the Danube, the other, Hucho blackistoni of

Japan.

Hucho blackistoni is found in mountain streams of the north where

it said to reach a length of 24 feet. Our specimen, received through the

kindness of Mr. Nozawa, is from Kushiro. We have examined others

from Nemuro, Chishima, Settsu and from Heigun River in Rikuchu.

7. The [wana or Rain Charr ; Salvelimus pluvius (Hilgendorf).

The genus Salvelinus containing the charrs, red-spotted and white-

74 DAVID STARR JORDAN.

spotted trout, is characterized by its very small scales and by the form of

its vomer, in which the shaft is depressed ont of line with the head of the

bone, and is without teeth at any age.

The species are all delicate, swift, brightly colored fishes living in

mountain streams and lakes, not descending to the sea except in the far

north. In general, the charr are the most vigorous of the trout, the

most gamy and most prized by the angler. ‘The pale spots are bright

red when the fish lives in dark pools, yellowish in lakes, and grayish

when the fish goes into the sea. All Salmonidw become silvery after a

stay in salt water. The /wana is the only pale-spotted species found in

Hondo. It may be known from Salvelinus kundscha by the smaller size

of its spots, always smaller than the pupil. In Salvelinus malma, the

spots are still smaller, and the mouth is smaller, not reaching beyond the

line of the eye.

Our specimens are from Lake Chuzenji, Ohata River at Aomori, and

Kawajiri in Rikuchu. The species is often called Amenouwo, but that

name is used also for the Yamabe.

S. The Malma ; Salvelinus malma (Walbaum).

This fine charr which is very close to the Iwana abounds in the

Aleutian Islands, and westward to Komandorski, the Northern Kuriles

and is recorded from Decastris Bay in Manchuria. It differs from the

Iwana in the smaller spots, and smaller mouth. We are, however, by

no means certain that the Iwana is more than a southern variety of tbis

species, as the differences are very slight.

9. The Amemasu or Yellow Spotted Trout ; Salverinus kundscha

(Pallas).

This charr may be known at once by the large yellowish spots,

larger than the pupil. It is a northern fish, abundant in Kamchatka,

rarely seen so far to the southward as Hokkaido. Our specimens are

from Petropaulsky, Tareinsky and Nemuro. ‘We have examined others

from Shinbeshi and Iturup Island. Brevoort records it from Hakodate.

THE SALMON AND TROUT OF JAPAN,

10. The Ayu or Sweet-fish ; Plecoglossus altielis Schlegel.

The Ayu is a dwarf Salmon, with the teeth of the jaws modified in

a very peculiar fashion, as serrated plates. It is found in all the rivers

of Japan throughout the four main islands. It is one of the most

delicious of food fishes, and from its abundance one of the most valuable.

It is of all the Japanese fishes the one which best deserves introduction

into other regions. It would donbtles thrive in the clear swift stream of

California and Chili, as well as in those of New Zealand and Mexico, and

probably in England aud France as well. Its white flaky flesh, similar

to that of the smelt or even finer is peculiarly delicate and nutritious.

Our specimens are from Same, Ishikari River, Aomori, Matsushima,

Tokio, Nikko, Morioka, Sendai, Gifu, Lake Biwa, Wakanoura, Kobe,

Hiroshima, Osaka, Kurume, Nagasaki and Taihoku, Formosa.

On a new Enteropneust from Misaki, Balano-

glossus misakiensis n. sp.

Hisato Kuwano.

Since the description by EscHscHoLTz, of Ptychodera flava, the

first Enteropneust mentioned in literature, thirty species of this

group have been reported up to the present date, from different parts of

the world by different writers. But our knowledge in regard to the

Enteropneustic fauna of our part of the world remains as yet very scanty.

Only two species, Glandiceps hacksi MARION and Dolichoglossus sulcatus

SPENGEL have been reported from the Main Island of Japan. Of these

the second species has been left with a rathar meagre description owing

to the lack of specimens. Under these cireumstances, I have thought it

worth while to devote myself to a study of the animals of this group

found in the vicinity of Misaki and I have succeeded in gathering fonr or

more different species.

The present species is one of these and appears to me to be new to

science. The animal is found in a sandy flat in the cove of Moroiso,

opposite the Marine Station.

The habitat is limited to a small area ot the flat exposed at the time

of spring tides. It is one of the gregarious species and lives in the same

locality with two different species of Balanogiossus DELLE CHIAJE, one

of which is smaller while the other is larger.”

Though the animal is one of the burrowing spacies, I have not found

it deeper in the sand than at the most two feet in depth, and have never

1.) I have been told by Dr. Oka and by K.?Awoki, Collector in the Marine Station’

this species is also fouud in the sandy beach of Tateyama, Province of Awa.

2.) The occurrence of different species of Euteropneusts in the same locality as co-

inhabitants have been frequently observed, as Ptychodera flava and Glossobalanus hedleyi in

Funafuti, Ptychodera flava and Spengelia porosa in Lifn and Glossobalanus ruficollis and

Balanoglossus carnosus in New Britain. (Hill, 1893, p. 205: Willey, 1899, p. 244.)

78 HISATO KUWANO.

found it among the meshes formed by the roots of Zostera which flour-

ishes here and there over the flat, unlike the above mentioned smaller

species which is often found among such meshes.

In natural condition, this species lies always horizontally and does

not make any building over the flat, while the larger co-inhabitant stands

vertically with up-turned anal end and makes a sand cone over the open-

ing of its burrow, on which castings are always found.

In a large percentage of specimens one, two or more of a small

commensal crab, yellowish brown in colour, are found in the same burrow

creeping about over the body of the worm.

Like some other known species, the anımal has an odour resembling

that of iodoform. This is so strong that by smelling the sand we are

able to know where the animal is to be found. Besides, the animal

emits readily a bright phosphorescent light making it a beautifal object

in the course of shore collection at night.

Though I am unable to say decidedly that the animal practises

autotomy, it is at least very fragile so that it breaks off easily to pieces

in spite of a cautious treatment and the fragments of the caudal or even

the hepatic region often turn themselves inside out.

They seem to have a power to regenerate the lost or injured portions of

the body, it is certain that at least the proboscis and the posterior portion

of the trunk are capable of being regenerated from the remaining portion.

The breeding season may be given as mid-summer (from the end of

July through August.)

Now I will pass into the descriptions of some important anatomical

characters of this species leaving the details to be published in a report

which I am preparing on this and other forms.

FAM. PTYCHODERIDA, Spengel.

GEN. BALANOGLOSSUS, Delle Chiaje.

Balanoglossus misakiensis, n. sp.

External form (Fig. 1):—Proboseis subconical ; proboscis-pore opens

slightly to the left of the dorsomedian line of the proboseis-neck ; collar

high cylindrical ; on the dorsal side of the anterior surface of the collar (in

ON A NEW ENTEROPNEUST FROM MISAKI. 79

front of the insertion ot the proboseis-

neck), a number of transverse epidermal

groove are constantly seen ; five epidermal

zones and collar grooves are distinct; genital

pleuræ are free from the posterior rim of

the collar and their free edges do not come

in contact with one anotber at the anterior

portion, while posteriorly they end abruptly,

just anterior to the hepatic region; in some

specimens they may extend themselves

along, and just outside, the anterior series of

the hepatic saccules; branchial tract is in the

shape of an elongated isosceles triangle with

a posteriorly turned apex, on each side of

which a deep abrupt depression of the epi-

dermis makes its appearance! (Fig. 1.); hepa-

tic saccules arrange themselves more or less

regularly one after another and present

cushion like epidermal thickenings richly

supplied with mucus glands; epidermal strips

paired and are limited into the abdominal

region; Epidermis presents an annular

markings in the caudal region and a semi-

annular in the ventro-lateral surface of the

branchiogenital region while in the hepatic

region, it makes fine cross markings, all

of which are interrupted by the sagittal

median lines of the body ; in the branchial

tract, epidermis is divided up to a series of

Fig. 1.

Dorsal view of a preserved specimen.

(3).

1.) We have similar cases in Glossobalanus hedleyi (Hill, 1898, p. 340.) and especially in

Glossobalanus ruficollis, in the later of which “ the depression is localised in the posterior end

of the branchial region” and Willey suggests its strong resemblance to the dermal pits of

Spengelia porosa (Willey, 1899, p. 263 and 275.)

80 HISATO KUWANO.

square areas ; anus dorso-terminal.! (Fig. 1.)

Colour :—the ground colour is of an uniform ochre-yellow, proboscis

and collar groove are of faint yellow; in sexually immature animals,

genital pleuræ are of a deep yellow while in the sexually matured they

are of dirty grey in female (due to the colour of eggs) and of yellow in

male ; the hepatic region commences with a few of brick-red saccules

followed by many of dark brown saccules which are again followed by a

number of large yellow saccules passing over into a long stretch of green

saccules.

MEASUREMENT: (in mm.)

Prob. Coll. Branch. reg. Gen. rev. Hep. reg. Caud. reg.

Length. - 7 9 40 62 140 125

Width 229 ih

Total length ca. 400 mm.

INTERNAL STRUCTURE.

Proboscis :—T'he circular muscle layer is interrupted by the dorsal

insertion of the pericardium ; longitudinal muscle fibres spring in the

proboscis wall near the bottom of each proboscis-canal and are inserted on

the anterior portion ot the wall, across the

circular muscle layer, with an attenuated high

refractile, difficultly staining terminal por-

tion, simple or branched; dorsoventral muscle-

plate extends beyond the front end of the

stomochord*; ventral septum stops short

behind the anterior extremity of the stomo-

Alveolar structure of the connective chord : right dorsal proboscis-canal ends

tissue in the proboscis coelom.

Cry imm.X4). blindly while the left opens to the exterior

through a proboscis-pore ; the parietal layer of the connective tissue

presents an alveolar structure’ (Fig. 2), while the splanchnic layer

1.) This is only case among known Enteropneusts.

2.) In this species I did not find any actual transition of fibres from the dorsal to the

ventral beyond the apex of the stomochord in accordance with the description of Glosso-

balanus hedleyi by Hill (1898, p. 335-335), and unlike that of Spengel, Delage and Hérouard,

(Spengel’s Monograph, p. 460; Delarge and Hérouard : Traité de Zoologie Conerète, p. 10.)

3.) Such a structure has not yet been described in any other species.

ON A NEW ENTEROPNEUST FROM MISAKI. 81

makes a thick envelopz over the central complex’; the anterior end of

the stomochord bends dorsally in a remarkable manner ; lateral pouches

extends themselves ventro-laterally ; pericardium is produced into two

short anterior horns?, each accompanied by the corresponding half of the

glomerulus ; Muscle fibre bundles along the inner surface of the peri-

cardial wall fall in four distinct layers*—a ventral (dorsal to the central

blood sinus), a dorsal and two lateral; nuchal skeleton has a prominent

carina, two parallel crests, and

short alary processes.

Collar : — Collar - canals

parallel to the body axis;

collar-pores open, on each side,

into the common cavity of the

most anterior four gill-pouches

(Fig. 3, cc.); dorsal septum

extends from the first root of

the collar nerve-cord up to the

posterior end of the collar

colom; perihæmal spaces

make together three dorsal

grooves—two deep lateral and

a shallow median—to clasp

the ventral surface of thecollar

nerve-cord. LE Re

Sagittal section through the common cavity of the most

Trunk 1 gill - pouches anterior four gill pouches with the collar-pore at the anterior

_ Wall of the cavity,

have no ventral cocum and c. collar-ccelom, cc. collar-canal, cgp. common gill-pouches,

cv. circular vessel, ge. gut cavity, pph. peripharyngeal space,

the most anterior four palrs spn. splanchnic nerve fiber layer, ¢. trunk cavity.

1.) The same structure occurs in Glossobalanus sarniensis and Balanoglossus australiensis

(Spengel’s Monograph, p. 101 and Hill, 1894 p. 9.)

2.) Bifurcation of the pericardium is one of the characteristics of the Glandicipitidæ,

and in the Ptychoderidæ we have only a parallel case in Glossobalanus ruficollis (Willey,

1899, p. 258.)

3.) They never make such a circular muscle layer as were suggested by Spengel

(Monograph, p. 169 and p. 511.)

82 HISATO KUWANO.

of them are confluent!, on each side, to a common gill cavity which

discharges to the exterior by a common gill-pore (Fig. 3, cgp.);

epithelium of the outer surface of the gill tongues is infolded into

the tongue cavity ; synapticula on each side of a tongue-bar is 15-17

in number; the respiratory and nutritive portions of the œsopha-

gus are nearly equal in size; gonads coextensive with the genital

pleuræ ; secondary gonads well developed ; postbranchial canal is directly

continuous with the respiratory por-

tion of the cesophagus and its ante-

rior end carries the last pairs of the

gill-slits’; the proximal origin of the

lateral septa passes abruptly over

the wall of the postbranchial canal

from the gut wall and thence, the

origin is again transferred on the

epidermis to terminate on with the

both origin and the insertion; in

the hepatic region‘, the epidermis

makes intersaccular involutions in

a remarkable manner’, ciliated

Fig. .4 grooves are paired and confined into

g. 4.

Transverse section through the anal region to the hepato-abdominal region 5 the

show the circular muscle fibres going round the gut x + A

wall. grooves appear in their typical form

emf. circular muscle fibres, ge. gut cavity, Imf.

only in the hepatic region, while

longitudinal muscle fibres, pch. py gochord.

1.) In Glossobalanus hedleyi, the most anterior two are confluent (Hill, 1898, p. 341).

2.) As far as its situation concerns, the common gill-pores correspond into the most

anterior pairs of the gill-pores in other species.

3.) Though this structure was described in some of the genera Ptychodera ESCHSCHOLTZ

and Glossobalanus SPENGEL, in the member of the genus Balanoglossus DELLE CHIAJE, it was

first recorded by Willey only in B. carnosa (Willey, 1899, p. 254.)

4.) From my own observation on the living specimens of this species, I may say the

vacuolation and bulging out of the epitherial cells lining the cavity of the hepatic saccules

and the gut cavity are not the result of the action of reagents as were supposed by certain

authors, but a normal physiological phenomenon accompanied with the secreting function

of the digestive fluid.

5.) The intersaccular involutions of the epidermis may be probably same with that of

Spengelia alba described by Willey (Willey, 1899., p. 283.)

ON A NEW ENTEROPNEUST FROM MISAKI. 83

in the abdominal region, they remain as only a thickening of the

ciliated epithelium at the bottom of the grooves finding in the dorso-

lateral angles of the gut wall; circular muscle layer makes its appearance

just external to the longitudinal muscle fibres standing on the gut wall

of the anal region! (Fig. 4, cmf.); pygochord dilates slightly along its

distal margin (Fig. 4, pch.)

Nervous system :—Splanchnic nerve fibre layer is well developed in

the buccal and pharyngeal walls; the lumen of collar nerve-cord dis-

continuous ; roots two solid and directed posterio-dorsally.

Vascular system :—at the anterior portion of the proboscis wall,

dermal capillaris makes four longitudinal stems, each lies respectively on

the sagittal median and mid-frontal planes ; ventral and dorsal medial

AUTO

SAS TA

Wd tt has W

Fig. 6.

Horizontal section passing through Transverse (slightly oblique) section through the anterior

the dorsal portion of the collar and the portion of the genital region showing the communication of the

genital pleure showing the circular ves- lateral vessel with the dorsal vessel and the capillaries of the gut

sel at the anterior portion of the trunk. wall.

c. collar-coelom, cva circular vessel, 9. dv. dorsal vessel, g. gonad, ge. gut cavity, ls. lateral septum, Jv.

gonad, £. trunk-ccelom. lateral pbe. post-brafichial canal.

1.) This muscle fibres do not make a complete ring around the gut canal but inter-

rupted by the pygochord.

2.) 'l'his arrangement of the dermal capillaries gives us a characteristic appearance in

the cross sections though the proboscis-tip of this species.

HISATO KUWANO.

glomeruli exist; just posterior to the circular vessel, another circular

vessel, thicker than the former, goes round beneath the epidermis of the

trunk making a loop in the genital pleure! (Fig. 5, cv,.); lateral vessels

stand in communication with the dorsal vessel and the capillarise of the

gut wall, directly behind the last pair of the gill-pouches.” (Fig. 6, lv.)

Characteristics by which this species is distinguished from all other

Enteropneusts are as follows :—

(L)

(2)

(3.)

(4)

(5.)

(6.)

(7.)

Definite arrangement of the longitudinal stems of the dermal

capillaries in the anterior portion of the proboscis wall.

Occurrence of an alveolar structure in the connective tissue of

the proboscis coelom.

Occurrence of another circular vessel at the anterior portion of

the trunk, immediately behind the circular vessel lying between

the collar and the trunk.

Confluence of the most anterior four pairs of the gill-pouches

into a common cavity, on each side, to discharge to the exterior

through a common gill-pore. So, the collor-pores find theml

selves at the anterior wall of the common cavities.

Communication of the lateral vessels with the dorsal vesse-

immediately behind the last gill-slits,

Occurrence of circular muscle fibres at the anal region, just

external to the longitudinal muscle fibres upon the gut wall.

Dorso-terminal position of the anus.

Finally, it may be of interest to mention that though this

species is to be ranked, from its general structure, under the

genus Balanoglossus Delle chiaje among the Ptychoderidæ, the

structure is associated, as already pointed out in foregoing foot

notes, with other structures characteristic for the members of

other groups of the Enteropneust (cf. p. 3. 5 and 6.)

1.) Occurrence of such a vessel is only case among known Enteropneusts.

2.) Such a communication does not seem to have been established in any other species.

NOTICE.

Terms of subscription—$2.00=8s=10 F= M8 per volume, postage

prepaid.

Remittances from foreign countries should be made by postal

money orders payable in Tokyo to M. NamıvE, Zoological Institute,

Science College, Imperial University, Tokyo.

All manuscripts should be sent to THE EDITOR, ANNO-

TATIONES ZOOLOGICA JAPONENSES, College of Science,

Imperial University, Tokyo.

All business communications should be sent to THE SECRETARY

OF THE TOKYO ZOOLOGICAL SOCIETY, College of Science,

Imperial University, Tokyo.

KH |

Ep |

Hint

QUE a

[Al

a bi

& AT

OCI 4

ANNOTATIONES

ZOOLOGICA JAPONENSES

Vol. IV, Part III.

PUBLISHED

The Tokyo Zoological Society.

August, 1902.

....

in the Saga

tus obtained

a and 8. Ixupa....

pecimen of Amphitre

With Plate IT.

st an

ne, e -

- Yo oe

= > / T

Notes on a S

Sea.

Notes on a Specimen of Amphitretus obtained

in the Sagami Sea.

I. Ijima and S. Ikeda.

With Plate II.

During February 1897, Kuma, the collector of the Misaki Marine

Laboratory, brought to Professor Isima a very remarkable octopod of a

beautiful and almost jelly-fish-like appearance, which was captured the

day previous near Okinosé in the Sagami Sea. The collector stated

that it had taken by mouth one of the hooks, baited with the flesh of a

Clupea, of a long-line which was lowered to a depth of about 572

meters. The specimen is now preserved in the Museum of the

Zoological Institute, Science College.

A water-color sketch was made of the animal while fresh. This is

reproduced in Plate II, not altogether satisfactorily, but to the best of

efforts, under the circumstances. We owe to the liberality of the

Science College authorities that we are enabled to give that plate in this

paper.

As the result of a careful examination of the specimen, conjointly

done by us, we have come to recognize it as a member of the rare and

interesting genus, Amphitretus HoYLE (Challenger Report, vol. XVI: p.

67; pl. IX, figs. 7—9), Perhaps it may with advantage be considered

even as identical with the single species as yet known of that genus, viz.,

A. pelagicus, which was described by HoyLE (l.c.) for the first, and to our

knowledge also the last, time from a single, much contracted specimen,

obtained off the Kermadec Islands in the South Pacific. True, there

86 I. IJIMA AND S. IKEDA.

exist certain discrepancies between HoyLE's description and the specimen

in our hand ; but these seem to us to be all such as may not improperly

be ascribed either to individual differences or to the different state of

preservation. At any rate, in the present state of our knowledge it is

difficult to discriminate distinctive points sufficiently material to regard

our specimen as specifically different from A. pelagicus.

The genus Amphitretus was made by HoyLE to constitute by itself

alone a distinct family, the Amphitretide. It possesses “the character,

unique amongst Cephalopoda, of having the mantle fused with the

siphon in the median line, so that there are two branchial openings into

the branchial cavity, on either side, whence the name.”

In view of this high systematic importance of the genus and of its

apparent rarity, we think no apology "is needed for putting on record

our observations on the specimen iu question, and that perhaps all the

less, since HoyLE's type seems to have been in a state much remote from

being natural and well preserved.

Here it may be remarked that the curious octopod with “ telescopic

eyes,” obtained by the “ Valdivia” expedition in the Agulhas-Stream

and figured, without description, by CHUN in his “ Aus den Tiefen des

Weltmeeres ” (p. 535), bears a striking resemblance to Amphitretus in

general appearance. In that figure the eye-bulbs are represented as

freely projecting while the peripheral gelatinous layer, so conspicuously

thick in our specimen while in the fresh state (see Pl. II), is scarcely

shown at all, If it be that this condition was brought about by the

shrinking of, and perhaps also by a partial damage to, the enveloping

gelatinous tissue, it would not be too far going to presume that the

“ Valdivia ” specimen is closely related to A. pelagicus, if not generically

or even specifically identical. Professor CHun’s forthcoming monograph

of the ‘ Valdivia’ Cephalopoda will likely contain matters that will

settle the question.

Description of the specimen in the fresh state.

When the animal was first brought in, it still showed some life as

NOTES ON AMPHITRETUS. 87

manifested in sluggish movements of the arms and other parts. It was

soft, transparent and, as already mentioned, almost jelly-fish-like both

in consistency and general appearance. The outline of the entire

animal was like that of an elongate bell, rounded at the blind end and

the edge of which ran out into the short free ends of the arms. The

appearance was strikingly similar to that of Alloposus mollis, as

figured by VERRILL (Bull. Mus. Comp. Zool., VIII; Trans. Conn.

Acad., V) and especially of a new species of the same genus (A. pacificus

Iy.*), known to us from the Sagami Sea.

The specimen was considerably compressed nearly but not quite

dorso-ventraliy ; the eyes were turned somewhat to the right and the

siphon to the left. This compression was however evidently due to the

fact that the animal had lain overnight and suffered to become morbid in

the position indicated in a flat vessel containing too little sea-water for

the soft body to have retained its natural, probably cylindrical form. In

this condition the total length measured 190 mm., and the breadth at the

middle, approximately 70 mm.

A thick layer of a delicate, colorless and perfectly transparent

gelatinous tissue covered up alike the body, the head and the umbrella

continuously, exactly as in Alloposus. So far as the external outline was

concerned, the layer exhibited no constriction or demarkation between

any of the three parts just mentioned. It was over 20 mm. thick at the

posterior end and 8—10 mm. at the sides ; but it must be borne in mind

* Alloposus pacificus Is. N.sp. ‘l'wo specimens have hitherto been obtained in the

Sagami Sea and are preserved in the Sci. Coll. Museum. One is small, measuring only

about 54 mm. in total length; the other is about 116 mm. long. Both without a

hectocotylized arm. They closely resemble A. mollis VERRILL, except in the fact

that the suckers are arranged in a single row for the greater part of the arm-length,

being biserially arranged only in the free tip. The body is unusually compressed

antero-posteriorly, its length (as measured from the posterior end to the mantle-edge)

being about + of the total length. There are two buttons at the siphon base, fitting into

grooves on the inner side of the mantle. It is apparently either this arrangement or the

depressor infundibuli that was called by VERRILL “the lateral longitudinal commissures,”’

while his “ median-ventral commissure ” seems to be simply a part of the ventro-median

septum, which, in this genus, is brought so far forwards that its anterior edge is partially

visible from the outside in the middle of the mantle-opening.—Isıma.

88 I. IJIMA AND S. IKEDA.

that the above-mentioned compression, to which the animal was

subjected, must have unnaturally augmented the thickness of the layer

in the marginal region at the expense of that both above and below.

The same layer, as already indicated, extended itself over the arms and

the webs in such a way that the external surface of the umbrella pre-

sented no elevations corresponding to the course of the arms below. At

about the middle of the umbrella the Jayer was nearly 5 mm. thick over

the arms, while the web measured as much as 12 mm. in thickness. It

somewhat decreased in thickness towards the umbrella edge, which,

nevertheless, presented a swollen appearance. On side-views of the arms

it could plainly be made out that the inner surface of these had the same

gelatinous layer as that on the outer and that the suckers lay buried

in it up to their external opening. The external surface of the layer,

and therefore that of the animal, was quite smooth and slippery.

Numerous small chromatophores of madder-brown to yellowish

colors besprinkled the surface of the otherwise perfectly colorless and

clear gelatinous envelope. In greater abundance aud larger sizes were

they found also in the deepest stratum of the layer, closely over the

muscular tunic of the mantle, along the surface of the brachial muscles

and on the eye-bulbs, According to the collector’s statement, the

animal, when in full vigor of life, should have displayed a vivid and

beautiful change of color, now “intensely red” and then “nearly

colorless.”

Through the gelatinous envelope were indistinctly discernible the

deeper outlines of the body, the head and the siphon. 'The body and the

head, in this sense, together formed a roundish or ovoid mass, nearly

transparent in the posterior portion but more opaque towards the head

and the siphon, Between the head and the crown of arms there was a

distinct neck-like constriction, which was not in the least shared by the

outer gelatinous envelope. The arm-bases were rather opaque ;

however, there could be seen through them the faintly colored buccal

mass, marking the fundus of the umbrella. For the rest the arms were

sufficiently transparent as to show beautifully their ganglionic chain.

NOTES ON AMPHITRETUS. 89

The ganglia corresponded in position with the suckers and, like these,

became smaller and more closely approximated together towards the

distal end of the arms.

The moderately large eyes are very conspicuous on account of its

brilliant coloring. The bulbous inner end had a shining dark purplish

or bluish ground-color, dotted with madder-brown chromatophores. The

conical outer end was marked with rings of reddish, purplish and of a

light color. The eye-bulbs projected out from the inner opaque head-

mass, but were in reality not freely outstanding. Around them existed

the peripheral gelatinous layer in nearly undiminished thickness as

elsewhere in the neighborhood, and they reached the external surface only

with the pupillar end.

The comparatively large siphon, as it lay obliquely anteriorly

directed, was similarly covered up by the peripheral gelatinous layer, so

that it nowhere projected outwards on the real external surface, except

perhaps at the edge of its external opening. But this opening, as also

those of the mantle-cavity, remained unascertained in the fresh state

of the specimen. The way both the eyes and the siphon lie buried in

the gelatinous tissue may be said as exactly comparable with the con-

dition obtaining in Alloposus.

Of the contents of the mantle-sac there were some that could be

oriented, though more or less indistinctly, from the outside. An ill

defined patch of deep purplish or bluish tint behind each eye probably

had some relation to the liver. A median brownish tract, between the

patches just mentioned indicated the stomach. On either side of the

body the gills showed themselves as oblong whitish masses. Two pale

yellowish spots behind and inside the gills were in all probability refer-

able to the branchial hearts.

So far, the appearance of the specimen in the fresh state. Our

further study was done after the preservation.

Description of the specimen in the preserved state.

For the preservation Flemming’s fluid and alcohol were employed.

90 I. IJIMA AND S. IKEDA.

Woodeut 1.

Ventral view of the specimen after preservation. Natural size. Siphon and edge of

the muscular layer of mantle, indistinetly indicated. Of the mantle apertures, only the

right-side one is visitor.

NOTES ON AMPHITRETUS. 91

Contrary to the desired result, the reagents have turned the specimen

into a shabby object, incomparable to its former beauty. The change

consisted in the loss of transparency, in the browning effect of the fixing

fluid and in the contraction of parts in general, but especially of the

peripheral gelatinous layer.

This layer is now reduced almost to the condition of an ordinary

soft skin. It has shrunk so as to present a wrinkled surface and is no

longer of such great moment in influeneing the general shape of the

animal as it was before. ‘The specimen may now be said to bear a

certain resemblance to the figure of Amphüitretus pelagicus given by

HoyLE in the Challenger Report.

The head and body together take up nearly one-third of the entire

lengtb. They are unnaturally flattened owing to the compression before

referred to ; otherwise they would have presented a nearly hemispherical-

shape, much like the Challenger specimen.

The following are measurements taken some time after the pre-

servation :—

Wotan... 0027... L48 mm.

Breadth of body . ese... AMM 45,

End of body to Janes margin ee MED >

» » 99 » the middle point sen Bi eyes.. 45 ,,

e: Ly Mouth en... 56 *.,

a Biphon'end A. FO yy

sì» » » the umbrella edge between the

ventralmost arm-pair ... ... 83 „

» » >» > the umbrella edge between the dor-

Balmost affine ee. 115 „

Peneretoramniet un. MOM n> a... ...86-96 ,,

Greatest thickness of arm . a Packen. Ars

Diameter of largest sucker à the Hier ond PTS DE SIRO

It may be observed that our specimen is more than thrice as large

as that described by Horvitz. If we are right in referring ours to the

same species as his, the latter was probably a young specimen.

The most characteristic point of the family, viz., the division of the

mantle-opening into two separate, right and left apertures, can now be

92 I. IJIMA AND S. IKEDA.

satisfactorily made out. Already in the fresh state we noticed on either

side of the siphon-base and laterally from the eyes two small spaces

surrounded by somewhat crowded chromatophores, being themselves

nearly free of these (see Pl. II). The spaces proved to indicate the

openings of the mantle-cavity. The openings may be described as

transverse, somewhat gaping, simple slits of about 10 mm. length. On

the left side of the body, the opening commences at about 7 mm.

distance laterally from the eye of the same side (see Woodcut 3, 1.m.o. ;

p. 97). The right opening is found on the other side of the flattened body,

evidently as the result of an assymetrical compression to which the

specimen had been subjected (see Woodcuts 1 and 2, r.m.0.) We should

say that in the natural cylindrical state of the body the openings occupy

x position midway between th: eye and the siphon-base on either side.

They are both freely open. We mention this, because HoyLE found

in his specimen one of the two openings closed, which closure be how-

ever regarded as abnormal.

The mantle-opening leads posteriorly into a compressed passage,

the outer wall of which is given by the mantle but the inner, by a

posteriorly directed, valvular fold, the lateral continuation of the

siphon-base (Woodcut 2, l.i.f.). This fold is anteriorly continuous with

the outer surface of the umbrella and has its thin free edge about 8 mm.

inside of the mantle-opening.

Between the ventral ends of the mantle-openings there is an inter-

space of over 30 mm. breadth, covered over by a sheet of the now

skin-like but originally gelatinous tissue, continuously connecting the

surfaces of the mantle, the siphon and the umbrella. Through that skin

is seen the mantle-edge or, more strictly speaking, the edge of the

muscular tunic of the mantle. The edge-line extends uninterruptedly

from one mantle-opening to another, and all along that line the muscular

mantle-edge seems to overlap, and is adherent to, the muscular layer

of the ventral wall of the siphon at base. It is apparent that, as was

pointed out by HoyLE, the originally single mantle-opening had under-

gone closure in the middle, leaving only its lateral ends open. If we

NOTES ON AMPHITRETUS. 93

suppose that in Tremoctopus or in Alloposus a fusion took place at the

buttons as well as medianly from these between the mantle-edge and the

ventral funnel-wall, much the same condition as that seen in Am-

phitretus will be the result.

The siphon in the preserved specimen is a flatly compressed tube,

measuring about 28 mm. in length. It is 22 mm. broad at base and

narrows anteriorly to a breadth of 9 mm. It lies flat on the umbrella,

is covered over by the common skin and extends forwards for about

one-third of the distance between the mantle-edge and arm-tips. The

lateral edges are vaguely visible through the skin. The external

opening is transversely slit-like, about 7 mm. long and bordered by

somewhat prominent lip-like edges. The flat shape of the siphon may

in a measure be due to the artificial compression. The head is indis-

tinguishable from the body, being indicated only by the eyes. It is

anyway small, that is to say, if we leave the outer gelatinous envelope

out of consideration ; otherwise it may be called large, being as broad as

the body. Just the same condition obtains in Alloposus pacificus.

So far as could be ascertained by feeling from outside, we are

inclined to assume the absence of a cephalic cartilage or of any other

cartilage in the entire body. We have also failed to discover aquiferous

pores.

The eyes are nearly pear-shaped with spheroidal base and rounded-

conical outer end. They are about 13 mm. long. ‘The bulbous base,

about 10 mm. in diameter, is in its outer half studded with chromato-

phores on a dark back-ground; the inner half is of a lighter color. The

outer rounded-conical or nearly hemispherical portion of the eye-bulb,

about 6 mm. in diameter at base and not more than 34 mm. in height,

is encircled basally by a light-colored zone and in the pupilar end by

another zone of a dark brownish color. The latter incloses a whitish

looking pupil of 11-2 mm. diameter. The two eyes are in contact with

each other in their broadest portion. Their axes somewhat diverge

outwardly and are at the outer end 12 mm. apart from each other.

Between the eyes and over their contact point, there are seen two rather

94 I. IJIMA AND S. IKEDA.

conspicuous muscular threads, running from the nuchal band to the

bases of the first arm-pair.

The jaws were not examined by us by cutting them out of the

buccal mass.

In the umbrella the arms are now traceable, on both its internal

and external surfaces, as ridge-like prominences owing to the shrinking

up of the enveloping gelatinous tissue. By the same cause the interpodal

edges of the webs have become much thinner and more deeply indented

than was the case in the fresh state. The indentation is deepest between

the ventral (4th) arm-pair, extending fully balf-way up the length of

these arms in a moderately outspread state of the web concerned. In

the dorsal median web the deepest point of the indentation lies about

one-third the arm-length away from the arm-tip. As regards the

remaining paired webs, there are indications of their having unequally

contracted on the two sides of the umbrella. We think we may say

that they are all equal to the dorsal median web as well as to one

another, both in extent and in the manner the ends of the free edge of

each insert themselves to the enclosing arms. It is difficult to deter-

mine for what fraction of their length the arms are free at the distal

end; we may however approximately estimate it at 4-4, without going

much wide of the mark. Through the skin and especially when seen on

the inner surface, the webs show numerous, thin, simple or branched

muscular bundles, which run between the oppositely standing sides of

every two adjoining arms (Woodcut 1).

The arms have unequally contracted in length (86—96 mm.). It

may however be safely said that they are all subequal or nearly equal.

They are about twice as long as the head and body taken together.

Being somewhat laterally compressed, the thickest part may measure up

to 9 mm. across in the lateral view and 6—8 mm. when seen from either

outside or inside. At base they are somewhat narrower, and in the

distal one-third of their length they again show a gradual tapering

towards the slender, pointed tip. The cirri are not present. Nor is

NOTES ON AMPHITRETUS. 95

there any sign of hectocotylization : the specimen proved to bea female

on dissection.

The suckers number in all 32—35 to each arm. For a short stretch

—say, of about 14 mm.—at the distal end of arms, they exhibit a

tendency to an alternate biserial arrangement. In the remaining, by far

the greater portion of the arms, they—fourteen or fifteen in number—

are arranged in a single row.

The first sucker, counting from the base of arms, is about 14 mm.

in diameter and is separated from the next following by an interspace

of not over 4 mm. In the middle one-third of the arm-length—which

space includes about five suckers, say the 7th—13th—the suckers attain

largest size. Here they are set 6—8 mm. apart from one another ina

series.

In the distal one-third of the arm they begin to become gradually

smaller and more closely approximated together, until near the arm-tip

they are half a millimeter or less in size and are placed in contact with

one another.

Though more or less cylindrical in shape, the suckers are broader

at base than at the truncated outer end and usually show a slight

constriction near the latter. The height of the largest sucker, leaving

the gelatinous envelope totally out of consideration, measures up to

4 mm. ; diameter near the base, 3 mm.; that at the outer end 2} mm:

The margin is thick and smooth ; the cavity is deep and pore-like.

Some points of internal organization.

In order to obtain an insight into the internal structure, incisions

and removal of parts have been made, so far of course as could be done

without destroying the rare specimen.

Our attention was first directed to the histological character of the

peripheral gelatinous layer. Studying this by means of both sections

and teased preparations, we have found, in the first place, the external

surface delimited by a fine, wrinkled, structureless membrane, on which

96 I. IJIMA AND S. IKEDA.

we have failed to observe epidermal cells. We can not but think that

the epidermis had been lost before the specimen was put into the fixing

reagent. In the second place, we have determined that it is by a

modification of the dermal connective-tissue that the gelatinous tissue,

almost similar in its character to that of a Medusa, is formed. We find

it traversed in all directions by innumerable, very fine, elastic fibrils,

amongst which are sprinkled stroma cells in large numbers. The

chromatophores seem to have their seat chiefly in the superficial portion

of the layer and also in its deepest part in direct contact with the

muscular tunic soon to be mentioned. The same distribution of

chromatophores is found in Alloposus pacificus. Here and there in the

layer are seen blood-vessels containing blood-corpuscles, fine nerves

sending branches to chromatophores and a few branching bundles of

fibers which we take for muscular. On the whole it seems the tissue in

question closely agrees both in structural and genetical respects with the

corresponding tissue of Alloposus mollis as described by JouBIN *; only,

in Amphitretus the muscles in the gelatinous layer does not show a

regular arrangement into definite layers as they seem to do in Alloposus.

Beneath the thick gelatinous layer and constituting the deeply

situated tunic proper of the mantle, there are two thin (about .15 mm.

thick) compact-looking sheets of tissues, which are well separated from

each other by a layer of a soft and loose consistence. All these can be

distinguished by the naked eye on the cut edge of the mantle.

The outer sheet exhibits on its external surface a number of large,

flatly apposed chromatophores. For the rest it is composed of densely

arranged muscle-fibers, of which we again distinguish an outer and an

inner stratum, composed respectively of longitudinal and transverse

fibers.

The inner sheet is somewhat firmer and again consists of two dense

layers of muscular fibers, the outer transverse and the inner longitudinal.

* Bull. Soc. Zool. de France. T. XX, p. 95.—Unfortunately, JouBIn°s “Contrib. a

l’etude des Céphalopodes de l'Atlantique Nord” in Rés. Sc. Camp. Prince Monaco, Fase. 9,

does not stand at our disposal for reference.

NOTES ON AMPHITRETUS, 97

2. Funnel and mantle-sac

cut open on the ventral side.

3. Mantle-sac cut open on

the dorsal side.

Both figures partially con-

structed.

an., anus.

ao., aorta cephalica.

a. v., abdominal vein.

b. a., branchial artery.

b. h., branchial heart with

small pericardial gland.

b. v., branchial vein.

d. i., depressor infundibuli.

f.o., funnel organ (Trichter-

organ).

ge gill.

9. s-, gill suspensorium.

i. b., ink-bag.

l. b., lateral band with gang-

lion stellatum.

l. i. f., lateral infundibular

fold.

l. m. o., left mantle opening-

l. v., lateral vein.

ov., ovary.

ovd., oviduct.

r., renal organ.

r. m. 0., right mantle open

ing.

r. 0., renal opening.

s. h., systemic heart.

st., stomach.

v.m. s., ventro-median sep-

tum.

Woodeut 2.

4 el Av.

PIA

Woodcut 3.

98 I. IJIMA AND S. IKEDA.

No chromatophores are found in connection with this sheet. The

internal surface is probably in almost direct contact with the epidermis

of the mantle-cavity.

The intervening space, between the two muscular sheets above

mentioned, may be one millimeter or more in width and is taken up by

an elastic connective-tissue, which is traversed crosswise by numerous,

fine muscular bundles.

The wall of the siphon is essentially of the same structure as the

mantle. In both sheets of its muscular tunic are found obliquely

running fibers in addition to the transverse and the longitudinal,

Exceedingly remarkable seemed to us the structure of the arms. It

may roughly be described as a thin-walled tube. The wall, not more

than 0.15 mm. in thickness, is a firm muscular membrane in which the

compactly arranged muscle-fibers form an outer circular and an inner

longitudinal system or stratum, exactly comparable to those of the inner

muscular sheet of the mantle. Apposed to the outer surface of the

membrane are a number of chromatophores, The wide internal cavity

of the arms is crossed diametrically by a large number of fine, white and

glistening muscular bundles, the interspaces between which are ap-

parently filled up by a gelatinously metamorphosed connective-tissue,

quite like that of the integument except that it contains no chromato-

phores. By making a longitudinal incision, the muscular membranous

wall of the arm may be laid aside and the internal ganglionic chain

exposed, as if we were dissecting an earthworm or a Nematode.

The muscular wall of the arms is of course continuous with that of

the suckers, The latter contain, inside the piston, a small space which

is shut off from the internal arm-space by a thin membranous partition

and which is traversed by numerous muscular bundles extending

between the piston and the partition just mentioned.

Cutting open the siphon and the mantle so as to expose the visceral

sac, we have ascertained the following points :

The ventral wall of the siphon and the ventral mantle, both as seen

on the inuer surface, are continuous but still show a boundary-line

NOTES ON AMPHITRETUS. 99

between them, in that there exist a certain difference in the appearance

of their innermost muscular layer. It can be made out on the cut edge

that the original edge of the former adheres to the inside of that of the

latter.

A short distance from the external siphon-opening is found the

W-shaped funnel-organ (“ Trichterorgan”) with wrinkled surface.

(See f. 0., Woodeut 2). It is of much the same shape as we know that

organ in Alloposus pacificus and in certain species of Octopus. The two

lateral ends of the organ are separated from each other by a space of 5

mm. on the ventral siphon wall. A valve is not present in the siphon ;

nor is there a trace of connective cartilages at the base. The depressores

infundibuli (d. %.), disposed in the usual way, is thin and membranous.

The visceral sac is enveloped in a thin, transparent, visceral mem-

brane, which is in many parts supplied with chromatophores. The

same holds good also for Alloposus pacificus. The development of

chromatophores in these parts evidently stands in relation with the

transparency of the animal.

The visceral sac stands in connection with the ventral mantle by a

ventro-median septum (v. m.s.). This is sagittally quite short and

antero-posteriorly about 10 mm. long. Its anterior edge lies only about

4 mm. behind the erstwhile free edge of the mantle. The lateral bands

(2. b.), on either side of the visceral sac, are thin and narrow. Near

their outer end is distinctly visible the ganglion stellatum.

Shortty in front of the visceral insertion of the ventro-median

septum and slightly to the left of the median line is situated the anus

on the end of a prominence and provided with a pair of small papilla-like

processes (an.). The rectum can be traced for a considerable distance

along the ventro-median septum on the left side.

On the right of the septum is seen a blackish patch, which is causéd

by the presence of the ink-bag (i.0.).

Behind this region and symmetrically situated on both sides are a

pair of small papillar prominences, on which the oviduct (ovd.) opens

externally. (In Woodcut 2, the two openings of the oviducts have been

100 I, IJIMA AND S. IKEDA.

placed too close together). Through the visceral membrane is visible

the terminal thickened portion of the oviduct. This runs from the

external opening laterally and nearly horizontally for about 5 mm., then

to bend backwards and mediad but soon to make a globular swelling,

whence posteriorly the oviduct can be seen as a fine thread for a short

distance before it becomes lost to the sight. On the dorsal aspect of the

visceral sac (Woodcut 3), both oviducts are seen to emerge from the

deep parts and to unite, right over the systemic heart (s. h.), into a very

short unpaired duct which soon joins the oval-shaped ovary (ov.),

situated just behind the stomach (st.).

For all the other visceral organs shown in Woodcuts 1 and 2, we

consider no special remarks are needed.

The gills, 18 mm. long, have alternately arranged, convoluted

lamelle, about 10 in number on each of their sides. The branchial

artery (0. a.) is thick and fleshy ; its surface, like that of the branchial

hearts, is sprinkled with chromatophores. It is suspended from the

mantle by a thin, mesentery-like band (g. s.), which is traversed across

by a number of fine muscular bundles.

It will be seen from the above cursory account, as well as from the

woodcuts given, that the arrangement of the organs in the mantle sac is

essentially as in Octopus and related forms. In no less degree does it

show resemblance to that of Alloposus pacificus.

Affinities of Amphitretus.

As regards the affinities of the genus, HoyLE assumed it to be most

nearly allied to Cirroteuthis, as agreeing with each other in having the

suckers arranged in a single row and the arms united by a broad web.

We greatly doubt if these points of agreement taken alone can be of

much weight in determining the affinity, especially in the presence of

such important differences as were pointed out by HoyLE himself, as, for

instance, the absence of cartilage, of fins and of cirri in Amphitretus.

NOTES ON AMPHITRETUS. 101

On the other hand, it seems to us that a very much nearer ally is

given in Alloposus, especially in A. pacificus. This species has in

common with Amphitretus not only the similarly arranged suckers (see

foot-note on p. 87) and the very extensive webs, but also many other

points of fundamental importance as regards the histology and anatomy,

to which attention has already been called. It may almost be said that,

so far as is known, the only and the most essential distinctive character

between the two genera, Amphitretus and Alloposus, consists in the

presence of the ventro-median connection of the mantle with the siphon

in the forıner.

VERRILL (/. c.), in diagnosing the family Alloposidæ and the genus

Alloposus, has given that the mantle-edge is directly united to the head

by “a median-ventral and two lateral longitudinal commissures,” in

addition to a large dorsal nuchal band. From this one might be led to

think that the “ commissures,” here spoken of, may be something

morphologically related to the ventro-median connection seen in

Amphitretus. ‘They may perhaps be considered as a forerunner of that

connection in so far only as they serve to bring close together the free

LE]

mantle-edge and the siphon-base ; but the ‘ comraissures ” themselves

are, in our opinion, nothing peculiar to Alloposus (see foot-note on p. 87 )

In fact, the ventro-median connection of Amphitretus is totally un

represented in Alloposus as in any other Cephalopod genera. Since now

the “ commissures ” referred to forms the chief point in the characters

of the family Alloposidæ of VERRILL, it may be questioned if that family

can be held up as distinct from either the Philonexidæ or the Octopodidæ,

to both of which it seems to show affinities in certain respects.

Be that as it may, we agree with HoyLE in regarding Amphitretus

as a representative of a distinct family, the Amphitretide. As already

indicated, this family probably has its nearest ally in the genus Alloposus.

In the next place, it seems to us the family is more nearly related to

the Octopodidæ than to any other family of the eight arıned Dibranchiata,

as shown in its organization.

n fe x

tres

‘

1 uit }

j x

pi Sa Mr

| MN,

rat ”

ba]

140

ash

e ah

EEE UT"

à Re re =

HUE dll "ART ee

NOTICE.

Terms of subscription—$2.00=8s= 10 F=M8 per volume, postage

prepaid.

Remittances from foreign countries should be made by postal

money orders payable in Tokyo to M. NAM1YE, Zoological Institute,

Science College, Imperial University, Tokyo.

All manuscripts should be sent to THE EDITOR, ANNO-

TATIONES ZOOLOGICA JAPONENSES, College of Science

Imperial University, Tokyo.

All business communications should be sent t THE SECRETARY

OF THE TOKYO ZOOLOGICAL SOCIETY, College of Science,

Imperial University, Tokyo.

Bet | RENE

ace

Lo HR = = AA

= x Bam

say

ANNOTATIONES

‘ZOOLOGICA, JAPONENSES.

Vol. IV, Part IV.

PUBLISHED

BY |

| The Tokyo Zoological Society,

TOKYO.

December, 1902.

po rt CORE a

II. On the Occurrence of u

aa x

Preliminary Notes on Coeloplana.

James Francis Abbott, A.B.

In the summer of 1901, while enjoying the hospitality of the

Misaki Marine Biological Station of the Tokyo Imperial University, I

was fortunate enough to find a number of specimens of a planarian-like

form very closely resembling KowALEvsKyY's Coeloplana. The next sum-

mer my renewed search was rewarded by the discovery of several more,

so that altogether some thirty or more specimens have been obtained.

As unavoidable circumstances will prevent me from publishing extended

results immediately, it has seemed best to briefly announce the discovery

and describe some points of superficial observation, reserving the details

of structure for a later paper. While doing so I must take the oppor-

tunity to express my gratitude to Dr. K. Mrrsuxurt, Dean of the Science

College of the University, and Direetor of the Laboratory, for the

unfailing kindness and courtesy with which he has aided me through

out my visits to Misaki.

The animal, which—tentatively at least—will be considered a Coelo-

plana, occurs as two clearly distinct species. It is littoral in its habits

and apparently wholly unfitted for life on the open sea. It possesses no

vibratile plates or swimming cilia and cannot swim, tho in captivity it

floats frequently on the surface of the water.

It is found principally on encrusting algae—Zostera, Melobesia, de.,—

which it resembles so closely as to make detection extremely difficult

When found on the rocks it is frequently stretched out to a thin mass

fo slime having little resemblance to anything animal. At such times

104 J. F. ABBOTT.

Explanation of Woodcuts. 3

Fig. 1. C. mitsukurti, in floating position.

‘l'he branches of the tentacle are re-

presented disproportionately thick,

for clearness’ sake.

Fig. 2. Diagram of gross anatomy of C-

mitsukurii. S.M., mouth of tentacle

sheath. D.T. dorsal tentacles. 7.,

tentacle within sheath. O., sense

organ. D.A., branches of digestive

canals.

Fig. 8. C. willeyi. Mouth of tentacle sheath,

showing the manner of partially

extruding the tentacle when the

animal is irritated.

PRELIMINARY NOTES ON COELOPLANA. 105

the larger species will frequently measure 5 or 6 cm. in longest diameter;

the same animal contracted, shrinking to one centimeter across. It

seems to adapt itself to its surroundings in color, and heavily pigmented

individuals as well as nearly transparent ones will be found in appro-

priate surroundings.

It is very sluggish and apparently inhabits a very limited field.

The ventral surface is heavily ciliated, the dorsal not at all. The ventral

surface is flat, and it never doubles upon itself as described for Clenoplana.

When dropped into the water the edges curl in, all around, and it falls

through the water in a shapeless lump.

It floats on the surface of the water with ease, and at such times

moves along the surface, suail fashion, on a film of slime. This frequent-

ly extends beyond the periphery so that it is possible to push the animal

all about the jar with a needle, without coming within a centimeter of

touching the edge of the body. I have never seen it floating except in

captivity and it tenaciously resists being taken from the rock on

which it is found.

Where one is found, others are quite sure to be, and the situations

in which they are found are sometimes strongly suggestive of

wultiplication by division, tho no evidence has been obtained yet as to

that point.

It would be as misleading as it would be useless to speak of front or

back or sides in connection with this animal, at least as regards locomo-

tion. It not only crawls in any direction whatever but it frequently

goes in more than one direction at once and the two halves, starting off

for opposite sides of the dish, often stretch the middle part to the break

ing point. At other times, it turns and twists upon itself until, tho

the outline may be fairly regular, the turns and coils of the internal

organs show that the interior is much confused.

When floating C. mitsukurii frequently drops down its long

tentacles which wave about ın the water in very graceful fashion (fig. 1).

The dorsal surface is ordinarily heavily pigmented, the pigment

being contained in large cells. Where the animal contracts, these cells

106 J. F. ABBOTT.

are approximated but when extended they are separated by wide

interstices so that the general color of the animal becomes much paler,

approaching transparent. About the periphery are thickly scattered

dull yellow or wbitish cells, sometimes forming a rather prominent

border.

The most prominent feature of the animal is the two long chalky

white tentacles which lie in great writhing snake-like masses on

either side of the sense organ, usually visible from the outside. These

are very long and very contractile, quite similar to those described for

Ctenoplana. The secondary branches are covered with batteries of

nettle cells. Except when floating as described above, they are usually

retracted in the sheaths, occasionally with the tip projecting like a

tongue. At other times, however when the creature is more animated,

it throws out the whole tentacle in a cloud of white filaments, while

crawling. And to watch it at such a time, shooting out and retracting

the tentacles, moving along the side of the aquarium like a battle-ship in

action, is truly a remarkable spectacle.

Touching with a needle will also frequently cause it to throw out

the tentacles. (fig. 3).

The sheath is raised in a log-like ridge along the dorsal surface,

ending freely a trifle back of a notch in the periphery. The mouth of

the sheath is separated from the body to forma short retractable tube.

Next to the tentacles the most prominent thing to be observed is

the network of anastomosing digestive canals, which ramify throughout

the body (fig. 2). The creature’s habit of constantly twisting itself out of

shape makes it extremely difficult to plot with any certainty the course of

the canals. There are three or four long channels that run more or less

paralle! to the axis of the tentacles and send out branches peripherally.

At a point about half-way to the edge these branch much more profusely.

The inner portion thus appears paler and clearer when viewed from

above. There is no peripheral cana]. The canals end blindly in finger-

like processes. I have not been able to make out with certainty any

PRELIMINARY NOTES ON COELOPLANA. 107

vertical canal running up to embrace the sensory organ as described by

KOROTNEF but what observations have been possible seem to point to

its existence in this form also.

The mouth is large, quadrangular, with colorless lips, lying

directly below the sense organ. It opens into a rather thin-walled

pharynx, which apparently is roughly four lobed. The canals branch out

from above it.

The circulation in the canals is very evident and may be observed

readily with a low power. The colorless fluid carries a quantity of

irregularly shaped bodies which frequently flow both ways in the same

channel. The circulation is apparently controlled solely by the motious

of the animal.

On the dorsal surface are a series of club shaped processes which

may be called dorsal tentacles. They are either entire or digitate and

fringed ; hollow and thin walled, and communicate with the digestive

canals. The circulating fluid enters into them and the particles carried

by it whirl about in eddies within the tentacle. The tentacle itself is

contractile and may be withdrawn. When the animal is extended they

usually kecome obliterated. They do not seem to be particularly

sensitive, however, and may be moved about with a needle without

being withdrawn. At the base there is usually an aggregation of

pigment cells, tho the tentacle itself is colorless. As a result of this the

position of the tentacles may be noted on an extended specimen by the

color, when there is no sign of the tentacle itself. The arrangement

is hard to make out, owing to the amoeboid habits of the animal

mentioned above, but in the smaller brown form, they follow

roughly the outlines of a figure 8, covering the dorsal surface,

with the intersecting lines crossing at the sense organ. When floating

on the surface the tentacles are nsually pendant and swinging in the

water.

The otolith is very small,—alniost insignificant in the larger form.

It lies in a depression closed by fleshy lips, but there are no accessory

sensory tentacles as described for Ütenoplana, and no external guide

108 J. F. ABBOTT,

to axial orientation, Two semicircular bands of yellow cells, probably

glandular, surround the otolith at the base,

Coeloplana willeyi. .«.

Average size :—one to two centimeters across. Very contiactile

and extensible ; amoeboid in movement. When killed, roughly circular

in shape. In life, no consistent body-shape is maintained. Dorsal

tentacles club shaped or cylindrical,—not branched or fringed. Color

scarlet or carmine red, fading towards the edges to a yellowish pink.

Chalky white spots about margins. Color deepest along the raised axis

of the tentacle sheaths. Dorsal tentacles frequently with yellowish

blotches at the bases.

Coeloplana milsukurii. -\. sp.

firmer, not amoeboid, consistently oval or almost circular. Prominent

notch opposite tentacles. Dorsal tentacles with two to five digitate

processes. Not nearly so contractile as C. willeyi. Arrangement of

dorsal tentacles approximately in a figure 8, with the extremities of the

figure interrupted, the intersecting lines at the sense organ. Color ;—

dirty brown or brownish yellow. Less transparent than C. welleyi,

frequently entirely opaque.

The two species are found together.

Oct. ist, 1902.

On Two New Species of the Family Maldanidæ

from the Sagami Bay.

Akira Izuka,

Science College, Imperial University, Tokyo.

“With Piane

Three species, representing as many genera, of the polychaetous

family Maldanide have been recorded from the Japanese waters by

M’IntosH in the Challenger Report. They are as follows :

Maldane sarsi MALMGREN.

Nicomache japonica M’INTOSH.

Praæilla (Clymene) lankesteri M'INTOSA.

The list may be augmented by two more forms, both apparently

yet undescribed. I propose to call these Maldane gotoi and Clymene

harai, in dedication respectively to Professors S. Goro and J. Hara, by

whom the type specimens were collected and kindly given to me for

study.

Both the new forms are from Yodomi in the Sagami Bay and a

depth of 80 fathoms. The descriptions follow.

Maldane gotoi. N. sp.

PI. III, figs. 1—8.

This is founded on two specimens. The larger one measures 42 m. m

in length and 1.8 m. m. in breadth. The other is 34 m. m. long and 1.5 m. m.

broad.

The worm, as preserved in spirit, is of an yellow ochre color with

light bluish iridescence, except in a few anteriormost segments which

present a brownish yellow hue.

In both specimens the cylindrical body is composed of 21 segments,

110 A. IZUKA.

The first or the cephalo-buccal segment is obliquely truncated in

front at an angle of about 60° to the body axis (figs. 1 and 2). The

truncated end, which faces dorsally and forms the cephalic plate, exhibits

a median elevation laterally bounded by two slightly depressed furrows.

This elevation is anteriorly continuous with a small protuberance at the

ventro-median point of the cephalic plate edge. The dorsal and the

lateral edges of the cephalic plate form a thin rim. This may be said as

being serrated in the dorsal part, the serration becoming more pronounced

as we follow it laterally on either side of the dorso-median line. The

lateral portions of the rim had better b> said as being furnished with

fringing cirri of unequal lengths, rather than to describe them as very

deeply indented (fig. 2). The mouth opens on the ventral side of the

head end. In the larger specimen the proboscis is seen slightly protrad-

ed (fig. 3).

The second to fourth segments are of about the same length. The

fifth and all the succeeding segments are slightly longer, though nearly

equally long amongst themselves.

The last or the anal segment (fig. 4) has the ventro-posterior margin

thinned out into a rim which shows four angular longitudinal edges.

The dorso-posterior margin is expanded into a petaloid plume having six

corners, each of which is prolonged into a slender cirrus. The anus lies

dorsal to the petaloid plume, in the dorso-median line of the segment in

question.

The cephalo-buccal and the anal segment are entirely destitute of

bristles. The remaining nineteen segments are all setigerous.

The second segment possesses on either side a flattened group of

short acicular sete, of which two kinds, the slender and the stont (fig. 8,

a and 5) are distinguishable. Neither uncini nor capillary sete occur

in this segment.

The third to twentieth segments have each, on either side, two

flattened groups of bristles representing the dorsal and the ventral ramus

of the totally or nearly totally suppressed parapodiam. The dorsal group

consists of capillary sete. These are present in two forms. Both are

TWO NEW SPECIES OF THE MALDANIDZE. 111

long and slender, aftenuating very gradually and uniformly towards

the finely pointed end; but the one is excessively fine and smooth

throughout (fig. 6, a), while the other, besides being much longer and

stouter at base, is provided with bipinnately arranged spins-like hairs in

the slender terminal portion, which on that account presents a plume-

like appearance when seen under a high power (fig. 6, b, and fig. 7).—

The ventral group consists of uncini arranged ina row. The uncinus

terminates sometimes in three, and sometimes in four, hooks.

The tube of the worm was found in a greatly damaged condition,

just enongh to show that it is thin and membranous. Fine particles

of bluish mud covered the surface.

The above described worm clearly belongs to the family Maldanidæ

as established by Saviany (System des Annelides, 1820). Further, it is

referable to the genus Maldane GRUBE, notwithstanding that it does not

quite agree with the definition of that genus as given by DE SAINT-

JOSEPH (Lies Annelides polychaetes des cortes de Dinard, 1894). The

descrepancies consist in the presence of acicular sete in the second

segment, in which respect it seems to approach Clymene, and in the

fact that the corners of the petaloid plume are elongated into cirri.

Clymene harai. N. sp.

Pl. III, figs. 9-12.

A single specimen lies before me. It is 92 m. m. long and 3.5 ın. m:

broad in the broadest region of the body.

The color, as seen on the alcoholie specimen, is a light yellowish

ochre with bluish iridescence all over the body.

The cylindrical body (fig. 9) consists of 23 segments, of which 19

are setigerous. It may b2 distinguished into the four following regions :

the fused cephalo-buccal or the first segment, the thorax, the abdomen

and the tail.

The cephalo-buccal segment is slightly obliquely truncated. The

truncated end or the cephalic plate shows a median fold which somewhat

broadens antero-ventrally and which is bounded on either side by a

1) RON A. IZUKA.

groove (fig. 10). The lateral and the dorsal margin form a thin and

narrow rim, which is smooth edged, exhibiting neither serration nor

cirri. The mouth opens on a gentle elevation on the ventral side near

the anterior end of the segment. No bristles are present.

The thoracic segments, 7 in number, are about 3 m. m. broad. The

anterior end of each surrounds the posterior end of the next preceding

segment somewhat in the manner of a collar. The seta are placed at

about the end of the anterior third of each segment.

The abdominal segments, of which there are 12, are marked off from

one another by simple grooves. The first 3 abdominal segments (3.5 m. m.

across) are of about the same size and shape, being broader than long

like the thoracic segments. From the fourth abdominal segment

backwards, the segments grow successively longer at the expense of

breadth, until the last abdominal segment is about twice as long as, but

considerably narrower than the first abdominal segment. The groups

of setæ in the first 2 abdominal segments occupy much the same position

as in the thoracic segments.. In the third abdominal segment they are

situated at about the middie, while in all the following abdominal

segments they lie approximately two-thirds the length of the segment

away from the anterior end.

The tail consists of the three last segments. These are of nearly

the same dimensions as the last abdominal segment but are destitute

of bristles. The very last or the anal segment has the posterior border

developed into a rim with finely and uniformly serrated edge. This

rim surrounds a flatly conical elevation, the summit of which is

occupied by the anus (fig.9).

The setæ on the first setigerous (i. e., the first thoracic) segment

are all acicular, comprising two kinds differing in thickness ; but neither

of these are capillary. Uncini are wanting.

In all the remaining setigerous segments, there are on either side

a dorsal flattened tuft of capillary sete and a ventral row of uncini. The

capillary sete are of two kinds (fig. 12, a and à); both are long, slender

TWO NEW SPECIES OF THE MALDANIDZ. 113

and tapering to a fine point. The uncini (fig.11) are curved and usually

terminate with four distinct hooks.

The tube of the worm was found in fragments. It is membranous

and covered with bluish mud particles on the outside.

The worm here described tallies well with the generic diagnosis of

Clymene as given by DE SAINT JosEPH, thongh the species is evidently

a new one.

In conclusion I beg to offer my thanks to Pror. IJima for the aid

rendered me in the preparation of this paper.

Oct. 234, 1902.

114

A, IZUKA.

Explanation of Plate III.

g. 1. Maldane gotoi n. sp. Lateral view. (about 5/,).

ig.2, Anterior portion of the body of the same. Dorso-lateral view.

(More enlarged).

. 3. Ventro-lateral view of the same.

. 4. Posterior portion of the same. Postero-lateral view.

ig. 5. Uncinus from an abdominal segment of the same. (?9/).

.6. The two kinds. of capillary bristles from an abdominal segment

of the same, (%/,).

. 7. The pinnate terminal portion of a capillary bristle from the

same, (7/,),

ig.8. Acicular bristles from the first setigerous segment of the same.

(29/1).

.9. Clymene harai, v. sp. Dorsal view. (about */,).

. 10. Anterior portion of the same. (More enlarged).

ig. 11. Uncinus from an abdominal segment of the same. (/).

ig. 12. The two kinds of capillary bristles from an abdominal segment

of the same, (I).

On the Occurrence of Phoronis australis

HASWELL near Misaki.

Iwaji Ikeda,

Sci. Coll., Imp. Univ., Tokyo.

As is well known, Phoronis australis HASWELL is a species hitherto

recorded only from Port Jackson in Australia. It has been known to

inhabit deserted Cerianthus tubes at a depth of 27 m. in the locality

mentioned.

Of late years a large and beautiful Phoronis, quite different from

P. ijimai OKA, was discovered as one of the commonest animals near

the Misaki Marine Laboratory. A remarkable fact it is that it so long

remained without attracting the attention of the workers in the Labora-

tory. Possibly it was seen by many but was taken for tentacles of

expanded Actinians, an illusion which, as will soon be seen, is not at all

unlikely to happen.

After a careful study the Phoronis in question was found to agree

quite well with the descriptions of P. australis as given by writers,

especially BENHAM.* So that, I take no heed in considering it as

specifically identical with that species. The fact is of interest as showing

the wide geographical distribution of that species.

The circumstances of the discovery of the Phoronis near Misaki were

as follows : |

On August 10th, 1901, Mr. Havara, then working in the Labora-

tory, obtained and brought in what he thought as a specimen of a gigantic

Actinian. On our examining it together, I was agreeably surprised to

find it to be simply a large gelatinous tube containing a large colony of

* Quart. Journ of Micros. Sci., 1890.

116 I. IKEDA.

a species of Phoronis which was quite unknown to me before. Ou

visiting at once the spot where it was obtained and which was only

about four feet deep at a distance of a few minutes’ rowing from th 4

Laboratory, we have found the very Actinian, a Cerianthus, which was

before supposed to have been captured, but which in fact remained

behind, having been divested of only its gelatinous tube together with

the Phoronis inhabiting it. The latter, being similarly colored as the

tentacles of the Cerianthus, might easily be mistaken for these.

About a month later, after I have left the Laboratory, Kuma, the

collector, reported that the same Phoronis turned out to be something

very common, almost every individual of the Cerianthus being found

in company with a more or less large colony of it.

During Jely in the following year (1902), I have had opportunities of

verifying the truth of Koma’s report and of myself observing the Phoronis

in its native habitat.

In the inlet of Moroiso, at the mouth of which is situated the Misaki

Marine Laboratory, as well as in that of Koajiro, next north to the

one just mentioned, the Cerianthus—a large species with deep reddish

brown tentacles—inhabits the shallow muddy bottom at places in such

abundance that one can not help treading upon it with every step. The

gelatinous tube, in which it lives, may be as thick as one’s arm. Ex-

cepting the smaller tubes, almost every one may be said to be tenanted

by the Phoronis, as was reported by Kuma. At low tides the bottom is

nearly exposed or is covered by only a few inches of water. The Cerian-

thus is then invariably retracted deep into the tube, but the upper end

of this remains visible above the mud surface, together with the Phoro-

nis colony, which radiates forth from the tube-opening simulating in a

way the appearance of the expanded Cerianthus tentacles. Moreover, as

before indicated, the Phoronis is of the same color as these; so that, we

seem to have here a case of mimicry in which the helpless Phoronis not

unlikely benefits itself, in an indirect way, of the protective influence of

_ the nestle-organs possessed by the host, which at other times of the day

will be found expanded at the identical spot. It is difficult to say if

ON THE OCCURRENCE OF PHORONIS AUSTRALIS NEAR MISAKI. 117

the host derives any advantage from the presence of the commensal

worms. i

These are found, usually close together in a large number, near the

upper end of the Cerianthus tube. "he greater part of the body, covered

by its own, loosely fitting, chitinous tubs, lies imbedded in the

gelatinous tube of the host in such a way that the posterior worm-end

which is free from that of the neighbors, is directed downwards and

somewhat obliquely outwards, while the anterior slender portion with

the tentacular crown projects into the lumen of the host-tube, and thence

outwards through the opening of the same

The worm may reach 90 mm. in total length and 4 mm. in breadth

at the broadest part. The tentacles are 12—15mm. long; they are

arranged in two and a half double coils on each side, That part of the

body which lies within the tube-wall of the host is of a pink color, while

the remaining parts are, as already indicated, deep reddish brown.

Closely behind the aboral lophophoral cleft are situated the anal

ridge and the nephridial ridges, both showing the same characteristic

features as were described by BENHAM. Over these ridges, as in fact

over the entire deeply pigmented portion of the body proper, the skin in

the fresh state exhibits closely set circular wrinkles.

The lophophoral chamber contains segmenting eggs and different

developmental stages reaching up to such larve as are in possession of

four pairs of tentacles. ‘These larva are of a moderately large size, being

on the whole relatively thick and short. The anal ciliated belt is dis-

proportionately large; the collar and the tentacular bases are shaded

with reddish brown pigments. Jn a few cases examined of free-swimm-

ing larvæ, evidently belonging to the species and which were obtained

by means of a surface-net in the neighbourhood of the spot inhabited by

the mother-animals, I hive found the number of tentacles increased to

8—10 pairs ; in other respects they were strictly comparable to those still

contained in the lophophoral chamber. And, in these peculiar larve

found at large, I recoguize a form which, long before the mother-animal

became known to me, once came under my obs2rvation in a limited

118 I. IKEDA,

number of specimens among the plankton captured near the Laboratory.

In the stage of development in which the free larve stand, it is unlike

any of the four types of Actinotrocha which I have described from the

waters near Misaki in 1901 (Jour. Sci, Coll, XIIT, p. 534); but whether

it should really be made into a distinct type, remains to be seen.

Note on Walteria leuckarti JJ.

Isao Ijima, Rig., Ph.D., Rig.-Hak.

That beautiful fir-tree-like Euplectellid, which i bave described

under the name of Walteria leuckarti, is seldom brought up from the

depths in a blameless state of preseı vation,— what is not to be wondered

at, if we consider its large size and the fragile nature of its parts. At the

time I have given its full description in the Studies on the Hexact-

inellida, Contribution I (Jour. Sci Coll. Tokyo, vol. XV), I have had

several specimens to base it on ; but there was amongst them only a

single specimen in which I thought the uppermost end of the tubular stem

was preserved intact. On the strength of the condition presented by that

specimen I have assumed that the stem is, as a general matter in the

species, closed at the attenuated apex, the lumen communicating exter-

nally by those oscula found scattered on the sides. While I still

maintain that individuals conforming to that assumption, especially as

the result of the healing of a mechanical injury received, may occa-

sionally be met with in the nature, I have been led to consider the

presence of a small terminal osculum at the superior extremity as the

normal condition. ‘Three specimens have recently been acquired by the

Science College, which all go to support that view.

In these specimens—all robust specimens, 2 ft. and over in height

and with well developed branches—the stem is about as thick as one’s

little finger in the thickest part ; in the upper part it gradually narrows

itself towards the apical end, which isin an almost perfect state of pre-

servation in every case.

One of the specimens is remarkable for the fact that the stem is

bifurcated, each stem after the bifurcation bearing the branches in the

120 I, IJIMA

usual manner One stem-end is injured, but the otber ıs preserved

intact without a sign of maceration. The latter is truncated, measuring

about 2 mm. across at the tip. The terminal surface is occupied in

the center by an orifice, about 1 mm. in diameter.

The second specimen shows the superior end shaped in much the

same way. The truncated apex is here about 23 mm. in diameter, the

central orifice measuring somewhat less than 1 mm. across.

In both the above specimens the small terminal orifice leads into

a canal, which, on account of its small caliber and the dried state of the

specimens, can scarcely be followed downwards to any considerable

extent, in order to determine whether or not it stands in a direct and

open communication with the lumen of the stem. I would probably

have felt no diffidence in assuming the open condition of the passage,

were it not for a circumstance in connection with the third specimen

now before me.

Especially well preserved is the superior end of the stem in the

third specimen. It is much broader than in the others. At a point

3 mm. from the tip, the stem is about 3} mm. thick; here is seen the

last short lateral branch. Above this point it is somewhat swollen in a

manner not unlike the end of the mouth-piece of a smoking pipe. ‘The

tip is here again truncated and, when seen from above, presents an oval

outline, measuring 5 mm. by 4 mm. across. The orifice in the center

is likewise oval, being about 24 mm. by 1} mm. in size. It is therefore

surrounded by an edge of a not inconsiderable breadth, the appearance

of which reminds one of the marginal cuff in certain other Euplectellids.

The moderately large size of the orifice in this specimen permits

the cavity it leads into to be examined under the hand-lens. The felt-

like inner surface presents no notable features ; but the one point, which

is rather against expectation, is the fact that the caveity is shut off from

the lumen of the stem by a partition of the same felt-like appearance as

the wall, although the lumen extends np right near to it with a diameter

of not less than 2 mm. in the uppermost part. By simply looking into

the terminal orifice, it at first seemed possible that the reflection of

NOTE ON WALTERIA LEUCKARTI IJ 121

ight from the desiccated, pure white tissues might cause an illusion

as to the real condition of the cavity in the deeper part. So a probe,

made of a soft paper rolled up into a flexible thread, was made use of to

ascertain the matter. From the side of the orifice. it could not be

introduced for more than 4 or 5 mm., and by inserting it in the contrary

direction from the stem lumen, it could never be managed so as to bring

its end into view through the terminal orifice. It was evident that the

terminal cavity really had a bottom,—that it was partitioned off from

the lumen of the stem by a tissue exactly like that of the wall. Finally

the sponge was carefully cut open at the part; but this did not help

much in making the matter any clearer, owing to the disturbance that

was inevitably caused to the brittle tissue by the scissors.

In the specimen in question, at any rate, I believe the partition was

there, But the partition should probably be considered as only an

apparent, not a real, one. For, it seems perfectly justifiable to assume

that the spaces on both sides of it represent parts of one and the same

‚excurrent space of the sponge, or in other words, that, in the natural

state before desiccation, they stood in open connection with each other

by one or more narrow excurrent canals running through it. ‘This

partition may then be considered simply as a thickening of the wall

inwards so as to narrow th> continuous internal passage at the part.

The relation between the terminal cavity and the stem-lumen should

have been just the same as that which obtains in certain pedunculated

Hexactinellids (f. i., Saccocalyz pedunculata F. E. ScH., among the Eu-

plectellidæ) betweeen the gastral cavity contained in the body proper and

the hollow of the stalk, which latter is only an extension of the excurrent

canal system. It of course remains to be seen whether this is to be

looked upon as an indication that Walteria leuckarti is a stalked form,

in which the hollow stalk is so extraordinarily developed as to represent

the essential part of the entire sponge and in which the body proper

forms so quite an insignificant portion at the apical end as to be said as

being in the verge of disappearance. Above all things it remains yet to

be established by futur» observations that the occurrence of the partition

122 I. LJIMA.

I bave mentioned is a normal and constant fact whenever the natural

apical end is found preserved. But it may here be mentioned in antici-

pation that neither the presence of the branches nor of the oscula on

the stem seems to stand on the way of the above speculation. For, the

former are known to occur sometimes even on the basal disc for attach-

ment, and the latter too seem to be something, which, as a general

matter in the Euplectellidæ, may occur almost in any part of the sponge

body. In Placosoma paradictyum, a new stalked Euplectellid which

will soon be described elsewhere, oscula are present on the stalk as well

as on the basal disc in addition to those on the body proper.

3e that as ‘it may, a stalk is always but a part of the sponge-body,

and there stands nothing against the indietment that Walteria leuckarti

is a tubularly developed form, with a small terminal osculam at the

superior end and a number of larger oscula on the sides.

The terminal osculum is in all probability the first formed in an

early stage of the post-larval development ; it should be strictly homolo-

gous with the similarly situated simple osculum of other Hexactinellids

(Malacosaccus, Saccocalya, the young ol Regadrella okinoseana,—to

take examples fro:n amongst the Euplestellids). On the other hand, I

hold the oscula on the sides of Walteria leuckarti as morphologically

equivalent to thos: openings in certain Euplectellids (Huplectella,

Regadrella, &c.), which were called “ parietal gaps” by F. E. ScuurLze

and “ parietal oscula” by me. Now, whereas it is a general rule with

the Euplectellids that the final outflow of water from the body is effected

principally, if not entirely, at the superior end either by a simple large

osculum or a congregation of separate orifices (sieve-plate meshes

simulating it to the best under the architectural cireumstances, and the

parietal oscula, if present, apparently play only a subordinate ròle in the

discharge of that function, —we seem to have in Walteria leuckarti a

peculiar exceptional case in which the matter is reversed. Here,

nainely, the terminal osculum is abortive while those on the sides are so.

much the more developed as to be indubitably recognizable as sach.

NOTICE.

Vol. [V, Part IV. Published Dec. 30th, 1902. Price, Yen 0.50.

This Journal is on sale at Z. P. Maruya & Co., Ltd.

Tori Sanchome, Nihonbashi, Tokyo:

All manuscripts should be sent to THE EDITOR, ANNO

"TATIONES ZOOLOGICA JAPONENSES, College of Science,

Imperial University, Tokyo.

All business communications should be sent to THE SECRETARY

OF HE TOKYO ZOOLOGICAL SOCIETY, College of Science

Imperial University, Tokyo. i

RARES R E l'ENA

GREAT | MIRE MÈRE KES SE

SHBIRHR | RER=TEREA

SRE (BSNS |

Ÿ. a

ES |

Re S BAB

HW à À BUN

REBEYLALEATZSRM

ANNOTATION ES

ZOOLOGICÆ JAPONENSES.

Vol. IV, Part V.

PUBLISHED

The Tokyo Zoological Society,

rok * ©:

September, 1903:

CONTENTS:

1, Bosminopsis in J apan. | With Plate T IV. >

By EDUARD KLOCKE i ears

II. On a New Polygordius from Misaki (P. Liimai n. sp)

By Atma Izuxa.... ce.

fior in Phoronis. With Plate V. é

AMY EN By Iwan TREDA ..... un aon 5

Bosminopsis in Japan.

Nebst Bemerkungen über einige andere japanische Cladoceren

und den Hakonesee.

(Nach einem Vortrage gehalten in der „Deutschen Gesellschaft für Natur-und

Völkerkunde Ostasiens ” am 28. Mai 1902.)

VON

Eduard Klocke.

(Hierzu Tafel IV.)

Im Jahre 1895 veröffentlichte Jules Richard im ,, Bulletin de la

Société Zoologique de France ” die Beschreibung einer bis dahin unbe-

kannten Cladocere, welche nicht nur eine neue Art, sondern auch

zugleich eine neue Gattung der Entomostraken darstellte. Zwei Jahre

später wiederholte Richard die Beschreibung in einem Aufsatze ,, Ento-

mostracés de l’Amérique du Sud, recueillis par MM. U. Deiters, H.

von Jhering, G. W. Müller et C.O. Poppe” in den ,, Mémoires de la

Société Zoologique de France pour l'année 1897.” Wegen der Aehnlich-

keit des äusseren Habitus des Tieres mit den Bosminiden nannte der

französische Forscher das neue Genus ,, Bosminopsis” und die Art

speciell nach ihrem Entdecker ,, Bosminopsis Deitersi.” Richard teilte

mit, dass die Cladocere in einem Süsswasser bei La Plata (Buenos-Aires)

von Herrn U. Deiters erbeutet worden wäre und zwar nur in einem

weiblichen Exemplare.

Seitdem ist, soweit der Verfasser der vorliegenden Arbeit wenigstens

übersehen kann, des Tieres in der Litteratur nicht wieder Erwähnung

gethan, bezw. ein neuer Vertreter der Gattung Bosminopsis noch nicht

124 EDUARD KLOCKE.

wieder entdeckt worden. Desto mehr dürften die Fachkreise von der

Nachricht überrascht werden, dass nicht nur Bosminopsis Deitersi in

Nord-Japan vorkommt, sondern dass sogar eine zweite Bosminopsis-

species in Mittel-Japan gefunden worden ist.

Der Autor dieser Schrift hatte sich, als er sich für einen mehr-

jährigen Aufenthalt in Japan entschied, dies auch in der Absicht gethan,

einige hier noch brach liegende Gebiete der japanischen Faunistik zu

bearbeiten, die ja in Hülle und Fülle vorhanden sind. Dass er dabei

besonders sich der interessanten Gruppe der Cladoceren annabm, lag um

so mehr auf der Hand, als einerseits noch verhältnissmässig wenig in der

Erforschung dieser niederen Krebse hier gethan ist, und andererseits der

Verfassser den grössten Teil der europäischen Formen von früheren

Arbeiten ® her kennt, so dass sich ihm interessante Vergleichungspunkte

boten.

Um systematisch vorgehen zu können, wurde zuerst Hokkaido,”

die nördlichste der vier grossen japanischen Inseln, in mehrmonatlichen

jährlichen Reisen seit dem Jahre 1901 besucht, und hier war es auch,

wo ,, Bosminopsis Deitersi” zum zweiten Male entdeckt wurde.

Natürlich wurde auch die nähere Umgebung von Tokio nicht ganz

vernachlässigt und besonders der in der Litteratur bereits bekannte,

wundervolle Hakone-See in den Bereich der Untersuchungen gezogen.

Dieser See lieferte die zweite Form der Bosminopsis-Gruppe, welche

Verfasser nach dem bekannten japanischen Zoologen, Professor Dr.

Ishikawa, dem langjährigen Schüler und Assistenten Weismanns in

Freiburg

1.) Ed. Klocke. Zur Cladocerenfauna Westfalens, mit Nachträgen. Westfälischer

Provinzialverein für Wissenschaft und Kunst. Jahresbericht der zoologischen Sektion.

Münster i/W. 1892 bezw. 1894.

Ed. Klocke. Beiträge zur Cladocerenfauna der Ostschweiz. Berichte der natur-

forschenden Gesellschaft Zürich 1894.

2.) Die wissenschaftlichen Resultate der Forschungen auf Hokkaido werden nach der

dritten Reise in diesem Sommer, wahrscheinlich schon im nächsten Winter als ,, Studien

über Hokkaido, (hauptsächlich behandelnd : Säugetiere, Vögel, Reptilien, Cladoceren und

die Süsswasserverhältnisse) in Tokio zum Abdruck kommen.

BOSMINOPSIS IN JAPAN. 125

Bosminopsis Ishikawai

nannte.

Das Verdienst, das Tier zuerst gefischt zu haben gebürt eigentlich

Herrn Dr. A. Fritze. Als derselbe Ende der 80-ger Jahre mit entomolo-

gischen Arbeiten in der Gegend des höchsten Berges Japaus, des fast

4000 Meter hohen Vulkans ,, Fujiyama ” beschäftigt war, machte er in

dem auf ziemlich halber Höhe des Berges liegenden ,, Kawaguchisee ”

einige Fänge mit einem feinmaschigen Netze und übergab die Beute dann

in Tokio Herrn Professor Ishikawa. Dieser untersuchte das Material

und fertigte einige Zeichnungen an, von denen zwei hier abgebildet sind,

liess die Sache dann aber wegen Überbürdung mit anderen Arbeiten

liegen.

Als nun der Verfasser im Mai 1902 indem am Fusse des Fujiyama

liegenden Hakonesee die neue Bosminopsis-Art gefunden hatte und ein

Praeparat derselben Herrn Ishikawa zeigte, erinnerte sich dieser der

mehr als 10 Jahre zurückliegenden Fänge des Herrn Dr. Fritze, die nun

aber leider nicht mehr existierten, und stellte dem Autor die damals

angefertigten Zeichnungen zur Verfügung. Dabei stellte sich heraus,

dass sich unter dem reichlichen Material auch der neue Bosminopsis

befunden hatte, als dessen Heimat also wohl vorlänfig die Seen des

Hakonegebirges anzusehen sind.

Einige Merkwürdigkeiten mögen dabei allerdings nicht unerwähnt

bleiben. Entomostraken aus dem Hakonesee sind bereits von Herrn M.

Schmacker gesammelt und den Herren J, Richard und S. A. Poppe vor-

gelegt worden. Unter diesem Material fand sich kein Bosminopsis.

Die in der ,, Note sur divers Entomostracés du Japon et de la Chine ” ”

von den Herren Poppe und Richard beschriebenen neuen Cladoceren-

Arten: Daphnia Schmackeri und Bosmina japonica habe ich beide an

der Fundstelle wieder konstatieren können.? Weiter ist bemerkenswert,

dass die starren borstenartigen Schalenanhänge der von Fritze im

1.) Bulletin de la Société Zoologique de France pour l'année 1890.

2.) Monospilus tenuirostris fand ich an anderen ‚Orten in der Umgegend Tokios,

ebenso Alona affinis.

126 EDUARD KLOCKE.

Kawaguchisee gefischten Bosminopsis nach den Zeichnungen Ishikawas —

und auf die kann man sich bei der peinlichen Sorgfalt, mit welcher Herr

Professor Ishikawa stets arbeitet und seinem grossen Beobachtungs-und

Zeichentalent aufs Sicherste verlassen—bedeutend geringer und kürzer

sind, als die von mir später im tiefer liegenden und grösseren

Hakonesee gefundenen. Es ist also die Annahme nicht ganz von

der Hand zu weisen, dass Bosminopsis Ishikawai später in den

Hakonesee eingewandert ist und dass unter den veränderten Ver-

hältnissen seine Schalenborsten mehr ausgewachsen sind. Das Tier

kommt im Hakonesee nur pelagisch vor; sollen ihm die Borsten viel-

leicht das Schweben erleichtern ? Wir hätten es dann mit einer Ober-

flächenvergrösserung zu thun, wie sie a. a. O. häufiger beobachtet ist.

Auch bei den meist in grösseren Gewässern vorkommenden Daphnia-

Arten, wie Daphnia cucullata, Hyalodaphnia Kahlbergensis und

Cederstrimii dürften die lang ausgezogenen Kopfschalen wie die Ver-

längerung der Schalenstachel auf Schwebegründe zurückzuführen sein.

Ich will hierbei bemerken, dass ich die der Hyalodaphnia Cederstrému

nahestehende Form von D. cucullata im ,, Sarumako,” einem salz-

wasserhaltigen Haff an der Nordküste Hokkaidos (Ochotzkisches Meer)

im vorigen August gefunden, von den übrigen gehelmten Formen aber

bisher in Japan noch keine entdeckt habe.

Die Gattung Bosminopsis charakterisiert sich folgendermassen :

Habitus äusserlich den Bosminiden ähnlich.

Körper rundlich, mit einem deutlichen Einschnitt zwischen Kopf

und Thorax.

Der niedrige Kopf ist vor dem Auge buckelartig stark gewölbt; er

ist nach unten in conischer Form lang ausgezogen.

Die Fornices sind schwach ; von oben erscheint der Kopf abge-

rundet.

Das Auge ist gross und mit zahlreichen Krystallkörpern versehen.

Das Nebenauge fehlt.

Die Stirnborste entspringt sehr tief, weit vom Auge entfernt, auf

einer kleinen Erhöhung des konishen Kopfendes. :

BOSMINOPSIS IN JAPAN. 127

Die Tastantennen setzen sich an die untere Verlängerung des

Kopfes fest an ; sie gehen im Winkel auseinander und tragen am Ende

ein kleines dreieckiges Glied, an dessen Basis die Sinnesfäden ent-

springen.

Die Ruderantennen sind zweiästig. Jeder Ast hat drei Glieder.

Der eine ist mit drei, der andere mit fünf articulierten Borsten versehen.

Die Schale ist rundlich, hinten in eine bald höher, bald tiefer liegende

Spitze ausgezogen, die sogar in der Medianlinie des Körpers liegen

kann.

Der hintere und untere Shalenrand ist mit mehr oder weniger

zahlreichen, längeren oder kürzeren, starren Borsten versehen.

Von den Beinen sind fünf Paare deutlich sichtbar. Das sechste

Paar ist ebenfalls im verkümmerten Zustande vorhanden.

Das Postabdomen ist ziemlich breit und verjüngt sich stark zum

Ende hin. Es erhält dadurch eien fast dreieckige Form.

Neben den Schwanzkrallen tritt wenigstens ein starker Dorn charak-

teristisch hervor.

Die Schwanzborsten sind kurz, aber ziemlich kräftig.

Eine Verschlussfalte des Brutraumes ist nicht vorhanden.

Der Darm wölbt sich im weiten Bogen nach vorne breit vor. Er

ist nicht geschlingelt.

Zwei Arten:

Postabdomen seitlich bewehrt. Tastantennen fast gerade, glatt.

Behalr-elaıs, quiche... ARMI lin ... B. Deitersi.

Postabdomen nicht bewehrt. Tastantennen nach hinten gekrummt

und mit kleinen Zähnchen versehen.

Schale reticuliert, granuliert, ..#2................. B. Ishikawaı.

Männchen beider Arten bisher noch unbekannt.

198 EDUARD KLOCKE.

Bosminopsis Deitersi. (Taf. IV., Fig. 1.)

Richard hatte nur ein konserviertes Exemplar (Abbildung 1.) zur

Verfügung und es ist

erstaunlich, eine wie

genaue und vortreffliche

Beschreibung er von

diesen einen Tiere

machen konnte. Es ist

hier nicht der Platz,

seine Beschreibung zu

wiederholen. Ich behalte

mir das für eine spätere

Arbeit, dieeine Uebersicht

über alle bis dabin in ci

ganz Japan gefundenen Cladoceren geben soll, vor. Hier möchte ich

nur die augenfälligsten Unterschiede zwischen den Zeichnungen und

Erklärungen Richards und meinen Befunden geben.

Richard bildet neben seinem schematischen Bilde noch die unterste

Schalenecke besonders ab, und zwar zeichnet er eine Borste und eine

zabnartige, weiter unten liegende Erhöhung (mucron). Verfolgen wir

dann das Hauptbild weiter, so sehen wir mehrere Caesuren in dem

untersten Schalenrande. Sowohl der Zahn, wie die feinen Kerben sind

in Wirklichkeit mehr oder weniger hervortretende Borsten von

verschiedener Grösse und Dicke, beziehungsweise Insertionsstellen

derselben. Der Schalenrand von Bosminopsis ist im Allgemeinen über-

haupt sehr variabel. Ich habe sowohl Exemplare gefunden, welche fast

vollständig dem Richardschen Bilde entsprachen, wie auch solche, bei

denen die Borsten noch weiter verkümmert oder noch mehr ausgebildet

sind. Einen systematischen Wert hat der von Richard gezeichnete Höcker

keinesfalls. Der Vorsprung von welchem das Sinneshaar (soie frontale)

BOSMINOPSIS IN JAPAN. 129

ausgeht, ist von Richard in der Einzelfigur, Abb. 2., genau ausgedrückt.

Bedenken erheben sich hier nur gegen den Endteil der en

Tastantennen, der von Richard auf Kosten der

Deutlichkeit vielleicht etwas zu gross gezeichnet ist. Bei

allen meinen Exemplaren, auch bei B. Ishikawai, ist

dieser dreieckige Endfortsatz sehr klein und hyalin,

öfters nur von der Innenseite der Antennen sichtbar.

An seiner Basis entspringen die von Richard nicht

gefundenen Antennenhärchen (soies sensorielles). Mir

scheint dieser Teil der Antennen uberhaupt nicht ein

wirkliches selbstständiges Articulum zu sein, sondern

der dreieckigen Schuppe zu entsprechen, unter welcher

auch bei den Bosmina-Arten die Tasthaare entspringen

(Abb. 3.). Dazu will ich noch bemerken, dass mir bei

keinem Exemplare eine solche Teilungsform aufgefallen

ist, wie sie Richard in seiner Figur 30 zeichnet. Die De

Tastantennen schieden sich vielmehr stets in

\ einfachen geraden Linien. Damit will ich

natürlich nicht sagen, dass Richard falsch

uy) gezeicbnet hat. Wenn sein Bosminopsis und

? die von Hokkaido auch zweifelsohne dieselbe Art

Abb. 3 vorstellen, so ist der räumliche Zwischenraum

zwischen den Fundorten der Beiden doch gross genug, um kleine,

unbedeutende Differenzen in die Erscheinung treten zu lassen. Auch

der Unterschied in der seitlichen Bewehrung des Postabdomens erklärt

sich so; ja er ist nicht einmal bei allen Hokkaido-Exemplaren vollständig

gieich. Die nebenstehende Zeichnuug (Abb. 4.) zeigt ein charakteristisches

Postabdomen. Die Grösse des Richard-

schen Exemplares stimmt mit der

Po > Durchschnittsgröss2 meiner Exemplare

nicht ganz überein. Die Länge der

japanischen Exemplare ist etwas gerin-

Abb, 4. ger und erreicht nur 0,38 mm. Man

130 EDUARD KLOCKE.

könnte aus diesen kleinen Unterschieden vielleicht eine besondere

Japan-Varietät formen; ich habe das nicht gethan, da dies bei anderen

Cladoceren ins Endlose führen würde.

Bosminopsis Ishikawai. (Taf. IV., Fig. 2. u. 6.)

Wollen wir nun noch die neue Art Bosminopsis Ishikawai

beschreiben, so ergiebt sich die Hauptsache schon aus den Abbildungen

und den vorher angeführten Bestimmungsunterschieden. Character-

istisch ist nicht sowohl die Krümmung der Antennen, sondern vor

allem die Bedornung derselben (Abb. 5.). Die Bewehrung erstreckt sich

nicht nur aufdie Antennen selbst, sondern auch

auf das untere Kopfstück. Arn stärksten sind

die Dornen am Innenrande der Antennen

ausgebildet,

Der Kopf erscheint deutlich retikuliert.

Die Schalen sind durch Einlagerungen (von

Kalk?) wenig bis stark undurchsichtig. Sie Abb. 5.

machen dadurch einen schmutzigen Eindruck.

Die ausgezogene Spitze des hinteren, oberen Schalenrandes liegt

meist bedeutend höher als bei B.

Deitersi. (Siehe Tafel).

Der hintere und untere Schalen-

rand ist im Allgemeinen bei den

Hakone-Exemplaren weitaus stärker

bedornt und erinnert beim ersten

Anblick fast an Ilyoeryptus.

Dasschwächstbedornte Kawaguchi-

Exemplar zeigt Abb. 6. nach einer

Abb. 6. Zeichung Ishikawa’s.

BOSMINOPSIS IN JAPAN. 131

ra: Das Postabdomen ist unbewehrt und weist nur

\ die charakteristischen Nebenkrallen auf. (Abb. 7.)

Die Länge des Körpers beträgt 0,38—0,42 mm.

Alles Übrige stimmt mit der vorhergehenden

RTS Art überein, so auch die Ruderantennen (Abb. 8.,

Ishikawa del.)

Als Fundorte von B. Deitersi sind

mir bis jetzt nur zwei kleine Seen

bekannt, die beide im flachen Lande

liegen und zur Niederung des Flusses

Ishikari, des grössten Stromes Hok-

kaidos und zugleich Japans überhaupt,

gehören. Im Ganzen habe ich in den

letzten 2 Jahren 52 Süsswasserbecken

in Hokkaido untersucht.

Die Fundorte von B. Ishikawar

beschränken sich ebenfalls auf zwei

Seen, den schon oben genannten

Kawaguchisee und den Hakonesee.

Immerhin sind die Fundorte der beiden

Cladoceren, ihrer Lage sowohl wie ihrer

Art und Weise nach, sehr verschieden.

Bosminopsis Deitersi lebt in schilf-

umstandenen, flachen Landseen, A

Bosminopsis Ishikawai dagegen in tiefen Bergseen, die von schroffen

Felspartien umgeben, eines starken Pflanzenwuchses ermangeln. Auch

die klimatischen Verhältnisse sind verschieden, doch scheint das Klima

bei der Verbreitung der Entomostraken keinen grossen Unterschied zu

bedingen—finden wir doch von Norwegen bis Italien fast die gleichen

Cladoceren; und heute kann ich schon sagen, dass die Cladoceren

Japans, besonders der nördlichen Teile dieses gesegneten Landes kaum

merkliche Unterschiede zwischen den mitteleuropäischen Formen

erkennen lassen.

132 EDUARD KLOCKE.

Von keinem See Japans wird im Auslande wohl mehr gesprochen

als von dem oben mehrfach erwähnten Hakone-See. Seine landschaft-

liche Schönheit, die wilde Gebirgsformation seiner Umgebung und die

Nähe des wundervollen, alles überragenden Vulkans Fujiyama locken

alljährlich Tausende von Fremden an diesen romantischen Platz. Auch

in der Geschichte Japans spielt der See eine grosse Rolle, wie er zugleich

eines der ersten Japanischen Wasserbecken war, das naturwissenschaftlich

in Europa und Amerika bekannt wurde. Es mag daher wohl am Platze

sein, in kurzen Worten die Verhältnisse dieses berühmten Sees zu skizzie-

ren. Um seine Erforschung hat sich besonders Herr Akamaro Tanaka

von der Adelakademie in Tokio verdient gemacht.

Der Hakone-See, oder vielmehr ,, Ashinoko ” (Schilfsee), wie ihn die

Japaner nennen, befindet sich südwestlich von 'Tokio, nicht weit vom

Schnittpunkte des 139 Längen-und des 35. Breitengrades. Er liegt in

vulkanischer Umgebung und verdankt wohl auch seinen Ursprung

vulkanischen Kräften. Rings von steilen, grösstenteils bewaldeten

Bergen umgeben, spiegelt sich bei klarem Wetter in seinem durchsich-

tigen Wasser das mächtige, fast stets mit Schnee bedeckte Haupt des

japanischen Riesenvulkans, des mächtigen Fujiyama. Der Wasserspiegel

liegt ca. 700 m. hoch ; er erstreckt sich in einer Länge von 6600 m. von

S.O nach S.W. Die breiteste Stelle befindet sich im Südteile des Sees

(2500 m.), die engste kurz von der Endausbuchtung weiter nördlich

(300 m.). Das Wasser bedeckt einen Flächenraum von 2,145,521 Tsubo.

(1 Tsubo=36 engl. Quadratfuss). Der Kubikinhalt des Seebeckens ist

auf 113,428,800 cbm. berechnet. Die Durchschnittstiefe beträgt 16 m.,

die tiefste Stelle 48 m. Im See befindet sich eine submerse Insel, “Naka-

shima ” genannt, mit einer Ausdehnung von 420 : 120 engl. Fuss. Ueber

ihr sinkt die Wassertiefe bis auf 2 m. herab.

Die Durchsichtigkeit ist sehr gross. Ich hatte leider noch keine

(Gelegenheit, genauere Untersuchungen nach dieser Richtung hin im

Hakone-See anzustellen. Im Allgemeinen traf ich aber in Japan, beson-

ders in den Kraterseen Hokkaidos auf erstaunliche Resultate. So ver-

schwand mein Fangnetz ans weisser Gaze in dem sehr tiefen “Shikotsuko”

BOSMINOPSIS IN JAPAN. 133

Ende Juli 1901 erst bei einer "Tiefe von 19,5 Meter. Forel hat die

Sichttiefe des Genfer Sees auf 6,6 m. als Durchschnitt im Sommer (Juli

5,6, August 5,3) und 12,7 m. als Durchschnitt im Winter berechnet. Dabei

gilt der Lac Léman als einer der klarsten, wenn nicht als durchsichtigst-

er See der Alpen !

Obgleich es in den Hakonebergen empfindlich kalt werden kann

und das Thermometer weit unter den Nullpunkt sinkt, ist eine Vereisung

des Hakonesees noch nicht festgestellt worden. Nur in dem sehr kalten

Winter des Jahres 1884/85 bildete sich etwas dünnes Eis an den Ufern.

Herr Tanaka fand während mehrjähriger Messungen eine Höchst-

temperatur von 24° C. im September und als niedrigste eine Temperatur

von 5°C im Februar. Ich selbst mass am 5. Mai des Jahres 1902 11°C in

einer Tiefe von 10 Metern. Warme Quellen, an denen das Hakonegebirge

reich ist, dürften der Grund sein, dass die Temperatur des Wassers nicht

tiefer steigt. Von unterirdischen Quellen wird der See auch gespeist.

Bemerkenswerte Zuflüsse sind nicht vorhanden. Dagegen fliesst das

Wasser durch zwei Abflüsse, einen Kanal und den Fluss Haiyakawa

(Schneller Fluss) ab. Der Fluss mündet nicht allzuweit von seinen

Ursprunge, bei Odawara ins Meer.

Der Name Ashinoko (Schilfsee) könnte zu der falschen Vermutung

Anlass geben, dass der Hakone-See stark mit Schilf bestanden wäre.

Dem ist nicht so. Im See wächst kein Schilf, desto mehr sind aber

einige Teile der Umgegend mit diesem hohen Grase bedeckt. Besonders

auf dem Tokaido, der alten Heerstrasse nach Tokio, wächst Schilfrohr in

Hülle und Fülle. Die hier vorkommende Art ist eine Varietät unseres

gemeinen deutschen Schilfrohrs, Phragmites communis var. longivalvis

Mig.

Schliesslich sei in Bezug auf den Hakone-See noch bemerkt, dass die

Ufer steil bis sehr steil sind und nur wenig Raum für Wege und Ansie-

delungen lassen. Die Europäer und Amerikaner, welche den See wegen

seiner Schönheit, der heissen Mineralquellen in der Umgegend und des

kühlen Klimas im Sommer jährlich in grosser Zahl besuchen, haben ihn

nach den beiden kleinen am Ufer liegenden Badestädtchen Motohakone

134 EDUARD KLOCKE.

bezw. Hakone “ Hakone See” getauft. Unter diesem Namen ist er

denn auch in der Wissenschaft bekannt geworden.

Der Erste, welcher im Hakone-See nach Entomostraken gefischt

hat, ist wohl, der schon erwähnte Herr M. Schmacker gewesen. Ich

bringe auf der beigefügten Tafel zwei Photograpbieen von Daphnia

Schmackeri und Bosmina japonica, Poppe u. Richard, deren Artkenn-

zeichen von diesen beiden Forschern bereits im Jahre 1890 vortrefflich

geschildert worden sind.

Bezüglich der Abbildung von Daphnia cucullata bemerke ich,

dass sich in meinem Material kein unverletztes Exemplar befand. Ich

machte die Photographie daher von einem wenigstens der Form nach

unbeschädigtem Tiere. Auffallend ist auch hier, wie bei Bosminopsis

Deitersi und manchen anderen japanischen Cladoceren, dass die Maasse

hinter den in anderen Ländern gefundenen ziemlich zurückbleiben. Die

vorliegende Daphnia bat von der Helmspitze bis zum Endrande der

Schale—also ohne Stachel—nur eine Länge von 0,58 mm. Exemplare

vou der Grösse der von mir selbst sonst in Europa Gefundenen kamen

mir hier nicht unter das Mikroskop.

BOSMINOPSIS IN JAPAN. 135

Erklärung der Tafel IV.

(Fig. 1—5 sind Microphotographien, aufgenommen mit einer mikro-

photographischen Camera der Firma E. Krauss & Cie., Paris, bezogen

von deren Filiale in Tokio. Fig. 6, ein Osmiumpraeparat, ist gezeichnet

mit dem Zeichenprisma. Als optische Instrumente wurden Krauss-

Zeiss-Instrumente benutzt).

Fig. 1.—Bosminopsis Deitersi Richard.

Fig. 2.—Bosminopsis Ishikawai Klocke. Stark behaartes Exemplar.

Fig. 3.—Bosmina japonica Poppe et Richard. Bei manchen Ex-

emplaren sind die Tastantennen noch mehr unter den Körper zurück-

gebogen.

Fig. 4+— Daphnia Schmackeri Poppe et Richard. Mit Krauss-Zeiss

Planar photographiert. Charakteristisch ist u.a. der gerade Schalenstachel.

Fig. 5.—Schale einer Daphnia cucullata aus dem Saruma-Haff am

Ochotzkischen Meere.

Fig. 6.— Verlauf der Nerven am kopfende und den Tastantennen

von Bosminopsis Ishikawai. Mit dem Zeichenprisma gezeichnet.

— To _—__T

On a New Polygordius from Misaki

(P. Ijimai n. sp.).

Akira Izuka.

Larvæ believed to be those of Polygordius have from time to time been

obtained in the neighbourhcod of the Misaki Marine Biological Laboratory

without the mature worm having been discovered. During the last

spring, bowever, I succeeded in obtaining, at a spot of the sea-shore close

to the Laboratory, a large number of adult Polygordius, probably the

generator of the larve mentioned and which I consider to represent a

species new to science. The discovery may be worth recording, since

all the species hitherto kriown * of the genus are from the coasts of

Europe.

I propose to call the Misaki species

Polygordius Iimai

in dedication to Professor Isao IJImA of the Science College, Tokyo.

The thin and cylindrical body measures 70-77 mm. in length and

0.6-0.8 mm. across in the middle part of the body, where this is broadest.

The general colour of females is yellowish pink, lighter in tone in the

* Six in number, viz.:

1. P. lacteus SCHNEIDER. (Helgoland).

2. P. apogon M'Intosa. (Shetland).

3. P. villoti PERRIER. (Roscoff).

4. P. erythropthalmus GIARD. (Concarneau).

5. P. appendiculatus Fraıront. (Naples).

6. P. neapolitanus Fraıront. (Naples).

138 AKIRA IZUKA.

anterior region and deeper in the posterior—say, in the posterior 3 of the

body-length. This deeper tone of the colour is evidently due to the eggs

contained in the region. The males are of a very light pink color, which

may bear a milky-whitish tint on attaining sexual maturity. Both

sexes show a slight iridescence all over the body.

In fresh specimens the segments are externally indistinct or scarcely

recognizable in tbe anterior region, though they are always plainly

indicated in the posterior half of the body. After preservation they are

marked by distinct ring-grooves all throughout the entire body.

The mouth, situated on the ventral side of the anterior end, is

subtriangular in shape. The cephalic lobe in front of it bears, as usual, a

pair of horn-like tentacles which are widest at base and are directed

antero-laterally. They are only about 1 mm. long. Eyes are wanting.

There exists, on each side of the head and at about the level of the

mouth, an oval-shaped vibratile pit, the greater axis of which is vertically

disposed.

The body gradually narrows behind in the part adjoining the

anal segment. This segment is broadest at base, where it is broader

than the part directly in front of it. It again narrows towards the

terminally situated anus. ‘The anal margin is nearly entire or is at most

wavy, forming obsolete lobes eight in number. In its anterior half the

anal segment bears numerous papille arranged in a number of longitudinal

rows or zones, —which arrangement, as also the peculiarities of the anal

orifice, forms one of the characteristie points by which the species may

be distinguished from all the others. Three filiform appendages occur

on the anal segment : one on the dorsal side in the median line and the

rest ventro-laterally on either side,

Observed on sections, a thick cuticle is found to cover the hypodermis,

beneath which I have failed to discover ring-muscle fibres. Neither do

these seem to exist inside the layer of the strongly developed longitudinal

muscles, this layer being directly lined by the peritoneum on the inner side.

A dorsal and a ventral mesentery suspend the intestine in the body-cavity.

Neither longitudinal nor ring muscles are present in the intestinal wall.

ON A NEW POLYGORDIUS FROM MISAKI. 139

The vascular system exhibits a general agreement with that of other

species—as for instance, of P. neapolitanus—except in one important

point. This consists in the fact that the consecutive lateral loops, present

in a pair in each segment and which connect the dorsal and th ventral

vessel, stand on either side in open anastomosis with one another by

means of lateral longitudinal commissures. These evidently correspond

to the blind vessels known to exist in certain species (f. i., P. neapolitanus)

arising from a point midway in the course of the lateral loops. In the

present species, they come into communication with the lateral loops

towards which they are directed, instead of ending blindly before reach-

ing these. On each lateral loop it is seen that the commissure coming

from the loop directly in front finds its insertion at a point a short

distance ventral to the origin of the rext posteriorly following commis-

sure. Therefore, the successive lateral commissures on either side of the

body do not form an uninterrupted longitudinal vessel, as does the

dorsal or the ventral vessel.—A pair of small vessels, arising from the

points where the ventral vessel sends forth the paired lateral loops,

seem to supply the intestinal wall.

The species contain nearly mature ova or active spermatozoa as

early in the season as the end of March.

The worms described live along the lowest tide-mark and in or on a

bottom consisting of shell-fragments, sand, pebbles and blocks of a rough

sandstone. They are exceedingly active creatures, constantly moving

the head-end and burrowing or creeping with remarkable ease. Not-

withstanling this nature, they are excessively frail and fragile, breaking

into pieces by the slightest cause or apparently even spontaneously when

kept in captivity.

July Ist, 1903.

On the Development ofthe Sexual Organs and

of Their Productsin Phoronis.

Iwaji Ikeda, Rigakushi.

With Plate V.

In Phoronis ijimai and P. australis, asin all other species of the

genus, the ovary and testis are situated in the posterior part of the body,

on one side of the stomach, When fully developed, they represent two

elongate and loose masses of 1-2 mm. in length, lying side Ly side—the

ovary dorsally to the testis—in close apposition with each other and

nearly filling up a lateral chamber of the body-cavity in the region

indicated. Clasped between the two organs runs the efferent blood-

vessel. This sends forth numerous, branched or simple, blindly ter-

minating capillaries, which may be distinguished into two series, the

ovarian and the testicular, according to the organ they penetrate into.

Of great importance are these capillaries in the genesis of the sexual

glands, as giving the fundament for their development.

In young individuals—say, in all those which have not yet attained

the full adult size— the sexual glands are without a trace. In their stead

there exist only the said capillaries, the so-called ccecal or contractile

capillaries, which, arising as slender thin-walled off-shoots from the

efferent od-vessel, lie free in the body-cavity. The capillary wall

consists of three layers, 7.e., (1) the endothelium, (2) the peritoneum and

(3 the connective-tissue layer lying between the above two. Of these

three layers the middle one is that which in the larger vessels bears the

142 I. IKEDA,

character of a tolerably thickly devel :p21 muscular layer. Both the

endothelium and the peritoneum are made up of the usual flat cells.

Needless to say that all the layers are directly continuous with the

corresponding layers in the wall of the efferent blood-vessel.

All the subsequent changes leading to the formation of the ovarian

or th» testicular tissue concern only the peritoneai layer of the contractile

capillaries. It may here at once be mentioned that that layer has to

serve for a time as reserve ground for certain nutrient matter and is

then absorbed, becoming finally completely replacel by the sexual cells

of likewise peritoneal origin. The changes begin to take place after the

animal has attained nea:]y or quite its full size. Among the specimens

of P. ijimai obtained and preserved in the months of December to May

and of P. australis kilied in September, I have found some individuals

with the sexual organs already fully developed, while in others they were

still in an earlier or a later stage of development.

As the prelude to the development of the sexual glands, a part of the

calls forming the peritoneal layer of the capillaries transforms itself into

a peculiarly characterised epithelial tissue, which has been called by

KowALEwsky the “ Fettgewebe ” and by Corı* the “ Gefässperitoneal-

gewebe.” That this tissue is derived from the originally flat peritoneum

of the capillaries has been correctly pointed out by Cori. The

peritoneal cells increase in number, although I have not been able

to see direct proofs of cell-division. At the same time the majority

of them, but not all, become bulky, growing more especially in

height, which is due in a measure to the accumulation of the yolk-like

inclosures soon to be noticed. This change, so far as I have seen,

begins at the basal part of the capillaries, that is to say, near

the junction of these with the efferent blood-vessel, and thence proceeds

distally towards their blind ends. The peritoneum of the efferent

vessel itself never participates in the above metamorphosis; it is

seen to pass gradually and continuously into the now greatly thickened

and much altered peritoneal layer of the capillaries at the base of these.

* Veitschr. f. Wiss. Zool., Bd. 51, 1891.

SEXUAL ORGANS AND THEIR PRODUCTS IN PHORONIS. 143

Directly beneath that layer, and therefore interposed between it and the

connective-tissue layer, there remain at wide intervals small and flat

peritoneal cells which may be said to retain their primitive condition.

These small cells may best be observed on longitudinal sections of the

capillaries (Pl. V., fig. 2, f.p.). Irregularly scattered though they are,

they may perhaps without impropriety be considered to form a fourth

layer to the wall. ‘Those cells are of great importance in that they give

rise to the germinal cells. Not that all the capillaries which later stand

in relation with the sexual glands undergo the above metamorphosis

nearly simultaneously, but there may remain for a considerable length of

time after some have completely gone through it, such others as are

or more less belated in the process or bave not even begun to show it.

A capillary after the completion of the change has increased in

external thickness, though the calibre of the lumen remains the same as

before. Or, it may be more proper to say that the capillary 1s now in-

vested all around by a remarkably thickened peritoneum, consisting of

columnar or pyramidal cells arranged in a layer. A cross-section (PI. V.,

fig. 1) presents a wheel-like appearance. In the centre is the capillary

lumen, lined by the usual endothelium which rests on a connective-tissue

layer. Externally to this are arranged the large and tall peritoneai cells

ina radial manner. As the result of this mode of Zrangement, the

enlarged cells are narrowed towards their inner end. The small perito

neal cells which should occur at their base are very difficult to observe

on cross-sections (and are therefore not shown in fig. 1).

The enlarged peritoneal cells exhibit a fine but distinct bounding

membrane. ‘The relatively small nucleus is always situated near their

outer and broader end—often almost in contact with the cell-membrane

at that end. The cytoplasm is of a clear appearance, being sparsely and

finely granular ; it is lightly coloured by staining reagents. In certain

preparations an indication of a reticular structure of the cytoplasm was

observable, reminding me of the “ Fadennetzwerk ” described by Cort

in the same cells of P. psammophila ; but that is certainly not a con-

stant phenomenon in the species studied by me.

144 I. IKEDA.

Inclosed in the cytoplasm are, as is well known, numerous yolk-like

spheres of various sizes (fig. 1, a, b).* They are not crowded together,

but are scattered throughout the cell-body. In the fresh state they are

colourless, opaque and very weakly refractive. The larger of them may

be nearly as large as, and sometimes even larger than, the blood”

corpuscle ; whereas the smaller may be as small as the nucleus of the

same. Allthe smallest (fig. 1, 5) are simple elemantary bodies, which

cohering in varying numbers, seem to compose the larger spheres (a).

These are then, as are in fact apparent from their structure, to be con-

sidered as conglomerates of the smaller elements. Osmic acid blackens

the spheres, as when the tissue had been fixed with Flemming’s fluid.

Iron-hematoxylin stains them so deeply that the conglomerate nature

of the larger ones becomes obscured, Very good preparations were

obtained by using corrosive-sublimate solution as the fixing reagent and

by staining with Delafield’s hematoxylin in combination with either

eosin or Congo-red. Eosin invariably stains the spheres very deeply.

As will be shown, the above described spheres disappear—are

evidently used up—during the development of the sexual cells, the epithel-

jum itself dwindling away and finally likewise disappearing to give place

to the genitial glands. Taken in all, it appears to me exceedingly prob-

able, if not ndubitable, that the spheres are to be looked at in the light

of a nutrient substance in reserve, which is needed for the development

of the sexual products. Whatever they may be in chemical respect,

physiologically they seem to be much the same as the fat. In accordance

therewith, the modified peritoneal cells containing them may without

impropriety be called the reserve-nutriment cells, or for the sake of

brevity, simply the nutriment cells. The layer formed of them may

conveniently be referred to as the nutriment layer.

Corr found in P. psammophila the so-called spindle-bodies, first

* Besides the yolk-like spheres here described, the cells in question contain in the

fresh state a number of minute, refringent and reddish-yellow granules scattered in the

cytoplasm. After the tissue is fixed and preserved, they are no longer to be seen. I am

not in a position to decide whether they are pigments or a fatty matter.

SEXUAL ORGANS AND THEIR PRODUCTS IN PHORONIS. 145

described by KowaLEwsKy, not only freely floating in the cœlomic cavity

but also contained in the nutriment cells. In P. australis I have met

with the same bodies in the cœlomie fluid, but never in the cells just

mentioned ; and in P. ijimai I have simply never and nowhere come

across them. ‘They are evidently somethiug of inconstant occurrence

and therefore probably of no great physiological importance, so far at

any rate as concerns the discharge of funetion by the nutriment cells.

Another noteworthy point mentioned by Cort is the fact that he found,

engulfed in the cells under question, bodies that appeared to him to be

the remains of degenerating blood corpuscles. It seems to me that this

matter requires confirmation before it can be accepted as a fact. For

my own part, I have failed to find any indication whatever tending to

support Cort’s observations on the point.

It now lies in order to describ> the formation of primary germinal

cells. BENHAM* rightly stated that the sexual products develop from

peritoneal cells of contractile capiliaries, but the details of the develop-

mental process have never become known; and as to the relation be-

tween the generative organs and the nutriment layer, nobody seems to

have as yet gone into the matter

According to my observations, the primary germinal cells begin to

appear on a capillary sooner or later after the full development of the

nutriment layer. As already indicaded, they arise from, and by prolife-

ration of, those small and flat peritoneal cells which remain at intervals

at the base of the nutriment layer. This cell-proliferation commonly

begins to take place near the base of the capillaries, thence proceeding

gradually towards the free distal end. This extension is due to the

addition of new cells which have arisen by multiplication of the smal,

and flat peritoneal cells situated on the road of advance. The newly

arisen cells—the oogonia or spermatogonia as the case may be—are

nearly uniformly small and approximately spherical in shape ; they form

at the region a layer, which is two, three or several cellsthick and which

is at first covered over by the nutriment layer.

* Quart. Jour. Mic:osc. Sci., Vol. XXX, 1889.

146 I. IKEDA.

It is henceforth necessary to treat of the ovary and the testis sepa-

rately. I willtake up the former first, as giving a clearer insight into

the developmental process.

The Ovary and Its Product.—Closely following their first formation,

the oogonia begin to differentiate themselves into the oocytes and the

follicular cells, the two kinds of cells which go to form the entire ovarian

tissue. Tbe former are distinguished by the larger size, which of course

beccmes more and more pronounced with the advance in growth ; the

latter remain small and are arranged around and between the former.

Each egg-follicle thus consists of a central cocyte enveloped in a single-

layered follicular epithelium. Around and along the capillary the follicles

range themselves ina layer. They show amongst them a considerable

difference in size from an early period of their development. A definite

rule seems not to exist as to the relative situation of the largest and the

smallest follicles, though at a later stage of the ovarian development it

is often noticeable that those more proximally situated on a capillary

are on the whole the more advanced in growth.

As already indicated, the ovarian tissue is at first covered over by

the nutriment layer. In inverse proportion as the follicles or the oocytes

in that tissue grow in size, the overlying nutriment cells diminish in

height (see PL.V., figs. 2 and 3); that is to say, they become smaller and

smaller, while the inclosed yolk-like spheres constantly diminish in quan-

tity. The smallest and simple spheres are the first that disappear, evi-

dently as the result of resorption. This leads finally to the total dis-

appearance of the spheres as well as of the nutriment cells themselves,

after which the ovarian tissue forms the outermost layer of the capillary

wall and is thus directly washed by the coelomic fluid (fig. 4). It may be

said in general that as the development of the ovarian tissue advances,

starting from the capillary base, the nutriment layer gradually dwindles

awrty before it.

At a certain stage in the genesis of the ovary, the numerous capil-

laries giving foundation to it may show the transformation carried out to

various degrees. While some capillaries or capil'ary branches may

SEXUAL ORGANS AND THEIR PRODUCTS IN PHORONIS. 147

already be invested by the ovarian tissue right up to the blind end, others

may exhibit it in the proximal portion only, the distal portion still retain-

ing the nutriment layer to a greater or less extent. Occasionally there

may exist even such as have just begun to bear nutriment cells and

therefore remain quite thin in the distal part. With full development of

the ovary, all the capillaries concerned in its formation should have the

nutriment layer entirely replaced by the egg-follicles.

To follow the growth of the oocytss, it is at first generally spherical

in shape, with the vesicular and relatively large nucleus situated in the

centre. As it grows it soon assumes an ovoid or flask-like shape, being

attached to the capillary by the narrower end. This is doubtless due to

mutual pressure exercised by the oocytes at base, as they find attachment

close together on the narrow axially running capiliary (see fig. 4). The

nucleus is now seen to occupy an eccentric position in the broader end

of the egg-body. The follicular envelope is visible for some time as a

flat epithelime showing compressed nuclei at rather long intervals; but

as the egg approaches full size it becomes so distended and thinned out

that it appears as a fine structureless membraue, in which the nuclei can

no longer be made out. In small oocytes, such as are represented in fig.

2, the vitellus is finely granular and is deeply stained by basic stains, such

as hematoxylin, gentiana-violet, cfc. In more advanced stages, it is

somewhat coarsely granular and then greedily takes up acid stains, such

as eosin, Congo-red, orange-G, etc. instead of basic.

The large clear nucleus exhibits a net-work of linin; it is wide-

meshed though densely arranged in immediate environment of one of the

nucleoli, of which there can be distinguished two (see fig. 4, E) Both

nucleoli are spherical and very distinct, their size growing in proportion

with the growth of the oocyt2s. The one nucleolus (x) is distinguished

by the fact that it consists of two parts: a small sphere which is but

very lightly stained by nuclear stains, and another larger sphere which is

deeply stainable. At first, while the oocyte is still very small, the former

lies simply apposed on the surface, but later sinks completely into the

substance, of the latter. In the stages shown in fig. 4, the nucleolus in

148 I. IKEDA.

question (x in A, C, E) is in a state in which the smaller and little stained

constituent is already completely inclosed within the other. The former,

after the inclosure, swells and produces within it a few vacuoles. The

second nucleolus (n’), in an early part of the ovogenesis, is of a homo-

geneous appearance, staining deeply by nuclear stains (fig. 4, A & B, w’).

However, after growing in size to a certain degree, it likewise becomes

vacuolized (fig. 4, D& E, n’).

Very remarkable is the occurrence within nucleus of a small lightly

stained body shaped like a comma; it is found attached by its narrower

end to the inner surface of nuclear membrane. It seems to be formed

after the oocyte has advanced considerably in growth (see fig. 4, D & E,

x). I know nothing to say about its nature or significance.

After the oocyte has grown to full size (0.2 —9.3 mm. in diameter) and

shortly before the ovulation by dehiscence of the follicular membrane,

the nucleus begins to show changes preparatory to the formation of the

first polar globule. The nuclear membrane is slackened, the linin net-

work becomes indistinct and the nucleoli no longer exist as such ; at the

same time a number of chromatic granules maks their appearance in the

nucleus.

The oocyte after liberation from the ovary is spherical; it is supplied

with a very thin vitelline membrane. Of the cytological changes which

now take place in connection with the production of polar globules, the

detail was given in my paper on the development of Actinotrocha (Jour.

Sci. Col., Vol. XIII, 1900). Suffice it to say here that so long as the

oocyte remains in the coelomic cavity, the nuclear division figure is in

the mesophase or in the metaphase. The chromosomes belonging to it

consist, in P. ijimai, of six pieces; in P. australis there are twelve of

them. The expulsion of the first polar globule takes place after the oocyte

is expelled to the exterior through the nephridial fanneis, at a time when

it is met by the free-swimming spermatozoa. This is soon followed,

without rest, by the formation of the second polar globule,—the well

known reduction division, by which the chromosomes are reduced to half

the normal number. Thus, in P. ijimai the fully mature ovum is in

SEXUAL ORGANS AND THEIR PRODUCTS IN PHORONIS. 149

possession of only three chromosomes of maternal origin. Exactly the

same reduction occurs during the formation of spermatids, as will soon

be pointed out.

The Testis and Its Products.—As to the development of the testis

I have not been able to obtain as clear an insight as in the case of the

ovary. However it may safely be said that the iuceptional condition as

well as the general manner oi development are essentially the same.

The small and flat peritoneal cells, remaining beneath the nutriment layer,

proliferates, thus forming a crowded assemblage of small spermatogonia

around a capillary at its basal portion. This is the first Anlage of the

testicular tissue. As the spermatogonia increase in number, the over-

lying nutriment layer is lifted up by them; at the same time its cells ba-

come smaller and are evidently on the way of becoming totaliy resorbed,

in much the same way as we have seen in the case of the ovary. Mean-

while, what is peculiar to the testis, the connective-tissus layer sends

out fibrous bundles outwards and radially into the asserablage of sper-

matogonia, which soon arrange themselves around and along each of the

said connective-tissue bundles. The spermatogonia thus form radial

strings, the testicular strings, of which the connective-tissue bundle con-

stitutes the axis. With the disappearance of the covering nutriment

layer, the strings are somewhat loosened in their relation to one another

and are seen to beset the capillary surface in a villi-like manner.

The above changes on a testicular capillary proceeds from the base

distally towards the blind end. We may say that the testicular Anlage,

consisting of a mass of spermatogonia traversed by radial fibrous

bundles and covered over by the degenerating nutriment layer, shifts

its position in the said directon, leaving behind it the testicular

tissue just developed in the form of uncovered cellular strings and

ever encroaching upon and lessening the domain of the nutriment

layer in front of it. The advance of the growing end of the testicular

tissue takes place in that the flat p2ritoneal cells remaining beneath the

nutriment layer undergo proliferation in site and thus give rise to new

spermatogonia. The development of a testicular capillary is complete

150 I. IKEDA.

when the entire vascular wall is covered by the loosend strings up to

the blind end.

Pl. V., fig. 5 shows a cross-section of a fully developed and

functional testicnlar capillary. (In the capillary lumen is seen a blood-

corpuscle). It will be observed that the spermatogonia lining the testicu-

lar strings are very small in size and approximately spherical in shape.

They contain a comparatively large and vesicular nucleus in which are

visible a few number of dot-like chromatin. Towards the outer end of

the strings they grow somewhat larger while the chromatin dots in the

nucleus become more numerous and distinct. The largest cells at the

outermost end may be called the spermatocytes. They are more or less

detached from connection with the strings and are loosely adhering to

one another in the periphery of the testicular capillary. Here the divi-

sion of the spermatocytes into spermatids and the development of these

into spermatozoa may without difficulty be studied.

In figs. 5 and 6, the same lettering has been used to denote the

same stages in the spermatogenesis. The cell indicated by a is a loosely

lying spermatocyte, in which the chromatin bas the form of a spiral

band (spireme). The cell d is another in a more advanced spireme stage.

Soon the nuclear membrane dissolves away and the karyokinetic figure

distinctly establishes itself. The rod-like chromosomes, in the equator-

ial plane of the spindle, number six in P. ijimai and twelve in

P. australis. Inc and d the karyokinesis is progressing; the division

ensues, which is directly followed with another without the nucleus

passing through a resting phase. The letter e indicates the daughter

cell produced by the first division of a spermatocyte and f, one of the

four spermatids given rise to by it after the second division. The sper-

matid is very small, corresponding in bulk to about one-fourth of the

original spermatocyte. Only half the normal number the chromosomes

enters into it, exactly as in a ripe ovam. Thus, in P. ijimai their num-

ber to each spermatid is three, which are at first so attached to one

another with their ends as to present a triradiate figure (see fig. 5, f).

This is soon followed by a stage in which the spermatid is supplied with

SEXUYL ORGANS AND THEIR PRODUCTS IN PHORONIS. 151

a small, spherical and compact-looking nucleus (fig. 6, f). In an early

stage of the spermatogenesis we see that nucleus greatly condensed and

trans/ormed into a slender elongate shape, curved after the manner of

the letter C (fig. 6, 9). The bent nucleus is then seen to grow somewhat

in thickness and greatly in length (h and è), eventually to stretch out as

the head of the filiform spermatozoa (7), of which the tail is about twice

as long as the head.

152 I. IKEDA.

Explanation of PI. V.

All figures relate to Phoronis ijimai Ik.

Fig. 1.—Cross section of a contractile capillary, externally covered over

by a layer of large nutriment cells (“ Fettgewebe ”) derived from

the peritoneal cells. a, d, large and small nutriment spheres. In

the center are seen two blood corpuscles (b.c.), lying in the

capillary lumen which is lined by the flat endothelium. (Zeiss,

oc. 2 and imm. syst. xx).

Fig. 2.—The efferent blood-vessel (ef.) in cross section and a piece of an

ovarian capillary in longitudinal section. c.l., capillary lumen,

lined by endothelium. .p., flat and small peritoneal cells

remaining at base of the nutriment cells. fol, egg-follicles.

og., proliferating oogonia. (Zeiss, oc. 2 and imm. syst. yz).

Fig. 3.—Cross-section of an ovarian capillary. Below and to the left, a

small oocyte over which the nutriment layer is degenerating.

Over the larger egg-follicles that layer has entirely disappeared.

l.c., capillary lumen. (Zeiss, oc. 2 and imm. syst. xy).

Fig. 4.—Same. Centrally, three blood-corpuscles in the capillary lumen.

A—H, oocytes in differet stages of growth. n and n’, two kinds

of nucleolus. +, comma-like body of unknown character, attached

to the inner surface of nuclear membrane. (Zeiss, oc. 4 and imm.

syst. Ir).

Fig. 5.—Cross section of a functional testicular capillary. Centrally, a

blood-corpuscle in the capillary lumen. Radially arranged sper-

matogonia-strings, each with an axis of a connective-tissue

bundle. a—b, stages in the spermatogenesis corresponding to

those denoted by the same letters in fig. 6. (Zeiss, oc. 6 and

obj. 1.5 mm. apochrom. imm.).

Fig. 6.—Stages in the spermatogenesis. 4—b, different phases of nuclear

change in the spermatocyte. c, one of the two daughter cells

formed by the first division of a spermatocyte ; the nuclear figure