Full text of "STUDIES FROM THE BIOLOGICAL LABORATORY"

See other formats

Google

This is a digital copy of a book that was preserved for general ions on library shelves before il was carefully scanned by Google as part of a project

to make the world's books discoverable online.

Il has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books

are our gateways to the past, representing a wealth of history, culture and knowledge that's often diflicult to discover.

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the

publisher to a library and finally to you.

Usage guidelines

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to
prevent abuse by commercial parlies, including placing technical restrictions on automated querying.
We also ask that you:

+ Make non-commercial use of the plus We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a b<x>k is in the public domain for users in the United States, that the work is also in the public domain for users in other

countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means il can be used in any manner
anywhere in the world. Copyright infringement liability can be quite severe.

About Google Book Search

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's hooks while helping authors ami publishers reach new audiences. You can search through I lie lull text of this book on I lie web
at|http : //books . qooqle . com/|

■ ■ ' c d
cc t e G

( 1 C (l
C( C (i

G<CC<

C C «

( 1

«.

,Vc« < * *

i c<

?<&, < 5 c

( CC

V ( '

■ c <

( «

c c*. j

c c <

( CC

,( '

« c . ..

« c<

( c(

(. i -

i c<

C 1

C ( ' C ' ' • ' ;

< i

i. t <

• >

( a i

(. CC I

< (C c

. i i

(■( < c
CC 1 C tec

1
C>C cJ c

t < C

C. CO

c c 1

ic C < *

•■ <■

, c« C'

, V C C CC

4t ' C

' c

c e t< «

•L ' c

(A 1

c t cc «

A- ' c

<.C '

1 «

:c <x< cc <

Xv l C

' «L

tc t t-C

^ v

CC ' c

■ CC

{ ' ••"

C CC C

C C < C v >. "^

( (

( CC (

C C< 1

< < < <

C CC c 't

C C( t CCC c-

C C< <. O- C '•

c " < c c ■

C ' < c ■ ■ c

>..«-«• 1 C

i i ■
<

(I '

*_ (1

-- — -~*

>■ c

1 ^

C<" ■ < < < "*.

i« < <

*■ ' r e .

I C

€ < (>

< • ' < " J >

C c

C ( C ((

C C

€ i 1 ( l

:c ( '• «

( (

C C I C <<

( ( i CC(

C C C "'

< < o

iCfC]

■

( <r • < «.

0 C'UO '

f i fCi

c 1 { C

i < ' ((C

( C (

<: (<m

ifC C c '

( ( C<?

< c

( (

C I

( c

C (

itw <

0 C

, c ( c

• « ' ' I

c Cf>< <~

«.' 1 I'

C ^r"( c

c «Lc: (

(£

( r c

t «.' <
c « c

t 1 S Q ( « C( (^

. < < ■ < ! (. C ■: < C<

1 |

C 1 '. <

> ! t

* (. ( c ' Cic

• «

( <.

€. 1

<. <t ^ •

c . <

c < 1

t < *

c< <«. l 1

<T<

re « CCC *-< i

«. ■< UK.

CC *(

■« CO

«fc '-<J-

CC «L

C C C ' ft

« if
«: ice

CC c«

C' c c< c

1 cc

c n c t «. c c

cc;

* V

(. , « C (

1 C < . : «: c (

«.

< CC

cC« '

C Cf

< ((

<ce ccccc c

c m

«« <x«r£

( < c

r ,, >

:< cc c « «

i c

t «r e ■« «

( ■ <

c cr ' tr <

c Ci

(

i fC

c $ t -,

c c <

( cc

'. 1

C 0

CTJf' t'»< <

< fC

c. <

' l(

■

< a
c a

S 1

< J

< <C f

( <<T(

C C(; ( i

(

( ( (

< 1 <

(.<■■<.«■

t a

IC<

CM 1 c fee

( (

c c • «r id

( 1 (

C < ' f'fl '*

i <i ■ ■ i

r (

c c c

U C i

l« i

c C <

(

CI i

c c *

8 c i

Ct i i

c c c

0 (

CC r

Cc ' I

C (c

• CC

C CC Cffl

c c

C Q

] CC

t U i

Cs,

('(( <c<

'. c

' ccc

< ( c

' C c

1 (Cc

CI V

1 ccc

C< (

( ^

'CCC

< < c

c c <

i . .

C IC I . ■

«<

CCC

c CC i

ia* • r? 7 / J. 'o 7

JOHNS HOPKINS UNIVERSITY,

BALTIMORE.

STUDIES

FROM THE

BIOLOGICAL LABORATORY,

EDITOR:

NEWELL MARTIN, M. A., D. So., M. D.

ASSOCIATE EDITOR:

W. K. BROOKS, Ph. D.

VOLUME II.

Published by N. MURRAY,

Johns Hopkins University.

July, 1888.

PreM of IflMtc medtnwsld,
Baltimore* Md.

STUDIES FROM THE BIOLOGICAL LABORATORY

OF THE

JOHNS HOPKINS UNIVEKSITY.

VOLUME II.

CONTENTS.

PAGE

I. A Contribution to the Study of Inflammation as illustrated by
induced Keratitis. By William Councilman, M. D. With
Plate IV. (Reprint from Journ. Physiol.) 1

II. Some further Observations on Heat-dyspnoea. By Christian

Sihler, M. D., Ph. D. (Reprint from Journ. Physiol.) . 13

m. The Influence of Quinine upon the Reflex-excitability of the

Spinal Cord. By Wm. T. Sedgwick, Ph. B. ... 2:5

IV. The Early Development of the Wolffian Body in Amblystoma
punctatum. By Samuel P. Clarke, Ph. D. With Plates I,

n and III 39

V. Notes on the Formation of Dentine and Osseous Tissue. By

Christian Sihler, M. D., Ph. D. With Plate V. . . . 45
VI. The First Zoea of Porcellana. By W. K. Brooks, Ph. D. With

Plates VI and Vn 58

VII. The Study of Human Anatomy historically and legally consid-
ered. By Edward Mussey Hartwell, M. A., M. D. G5
VIII. Alternation of Periods of Rest with Periods of Activity in the
Segmenting Eggs of Vertebrates. By W. K. Brooks, Ph. D.

With Plate Vm .... 117

A New Method of Studying the Mammalian Heart. By H.
Newell Martin, M. A., D. Sc.. M. D. With Plato IX. . Ill)

X. A Note on the Processes concerned in the Secretion of the Pep-
sin-forming Glands of the Frog. By Henry Sewull, Ph. D. 131
XI. List of Medusae found at Beaufort, N. C. , during the summers

of 1880 and 1881. By W. K. Brooks, Ph. D. 135

XII. On the Origin of the so-called " Tt«t-Cells" in the Ascidian

Ovum. By J. Playfair McMurrich, B. A. With Plate X. . 147

A CONTRIBUTION TO THE STUDY OF INFLAMMATION
AS ILLUSTRATED BY INDUCED KERATITIS. By
WILLIAM COUNCILMAN, M.D. (Plate IV.)

(From the Biological Laboratory of the Johns Hopkins University.)

It would be useless to attempt to give here any but the briefest sketch
of the views which have been held concerning the origin of pus and
the nature of the cellular changes occurring in inflammation since the
time of the establishment of the cell theory. A mere recapitulation
of the articles written on this subject in the decade of '60 — '70 would
fill pages. I will, however, briefly glance at some of the more important
ideas which hare been or are held with reference to it.

Rokitansky was one of the first to appear in this field of litera-
ture. He, in accordance with the cell theory of Schwann (that is,
the free cell formation doctrine), assumed that the pus corpuscles were
formed in the exudation which played the part of the blastema. These
ideas prevailed generally until 1855, when Virchow was led, from his
knowledge of the conuective tissue corpuscle, to dispute the free cell
formation, and to apply the law " Omnis cellula e eelluld " to patholo-
gical new formations. Virchow held that the pus cell was the direct
derivative of the connective tissue corpuscle, because, wherever he found
pus he also found connective tissue in some of its forms ; and since he
was obliged from his views to have some cell as the parent of the pus
cell, he took the connective tissue corpuscle.

Strieker appears as the latest, and certainly the ablest, defender of
these views, though they have undergone essential modifications at his
hands. He says that the cells of a tissue, when inflamed, return to their
former undifferentiated embryonic condition, become amoeboid, and
possess the power of dividing indefinitely. This property holds good for
all tissues equally ; no matter whether muscle, gland, or ganglion cell,
they all can undergo this change and become converted into pus. He
holds also that, when the cells return to the embryonic condition, they
again become capable of differentiation, and that blood vessels, and even

•1

2 W. COUNCILMAN.

blood corpuscles, are formed in an inflamed part in the same manner as
in the embryo. Against these views we have what is known as the
" wandering cell " theory, which assumes that the pus cells are white
blood corpuscles which have escaped from the vessels. Waller had, as
early as 18481, observed the passage of the white corpuscles through the
walls of the vessels. At that time his observations attracted but little
attention, and were generally distrusted. Cohnheim in 18682 firmly
established the fact that the white corpuscles did pass through the vas-
cular walls, and, as the result of his study of the inflammatory processes
in the frog's cornea, tongue, and mesentery, asserted that the pus cells
are white blood corpuscles.

From his study of keratitis, principally induced by cauterizing the
centre of the frog's cornea with silver nitrate, he found that, however
great the number of pus cells in the inflamed tissue might be, the fixed
corneal corpuscles with their processes were unchanged ; that the nuclei
of the corneal corpuscles did not increase ; that the clouding of the
cornea always began at the periphery and from there advanced to the
centre ; that after the injection of pigment granules into the blood some
of the pus cells in the cornea were found with similar granules in their
bodies. From these four circumstances, supported by the direct know-
ledge that the white corpuscles in inflammation did escape through the
vessels in large numbers, he concluded that the pus corpuscles were not
derived from the fixed cells of the cornea, but had wandered in from
without. Strieker, as the result of observations made on the frog's
cornea and on the cornea of the cat, asserts that the three first argu-
ments are based upon imperfect observations, and that the conclusion
formed from the fourth is illegitimate. According to Strieker, the
fixed corpuscles do undergo change, their nuclei increase, and the
clouding always begins where the injury was inflicted. With regard to
the presence of pigment-bearing pus cells in the inflamed cornea after
the previous injection of pigment into the blood, he thinks the granules
could have passed through the walls of the vessels as easily as the blood
corpuscles and have been carried by the lymph streams into the cornea.
There they could easily have been taken up by pus cells which were
already produced by multiplication from the corneal corpuscles.

Here I may remark that the passage of solid dead particles through
the walls of a blood vessel without being carried through by the white blood

4 Phil. Jfay., Vol.xxix.

1 Virohow'i Arch., Bd. xl.

INFLAMMATORY CHANGES IN CORNEA. 3

corpuscles, easily as Strieker thinks it could happen, has up to this
time been seen and described by no one. That Cohnheim's descrip-
tion does not hold good for all cases of induced keratitis, even on the
frog's cornea, is certain ; but the differences can be easily reconciled.
Strieker bases nearly all his views of inflammation and of inflammatory
new formations on his study of keratitis. I shall, I think, be able to show
in this paper that these views, certainly as far as keratitis is concerned,
are erroneous, and may possibly be due, even in his case, to imperfect
observations.

I can only excuse my temerity in entering upon a field of research in
which so many and distinguished investigators have laboured, by the
fact that when endeavouring to satisfy myself of the correctness of
Strieker's views on the subject, I obtained, after nearly a year's steady
work, results which lead to conclusions utterly at variance with his, but
which I think go far towards clearing up some of those points in the
pathology of keratitis over which there has been most contention.

The corneas of the frog and of the cat have been principally used in
my investigations ; the latter animal being chosen for studying the pro-
cesses in the mammal from the advantages its cornea offers over many
others for investigation, especially in the readiness with which it can
be split into layers.

The structure of the normal cornea has been thoroughly investigated
by various observers in recent years. We know that its proper tissue
is lamellated, and consists of flattened branched cells embedded in inter-
communicating centres (the serous canaliculi) hollowed out in an inter-
cellular fibrillated ground substance, which makes up the larger portion
of the corneal mass ; that the tissue is well supplied with nerves arranged
in plexuses which become finer towards the conjunctival surface ; that
with hematoxylin or gold the cells stain and are seen to communicate
by fheir branches ; and that with silver nitrate the ground substance
is tinted, while the cells and cell spaces are left unstained. Hema-
toxylin also stains the nerves, while with silver preparations the lymph
channels in which the larger ones run are seen as colourless lines.

We also find, even in the normal cornea, another set of cells, which
cannot be considered a part of its fixed histological elements. Their
numbers are variable ; in some corneas very abundant, in others few : in
animals of the same species sometimes they are found in greater numbers
at one portion of the tissue, sometimes at another. In -fresh preparations
they can be seen to pass by active amoeboid movements from one place
to another, and they never, so far as we can see, stand in any fixed his-

•1—2

4 H". COUNCILMAN.

tological relation to the other elements of the tissue; these are the " wan-
dering cells." Their position is not at all constant ; sometimes we find
them lying in the cell space along with the branched corpuscles, sometimes
in the narrow communication between two spaces, sometimes as long
drawn out rods in the tissue between the fibres (b, Fig. 1, PL IV.), some-
times in the nerve lymph channels, and in- one preparation I have been so
fortunate as to get one seemingly in the act of passing from the nerve
channer into a cell space communicating with this, half of its body
lying in the channel and half in the space. They can be clearly dis-
tinguished from the branched corpuscles both in the fresh condition and
when stained ; they are much smaller, and with the usual reagents they
stain more brilliantly than the others. In fresh preparations in aqueous
humour they are easily recognized by their amoeboid movements, their
greater index of refraction, and their granular contents.

So much for the normal cornea. We will now take up the patho-
logical changes which occur after an acute keratitis has been induced,
commencing with those seen in the frog's cornea.

I have employed various means for exciting inflammation here The
passing of a thread through the centre of the cornea and bringing it out
through the sclerotic, the application of various caustics, such as croton oil,
silver nitrate, caustic potassa, and the hot iron (actual cautery). With
few exceptions they produce results relative to the severity of the sti-
mulus used. Agents such as the hot iron, which at once kill the tissues
with which they come in contact, will, of course, produce less inflammation
in surrounding parts than those like the thread, whose action is more
or less gradual. A method which I have used on the frog's cornea
with excellent results has been to pass a thread through the membrana
nictitans and then make several pricks in the cornea with a needle. The
inflammation produced by this method will be discussed Separately, since
results are in this way obtained which at first seem perplexing.

As one of the most typical, I will take a cornea which has been
inflamed by touching it at the centre with a crystal of silver nitrate.

This may be examined after various intervals of time have elapsed,
both in the fresh condition and after staining. About twenty hours
from the application of the caustic the most important changes can be
seen. To examine fresh, it is necessary to puncture the sound eye and
collect the aqueous humour on a slide ; the inflamed cornea is then care-
fully excised and spread out in this, with the posterior surface uppermost

INFLAMMATORY CHANGES IN CORNEA. 5

To avoid folds in the tissue it is better to make three or four incisions
at the edge, extending for some distance towards the centre, before
putting on the cover slip. The powers I have found most satisfactory
to use have been the No. 2 immersion of Zeiss (yf) and the E of his
dry system (£).

The first thing noticed here is that the large branched cells are
visible ; in the normal they cannot be made out at all directly after the
cornea is cut out, and only appear after an interval of half an hour to
an hour. They are no more granular than in the uninfiamed, and pre-
sent no changes from the normal an hour after the excision of the latter.
Why they become at once visible I do not know ; it may be due to
some change in the refraction of the ground substance caused by the
greater amount of fluid now in the tissue, or to some change having
taken place in the corpuscles and only revealing itself in this way, or
to both.

The wandering cells are present in vast quantities, exhibiting the
most active -and varied movements ; while in the normal cornea, as
before remarked, we only occasionally see them. Spmetimes one
may be seen to send out a long process, at the end of which a knob
presently appears, which, growing larger and larger, finally becomes
the main body of the cell : as though in this way it had passed from
one space to another through a narrow communication. Sometimes we
see them as more or less irregular bodies, undergoing changes of form
and not of position ; again as the long, staff-like bodies spoken of in
the normal cornea. They are present in the greatest numbers at the
edge, becoming fewer as we proceed to the centre. Since in the fresh
specimens our observations must be made on the whole thickness of
the cornea, all these changes become much more clear and can much
better be studied after it is stained and split up.

For staining I always use the double staining in silver with haema-
toxylin or carmine, the former being much preferable for the frog.
The cornea is exposed by pushing the eye upward from the roof of the
mouth, and rubbed smartly with the solid crystal of silver. At the
expiration of ten minutes it is cut out, and exposed in glycerine to the
action of diffused daylight ; when it becomes of a light brown colour it
is split up and stained in one of the two reagents mentioned. With
care the cornea of the frog can easily be split into eight or nine layers.
I vastly prefer this method of staining to the gold chloride method,
which has hitherto been almost exclusively used in these investigations.
It has the great advantage of being always certain in its results ; while

6 W. COUNCILMAN.

gold, although sometimes giving us beautiful preparations, is the most
uncertain of reagents, and its success depends for the most part on
unknown circumstances. Another great advantage is that we have
both the negative and the positive picture at once ; the cell space shown
with the cell within, and the relation of the one to the other always is
kept in view. The preparations are mounted in slightly acidulated
glycerine.

In preparations of the twenty-hour cornea examined after this
treatment we can easily make out three distinct parts : — A central one,
on which the caustic was applied, and which is now represented by a
black scar, in which the cell spaces are imperfectly seen. Around this
is a zone of variable width, in which absolutely no change from the
normal can be made out ; here we see the sharply-defined cell space,
with the nucleus, or, in deeper staining, the body of the cell within.
The width of this zone is dependent on the extent of the injury, the
length of time which has elapsed since its infliction, and on the general
irritability of the tissues of the animal used. Without doubt, from the
same amount of irritation, the extent of the pathological changes in
some animals of the same species is different from that seen in others.

This zone passes, separated by no well defined line, into the outer-
most one. In this, along with the corneal corpuscles, other elements can
be seen in numbers far in excess of the branched cells and always in the
greatest quantity at the outer edge. These other elements stain in all
respects similarly to, and are always of the same size as, the wandering cells
previously described in the normal cornea. They can always be distin-
guished from the branched cell, even when lying in the same cell space
with it. In one place we see the nerve channel filled with them, in
another we see them lying in the tissue between the fibres, and elongated
until they have the appearance of rods. Again we see them in the cell
spaces or in the narrow interspace between two cells ; their form always
influenced by the dimensions of the cavity in which they lie. Often
where they are most numerous in the tissue the branched corpuscles
cannot be made out at all. It may be that these are simply concealed
by the vast numbers of the others, or it is possible that the fixed cor-
puscles have then been absorbed or destroyed by the young and
vigorous strangers.

In no case do we see in the corneal corpuscles proper, any indications
which would lead us to suppose that multiplication had occurred or was
taking place. They stain with reagents as did the normal, and the
nucleus always has the same shape as this, except in instances where it

INFLAMMATORY CHANGES IN CORNEA. 7

may be indented by pressure from a wandering cell lying in tbe same
space, as represented at a, in Fig. 1, taken from the normal cornea. If
the cornea be examined at an earlier period, say twelve hours after the
injury, these wandering cells will be confined to a small area at the outer
edge ; if later than twenty hours, forty, for example, they will be found
to fill almost the entire cornea, completely obliterating the unchanged
zone in some cases.

If we examine the surrounding portions of the sclerotic and conjunc-
tiva we find the blood vessels full of cells just like these, and the whole
tissue there also infiltrated with them.

A still further proof that these wandering cells enter the cornea from
without is furnished by the result of the injection of finely divided
colouring matter into the blood, according to the method of Cohnheim,
whose results in this respect I can completely confirm.

If the cornea is cauterized shortly after the injection of cinnabar into
a lymph sac or the anterior abdominal vein and examined after the usual
time, we find among the wandering cells a great many in which pigment
granules are plainly visible, though they differ in no other respect from
the others. Sometimes a few granules can be seen in the tissue not
inclosed in the cells. These may be accounted for by supposing that
they were here dropped by the wandering cell which brought them from
the vessel. Strieker himself says that he and N orris have seen one
wandering cell transfer to another cell of the same nature some of the
vermillion granules contained in its substance. Since the vermillion
granule can in nowise contribute to the nutrition of the cell, and forms
rather a heavy load to be carried round, we can see excellent reasons
why the cell^vould be willing to throw it away. The number of cells
containing these granules is far too large to suppose they could have
gotten them in any other way than by taking them up in the blood
vessels.

The inflammation produced by methods involving a laceration of the
corneal tissue gives some results differing from those last described. Here,
as in the^last case, we see the peripheral portion of the tissue infiltrated
with wandering cells ; but we see them also elsewhere. Around the spot
where the injury was inflicted we see cells of the same appearance and
offering the same variety of form and position as those at the outside,
and here narrowing the zone, which in the cauterized corneas we have
described as free from them, very materially. How came these cells
here ? From the outer edge they could not come, for we have lying
between this and the centre a zone which, in the earlier stages of the

8 W. COUNCILMAN.

process certainly, is free from them. If now we combine both methods
of producing the inflammation, and having cauterized two corneas in the
centre, we make a prick at the outer edge of the cauterized spot of one,
and examine the two after the usual interval of time, we shall find plenty
of wandering cells around the laceration in the cornea whose tissue. was
punctured, and none at the same spot in the other. Only one conclusion
is possible, that they have entered the cornea where its substance was
broken. This is easily comprehensible, since a keratitis can scarcely be
produced in this way without involving at the same time an extended
conjunctivitis, and as a consequence of this having quantities of white
blood corpuscles in the conjunctival secretion. From this source they
could easily enter the tissue where broken.

The results obtained after passing the ligature through the membrana
nictitans point clearly to this. Here a violent conjunctivitis is necessarily
set up ; many blood vessels in the membrane are ruptured and plenty of
white corpuscles poured out. As a consequence, in these preparations
we have a very large number of wandering cells at the point where the
prick was made; in some cases they are so plentiful that everything else
is obscured. After the injection of pigment granules these wandering
cells also contain them. No change is seen in the branched
corpuscle at either place.

Proceeding now to the cat's cornea, we meet here, even in the normal
state, some difference from that of the frog. The corpuscles (Fig. 2), are
smaller, are more numerous, and the cell spaces communicate by larger
passages than in the frog. The brightly-staining wandering cells in the
normal cornea are fewer in number than in the frog's confea, and mostly
found in the cell spaces. Their special characteristics will be described
when we come to speak of the pathological changes.

As a means of exciting inflammation I have, following Strieker,
used the solid stick of caustic potassa, and found it vastly superior to any
other agent. A young cat is preferable to an old one, from the fact that
the cornea of the former is much more easily split into its lamellae than
that of the latter. The animal is first etherized and the cornea touched
with the caustic ; particular care must be exercised in doing this, as the
potassa melts so rapidly on contact with the moist surface that there is
great danger of its involving too great an extent of tissue. To avoid
this the caustic stick must be pointed (which is easily effected by
holding it in a stream of water), and the cornea previously carefully
dried with filter paper. By varying the period of contact, an eschar

INFLAMMATORY CHANGES IN CORNEA. 9

extending only a few lamellae in depth or one involving the whole
thickness of the tissue can be produced. The animal is then left in quiet
and the cornea cut out and examined after periods of from 14 to 60
hours. The silver staining, before removal, and the after staining, with
carmine, are used. If we examine such a cornea, say 40 hours after
cauterization, and as yet unstained by carmine, the changes found can
be divided into two heads. Of these the first will comprise the changes
around the outer corneal edge, and the second those in the immediate
neighbourhood of the eschar. In the first we find the cell spaces some-
what larger and the communications between them wider than in the
normal cornea. Scattered about through the tissue we find the strongly
refracting rod-like cells, appearing very similar to those we have seen
in the frog. If the silver staining has been very deep we find the silver
precipitated in the substance of the corneal corpuscle as well as in
the ground substance, leaving a clear unstained nucleus in every
space.

In the immediate neighbourhood of the eschar the change is more
pronounced, and different from anything we have hitherto seen. These
changes are all the more important to us, since it is here that Strieker
says the corneal corpuscles are undergoing the most rapid proliferation.
In the silver preparations we see, lying in the coloured ground, groups
of small white spaces with dark brown lines separating them from one
another (Fig. 4) ; these groups correspond in shape to enlarged cell
spaces. Strieker seems to have confined his observations to this spot,
and explains the picture by supposing the corneal corpuscle has here
broken up into a number of smaller cells, and that the brown lines mark
off the new cell limits.

Let us now see what the carmine staining shows in the two parts.
In the outer ring we have (Fig. 3) in each of the slightly enlarged cell
spaces the large oval nucleus of the branch cell totally unchanged, and
staining in all respects like the normal. In rare cases we find (as is
also the case with the normal) two of these nuclei in a space. In
addition to these there are other cells, which, from their characteristic
appearance, merit a mqre detailed description. These have a differ-
ence in shape according to whether they are found in the cell spaces
and nerve channels, or in the proper corneal substance, there lying
between the fibres. In the former they are round, with a brightly-
stained granular nucleus of the shape of a horseshoe, and correspond
to the wandering cells in the normal cornea. Under high powers
(800 — l'OOOx) the apparently single nucleus is usually found to be

10 W. COUNCILMAN.

composed of three or four small bodies lying in juxtaposition, the
mass being always arranged in the shape of a horseshoe.

When lying in the tissue between the fibres they are elongated,
and then appear as jointed rpds, each joint having the highly-stained
granular nucleus. At first sight these rod-like bodies would seem to
be entirely different from the round cells in the spaces ; but, on closer
inspection, at different places every variation can here be seen, from the
slightly elongated cell with a horseshoe nucleus to the long rod-like cell.
If we now stain some of the blood of the cat, we find that the white
blood cells have a nucleus of this horseshoe shape and stain in all respects
like these wandering cells.

Proceeding now from the corneal edge towards the eschar, we come
to a region where the corneal corpuscles are wanting, passing on the way
through a district where they have taken on changes which will occupy
our attention presently. Beyond this line, which can be seen by even a
simple lens, the corneal corpuscles are dead — have been destroyed by
the caustic. The cell spaces can be seen, most of them much shrunken,
but no nucleus in them, or anything which would afford us proof of the
presence of a corneal corpuscle. Lying in these cell spaces, but still
more in the tissue between them, are seen multitudes of cells before
described, at the scleral edge. These cells become more numerous as we
proceed, until we reach a territory where the cell spaces are filled with
them (Fig. 5). The spaces here are enlarged, and the communications
between neighbouring ones are wider ; spaces and communications are
all full, and no one comparing these cells with those at the outer (i.e.,
the scleral) edge can doubt for a moment that they are similar.

Beyond this line of general infiltration the tissue is totally destroyed.
By this I mean that not only its living protoplasm is killed, but its
physical properties are also altered. Nothing of the cell spaces can be
seen, and apparently the wandering cells can make their way no further.
At the point of general infiltration the tissue sloughs.

In corneas examined 10 to 14 hours after cauterization this district of
general infiltration is wanting ; no wandering cells are seen there. In
the other district, however, that around the outer corneal edge, the wan-
dering cells are numerous ; sometimes so many will be seen that the
feintly-stained nucleus of the branched cell is entirely obscured, the wan-
dering cells filling up the space. From this edge they become fewer and
fewer as we proceed towards the centre. The line of corneal corpuscles
marking off the portion of the cornea in which the corneal corpuscles
were destroyed by the caustic from that portion of the cornea where the

INF LAMM A TOR Y CHA NOES IN CORNEA . 1 1

corpuscles were uninjured, is not now so well seen, as these corpuscles
have as yet taken on no change by which we can distinguish them. We
readily see, however, even here, where the living tissue ends. Now it is
beyond this line that we get from the silver preparations of a later
period of inflammation the appearance as though the corneal corpuscles
had proliferated. Here were the colourless areas subdivided by brown
lines. From this place Strieker's drawing was made, and here he,
judging merely from silver staining of corneas, taken always at a fixed
time after the cauterization, supposed the proliferation to have been most
rapid. Further examination by better methods, and at different periods,
after cauterization, shows us that there is here nothing to proliferate.
The tissue is as bare of living corneal corpuscles as a sheet of paper. In
48-hour preparations the line of demarcation is more evident and the
tissue beyond more infiltrated with cells than in the 40-hour preparations.
In all the portion first described, that along the edge of the sclera, no
change can be seen in the nuclei of the branched cells. In corneas
examined 60 to 80 hours after cauterization, that portion of the tissue
surrounded by the infiltration is converted into a slough, which easily
comes away, and the peripheral portions, the district around the sclera,
still contain wandering cells.

In the corneal corpuscles which form the line outside the zone of
infiltration, and which indicate the separation of the dead from the
living proper corneal tissue, we find changes as early as twenty hours
after cauterisation. These changes are at this period only shown by a
brighter staining ; the whole substance of the cell here stains and else-
where only the nucleus. At a later period (30 to 40 hours) the nuclei
can be seen in different stages of division, and at the same time long
processes are sent out from the cells into the dead tissue. These
processes become longer (Fig. 6), nuclei pass from the old cell up into
them, and thus they form in the dead tissue new corneal corpuscles,
but never pus. These processes and new cells stain in all respects like
parent cell from which they originated, and the nuclei have the same
shape as in the old cells, though they stain more brightly, and are more
granular.

The appearance of a segment of the cornea taken three or four days
after injury, in which the branched corneal corpuscles are undergoing
this proliferation, is most beautiful. The nuclei of the new corpuscles
divide rapidly, and in some as many as four can be seen. Even if the
whole cornea is destroyed with the exception of a small strip along the
outer edge, the corpuscles limiting this take on this renewed activity'.

12 W. COUNCILMAN. .

The difference between these two processes — the suppurative, on
the one hand, in which the wandering (ells are the agents,
and the regenerative, on the other, by which new corneal
corpuscles are produced from corneal corpuscles — is so clear
that no one seeing them side by side could mistake them. In no tissue
in the body can the processes of repair be so clearly studied as in the
cornea ; and in no other tissue can the wandering cell theory as to
the origin of pus corpuscles be so clearly proven to be correct.

DESCRIPTION OF THE FIGURES. PL. IV.

Fig. 1. — Normal cornea of frog, stained with hematoxylin. Two of the branched
corneal corpuscles are shown with a wandering cell, a, lying in the cell space with one of
them, b b represent two of the wandering cells in the subitanoe of the cornea ; these have
taken the elongated form.

Fig. 2. — Normal cornea of a oat, stained with silver and carmine. The ground sub-
stance is stained brown with the silver, leaving the oell spaces unstained. In these are
seen the nuclei of the branched cells stained with carmine, b b, two wandering cells in the
cell spaces.

Fig. 3. — Scleral edge of cat's cornea fourteen hours after central inflammation. The
wandering cells, b b, are increased in number, and the communications between the spaces
are larger than in No. 2. Silver and carmine.

Fig. 4. — Area of general infiltration forty hours after central inflammation. The oell
spaces are greatly enlarged, and broken up into small areas by the brown silver lines.
The ground substance is reduced in amount, in some places represented only as small
islands.

Fig. 5. — Innermost limit of area of general infiltration. Here, as in No. 4, the cell
spaces are greatly enlarged, and divided into small areas, in each of which the brightly-
stained horseshoe nucleus is seen. From this point to the centre no cellular elements are
found. Silver and carmine.

Fig. 6. — Two corneal corpuscles, which have taken on regenerative changes. The
nuclei have increased in number, and long processes which are much branched have
grown out from the parent cell.

directly on the centres in the medulla, though, if bo, producing leas
effect than the peripheral stimuli.

In the following short communication, I propose to give further
support to the first statement, and discuss the second as well as another
which I touched in the published essay, namely, the action on the
medulla of higher temperatures than those used in my former investi-
gations.

I feel the more inclined to add further proofs to support the con-
clusions which I reached, as views contradictory to them and based on
Goldstein's experiments, which I have shown to be imperfect and
^conclusive, are taught in several text books of Physiology, and
apa gaining ground in the medical profession.

Foster says, on page 377, 3rd Edition: "If the blood in the
carotid artery in an animal be warmed above the normal, dyspnoea
18 at once produced. The over- warm blood hurries on the activity

12 W. COUNCILMAN. .

DESCRIPTION OF THE FIGURES. PL. IV.

Fig. 1. — Normal cornea of frog, stained with haematoxylin. Two of the branched
corneal corpuscles are shown with a wandering cell, a, lying in the cell space with one of
them, b b represent two of the wandering cells in the substance of the cornea ; these have
taken the elongated form.

Fig. 2. — Normal cornea of a oat, stained with silver and carmine. The ground sab-
stance is stained brown with the silver, leaving the cell spaces unstained. In these are
seen the nuclei of the branched cells stained with carmine, b 6, two wandering cells in the
cell spaces.

"Fig. 3. — Scleral edge of cat's cornea fourteen hours after central inflammation. The
wandering cells, b b, are increased in number, and the communications between the spaces
are larger than in No. 2. Silver and carmine.

Fig. 4. — Area of general infiltration forty hours after central inflammation. The cell
spaces are greatly enlarged, and broken up into small areas by the brown silver lines.
The ground substanoe is reduced in amount, in some places represented only as small
islands.

Fig. 6. — Two corneal corpuscles, which have taken on regenerative changes. The
nuclei have increased in number, and long processes which are much branched have
grown out from the parent cell.

directly on the centres in the medulla, though, if so, producing less
effect than the peripheral stimuli.

In the following short communication, I propose to give further
rapport to the first statement, and discuss the second as well as another
*hich I touched in the published essay, namely, the action on the
ntedalla of higher temperatures than those used in my former investi-
gations.

I feel the more inclined to add further proofs to support the con-
clusions which I reached, as views contradictory to them and based on
Goldstein's experiments, which I have shown to be imperfect and
inconclusive, are taught in several text books of Physiology, and
we gaining ground in the medical profession.

Foster says, on page 377, 3rd Edition : " If the blood in the
carotid artery in an animal be warmed above the normal, dyspnoea
*• at once produced. The over- warm blood hurries on the activity

14 C. SIHLER.

of the nerve cells of the respiratory centre, so that the normal supply
is insufficient for their needs. The condition of the blood then affects
respiration by acting directly on the respiratory centre itself."

Fick says, on page 266 of his Physiology, 2nd Edition : " If au
animal is artificially heated several degrees above its normal tempera-
ture, the respirations become deeper and very much more frequent, even
if the quality of the blood is in nowise changed ; yes, even when by ener-
getic artificial inflations arterialisation of the blood is insured ; indeed,
it is quite impossible in an animal thus super-heated to produce the state
of apnoea. That reflex influences do not come into play here — e.g., from
the heated skin — can easily be proven by the following experiment.
By application of the proper apparatus one can succeed in heating
nothing but the blood flowing in the carotid arteries. As soon as that
takes place the frequency of the respirations rises just in the same way
as if the whole animal had been heated. From that one must conclude,
that it is the increase of the temperature in the respiratory centre itself
which increases the irritability, and at the same time diminishes the
resistance, so that the exciting agent produces in the same unit of time
deeper and more frequent respirations."

In an article on Progressive Pernicious Anaemia, by Herbert
Jones, published in the Practitioner, February, 1880, we read this :
1 Heat is also a stimulant to the respiratory centre in the medulla
oblongata, by which the movements of respiration are regulated, and
as#Fick and Goldstein have shown, when warm blood is supplied
to this centre the respiratory movements become quicker and deeper
until marked dyspnoea takes place, although the blood which is circu-
lating in the rest of the body still retains its normal temperature."

I let Exp. 1, see Table I., precede the remarks which I wish next to
make. The observation it records, like the rest of my experiments, was
carried out on a dog. The temperatures, during my observation, were
taken in the rectum or vagina.

I had various reasons for undertaking this experiment. In the former
investigation I had found that one animal might breathe 200 to 300 times a
minute without its temperature going up, and vice versd, the temperature
of another animal might go up several degrees while the respirations
went up from 26 to 62 per minute only, the cord in the latter being
divided in the lower cervical region.

HEAT-DYSPNCEA.

Table I.

0
O

4
5
6
7
8
9
10
11

lime.

10.05
40

48
53
58
11.05
10
17
20
29
35

12.80

1.00

1.01

3.15

4.27

5.10

6.21

•a

24
41

44
42
44
44
46
48
48
49

85
87
38
87
38

50
58
49

60
60

38*9

88-9

88'9

88*9

891

39-1

39-1
88-9
389

891

89-2

898

89-8

376
87-8
38

38*5
38*9
89*6

is
Si

P4 M

86
120
132
200
204
268
280
280

216
87
30

152
240

228
810

280

18
18
20

18
22
24

Wednesday.

Tracheotomized.

Head only in apparatus, breathing warm air
through tube.

Panting.

Artificial respiration for 2 minutes with cool
air, apnoea for J minute ; shallow respira-
tion for 1 minute ; out of apparatus.

Artif. resp. for 2 min. ; no apnoea.

Artificial respiration ; apnoea of 1 J minute.
Placed in apparatus ; head free.

Artif. respiration for 2 minutes ; no apnoea.

Panting.

Artif. respiration for 2 minutes ; no apnoea.

Taken out of apparatus.

Cord cut.

Placed in apparatus ; head and arms free.

Artif. respiration ; apnoea of over 1 minute.

C. SIHLER.

86
87

88
89

40
41
42
48
44
45
46
47
48

Time.

6.28
86
46
58

9.20
25
10.80
85
88
48
47
50
55

60
60
60

60
68

50
49
49

89-5
39-7
40
40-1

87
87
40*8

41-4

& 2

24
21

19
26

100

425 156
42-5

Wednesday.
! Artificial respiration ; apnoea of } minnte.

Artif. reap, for 2 min. ; apnoea over 1 min.

Thursday.

Artif. resp. of 2 min. ; apnoea of 1 J min.

Placed in apparatus.

Artificial respiration ; apnoea of £ minnte.

Artificial respiration ; apnoea of £ minnte.
Begins to pant.

Artificial respiration ; no apnoea.

These facts being brought oat on different animals, one object
here was to try one and the same animal. That is, to take account
of increased temperature, if there was any, and increased number of
respirations before the cord was cut and after the cord was cut in
the same dog.. It will be seen that the present results agree with
the former conclusions. The same dog is made to breathe 240 times
a minute (Obs. 21) while haying a temperature of 39 (38*9 had been
the temperature of the dog when the experiment began). When
we look, however, for the respiratory rate at the temperature of 39,
after the dog's cord had been divided, we find it (Obs. 34) 22. And
if the objection were to be made that the dog was unable to breathe
rapidly on account of the section of the cord, by looking towards the

HEAT-DYM>AT(EA. 17

end of the table it will be found that this is not the reason, for the dog
can make as many as 156 respirations in a minute.

We see, then, that in the same dog, when exposed to warm air acting
on a large surface, connected by afferent nerve paths with the medulla,
the respirations may go up enormously without the animal's temperature
rising ; and, on the other hand, the respirations may go up less than
25 per cent., while the temperature increases over one degree Celsius ;
in this latter case the greater part of the skin being thrown out of
nervous connection with the medulla by previous section of the cord.

It will further be observed, by glancing over the table, that artificial
respiration was carried on several times. Some of these produced
apnoea, others did not. At 12.44 (Obs. 17), while the animal was at
38*9°, and not in the warming apparatus, apnoea of 1£ min. was pro-
duced. At 1.15 (Obs. 22), while the animal was at 39°, i.e., only 0*1°
higher than before (practically not higher at all), the same amount of
inflation was not successful. That the 0'1° of temperature was not the
cause for this condition is shown further on. When the cord had been cut
apnoea was successfully produced, although the temperature of the animal
had* risen not 0-1°, but 1*1°. Again, when finally (Obs. 47) the tempera-
ture had reached 42*5, and the respirations 156, the efforts at producing
apnoea were again fruitless. It is clear from this that it is not the
temperature of the blood per se which makes apnoea impossible. We see
apnoea may be possible both at normal and at elevated temperatures ; it
may also be impossible both at normal and at elevated temperatures; the
reason of the difference being that the dog cannot be made apnoeic if he
pants vigorously. Of course there is a limit when artificial respiration will
at times be successful and at times not : just when the respiration begins to
grow rapid and take on the character of panting, as is shown in Table II.,
when the dog had the head only in the apparatus. Here, then, we have
another support for our conclusions. In the last paper it was shown
that it was not the heat acting on the centres which produces this con-
dition of the animal, in which it cannot be made apnoeic. The present
observations show the other side of the same fact, and make it evident
that peripheral influences, due to exposure of the skin only, may be so
strong that they do not allow the centre to come to rest, although there
is no venosity of the blood to act as a stimulus, nor has the animal's
temperature risen more than a degree.

In the third place, it will be seen (Obs. 46 — 48, Table I.), that the
dog did commence to pant — with the cord cut — after he had reached a

temperature of 42.

C. SIHLER.

Let me recall now one of the conclusions of my previously-published
paper : " The increased respirations .... are due to two causes, skin
stimulation and warmed blood." A somewhat closer consideration
makes it evident that the experiments there given were not sufficient to
show that the warmed blood has any direct central effect : for although
by section of the cord in the lower cervical region a large part of the
skin was thrown out, yet the fore limbs, neck, and sensitive head,
mouth/ and tongue remained in connection with the medulla ; and
although in the experiment the direct action of the heated air from
without was prevented by keeping the animal's head, &c, out of the
warm chest, yet this did not preclude the heating of the nerves of the
skin of those parts from within by means of the blood which had been
heated in the other parts of the body flowing into them.

To show how sensitive the mucous membrane of the mouth and
the tongue is, I add Exp. 2, Table II.

Table II.

December 3rd,

1879.

No. of
observation.

Time.

•

7.53

Head and fore-feet placed

8.01

88*9

apparatus.

89-1

39'1

152

89-1

Nose free.

Nose back in oven.

39-1

160

89-1

Dog pants.

In this experiment it was the aim to have the surrounding air which
the animal took into its mouth not very hot, not warmer than the blood
was when the dog began to pant in the experiment above referred to.
The experiment shows that exposure of a small part of the body, mouth,
neck, and fore limbs, to this not very high temperature is sufficient to

HEAT-DYSPNCEA.

produce quickened breathing and even panting, although the animal's
temperature is not raised. Human experience agrees with this ; if in
the effort of getting into perspiration by means of a hot-air bath one
keeps the head under the sheet and thus breathes air of about the body
temperature one finds the respirations similarly increase in frequency.

In the former paper it was shown, that the temperatures there
employed (41*3) did not produce the panting when the cord had been
cut, and it was left for further investigation whether higher blood tem-
peratures might produce such an effect by action on the centres directly.
The setting in of panting in Exp. 1 when the dog had reached the
temperature of over 42 might be adduced to support the view, that the
heat in conditions like the above acts centrally, the cord having been
cut. But the foregoing remarks show that such a conclusion would not
be justified, as the peripheral influences from mouth and head are not
excluded ; nor were those from the lung nerves. I cannot see how to
throw #ut these peripheral influences altogether, and the question,
possibly, must remain an open one, although there cannot be adduced
any fact showing a direct action of heat on the centres.

A third experiment, see Table III., however, was devised in which
peripheral influences were eliminated as much as possible.

Table III.

January 9, 1880.

2
3

5
6

9
10

Time.

.3 J

«H O

9. fi

10.80

36*7

11.40

12.48

12.52

1.35

2.53

3.03

| 3.10

3.24

8.29 i

38-5 ,

7
6
6

6
9
9

Cord and pneuinogastrics are cut.

Placed in apparatus; head and fore limbs
free.

Ice in cloths placed around bead.

2—2

C. SIHLER.

. in

•♦-1

•

«M O

•8

Time.

Temp
appars

Temp
anim

3.45

88*9

4.14

40*5

41-8

Artificial respiration necessary.

Respiration shallow and weak.

Artificial respiration necessary.

41-4

Artificial respiration necessary.

Respirations shallow.

41-7

Muscles twitching.

Efforts at respirations rather than respira-
tions.

•

Artificial respiration.

Dog died.

Table III. then shows that when cord and pneumogastrics are cut
the increase in the number of respirations is very low indeed. This
certainly does not look as if the hot blood had the power to directly
diminish resistance and increase the irritability of the respiratory centre.
It is not without interest to observe how the panting can be produced if
the cord is cut and the pneumogastrics preserved — in that case, however,
the temperature must be. raised considerably — and how it can likewise
be produced when the pneumogastrics are cut and the cord left intact,
in that case the temperature need hardly be raised at all. But when
both cord and pneumogastrics are cut panting is not seen, excepting
under certain artificial conditions.

The next question, then, would be how much is due to the peripheral
stimulation of the vagus-endings in the lungs by the increased tem-
perature, and do they act just like the nerves of the skin ? Are they
sensitive to warmth ?

Exp. 4, Table IV., may help to answer this question.

HEAT-DYSPNCEA.

Table IV.

March 4th, 1880.

of
tion.

■M

of
lion.

No.
observe

Time.

■8 5

7.35

39-2

The temp, in this exp. from No. 9 onwards
refers to the heated air in the can. The
evening was very warm and close.

89'2

39-1

8.00

89-2

184

Cord cnt.

88*8

38*8

38'5

Dog's trachea-tube connected with a large tin
containing water at elevated temp, and Ba.
(O.H.),.

38-4

9.05

88-4

38-8

Placed in warm apparatus.

10.06

88-8

39*2

895

39-9

•

40-4

160

40-5

176

Pneumogastrics cnt.

40-9

232

March 5th.

8.40

345

•

Placed in apparatus.

9.08

10.01

89-2

41-8

Artificial respirations for two minutes.

42-5

11.00

42-6

48-2

43*6

156

22 C. SIHLER.

We can gather, then, from Table IV. that giving the animal warm
and moist air to breathe did not seem to have any effect on the peripheral
vagus fibres, the animal was not made to pant thus; and, again, cutting
he nerves did not stop the panting after it had once been set up. The
same observation was made on a dog in which the cord was intact, the
animal breathing hot air. The respirations were not permanently
diminished by cutting the vagi.

But why did the dog not pant the next day after reaching a
temperature of 41 ? Or why not in Table III. ?

I may add here that the dog would not have reached before dying
the high temperatures which it did in Table IV. if artificial respiration
had not assisted him ; and, further, the observation has repeatedly been
made, that the respirations go up in frequency during artificial respiration
and remains high a little time afterwards.

Regarding the depth of the respirations, I cannot support the state-
ment that they grow deeper. Tracings which I have taken show that
they grow more shallow, as it also appears to ordinary observation.
Accidentally I found out, I think, how Fick's statements, that they
grow deeper, came to be made. In an experiment which I made the
board on which the dog rested got a little too hot accidentally, and then
the respiratory movements grew deeper. As soon as the animal was
protected from pain they went back to their normal character, showing
more limited excursions than the respiration at the normal temperature.

THE INFLUENCE OF QUININE UPON THE REFLEX-
EXCITABILITY OF THE SPINAL CORD. By Wm. T.
SEDGWICK, Ph. B., Fellow of the Johns Hopkins University,
Baltimore, U.S.A.

It is the object of this paper to describe a series of experiments which
seems to indicate a different explanation from that commonly accepted for
the influence of quinine upon the reflex-excitability of the spinal cord.
A knowledge of the real action of this drug is particularly desirable at
the present time, because Setschenow's theory of special reflex-
inhibition centres has been so often and so successfully attacked that the
arguments drawn from the marked effect of quinine upon reflex-irrita-
bility are to-day, perhaps, among the best reasons for retaining it.

The theory of Setschenow1 was originally offered to explain the
great loss of reflex-irritability which is the uniform result of stimulating
with sodium chloride the optic lobes or optic thalami of the frog's brain.
It has also been looked upon with favour as accounting most easily for
that singular rise in reflex-irritability which follows division of the
medulla in the normal frog.

Herzen2 weakened the argument for the existence of these centres
by showing that a depression of irritability was not limited to stimula-
tion of the optic lobes and thalami, but might be induced by stimulation
of the cord itself. Goltz8, sines that time, has removed the necessity of
retaining the theory of Setschenow to explain the increased irrita-
bility of the normal frog after division of the medulla by bringing
forward his theory of simultaneous stimulation. Besides these investi-
gators, Freusberg4 and others have tested and finally abandoned the
doctrine of special reflex-inhibition centres. Nevertheless, this doctrine
still offers the readiest explanation of numerous phenomena in physio-

1 Ueber die llemm wig 'smeehanismen fur die Reflex 'that'tglieit dea Ruckonnarksy 1863. Set-
schenow und Paschutin, Neue Yersuche, 1865.

8 Exp. sur leu Centres moderators de r action re/fere, 1864.

3 Beitrage zur Lehre von den Ftmctionen der Nervenccntrcn des Frosche*, s. 39, u.s.W. Berlin,
1869.

4 Pflugcr's Archir, x. (1875), 174.

24 W. T. SEDGWICK.

logy, one of which is the remarkable loss of reflex-excitability following
the administration of a small dose of quinine to a normal frog.

Except Meihuizen1, whose work I shall review further on, no one,
so far as I know, has offered any other explanation of the action of this
alkaloid than that it stimulates the so-called centres of Setschenow.
Chaperon2 suggested that it probably acted in this way, and believed
that he had proved it beyond a doubt by experiments which are bo simple
and yet seemingly so conclusive that they have been widely adopted for
demonstration purposes.

Thus, if a small dose of some salt of quinine be thrown under the
skin of a normal frog, or one from which only the cerebral hemispheres
have been removed (and which we may conveniently call an " optic-
lobe " frog), a great loss of reflex-excitability occurs. If now we divide
the medulla, the excitability returns quickly to the normal. Conversely,
if the medulla is divided before the dose is given, no loss of irritability
can be detected.

It has also been found that, with large doses, the reflex-irritability
may be depressed after a time even in the pithed frog.

Having thus rapidly sketched our present knowledge of the working
of this drag, I shall next describe my methods of experimenting ; then
will follow a discussion of my work, a few words concerning the case
of sodium chloride, and, finally, the application of my results to the
general theories of reflex-inhibition.

Methods of Experimenting.

In spite of the objections which have been made to it, Tiirck's
method for measuring the reflex-irritability was used throughout.
Cyon's objection that, what one measures in such cases is not reflex-
excitability, but only the duration of the reflex-time, seems groundless.
We know that stimuli in the nervous central organs are cumulative, and
if a longer time elapses between the application of acid to the skin
and the occurrence of a reflex movement, this can only mean that the
stimulation had to attain a greater height before it gave rise to an
efferent discharge. The objection has also been raised that, sooner or
later, the acid employed must act harmfully upon the bit of skin to
which it is again and again applied. It is claimed, too, that the suspen-

1 Pfliiger's jirchiv, vh., 216.
1 Pfliiger's Archil y n., 293.

ACTION OF QUININE ON SPINAL CORD. 25

sion of the frog pats him in such an abnormal position that the results
obtained are not trustworthy. These objections, and many more, dis-
appear in the light of experience gained by observations made with a
careful attention to details.

I have worked only upon frogs. Except in a few cases they were
hung up by a large pin (passing through the head between the nares),
from either end of a horizontal wooden bar. This bar was supported by
having its middle portion nailed to a tall block, so that no other part of
the frog's body was in contact with any solid object. A reservoir of
water above communicating with a flexible rubber tube closed by a pinch-
cock gave abundant and ready means for washing the frogs and keeping
them in good order. They were constantly watched, and frequently
bathed by immersion in a basin of water lifted from below. This basin
of water is also a quick means of removing the acid after the reflex
movement has occurred. Draughts of air were found very irritating, and
were, therefore, avoided.

Dilute sulphuric acid was employed, and was made by diluting to a
litre two c. c. of commercial " pure sulphuric acid." It had a quite
distinctly acid taste. The time was marked off by a metronome, beating
one hundred strokes a minute. At first the reflex-irritability was esti-
mated every ten minutes ; however, if the conditions are good, five
minutes is a sufficient interval, and my later observations were made
five minutes apart. No comparative experiments were attempted until the
record of several consecutive reflexes showed only such variations as
would fall within the limits of observation errors.

Perhaps the greatest difficulty met with in using the method of
Tiirck is to be sure thatthe toe of the frog dips into the acid equally far
every time the reflexes are determined. Carelessness in this respect may
produce great variations in a record, and for this reason Meihuizen's
plan of holding the frog in the hand is objectionable. Again, the acid
must be removed with all possible speed after the reflex movement has
taken place.

As I employed it, Turck's method gave satisfactory results; for
frogs could usually be kept in good order as long as was needful. A
test experiment, in which two frogs had their medullas divided, and
soon after were hung up as I have said, showed a record of reflexes
which hardly varied for six hours. The irritability was taken every five
minutes. So that they were suspended in an abnormal position for nearly
six hours ; they had sixty-six applications of dilute acid to the same bit of
skin ; these sixty-six stimuli set up as many reflex movements ; yet at

26 W. T. SEDGWICK.

the end of the trial the reflex excitability was precisely the same as at the
beginning, and observations ceased only from my own weariness. In
the face of such experiments it seems absurd to claim that, under proper
precautions, repeated applications of acid of the strength indicated, or
repeated demands upon the spinal cord, will lead to serious errors.

It is a matter worthy of close attention, especially in view of the
results which I have reached, to consider the form in which quinine shall
be given. As Hermann1 points out, the use of acid to dissolve the
sulphate is not to be recommended ; for the acid may set up stimuli
which will depress the reflexes like other stimulation of sensory nerves.
Yet, if sulphate of quinine is to be used at all, acid must be added, for
it is little soluble in pure water. For these reasons it seems best to reject
the sulphate, and to use the chloride, which is quite soluble in pure water,
and weight for weight, contains much more quinine. There is, however,
one danger in using this salt which must be borne in mind. If given
in doses of a rather concentrated solution it behaves as an irritant.
Commonly the drug is injected under the opaque skin of the frog's back.
Thinking that less danger of losing any of the dose was incurred by
putting it under the abdominal skin (as the frogs sometimes jump about,
and by arching the back squeeze out a few drops), I have lately thrown
the drug in at a small incision on the abdominal skin near one of the
arms. I have noticed, with some surprise, that, after a time, there often
appears a large congested area just over the part where most of the
solution is lying. m If the aqueous solution of quinine chloride may act
in this way it suggests that it should always be dilute ; for if irritating
it is quite as objectionable as the acid solution of the sulphate. For
ordinary work I have used a freshly made solution, '06 grams of quinine
chloride in 10 c. c. of distilled water. This does not appear to irritate ;
and using ^ c. c, which is a convenient quantity, a dose of *003 grams is
given. The solution of atropia which I employed had '005 grams of the
sulphate dissolved in 10 c. c of water ; this gave for each dose of £ c. c.
only '00025 grams, yet this minute quantity proved ample.

Quinine Salts.

As has been said above, so far as I know, the'only attempt to explain
the action of these salts on any other theory than that they stimulate
the so-called centres ofSetschenow has been made by Meihuizen2,

1 Lehrbuch der Experimentellen Toxicologie, s. 366. Berlin, 1874.

2 Pfliiger's Jrchir, vn., 216.

ACTION OF QUININE ON SPINAL CORD. 27

and by him only indirectly. He worked only with frogs whose medullas
had been divided, so that these particular centres were out of the
question. Still, he advanced a theory for the action of the chloride of
quinine on such frogs which, if true there, might also be true perhaps
in the entire or optic-lobe frog. It was thought best, therefore, to test
his theory. Meihuizen found — and I agree with him in this — that
although in the frog whose medulla has been divided small doses of
quinine do not seem to affect either the heart-beat or the reflex-excita-
bility, large doses do, on the contrary, affect both. They slow the
heart-beat and depress the reflex-excitability.

In his other work I have not been able to confirm Meihuizen' s
results. Under large doses of quinine I have repeatedly seen the
reflex-excitability grow feebler and feebler, till it finally disappeared
altogether. In such cases I have almost invariably found the heart
still beating, though the circulation in the web-vessels was usually
stopped. Meihuizen, on the other hand, finds no loss of reflex-
excitability until the heart has wholly stopped beating ; then, he says,
the reflexes disappear in from fifteen to thirty minutes, or often even
sooner — that is to say, a great loss of reflex-excitability never precedes
a cessation of the heart-beat. On this observation he builds his theory,
which is, that in frogs with divided medullas quinine depresses the
reflexes by producing grave disturbances in the circulation. I can only
reconcile my own results with his by supposing that the exposure of
the heart which he resorted to in some way causes it to stop sooner
than it otherwise would. Different as the case is from that of the
ordinary frog supposed to have inhibition-centres, it might be Jhat
in the latter the circulation was affected even when no obvious change
was seen ; and, as a consequence, by virtue of these centres, quickly
depressed the reflexes of the spinal cord. Experiments were therefore
begun both with quinine and with sodium chloride, in order to settle the
point upon frogs having the so-called centres ofSetschenow. The
heart having been exposed in an optic-lobe frog, and a crystal of sodium
chloride laid on the cut ends of the thalami, no change in the heart-
beat is seen for a short time ; very soon, however, the heart beats
slower, becomes dilated, and stops in diastole, with all the phenomena of
vagus-inhibition. Almost at the same time convulsions usually begin,
and when they are over the heart is found beating again. If the vagi
are cut beforehand the heart cannot be stopped in this way ; and so,
too, if a minute dose of atropia is given before beginning the experiment
it always fails; hence we are probably safe in concluding that the phe-

28 W. T. SEDGWICK.

nomenon is due to vagus-inhibition of the heart-beat, brought about by
stimulation of the thalami with sodium chloride. This is a fact of some
interest, perhaps. So far, support seemed likely to be given to the
theory of Meihuizen. Accordingly, the work upon quinine chloride
was begun with special interest, for I did not then know that quinine
forbids vagus-inhibition. I soon found, however, that in the entire or
optic-lobe frog the heart was not stopped in the same way by small
doses of quinine. Moreover, a dose large enough to slow the heart-beat,
or to stop it, continues its effect even after that organ has been separated
from its extrinsic nerves. Clearly, the cases of quinine and sodium
chloride are very unlike, so far as the heart is concerned. I next pro-
ceeded to estimate directly the influence upon the reflexes of profound
disturbance of the circulation. The reflex-time in frogs with divided
medullas having been carefully recorded and found fairly constant, the
heart was exposed, and a ligature passed tightly around it, so that all
circulation stopped at once. This experiment seemed to show that in
no case did the reflex-time change much within half an hour ; and this,
it will be remembered, was the extreme period during which, according
to Meihuizen, the reflexes lingered after total stoppage of the heart-
beat by quinine. Table I. records some experiments made in April on
frogs in good order, and under the same conditions. All were tested at
the same time, the animals being hung up side by side, and observed
one after the other at equal intervals. When it is recollected that,
although the incision to expose the heart does not perceptibly affect the
reflex-time, ligaturing-off the heart is a more profound operation, the
moderate variations which the records indicate may perhaps be well
accounted for. On the average, about forty minutes elapsed before the
reflex-irritability suffered any great change ; even then the reflexes
seemed to fail rather from stiffening of the muscles than from any change
in the nervous elements. From the fact which these experiments seem
to prove, that a total stoppage of the oirculation has less rapid effect
upon the reflexes than even large doses of quinine, we must conclude
that quinine does not act primarily upon reflex-excitability by diminish-
ing the blood-flow.

Experiment 1.

The observations were made ten minutes apart. Frogs A, B, C, D,
E, F, with heart ligatured, show the effect of a total stoppage of circu-
lation upon reflex-irritability : their medullas had been divided one

ACTION OF QUININE ON SPINAL CORD.

hour before experiments began. E was an optic-lobe frog whose hemi-
spheres had been removed several hours before. Frogs G, H, and Z
give an opportunity for comparing the effect of large doses of quinine
with a complete stoppage of circulation- Z had a dilute, and G and H
had a concentrated, dose of quinine.

Table I.

Fboo

Faoo

Pmoo

FRO0

Fboo

Faoo

Fboo

Tra.

BEMARKS.

10.40

10.50

u 50+ " means that

11.00

the metronome beat
over 50 times, and no

11.10

reflex movement was

11.20

Heart

♦

seen.

tied.

tied

11.30

The ligatures were

11.40

put on lust as soon as
the observations
given under- 11.10

11.50

50 +

12.00

50 +

were over, although
the Table shows 11.20

12.10

50 +

etc.

as the real time* The

12.20

50 +

average time which
elapsed between the

12.30

50 +

etc.

50 +

application of the
ligature and the loss

12.40

etc.

12?

etc.

of all reflex indicated

12.50

50 +

by the second 50+
was not less than 46

1.00

50 +

minutes.

1.10

etc.

Having in mind the remarkable inhibition of the heart-beat by
sodium chloride applied to the mid-brain of the frog, which seems to
point clearly to a distinct efferent impulse proceeding from the stimulated
part, and recollecting the various phenomena of simultaneous stimulation,
such as the diminished irritability in one leg when the sciatic nerve of
the other is stimulated by a strong electric current, it is not difficult to
suppose that all the reflex-inhibitions produced by applying sodium
chloride to the nervous apparatus of the frog are special cases of
simultaneous stimulation.

Turning next to quinine, and attempting to apply, in this case, the
same theory, our attention is at once drawn to the important fact that
quinine has a decided effect upon the heart itself. Something is certainly
going on here, for the heart beats slowly under a moderate dose and
ceases to beat altogether under a large one.

If counter nervous stimulation occurs in this organ it must be

30 W. T. SEDGWICK.

through the vagus nerve. If this acts as the afferent nerve, whose
stimulation is to depress reflex-excitability, then division of the medulla
below the nerve must forbid that depression, as it does. By reviewing
the subject and by this train of thought I was led to believe that such a
theory would account well for the facts and do away with the necessity
for supposing the existence, in this case, of special inhibition centres ;
it would be this : quinine salts acting upon the nervous network of the
heart, stimulate the vagus nerve, and so depress general reflex-irritability
in a way similar to that in which electrical stimulation of one sciatic
nerve may depress the reflexes in the other leg.

This theory accounts well for facts which have long been thoroughly
established, and, if true, need meet with little objection, for its depressing
effects upon the reflex-excitability of the cord are only simplified and
placed alongside of many other cases of simultaneous stimulation which
are unquestioned. The return of that excitability after division of the
medulla is accounted for, since the source of the depression — the
stimulated nerve — is no longer connected with the cord ; and, con-
versely, if the medulla is divided beforehand no depression can occur
for the same reason. Moreover, the effects of small and large doses
upon frogs with divided medulla should be, as they are, totally unlike
the effects of the same doses upon normal or optic-lobe frogs.

If my theory is true, section of the vagus nerves ought to be, so
far as the reflexes are concerned, equivalent to dividing the medulla.
Accordingly I divided both vagi close to the medulla, but the results
were not constant. Owing to paralysis of the laryngeal muscles the
frogs no longer breathed normally and always bore the marks of a too
severe operation. No absolutely contradictor}* results were obtained;
still sometimes, after quinine-giving, the reflexes fell, but as it seemed
rather from general exhaustion of the animal, and in others the reflexes
continued as if no quinine had been given. Division of the visceral
branches of the vagi below the origin of the laryngeal was a less severe
operation, and was correspondingly more successful. A great trouble
in this mode of experimenting is that a very considerable number of the
frogs after the operation swell up enormously and utterly fail to expire.
The phenomenon is described by Heinemann, and frogs which show
it are no longer available for experiments upon reflex-irritability. Then,
too, even of those which do not seem affected in that way, it may be true
that there is something going on which, while it is not conspicuous, may,
notwithstanding, affect the reflexes. Those frogs which showed no gigns
of Heinemann' s phenomenon gave very fair results.

ACTION OF QUININE ON SPINAL CORD. 81

Exp. 2, see Table II., records some of these results and affords an
opportunity for comparing them with the effects of similar doses upon
the normal frog. It should not be forgotten that an animal which has
undergone a severe operation cannot be expected to retain irritability so
long as an animal unoperated upon.

Experiment 2.

The observations took place five minutes apart. For convenience
they are arranged as if they were made simultaneously. They repre-
sent some of the best cases for the theory.

A, B, C, D had had the visceral branches of their vagi divided below .
the origin of the laryngeal several hours before.

E was an optic-lobe frog. F, G, and H had undergone no operation.

The experiments occurred in April and May, 1880, and the weather
was favourable for the work. Frogs were chosen with special reference
to the apparent absence of Heinemann's phenomenon.

The results thus far obtained, though very encouraging, were not
perfectly satisfactory. It was therefore decided to make use of atropia,
in the hope that, since it is believed to paralyse the inhibitory vagus-
endings in the heart, it might also paralyse the ends of the afferent
fibres, and so prevent the action of quinine, which, the theory supposes,
stimulates those endings. This test proved perfectly satisfactory. Table
III. shows several cases, which may be compared with others taken at
the same time when no atropia was given. They are representative
examples. The dose of sulphate of atropia ("00025 grams) is very small,
but after it has been given the usual amount of quinine seems to have
no effect at all.

It will be remarked that the small quantity of atropia does not itself
affect the reflexes. This I have proved by separate experiments, see
Exp. 3, Table III.

W. T. SEDGWICK,

•

Table II.

FftooA

FbooB.

PmooC

FrooD

Tina.

FbooB

. FmooF

FbooG

. FbooH

REMARKS

y^"

With rafl divided.

3.10

3.15

8.20

3.25

3.80

3.35
3.40
3.45

For the signi-

ficance of "60+"

—

see Table I.

3.50

8.55

4.00

4.05

4.10
4.15

12
10

9
11

15
14

16
17

12
18

"M. d." means

More.

M. d.

"medulla di-

4.20

^■n_

vided."

4.25

—

4.30

—

50 +

—

etc.

4.85

4.40

4.45
4.50

—

12
12

—

4
5

50 +

15
18

—

■

M.d.

The dash is

20
21

used instead of

4.55
5.00

12
29

More.

♦
Q

18
18

18
13

•

the words, "No
observation."

5.05
5.10
5.15

10
12
11

—

M.d.

—

D had, in all,

5.20
5.25

11
10

—

•^—

three large doses

5.30

reflex.

—

of quinine.

5.85

5.40 !

50 +

82 i

5.45 :

5.50 ,

50 +

•

Deed.

ACTION OF QUININE ON SPINAL CORD. S3

If proof for my theory depended solely upon, the action of atropia,
it might properly be argued that we know too little of the action of
this drug to base upon its effects any explanation of the working of
quinine ; but when taken in connection with the effects produced by
vagus-section, it becomes a valuable ally for the theory. Owing to the
sudden advent of warm weather I have made but a single experiment
to see if atropin was a general paralyser of inhibitory fibres. An optic-
lobe frog, to which a large dose of atropia had been given, showed the
ordinary loss of reflex-irritability when his lobes were stimulated with
salt. Moreover, it is hardly possible that the small dose which I have
used could prevent general reflex-inhibition.

These experiments seem to me to show that quinine salts, when
given to the normal or optic-lobe frog in small doses, depress the reflex-
excitability by stimulating the vagus nerve through its endings in the
heart. It is not unlikely that the pulmonary and gastric endings may
also be influenced, but I have no proof of their action.

If my work shall be confirmed, it must be admitted that in the frog
with divided medulla we have a different problem to solve. Small doses
are here ineffectual ; and when we recollect that quinine is a proto-
plasmic poison, and in large or concentrated doses may become an
irritant, several possibilities arise. Quinine may poison the cord directly,
or have some other equally obscure action ; but from some experiments
which I have begun but have not yet completed, it is possible that the
depression in these cases is due to intense simultaneous stimulation ; the
irritating quinine solution being a stimulus comparable to the electric
stimulus applied to a sciatic nerve, and, like that, affecting materially
the general reflex-excitability. That it acts more feebly in case the
brain and great nerve-centres are gone is to be expected ; it has less to
work with and upon.

Sodium Chloride.

My work upon the behaviour of this substance has not perhaps gone
beyond that of other observers. Their accepted results I have been able
to confirm in most cases. Herzen's observation that stimulation of
the cord could cause a depression of excitability I have fully confirmed
by dividing the medulla, estimating the reflex-time before and after
placing salt upon the section. On cutting across the cord again below

•3

W. T. SEDGWICK.

Experiment 3.

Observations occurred five urinates apart. Frogs A, B, C, D, E
show that atropin does not in such doses affect the reflexes ; also that
after atropin-giving quinine is ineffectual. F, Q, H, I show the effect
of the same doses of quinine when no atropin has been given. In no
case after atropin-giving have I seen quinine have its ordinary effect.

-■

M.A.F.1

Fin C.I Tmoo D-fivoE

^gajg^g^.,

.„.,„

£8

«J^.,

W-J^Wb.

***>*-«■ '

9.25

9 7

—

10 1

s.ao

9 4

'■>

9 '

9.35

9.40

10 |

9.45

For the nlgnMcmnrt

(

■m

va ' an

■tm

9.50
9.56

A. 80

A Ml,

ABO

ISO,

A BO

irJt*

J of -IO+»Md tW
diuh, «<e T»blM I.

17 sad It.

10.00

10.05

M +

1* i

10.10

50+

10.15

etc

Owing to •miitikt

H bad a double flow.

10.20
10.25

$a.

5«

$a.

$CL

to.

20
17

18
16

•!

—

1-1

10.30

50+

'•(> +

A 80*1 Frog V shown bow [

10.35

50+

-iy tbe amrags eflset o(

16
17

10.40
10.45

6
7

7
9

8
8

etc.

etc

13
25

UU1 1'IWb |nnj OT
quinine chloride » I
bavo freqnentlj ob-

10.50

icrvedlt.

10.55

:"::;:;.

Lkm

11.00

ST"l

11.05

—

50 + !

11.10

j0+ FroB 1'* record

11.15

etc. i ^? th" *" dM

24
2ft

11.20
11.25

etc.

7
6

(■OWiiJ of atropin
■tors tbe reBBi trrlu-

11.30

billtT-

11.35

11.40

etc.

11.45

11.50

etc.

Etc

ACTION OF QUININE ON SPINAL CORD. 85

this point the reflexes may often be restored; they may then be again
depressed by salt and restored by section.

I wish merely to call attention to the evidences of simultaneous
stimulation in the case of sodium chloride, as bearing upon theory.
These evidences are, first, the fact that the heart may be stopped by
applying salt to the thalami; second, that at about the same time
convulsions occur ; and, third, these are not due to stimulation of the
so-called " convulsive centre," since they occur almost as well if the salt
is laid upon the cut cord from which the medulla has been removed.
These facts, if they show anything, show that the salt may act as a direct
stimulus of considerable power.

Theoretical Considerations.

I have not overlooked the difficulties which seem to arise from the
strange behaviour of atropia towards quinine depression of reflex-
excitability. It is not easy to understand how two drugs which have,
apparently, the same effect upon the inhibitory function of the vagus
shall nevertheless .act precisely unlike upon the vagus nerve in respect
to reflex depression. At first sight we can only escape by saying that it
(quinine) acts as a paralyser of inhibitory endings and as an excitant of
afferent endings of the vagus nerve, while atropin paralyses both. This
hypothesis, however, assumes a distribution of function in the vagus
fibres which we are hardly justified in making. In view of the
discovery by Prof. H. Newell Martin1, that special reflexes may
be inhibited by the stimulation of the central ends of efferent fibres, we
may have to change all our ideas of reflex-inhibition, and it may be
that quinine merely stimulates the ends of the efferent cardio-inhibitory
fibres, and these act back in the centres.

Since atropin is known to paralyse the peripheral organs of the
cardio-inhibitory fibres, we would then get an explanation of the fact
that after its administration small doses of quinine are without effect on
the reflexes. Otherwise it would appear that we must assume that
atropin paralyses also the ending of afferent vagus fibres in the heart,

1 Johns Hopkins, University Circular, May, 188Q. A preliminary account of some
experiments tending to prove the existence of a new function in the anterior roots of the
spinal nerres.

86 W. T. SEDGWICK.

which are stimulated in the organ under the influence of quinine and
depress the reflexes.

The slowing of the heart under quinine, and at the same time the loss
of cardiac inhibition on direct vagus stimulation, show that the cardiac
action of the drug still needs much more investigation.

If the statement which was made at the outset, that the theory of
Setschenowis better sustained by quinine than almost anything else,
is true, then it must be granted that that theory now rests on a weak
support. If my results secure confirmation, quinine does not depress the
reflexes by the mediation of any special inhibitory centres. Moreover, it
seems to me that all the phenomena found in using common salt to
demonstrate the existence of these centres may be better explained by
looking at them as particular cases of simultaneous stimulation, com-
parable to the general inhibition of reflexes accompanying the powerful
stimulation of a sensory nerve.

Sodium chloride, although its first cause, has for some years been
a stumbling-block in the way of the theory of Setschenow, while
quinine has been one of its most important supports. Goltz's theory,
on the contrary, has been made more probable by the action of salt, and
has hardly accounted for the effect of quinine. It will be seen that the
results of my work support Goltz and render highly improbable the
theory of Setschenow.

The general results of this paper may be stated thus : —

1. Quinine salts in small doses seem to depress the reflex-excitability
of the cord by stimulation of the vagus nerve ; mainly through its end-
ings in the heart.

2. This places the quinine action alongside other stimuli of sensory
nerves, and explains it action by saying that it is a special case of reflex
depression by simultaneous stimulation.

3. Goltz's theory is supported, and that of Setschenow much
weakened by these phenomena.

4. Reflex depression under quinine salts, in the pithed frog, is a case
wholly different from the same depression in the entire frog. Larger
doses are required, and the drug possibly acts as a direct poison on the
cord.

It is not unlikely that other drugs may act like quinine upon the

ACTION OF QUININE ON SPINAL CORD. 37

•

reflexes. I propose to continue my work and shall especially examine
digitalis, and others which act upon the heart.

The materials for this paper were accumulated in the Biological
laboratory of the Johns Hopkins University, in charge of Prof. H.
Newell Martin. I am glad of an opportunity to express my feeling
of deep indebtedness to him for the constant encouragement and wise
counsel with which he has favoured me.

THE EARLY DEVELOPMENT OP THE WOLFF-
IAN BODY IN AMBLYSTOMA PUNCTATUM.

By SAMUEL F. CLARKE, Ph. D., Late Fellow and Assistant in
Biology, Johns Hopkins University. With Plates I, II and III.

The first indication of the urinogenital system in Aniblystoraa
is found at the period of development represented in Figure 1.
At this stage, as 6een in cross sections, Figures 4iV to 13JV, the
mesoderm extends entirely around the body forming a two-celled
lamella. In the region which is to become the intermediate cell
mass, both layers of mesoderm are much enlarged, see Figures 4N
to 12 N. This enlargement of the mesoderm is produced by a
growth of and not a multiplication of the cells, as is seen in the
Figures of series "N" This beginning of the Wolffian blastema
was found extending through a few sections only in a consecutive
series. In the next later stage from which a series was obtained,
Figure 2, the somatopleure cells of the blastema have become very
much larger than those of the splanchnopleure ; the former divide
transversely and then become differentiated from the rest of the
mesoderm by a definite outline. This blastema now consists of a
solid mass or rod of cells lying just ventral to the lateral plates,
bounded on the inside by the splanchnopleure, on the outside by
the epiderm and formed from the outer layer of mesoderm. At
its anterior end through six or seven sections it is of considerable
size, then it suddenly becomes much smaller and continues without
change through ten or twelve sections farther backward. In the
next succeeding series of sections, taken from a specimen repre-
sented in Figure 40, one finds that the body cavity is beginning to
be formed and the Wolffian blastema is seen to be entirely in the
somatopleure; no anterior opening has yet been formed in the
segmental duct, as Balfour has called this structure in the Elasmo-
branchs, as is demonstrated by the first section. The next two
sections show that a lumen is being formed within the previously
solid rod, while the three sections following these two indicate a
partial differentiation of the blastema into a dorsal and ventral
part. After one or two sections more, the dorsal portion termi-
nates and the ventral part continues posteriorly as a solid rod.
4 • 39

40 8. F. CLARKE.

The next or fourth series are from an embryo represented in
Figure 41 Y. In this stage one finds from the sections that the
dorsal duct now opens anteriorly into the body-cavity ; the split
has worked its way forward to the anterior end of the blastema,
separating the anterior end into two quite separate parts or ducts,
each with a lumen, but the ventral one ends blindly while the
dorsal one communicates with the body-cavity. Below the ventral
duct is a small solid rod of cells which was, I believe, not formed
from the blastema. In section number 37 Y of this series the
dorsal and ventral ducts have united into one which possesses a
single large lumen. The next succeeding section shows this single
duct opening into the body-cavity.

The Wolffian body then, arises from the outer layer of the
mesoderm as a solid rod of cells, and is at first largest anteriorly ;
a split then occurs in the larger portion which begins at the pos-
terior end of the smaller part and travels anteriorly, and at this
time a lumen has appeared in the anterior end of the blastema ;
finally, the split reaches the anterior end thus dividing that
portion into two ducts; the lumen is extending itself backward, a
small rod of cells has been formed below the anterior end of the
ventral duct, the dorsal and ventral ducts are united at one point,
and a second opening into the body-cavity from the dorsal duct
has been made. This method of development seems to be quite
different from that in any allied forms in which the development
has been worked out. As it is most like that of the Elasmo-
branchs, I will add a brief account of the development of the
urinogenital system in the latter group as given by Balfour. It
first makes its appearance as a solid knob of cells springing from
the intermediate cell mass. From this knob a solid column of
cells grows backwards to the level of the anus. The knob then
acquires an opening into the body-cavity which is continuous with
a lumen that makes its appearance in the rod of cells. Solid out-
growths of the intermediate cell mass then appear which soou
become hollow and open into the body-cavity. Their blind ends
curl obliquely backwards and open into the segmental duct.
After all this has taken place the segmental duct splits longi-
tudinally into two ducts in the female, and into one duct and
parts of another in the male.

In comparing this with Amblystoma, one notices that the origin
of the primitive rod of cells is very similar in both, they agree

AMBLYSTOMA PUNGTATUM. 41

in the anterior opening into the body-cavity and in the lumen
appearing anteriorly and working its way backward. Beyond
these points they are unlike. The splitting of the segmental duct
in Amblystoma takes place at a much earlier period and proceeds
in a different way. The second opening into the body-cavity is
also peculiar to Amblystoma as is the small rod of cells lying
ventral to the two tubes which are derived from the blastema. It
is possible, however, that this small rod is not a part of the urino-
genital system; and this second opening into the body-cavity is
probably the beginning of the first segmental tube.

It is a matter of great regret to me that I have not sufficiently
complete results to allow of any theoretical considerations, and I
have concluded to publish this short descriptive paper because
there is enough to show that the method of development of the
Urinogenital system in Amblystoma is quite different from that of
allied forms, and indicates a promising field of work, if the sections
can be obtained. I have worked many months to obtain the few
results here recorded, so difficult is it to obtain workable material.
Many thousands of sections have been prepared and mounted,
nearly all of which from one cause or another are valueless; many
are utterly worthless, while a large number, though partly good,
are not reliable. I have had the best results with Picric acid
specimens, and find that they work better a few days after they
have been transferred to absolute Alcohol, than when longer kept.

EXPLANATION OF PLATES.

The figures are numbered from 1 to 41 and the different series of

sections are indicated by letters annexed to the numbers of the figures.

All of the figures were outlined with the aid of the camera lucida.

PLATE I.

Figure 1. — A side view of the specimen from which the series of

sections marked "N" were obtained, nc, neural canal;
e, eye ; t, throat ; a, future position of cloaca. Mag-
nified six diameters.

Figure 2. — A side view of the specimen from which the series of

sections marked "P " were made, e, eye ; nib, mid-

42 S. F. CLARKE.

Figure 2. — Continued.

brain; bn, branchial lobe; ba, brachial lobe, from
which the anterior limb is developed; pr, protover-
tebrae. Enlarged six diameters.

Figure 3. — A diagrammatic figure of the developing Wolffian body

of Amblystoma made from series "Y." pp, body-
cavity; 61, the dorsal duct, opening anteriorly into
the body- cavity ; x, its second opening into the body-
cavity; 62, the ventral duct which unites with the
dorsal duct just in front of the second opening of the
latter; 63, the small rod of cells which appears just
beneath the ventral of the two large ducts.

Figure 4N. — A cross-section through the body at the anterior end of

that enlarged portion of the mesoderm from which the
Wolffian blastema is formed. The hypoblast cells are
very large and filled with very coarsely granular proto-
plasm ; ac, the alimentary canal ; nt, the notochord
which appears to be formed from the hypoblast ; w>6,
the enlarged part of the mesoderm from which the
Wolffian blastema is formed. The mesoblast at this
stage extends entirely around the body, forming a two-
celled lamella, ep, epiblast.

Figures 4 N to 8 N, are consecutive and show that this enlarged area

of mesoderm extends through these five sections without
any marked change.

PLATE II.

Figure 9 N. — This is not the next section to 8 N, but is next but one.

The enlarged portion of mesoderm wbt still per-
sists.

Figure 10^. — This represents the next section but one to Figure 9 N,

and shows no marked change.

Figure ION, to 13 N, are consecutive. Figure 13 indicates the pos-
terior termination of the mesoderm marked wb.

Figure 14P, to2lP, are consecutive, and are taken from the speci-
men represented in Figure 2. The series- is com-
pleted with Figures 22 and 23 on Plate III.

Figure 14 P. — A section through the anterior end of the Wolffian

* blastema, wb.

AMBLY8T0MA PUNCTATUM. 43

Figures 15 P, to 20 P, are essentially alike, showing the Wolffian blas-
tema, 6/, extending backward without marked change
in size or form.

Figure 21 P. — In this section the blastema, bl, suddenly diminishes in

size.

It will be seen from a comparison of sections, 14 P and 15P, that the
Wolffian blastema is found from the outer layer of cells of the mesoderm.

PLATE III.

Figure 22 P, is next but one in the series to 21 P. There is not much

change ; the intermediate cell mass with the blas-
tema bl, is more distinctly separated from the pro-
tovertebrse.

Figure 23 P. — This is five sections further backward in the series than

22 P, and shows the blastema reduced to a small rod
of cells. It occurs in one or two more sections only
and then terminates.

Figures 26TFto 32 W, form the third series, and were made from the

specimen represented in Figure 40.

Figure 24 W. — The anterior end of the blastema is shown at bl. The

body-cavity pp, is beginning to be formed.

Figures 25 IT, and 26 IF. — The blastema is larger than in 24 IF, and

the body-cavity is still present.

Figure 27 W. — The blastema is here much enlarged and is being

divided by a median transverse division.

Figure 28 IF — The split is here indicated also, but the upper or dorsal

portion is much the largest. The body-cavity pp,
is present but disappears in the next section.

Figure 29 IF. — There are now two distinct ducts, a dorsal bllt aud a

ventral bill.

Figure SOW. — The two ducts are still present but their lumena have

disappeared.

Figure 31 W. — The dorsal duct bll, here terminates while the ventral

one persists.

Figure 32 W. — This is next but one in the series of sections. The now

single rod of cells extends only a few sections farther.

In studying this series "IF" it appears that the blastema in its en-
larged anterior part becomes longitudinally divided by a split which
starts at the posterior end of the swollen portion aud travels anteriorly.

44 S. F. CLARKE.

Figures 33 Y, to 39 Y, comprise the last series, and were obtained from

an individual shown in Figure 41 Y.

Figure 38 Y. — A section through the anterior end of the developing

Wolffian body ; be, body cavity, into which opens the
dorsal duct 61; 62, the ventral duct and 63, a small
rod of cells which is found only in this and the fol-
lowing section.

Figure 34 Y. — The dorsal duct 6 1, is here distinct from the body cavity,

6c. There is a peculiar collection of cells about the
ventral duct which may be a trace of the primitive
connection of the dorsal and ventral ducts, the split
not being quite completed at this point. There is
a small lumen in each of the two ducts. The small
ventral rod of cells is also present.

Figure 35 Y. — The dorsal and ventral ducts hold the same relative

positions and have the same characters.

Figure 36 Y — The two ducts have united, forming one large duct

with a large lumen.

Figure 37 Y. — The single duct here opens into the body-cavity.

Figure 38 Y. — The single duct has become a solid rod of cells, and in

this condition stretches away toward the posterior
end of the body.

Figure 39 Y. — This is six sections posterior to 39Y, and beyond this

the "rod " does not extend.

Figure 40. — A side view of the individual from which the series of

sections marked "W," were made, e, eye; 6at
branchial lobe ; 6n,.brachial lobe; pr, protovertebrae.
Enlarged six diameters.

Figure 41 Y. — A side view of the specimen from which series " Y" were

obtained ; r?p, nasal pit ; e, eye ; bal, balancer ;
6n, branchial lobe ; 6a, brachial lobe. Enlarged
six diameters.

Figure 3, on Plate I, gives a diagrammatic side view of the develop-
ing Wolffian body of Amblystoma constructed from this series of
sections marked "Y."

NOTES ON THE FORMATION OP DENTINE AND
OP OSSEOUS TISSUE. By CHRISTIAN SIHLER,

M. D., Ph. D., Late Fellow and Assistant in Biology, Johns
Hopkins University. With Plate V.

I. Dentine.

There are two views held regarding the formation of dentine :
one supported by Waldeyer, in Strieker's Handbook, the other by
Rolliker, in his Histology. According to the former all the cells
of the tooth pulp are used up in the formation and are actively
engaged in the production of dentine. According to the second
view the odontoblasts only are the elements whose function it is to
deposit dentine. Waldeyer believes that osseous tissue and enamel
develop in quite an analogous way.

I shall now bring forward the observations which I have made
on the tissues coming into play in the process, and then consider
which view they support.

(1). Cracking with a vice the incisor of a calf, or splitting the
root with a knife, one finds that the pulp is removed from the
dentine very easily indeed; great difficulty is often experienced
in keeping it adherent to the dentine in order to make sections
through pulp and dentine, both remaining in their natural position
with reference to each other. This behavior of the pulp towards
the dentine is in striking contrast to that of the pericementum
towards the ceraentum, and seems to me to throw some light on
the difference in their respective modes of growth. Although this
fact is one readily observable without the microscope it seems not
less important on that account.

(2). Before enumerating the facts brought out by the microscope,
I shortly describe the method. The materials used were, princi-
pally, the incisors of the calf, the roots of which were split longi-
tudinally that the stainiug fluid might have access to all the parts,
including the dentinal canals; and care was taken to disturb the
relation between the pulp and dentine as little as possible. In the
staining fluid, Beale's carmine, the teeth remained, until the pulp-
cells were deeply stained; after washing with acidulated glycerine,
they were transferred to dilute alcohol, from this into strong
alcohol, and then allowed to dry, the pulp applying itself closely —

46 C. SIHLER.

in some parts at least — to the dentine. Sections were then made
with a hard and fine scalpel, through all the parts, pericementum,
cementuni, dentine and pulp, or through parts of these layers, as
was desired. The sections were then treated with glycerine and
acetic acid, which swells them out and brings them back to their
natural condition.

Figure 1, Plate V, shows such a section passing through the
root, a, pulp — near the dentine the darker red shows that the
cells there are either large or more numerous. 6, the pink zone,
the newest layer of dentine which is not yet ossified, c, the fully
formed dentine. /, the pericementum. e} the uncalcified cemen-
tum (again a pink zone), rf, the calcified cementum.

This drawing is intended to show only the general arrangement
of the parts, and is but little magnified.

Figure 2, Plate V, more highly magnified, shows the fully
formed dentine and the adjoining soft tissues where the growth of
the dentine must be in progress. We observe here — a, the calci-
fied deutine. 6, the uncalcified dentine, (the pink zone already
mentioned). The newly laid down semi-solid material absorbs
some of the carmine, but is not stained as deeply as the protoplasm
of the cells. Next comes a layer of large and long cells, reminding
one of columnar epithelial cells, with a dark red nucleus situated
generally towards the blunt end of the cells which is directed
towards the pulp. There has never come such an odontoblast
under my notice with more than one nucleus, d, the pulp proper
showing oval and roundish masses of protoplasm imbedded in
formed matter of a finely fibrillated character.

(4). The elements making the pulp can readily be examined, by
teasing and scraping a pulp which, after having been removed, has
been kept in bichromate of potass solution. Figure 4, Plate V,
shows such cells. They form very irregular, Branched, and varied
figures, their processes evidently running out into and continuous
with the fibrous network of the pulp. The naked eye shows, and
this must be borne in mind, that the pulp is exceedingly vascular,
and, upon teasing, larger vessels with unstriated muscular fibres,
and smaller ones richly nucleated, are observed pervading the
whole pulp.

(5). The odontoblasts in a very natural condition can be pro-
cured by scraping the freshy formed dentine or walls of the pulp
cavity, after removal of the pulp. For when the pulp is drawn

DENTINE AND OSSEOUS TISSUE. 47

out of the tooth, the line of separation takes place as a rule between
the odontoblasts and the pulp, the former remaining in connection
with the dentine.

Figure 5, Plate V, gives a number of forms, not unfrequently
observed. A very typical one is a, where we observe a large
nucleus near the inner rounded end, while the other extremity of
the cell looks squarely cut off, with a process or fibres attached to
one corner. In other odontoblasts a large process runs from the
outer extremity of the odontoblast, evidently pulled out from a
dentinal canal, (and as we shall notice afterwards, continuous with
the dentinal tubule and its contents), I have never observed two
odontoblasts joined end to end.

(6). In a well-stained specimen not only the odontoblasts are of
a red color but also the contents of the dentinal canals ; just as in
the cornea we have the nucleus, the body of the cell, and its pro-
cesses by which the protoplasm of the different cells is put in
connection, so in the living and growing dentine we have the
nucleus in the odontoblast, the body of the cell and its processes
permeating the dentine.

(7). If a section is made with a scalpel through the root of the
tooth, or more accurately through the dentine with the odonto-
blasts attached and in place, and such a section is treated with
strong hydrochloric acid, the ground substance of the dentine is
destroyed, and there are left behind the cells and their main
processes, corresponding to the tubules. I may just recall here,
that when dead and dry dentine which has been boiled, and where
the protoplasm has been destroyed, is treated with strong hydro-
chloric acid, the tubules remain behind. By taking these two
observations together, we see that the odontoblast and the dentinal
tubule with its contents are one thing. (It is hardly necessary to
mention that the former observation has been made also before by
Lent, and the latter by every body.) Jt is found further that the
odontoblasts do not separate readily laterally but are evidently
united one with another along their sides, although the con-
necting fibrils or tubules cannot be distinguished; but the short
processes apparent on the isolated odontoblasts seem to be these
connecting threads.

(8). Treating a section prepared as described under (2) with
dilute hydrochloric acid and pressing it with a coverglass one
often succeeds in separating the odontoblasts — adhering then to
5

48 C. SIHLER.

the pulp — from the dentine, in such a way that fibres are seen
across the interval between dentine and soft parts; and in favor-
able specimens it can be made out that these threads corres[>ond
to the large and thick external processes of the odontoblast de-
scribed above. That they can undergo so much stretching, as they
do, without tearing, seems to show that they are not protoplasm
pure and simple, but that their outer part is a thin dense envelope,
in fact the dentinal tubule (or Neumann's sheath) of the dead and
dry dentine. I think some authors confound this elastic tubule
with its protoplasmic contents.

Taking all the observations together we would have then — the
vascular pulp with its branching cells, the processes of which have
no definite arrangement but pass into a fibrous texture, the meshes
of which are filled up with a mucous ground substance; and out-
side the vascular pulp the odontoblasts, the end processes of which
pass into the walls of the dentinal canal, i. e., the dentinal tubule.
The odontoblasts themselves staining readily and carrying a large
nucleus are evidently in great nutritive activity, and their proto-
plasm is continuous with that lying in the dentinal tubule. The
newly formed dentine we find as an apparently homogeneous, semi-
transparent coating, covering the calcified dentine; it is not found
between the odontoblasts, but only at their outer extremities.

Now taking the case before us, i. e., a root of a tooth which is
growing, and waiving at present the question as to the method of
the first beginning of the growth of dentine, in what way does the
increase in the thickness of the dentine take place?

Taking all the facts into consideration, the most probable view
seems to be this: The odontoblasts absorb from the pulp the
necessary nutriment and form a secretion ; they pour this out in
such a way that the portion produced by the single cell cannot be dis-
tinguished from that produced by its neighbors, and this new layer
stains pink if the lime-salts have not yet been deposited in it. As
the odontoblasts form this secretion on their outer ends, they move
necessarily inward, and at the same time spin 'in their wake the
dentinal tubule. The side branches of the main tube correspond
to the lateral processes (spoken of above) holding the cells together.
Of course we must conceive that new lateral processes are con-
tinually being formed by the soft anterior part of the odontoblast
as this moves and grows inward. In moving onward thus, the
odontoblasts must of course remove the pulp, and we may imagine

DENTINE AND OSSEOUS TISSUE. 49

this to be done in two ways: either the odontoblasts being very
active in their nutrition take away the pabulum from the other
pulp-cells — the latter shrivelling and disappearing, or the odonto-
blasts live on the pulp-cells directly just as the tooth-sac of the
second tooth absorbs the roots of the deciduous tooth.

Waldeyer has come to different conclusions with reference to this
process. In Strieker's Handbook, (p. 337), we find the following
passage: " Whilst the peripheric portions of the odontoblasts con-
tinually undergo metamorphosis, with disappearance of their nuclei,
into a gelatinous matrix which subsequently undergoes calcifica-
tion, their centric portions penetrate the hardened mass in the form
of longer or shorter threads, and represent the first rudiment of
the dental fibres. The lateral processes of the odontoblasts occa-
sion the numerous anastomoses of the dental fibres or of the dental
tubule. Every odontoblast communicates with the nipre deeply
situated and successively enlarging cells of the young pulp, by
means of its pulp process, so that when an odontoblast is calcified
up to the base of the fibre another occurs in its place without any
interruption to the continuity of the fibre. Hence every dental
tubuje with its anastomoses must be regarded as formed of several
continuous odontoblasts. The layers of matrix immediately sur-
rounding the fibres undergo conversion, as appears from their
chemical character, into elastic tissue and form the dental sheaths
of Neumann. It has not yet been ascertained whether they also
undergo calcification. Thus it appears, that the dentine with all
its constituents proceeds from odontoblasts that have become
metamorphosed in their form and chemical composition. "

There seem to be several objections to this view.

In the first place if we do what Waldeyer asks us and imagine
the process to take place as he describes it, let every odontoblast
have a pulp-process analogous to its dentinal process — (which I have
failed to find and others fail to mention) — imagine the numerous
nuclei to disappear, the rearrangement of the eel I -processes of the
pulp-cells into the tubule-network of the dentine, the metamorphosis
of the bodies of the pulp-cells into dentinal matrix, having done
that, would we then after all have such a tissue as we find dentine
to be? No, we would have a hard tissue, with cauals, (but could
they have the regularity of the dentinal tubules?) and supplied very
richly with bloodvessels something like very vascular bone. Wal-
deyer quite forgets to dispose of his vessels and they are present in

50 C. SIHLER.

•

great abundance. Nor would they disappear by mere conversion
of the pulp-cells iuto gelatinous matrix and Neumann's sheaths.

Further, if such a direct transformation of the pulp took place
one might expect to find evidences of the former pulp-structure in
the final dentine ; this has so far not been demonstrated.

In the third place, if the outgrowths of the dentine took place
ought we not to see the deposition of dentine between the odon-
toblasts along their sides, and ought we not to find dwindling
odontoblasts or evidences of disappearance of their nuclei, as well
as pulp-cells which were being changed into odontoblasts to take
their place; finally, ought there be that tendency to separation
between odontoblasts and the remainder of the pulp which cer-
tainly exists?

What we really find is the newly produced dentine deposited
as a homogeneous coating on the calcified dentine, without any
evidence whatever of one portion being the metamorphosed odon-
toblast and of the other being another chauged cell.

If it could however be shown that there was a very intimate
union between the odontoblasts and pulp, and if the odontoblasts
which Waldeyer figures were the typical ones, this would speak in
favor of his view.

It is of course difficult to give convincing proof of such processes
as we cannot watch during their occurrence; all we can see is
the machine at rest. It seems however to me that there are
more difficulties connected with Waldeyer's than with Kolliker's
theory.

There is one point however in Kolliker's description with which
I cannot agree, namely, the formation of the side tubules. He
says: "The finer processes of the dentinal fibril are not present
when the dentine is first formed and must be looked upon as
secondary formations, just as those of the lacunae of bone."

This point will be better discussed when the formation of the
osseous tissue is under consideration.

II. Osseous Tissue.

In the investigations of the formation of osseous tissue, the long
bones of kittens, newly born and of more advanced age, were
chosen, and the calf's teeth illustrating the formation of cementum,
which I include here under bone. Embryonic bones of sheep

DENTINE AND OSSEOUS TISSUE. 51

and calves were also used, and the tissues were treated essentially
as were the teeth for the study of the development of dentine.

Figure 5, Plate V, represents a longitudinal section of a kitten's
femur, passiug through bone and the outside periosteum. The
following are the points that are to be distinguished and taken into
consideration, (a) is the fully developed bone substance; in it we
recognize the lacunae and canal iculi. The latter (the canal iculi)
we cannot see in the pink zone (6) although two lacunae happen to
be therein, in this specimen. In the soft parts on the outside of
the bone we find an outer part, which is distinctly fibrous, — (treat-
ment with strong acids indicate that the fibres are elastic enveloped
in a gelatinous homogeneous substance) and an inner part (c) which
abounds in (young) cells and which shows but faint fibrillation:
(e) is a pink zone similar to (b) and adjoining an Haversian canal,
in the lumen of which there appear also a group of cells similar to
(o). The soft tissue surrounding the bone, blends with it, merges
or passes into it, and we fail to see here such a schematic arrange-
ment of cells (typical osteoblasts as they are described in the books)
and which we are led to expect. The specimen was magnified
about 500 diameters (Gundl. V. Eyep. Ill), and reduced in the
drawing.

(2). Scraping the surface of such a bone, which has been kept in
bichromate of potass solution after the periosteum is removed, we
get these covering cells off the bone in a very natural condition, and
Figure 6, Plate V, shows some of them. They are all nucleated,
which can be demonstrated by the aid of acetic acid. The nucleus
was not apparent on all when the drawing was made. Generally
short processes are seen, and the drawings show the coarser
ones. Finer processes would of course be very apt to be broken
off.

(3). Figure 7, Plate V, is a section through part of the root of
a calf's tooth showing the cementum and pericementum. We
observe here, as in the kitten's femur, the calcified tissue with its
lacunae and the processes from these and a very broad zone of
uncalcified cementum with numerous lacunae, no canaliculi how-
ever are apparent to the eye; just as in the layers b and e from
the kitten's femur. As in the periosteum, we find an inner finely
fibril la ted part of pericementum, rich in cells, and an outer with
coarser fibres. The union of the enveloping parts to the tooth is
also very intimate. (In parenthesis I may remark, that in speak-

52 a SIHLER.

ing of fibri Hated or fibrous tissue I am using only descriptive
language; fibres and tubules are not so easily distinguished).

(4). Figure 8, Plate V, is a highly magnified drawing, a por-
tion of a transverse section of a femur of a kitten some months old.
The bones had been remaining in the staining fluid a long time, and
thus one point of importance is brought out plainly. While in
Figure 5, as well as in Figure 7, we see the pink zone quite
homogeneous, we perceive here that darkly stained lines pass
through it which we may be allowed to interpret as the future
canal iculi.

(5). Treatment of such a section as shown in Figure 7 through
the tooth, with strong muriatic acid brings out other important
facts. The strong acid will here as elsewhere dissolve the homo-
geneous gelatinous ground substance, it does this in the calcified
part as well as in the pink zone, aud in so doing brings to light
in the pink zone a network of fibres and tubules corresponding to
the canalicular network of the calcified cementum. In fact after
treatment with acid the calcified and uncalcified layers become
one; the walls of the tubules, as in dentine, evidently correspond
to a substance of the nature of elastic fibre.

The same observation can be made on the bone and periosteum.
Acid shows that the pink zone is not homogeneous, although it
appears so to the eye. In the bone and periosteum another fact
is brought out by this reagent. After it has acted some time
glistening fibres make their appearance in the periosteum and, by
pressing on the ooverglass, one can make out that some of these
periosteal fibres enter the bone.

Taking all these facts into consideration one may form the fol-
lowing conception of the process taking place here on the outside
of the bone, or wherever bone is formed, and on the root of the
tooth. The cells in the deep layer of the periosteum, or of the
pericementum, multiply and form blood vessels ; as they do so they
remain in connection with their mother cells and in all probability
form new connections with neighboring cells; these connecting
processes afterwards become the canal iculi. In their vital processes
these cells jointly excrete a gelatinous material and the elastic
membranes, which partly if not altogether produce the striation
observed in bone and so plainly visible in cementum ; the newest
layer presents itself, when treated in the way indicated, as
the pink zone; as the cells secrete layer upon layer they, as a

DENTINE AND OSSEOUS TISSUE. 53

whole, are carried outward further away 'from the finished bone-
substance. Some of the cells however, get entangled— so to speak-
in the secretion, and come to be, in the fully formed bone, the
lacunae, (or at least their contents). At the same time as the layer
of plastic cells moves outward, secreting the basis-substance, they
spin out, or draw out their processes thus giving rise to the canal i-
culi. Although these under ordinary circumstances are not easily
recognized, they are already present from the beginning and are
formed pari passu with the ground substance of the bone. All it
needs to make them apparent is the infiltration of the newly formed
tissue with lime salts.

One may compare the surface of a growing bone with that of a
granulating ulcer; on the surface proliferation of new cells and
formation of new blood vessels takes place (only in the bone they
are a wide network while in an ulcer they form loops), and a little
deeper in the deposition of new substance takes place ; in the one
case the typical osseous tissue and in the other the cicatricial
substance.

The view presented here on osteogenesis allows also enough
liberty for the formation of the different varieties of bone, which
vary, e. g., not unmarkedly in the young and the old, the character
of the bone depending on the nature of the fibrillar or connective
tissue forming it.

Kolliker and Virchow offer a different explanation of the form-
ation of the canaliculi. Kolliker says, p. 222, 5th Ed., 1867, of his
histology: "According to Virchow's discovery, which I can fully
confirm, these cells [the periosteal cells] become stellate gradually,
and are thus changed directly into the stellate bone corpus-

cles."

Virchow gives the following explicit account, p. 469, Cell. Path.,
7th Am. Ed.: "The cartilage cells (and the same holds good of
the marrow cells) during ossification throw out processes (become
jagged) in the same way that connective tissue corpuscles, which
we also originally found, do both physiologically and pathologically.
These processes which in the case of the cartilage cells are generally
formed after, but in that of the marrow cells frequently before, cal-
cification has taken place, bore their way into the intercellular
substance like the villi of the chorion do into the mucous mem-
brane and into the vessels of the uterus, or like the Pachionian
granulations (glands) through the calvarium."

64 C. SIHLER.

" The cells which thus result from the proliferation of the peri-
osteal corpuscles are converted into bone corpuscles exactly in the
way I described when speaking of the marrow. In the neigh-
borhood of the surface of the bone the intercellular substance grows
and becomes almost cartilaginous. The cells throw out processes,
become stellate and at last the calcification of the intercellular
substance ensues."

A view on the formation of osseous tissue differing from the one
above worked out, is that of Waldeyer, which is gaining favor
among histologists.

"The osteoblasts/' says Waldeyer, "are the embryonal cells
forming the osseous tissue, a portion of the same (the nucleus dis-
appearing) is changed into a gelatinous more or less fibrous texture,
which during normal ossification takes up lime salts almost at the
same time; of a certain proportion of these osteoblasts only the
peripheral part of the protoplasm is thus changed, what is left
remains behind as the nucleated bone corpuscle, imbedded in the
intercellular substance, like a connective tissue corpuscle in the
substance of tendon."

After describing the calcified cartilage and the changes it under-
goes, Waldeyer gives a description of the parts in which the first
deposition of bone takes place (the crypts of calcified cartilage with
the medulla), which seems to m#e true and to fit into my theory
fully as well or better than into his.

On p. 365 of the Archiv fur Mikroscopische Anatoraie, I, 1865,
he makes a statement which I cannot bring in harmony with his
Figure 2. He says: "At the time when the first bone substance
is deposited upon the cartilaginous framework, there is not the trace
of a separation to be observed between the osteoblasts and the
medullary tissue. This occurs later, wheu a very distinct stratum
of bone is deposited."

But on looking at the drawing we see a marked difference
between the fibrous tissue in the centre of the cavity and the layer
of "osteoblasts" lining the walls of calcified cartilage, no bone
having as yet been laid down there.

Waldeyer continues his argument thus: "It is not difficult to
ascertain here already the correctness of my view regarding the
formation of osseous tissue, as I expressed it above. While the
first bone substance is formed the medullary spaces are closely
filled with osteoblasts, there is no room left for any excretion,

DENTINE AND OSSEOUS TISSUE. 55

excepting that at the same time a number of osteoblasts perish, which
cannot be assumed, or a bone substance ought to be formed studded
so thickly with lacunae as it is never found to be the case. This
fact makes it to me very improbable that the ground substance of
bone is a mass excreted by the osteoblasts."

The examination of these regions has never roused this difficulty
in my mind, and if we again turn to the figures of Waldeyer
himself, we find no lack of space, there is amply, room in the
spaces produced by the openings of the cartilage cells for six
times as many cells as are figured — hence also for a thin coating
of bone.

Waldeyer continues : " One observes further that the peripheral
parts of the single osteoblasts are changed, loosing their darkly
granular appearance and applying themselves closely to the sinuous
walls of the medullary spaces. Other osteoblasts in the neighbor-
hood are in connection with these modified peripheral layers, they
also with their metamorphosed outer layers approaching the former.
The portion of protoplasm around the nucleus only remains un-
changed. I take this change of the peripheral strata for the ex-
pression of a metamorphosis into glue yielding substance, which at
once takes up the lime salts/'

Looking at the figures, I fail to see any indications of such
processes, especially can I not distinguish between unchanged
osteoblasts and such as are dwindling, their bodies being
changed into gelatinous substance. Heading this description one
would also think that the deposition of new osseous substance was
taking place all around the cells, around each individual cell, and
that the substance was immediately calcified ; in fact one would
expect a tissue somewhat like cartilage. But neither is the case,
neither the laying down of bone substance around individual cells
nor the immediate calcification. The new layer is in sections and
always forms an uncalcified seam, in which here and there a "cell"
is found.

Although there are disadvantages connected with the carmine
method, as above described, yet there are some facts brought out
by it very well; and the individual steps of the processes can I
think be better followed than by the chromic acid method. Using
chromic acid and decalcified tissue there are certain differences
necessarily obliterated, which the other method brings out, and
which are apt to make a strong impression.
6

56 C. SIHLER.

That the cells alter in character would be not easy to prove,
as it depends on very slight changes in size, and the greater or less
amount of granules, but if it should be the case, this fact would
favor the theory supported in this paper very well indeed, as I
shall show further on.

So does also the fact, that the cells in specimens which have
been brushed, are partly free, partly adherent to the bone by their
processes.

Waldeyer further points out, that the cells inclosed in the
osseous substance are smaller than the osteoblasts, and says: "If
we find however cells in formative action, it is difficult to conceive
how they can effectually perform their functions, and at the same
time undergo atrophy ."

This difficulty can easily be overcome by the examination of
gland cells which have been resting, and such as have been made
to secrete very actively; the former we find large, plump, with
sharp outlines; the latter small, shrunken, shrivelled, their out-
lines difficult to make out. We see here that cells which are
secreting actively shrink very markedly; and might such a change
not have been expected ?

I said above that it was not easy to distinguish with certainty
minute differences in size. To support this statement, I would
call attention to a figure, by Klein, in Sanderson's Handbook, a
transverse section of a femur from a human foetus, treated with
chromic acid ; I do not think that any one can perceive any differ-
ence in size between the cells lining the medullary spaces and
those inclosed in the lacunae.

To explain the process as taking place beneath the periosteum,
Waldeyer adduces a cross-section of a foetal Tibia, (Plate XXII,*
Figure 5) — at (a) we are to see osteoblasts. I have serious doubts,
however, if at this place, and others, where Waldeyer thinks deposi-
tion of new osseous tissue is taking place, the opposite is not occurring y
namely, the absorption of the bone. There is a good deal of evi-
dence that the latter is the case, — the jagged outline and character-
istic excavations, — while there is no evidence at all that formation
is just now taking place there. First there ought reasons to be
given that formation is going on there at all before the specimen
is used for demonstration. Treatment of the material with carmine

* Arch. f. Mik. Anat., 1865.

DENTINE AND OSSEOUS TISSUE. 5T

in the way described shows at once where excavation and absorp-
tion is going on, where deposition of new bone is going on, and
where the soft parts covering the osseous tissue are at rest.

The view here favored agrees with Kolliker's, excepting as
regards the formation of the canaliculi, and probably' agrees with
Gegenbaur's, (whose writings I have not had opportunity to ex-
amine) if I may form a judgment from scattered references.

DESCRIPTION OF PLATE.

Fioubb 1. — Section through root of calf's incisor, a, pulp ; 6, uncal-

cified dentine; c, dentine ; d, cementum ; e, uncalcified
cementnm ; /, pericementum.

Figure 2. — Section through pulp, odontoblasts and dentine; calf's

incisor, a, dentine ; b, uncalcified dentine ; c, odonto-
blasts; d, pulp.

Figure 3. — Odontoblasts from calf's tooth.

Figure 4. — Pulp-cells from same.

Figure 5. — Section through femur and periosteum ; kitten at birth.

a, fully formed bone; b and e, uncalcified bone ; c, layer
of cells forming bone ; d, outer periosteum.

Figure 6. — Osteoblasts ; cat's bone.

Figure 7. — Section through cementum and pericementum of calf's

tooth, a, cementum ; b, uncalcified cementum ; c,
cementum forming cells ; d, outer part of cementum.

Figure 8. — 8ection through femur, kitten 3-4 months, deeply stained.

All drawings except Figure 1 were made under Gundlach V. Oc. 3,
and reduced.

THE FIRST ZOEA OP PORCELLANA. By W. K.

BROOKS and E. B. WILSON. With Plates VI and VII.

Since 1835, when Thompson obtained the larva of a British
species of Porcellana from the egg, this very remarkable zoea has
frequently attracted the attention of natural ists, and we now have
quite an extensive list of papers, giving a satisfactory account of the
structure of the advanced zoea, and of its transformation into the
adult crab. The bibliography of the subject is given, at length,
in a recent paper by Faxon, (On some young stages in the devel-
opment of Hippa, Porcellana and Pinnixa. Bulletin of the
Museum of Comparative Zoology, at Harvard College, Vol. V, No.
11,) and it seems unnecessary to duplicate it here.

Most of the observers who have studied it started with the ad-
vanced zoea which is frequently captured with the hand net at the
surface of the ocean, and the few papers which notice the early
stages of the larva were published so long ago, that a minute
account of the young, as it leaves the egg, is still lacking.

During the latter part of June, 1880, we obtained, at the marine
laboratory of the Johns Hopkins University, at Beaufort, N. C,
a female specimen of Porcellana ocellata, Gibbes, with eggs, which
we succeeded in keeping alive and in good condition until the
eggs hatched, and we were thus supplied with an abundance of
material for studying the early stages.

As all the members of the party were at the time fully occupied
with other work, we undertook to study the larva together, and
to make as many notes and drawings of the early stages as
possible.

This paper is therefore the result of our combined observations,
but the work of copying the original drawings, and of preparing
the description has been done by W. K. Brooks. In the explana-
tion of the figures the author of the drawing which was copied is
named in each case, although in nearly every case, the accuracy of
the observation was verified by an independent drawing by the
other observer.
58

FIRST ZOEA OF PORCELLANA. 59

The larva immediately after its escape from the egg, is shown
in Plate VI, Figure 1. It is able to rise from the bottom and to
swim a little by flapping its abdomen, but until the next moult it
spends most of its time lying nearly motionless upon the bottom.

The carapace makes a little more than two-fifths of the total
length of the body, and is folded upon itself in such a way as to
form a well defined transverse band running across its dorsal
surface near the posterior edge. The posterior spines of the cara-
pace do not seem to be at all invaginated, but they are very much
convoluted and wrinkled, and their free extremities are bent
forwards under the posterior edge of the carapace. Between the
eyes the anterior end of the carapace forms a protuberant rounded
front, and the convoluted and wrinkled rostrum is bent down
towards the ventral surface. The eyes lie in deep notches on the
anterior edge of the carapace, and they appear to be movable,
although the stalks are very short.

The third pair of maxillipeds are small and rudimentary, while
the first, Mpj and second, JHp', pairs are well developed, although
their locomotor setse are not yet protruded, and the limbs are not
moved but remain constantly in the position which is shown in the
figure. The abdomen has five free movable somites, besides the
sixth which is not separated from the telson, T.

The pigment is more conspicuous at this time than during the
stages which follow, and consists of a number of pretty constant
bright red spots. One of them is on the basal portion, and one on
the flagellum of the second antenna, one on the mandible, M, one
on the basal joint of the first maxilliped, two on the basal joint of the
second and one on the third, as well as one about half way between
the base and tip of the secoud; there is a rong dendritic spot on
the posterior edge of the first, the second, the third, and the fourth
abdominal somite, and a pair of spots on the telson.

The whole surface of the body is covered by a delicate embryo-
nic cuticle, which is too transparent to be visible with the magni-
fying power under which Figure 1 was drawn. This cuticle
conforms to the outline of the body except on the two pairs of
antenna? and the telson. It will be described, in detail, later, in
the account of the appendages.

Some of the larvae free themselves from it within a couple of
hoars, and assume the form shown in Plate VI, Figure 5, while
others do not escape from it until nearly or quite twenty-four hours

60 W. K. BROOKS AND E. B. WILSON.

after they leave the egg. After this first moult the stalks of the
eyes, (see Figure 6), elongate, the fold at the posterior edge of the
carapace is stretched out so that the latter is now about half as
long as the whole body ; the rounded front disappears, and the con-
volutions and wrinkles of the rostrum and spine are no longer seen,
although these processes are still rolled up, as shown in the figure.
Figure 5 shows them as they appeared in the zoea which was
drawn, but the form of the bends is not at all constant.

The swimming hairs on the first and second maxillipeds, Mp>
Mp'} are extended, and these appendages, as well as the telson,
are now used as locomotor organs. Spines have now made their
appearance upon the posterior edges of the third, fourth and fifth
abdominal somites, and the rostrum and processes of the carapace
are covered with short hairs.

In from one to two days after hatching the rostrum and pro-
cesses become extended, as in Plate VII, Figure 8, and the zoea
assumes the familiar form which has been described and figured by
many observers.

The Appendages:

The first antenna of the newly hatched larva is shown in Plate
VI, Figure 2, and that of the fully developed zoea in Plate VII,
Figure 3.

In the first stage it is covered by the delicate embryonic skin,
which follows the outline of the appendage very closely, except at
the tip where it is produced into two long, broad, flattened, pointed
setse," which are fringed with smaller hairs. These structures,
which seems to be swimming hairs, are not present in the zoea
after the moult, but ki the first stage the antenna carries a single
stout sensory hair which, as shown in Plate VI, Figure 2, extends
into one of the swimming hairs, more than halfway to the tip. After
the moult, Plate VII, Figure 3, the appendage ends in a number of
long blunt sensory hairs, from the bases of which fine fibres run
downwards to a large club-shaped granular mass, which appears
to be ganglionic in nature.

The second antenna is shown before the moult, in Plate VII,
Figure 1, and after the moult in Plate VII, Figure 2. It is essen-
tially alike in both stages, but before the moult is loosely invested by
the embryonic skin, which is loose and much larger than the true
appendage. It consists of a swollen basal portion d, which carries
a short pointed external branch, and a longer internal branch.

FIB ST ZOEA OF PORCELLANA. 61

The mandibles and maxillae are shown before the moult in
Plate VII, Figure 7, and after the moult, in Plate VI, Figures 3,
4, and Plate VII, Figure 5.

In the first stage, Figure 7, Plate VII, these three appendages are
folded together, and covered by the embryonic skin which is nearly
conformable to their surface, although, as shown by the light outer
line in the fieure, it does not follow all the folds. No trace of a
palpus could be discovered on the mandible, and the hairs at the
tip of the maxillse were almost completely invaginated into the
appendages.

After the moult these three pairs of appendages become func-
tional, and have nearly the adult character. The mandibles,
Plate VII, Figure 5, and Plate VI, Figure 6, M> are not exactly
alike, but exhibit that slight departure from bilateral symmetry
so frequently found in these appendages. No trace of a mandibular
palpus could be found, although there was a small area where the
integument had been broken in each of the two specimens which
were dissected ; and as this area, shown in the figure, was at the
same place in both cases, the fracture may have been produced by
the removal of a palpus.

The first maxilla, Plate VI, Figure 3, and Figure 6, Mx9 consists
of a two-jointed basal portion, a, 6, with stout cutting hairs, and a
slender endopodite c, which in one specimen ended in two, and in
another specimen in three long, slender, irregularly plumose hairs.
The distal joint, 6, of the basal portion carries upon its cutting
edge, one row of five stout spines and a second row of four slender
8 pines parallel to the larger ones. In the specimen figured, the
proximal joint, a, was twisted so that its inner surface was shown,
and the posterior edge is therefore the one at the left of the figure.
It carries five long, stout, plumose spines, and at the posterior
angle of its cutting edge a single spine without secondary hairs.
No trace of an exopodite or scaphognathite could be detected in
this appendage.

The second maxilla, Plate VI, Figure 4 and Figure 6, Mx\ con-
sists of a three-jointed basal portion with short stout hairs; a two-
jointed endopodite, 4, with longer hairs ; and a long flat exopodite,

c, with five long hairs at its distal, and a long plumose flagellum,

d, at its proximal end.

In the first stage, the first and second maxillipeds, Plate VI,
Figure 1, Mp9 Mp'y are fully developed, although the presence of

62 W. E. BROOKS AND E. B. WILSON.

the embryonic skin prevents the extension of the locomotor
hairs.

In Figure 1, the rudimentary third maxilliped is shown behind
the base of the second.

In Plate VII, Figure 4, the third maxilliped, c, is shown, more
highly magnified, lying in the same series with the bases, a and 6,
of the first and second. A fourth appendage, no doubt the first
pereiopod, is also represented at this stage by a bud or rudiment,
rf, and the appendages, 6, o, and d, are furnished with little buds,
which would seem to be rudimentary gills. After the moult we were
not able to detect either the appendage, d, or the gill-like processes.

After the embryonic skin is moulted, the locomotor hairs of the
first and second maxillipeds lengthen and these appendages become
functional, while the third pair remain rudimentary. Figure 6,
Plate VII, shows the first and second maxillipeds soon after the
moult, and hardly calls for explanation.

The embryonic skin conforms closely to the surface of the ab-
domen and telson, although it appears to have no trace of a
division into somites.

Figure 7 of Plate VI shows one-half of the telson of Figure 1
before the embryonic skin is shed. A comparison with Figure 6,
T9 will show that the great difference which has been pointed out
by Faxon and others between the telson of the embryonic skin and
that of the zoea in the ordinary crab, does not occur in Porcellana,
but that the two are here nearly alike.

The five pairs of long swimming hairs of the zoea are, before the
moult, about half invaginated, and the extended portion, Plate VI,
Figure 8, is finely plumose. The hairs of the embryonic cuticle
are much stouter, and their edges are not plumose, but they agree
with those of the zoea, in number and arrangement.

The outer hair, or marginal spine of the telson, has the same
appearance before the moult that it has afterwards.

EXPLANATION OF THE FIGURES.

PLATE VI.

Figure 1. — Zoea immediately after its escape from the egg, seen from

the left Bide. From a drawing by W. K. Brooks.

FIRST ZOEA OF PORCELLANA. 63

Figure 1. — Continued.

At first antenna; An, second antenna; Mf mandible;
Mp, first maxilliped; Mp', second maxilliped; R,
rostrum ; T, telson.

Figure 2. — First antenna of the same larva, more highly magnified.

From a drawing by W. K. Brooks.

Figure 3. — First maxilla of the larva shown in Figure 5. From a

drawing by W. K. Brooks.

a, proximal joint of basal portion; b, distal joint of
basal portion; c, endopodite.

Figure 4. — Second maxilla of the larva shown in Figure 5. From a

drawing by W. K. Brooks.

a, three-jointed basal portion; 6, two-jointed endo-
podite; c, scaphognathite ; d, flagellum.

Figure 5. — Zoea, seen from the right side, immediately after moulting

the embryonic skin. From a drawing by E. B. Wilson.
A, first antenna; An, second antenna; Mp} first max-
illiped; Mp' , second maxilliped; R, rostrum.

Figure 6. — Ventral view of the same zoea, one day after moulting the

embryonic skin. From a drawing by E. B. Wilson.
A, first antenna; An, second antenna; L, labrum, M,
mandible; Mp, first maxilliped; Mp' , second maxilli-
ped; Mx, first maxilla; Mx\ second maxilla; R, ros-
trum ; T, telson.

Figure 7. — Dorsal view of right half of telson of the larva shown in

Figure 1. From a drawing by E. B. Wilson.

Figure 8. — One of the setae of Figure 7, more highly magnified.

From a drawing by E. B. Wilson.

PLATE VII.

Figure 1. — Second antenna of the larva shown in Plate VI, Figure 1.

From a drawing by W. K. Brooks,
a, embryonic skin ; 6, external branch ; c, internal
branch ; d, enlarged basal joint.

Figure 2. — Second antenna of the zoea shown in Plate VI, Figure 5.

From a drawing by W. K. Brooks. Letters of refer-
ence as in Figure 1.

Figure 3. — First antenna of the zoea shown in Plate VI, Figure 5.

From a drawing by W. K. Brooks.

64 W. K. BROOKS AND E. B. WILSON.

Figure 4. — Basal joints of the maxillipeds of the lar?a shown in Plate

VI, Figure I. From a drawing by W. Brooks.
a, base of first maxilliped ; 6, base of second maxilli-
ped; c, third maxilliped; d, first pereiopod; e, edge
of carapace.

Figure 5. — Mandible of the zoea shown in Plate VI, Figure 5. From

a drawing by E. B. Wilson.

Figure 6. — First and second maxillipeds of the zoea shown in Plate

VI, Figure 5. From a drawing by W. K. Brooks.

Figure 7. — Mandible and maxilla of the larva shown in Plate VI,

Figure 1. From a drawing by W. K. Brooks.
M, mandible; Mx, first maxilla ; Mx', second maxilla.

Figure 8. — Side view of the zoea, one day after moulting the embryo-
nic skin. From a drawing by £. B. Wilson.

ERRATA.

Page 60. bottom line,
fir "external," read "internal."
for " internal," read " external."

THE STUDY OP HUMAN ANATOMY, HISTORI-
CALLY AND LEGALLY CONSIDERED.1 By

EDWARD MUS8EY HARTWELL, M. A., Fellow of the
Johns Hopkins University.

Part First.

" Practised architects, before they venture in thought to build a
new edifice, to strengthen an old one, or restore a ruined one, first
consider carefully and examine closely all the minute parts of such
structures. So, physicians, indeed, before they endeavor to care
for the human body and preserve it from the diseases which
threaten it, ought to know very accurately, and to a nicety, all the
parts of that body. Anatomy, the eye of medicine, furnishes such
knowledge. Verily, the beginnings, the foundations, and the
sources of origin of the medical art are, without the light and
vision of anatomy, shrouded in thick darkness; wherefore, it is
not inaptly called by Johannes Montanus, the alphabet of medi-
cine." So wrote Rolfincius, in his " Dissertationes Anatomic©,"
published at Nuremberg, in 1656.

When we of to-day seek the origin of this "alphabet of medi-
cine," we turn to the East, whence we are accustomed to derive
the beginnings of all our arts; but we find the history of ancient
anatomy to be almost a blank page. Priest, and law-giver, and
people were all averse to anything like the dissection of the
human body. The Egyptians, Hebrews, Greeks, Romans, and
Arabs, alike regarded with abhorrence the mutilation of the dead.
There is abundant proof of this in their laws and customs touching
burial and defilement.

It is said that Democritus, of Abdera (460 B. C), the friend of
Hippocrates, was the first to dissect the human body. However

1 Portions of the following paper have been printed already in the Journal
of the American Social Science Association; the Boston Medical and Surgical
Journal and the Brooklyn Annals of Anatomy and Surgery. In its present
form it contains much new material ; and embodies the result of the latest
•statistics and most recent legislation so far as I could ascertain.

66 E. M. HABTWELL.

that may be, it is as the Laughing Philosopher, and not as the
Father of Anatomy, that be has influenced mankind. It was in
what we fondly call " Egyptian darkness," and through the favor
of an enlightened despot, that the first school of anatomy was
founded at Alexandria, three hundred years before Christ, by
Ptolemy Soter. "Braving," says Bouchut, "all prejudices, and
considering that the interests of science ought always to outweigh
those of the individual, Ptolemy authorized the dissection of
human dead bodies, and himself set the example by beginning to
dissect with the physicians gathered around him." Herophilus,
and Erasistratus, his pupil, made the school of Alexandria famous
and influential ; their contributions to anatomy were genuine and
considerable. No name worthy of mention, beside theirs, is to be
found in the history of anatomy, until we come to that of Mondino,
Professor at Bologna, who first publicly dissected in Europe, early
in the fourteenth century. Yet, in the interval between the deca-
dence of the Alexandrian school, which followed hard upon the
death of its founders, and the rise of the Italian schools of anatomy,
Aristotle, Galen, Celsus, and the Arabists, lived and wrote.
George Henry Lewes declares that "Aristotle has given no single
anatomical description of the least value." Daremberg, Galen's
editor and translator, who says he has repeated every one of
Galeu's dissections, is convinced that he used only the lower
animals. Celsus expressed himself as a believer in the utility of
human dissection. The medicine and surgery taught by the
Arabs, at least so far as its anatomy was concerned, was borrowed «
from the Greeks.

Previously to the rise of human anatomy in Italy; Galenism,
founded on the dissection of the lower animals, notably the ape,
dominated the known medical world. Galen had written his "De
Usu Partium Aniraalium," as a prose hymn to the Deity. The
hierarchy commended his system which was upheld as scientific
orthodoxy, alike by political and religious authority; all research
capable of contradicting his views was condemned. The first
Italian anatomists were quite content to expound Galen. One of
the Arabists, Abdollaliph, criticised the slavish dependence of his
contemporaries on books. He commended those who, like him-
self, repaired to burial grounds to study the bones of the dead;
but he seems never to have dreamed that anything could be
learned from a like scrutiny of the soft parts.

THE STUDY OF HUMAN ANATOMY. 67

Galenism died hard, even in Italy where it was first attacked.
How tenacious it was of life is well shown by Malpighi, who was
born in 1628, the year that Harvey first published his "Essay on
the Motion of the Heart and Blood." Harvey never saw the
passage of the blood through the capillaries; Malpighi discovered
those vessels and first demonstrated the flow of blood from the
arteries into the veins. Malpighi writes: "In the meantime, con-
tentions being raised among studious men, especially the younger,
both practical and theoretical, and the new doctrines growing daily
into more credit, the senior professors of Bologna were inflamed to
snch a pitch, that, in order to root out heretical innovations in
philosophy and physic, they endeavored to pass a law whereby
every graduate should be obliged to take the following additional
clause to his solemn oath on taking his degree, viz: 'You shall
likewise swear that you will preserve and defend the doctrine
taught in the University of Bologna, namely, that of Hippocrates,
Aristotle, and Galen, which has now been approved of for so many
ages; and that you will not permit their principles and conclusions
to be overturned by any person, as far as in you lies.'" "But/'
says Malpighi, "this was dropped and the liberty of philosophizing
remains to this day."

Practical anatomy was taught at Padua it is said, as early as
1151 ; Haeser, in the third edition of his Geschicte der Medecin
says that: " in the year 1238, Kaiser Frederick II. ordered, at the
suggestion of Marcianus, chief physician of Sicily, that every five
years a corpse should be dissected publicly, and that physicians
and surgeons should be admitted, according to their rank, to the
dissection." It is elsewhere stated that Frederick forbade, in the
code for Sicily, any one to practice surgery unless he had been in-
structed in anatomy. There is no dispute, however, that Mondino
publicly dissected two subjects as early as 1315; and. some writers
give 1308 as the date.

We find many bulls of Popes and canons of Councils regarding
the study and practice of physic and surgery by monks; from the
time of the Council of Laodicea, in 366 A. D., when the priest-
hood were forbidden to study enchantment, mathematics, and
astrology, and the binding of the soul by amulets, till 1215, when
Pope Innocent III. is said to have fulminated an anathema against
bloody operations in surgery. Although these utterances of the
Church are interesting, we pass them by as being outside the scope
of this paper.

68 & M. HARTWELL.

The edict of Boniface VIII., however, published in 1300,
affected the progress of practical anatomy, and is worthy of note.
In 1299, Pope Boniface VIII. forbade, under pain of excommuni-
cation, any one to boil, cut up, or dry the bodies of the dead.
Such an act he characterized as barbarous and abhorrent to
Christian piety. Raynaldus, in whose " Annates Ecclesiastici,
Lucae, 1749," the edict of Boniface is found, says that such cus-
toms "had prevailed in regard to those who, having undertaken a
pilgrimage to the East, died in foreign parts; and in order that
their bones might be freed from flesh, and so easily carried about
without the fear of corruption. And yet we know," he adds,
" that the body of Saint Luke was boiled by his friend8.,, It is
hardly probable that Pope Boniface directed this edict primarily
against anatomy. Edward I., of England, directed that the flesh
should be boiled from his bones and that they should be carried to
battle in a bag by his successor, in order to terrify his enemies.
The story of Douglas and the heart of Robert Bruce is familiar
to all. It is quite likely that Boniface launched his anathema iu
order to restrain such practices as these; nevertheless, his edict
proved an obstacle to anatomical studies. Mondino apologizes
for not making the most exact study of the bones of the skull,
saying : " the bones beneath the basilar bone are not to be clearly
distinguished, unless they be boiled'; a sin which I have .been
accustomed to shun." Hyrtl, the famous German anatomist,
holds that the edict of Boniface was in force till 1556, when the
Emperor Charles the Fifth, the patron of Vesalius, ordered the
question to be put to the theologians of the University of Sala-
manca, " Whether or not it could be allowed, without violating
one's conscience and incurring the suspicion of criminality, to cut
up human dead bodies ? " " Et respondisse Universtiatem, Licere"
says Rolfincius, quoting a still earlier writer.

That dissection was not universally banned by the Church before
the Divines of Salamanca pronounced it lawful, may be seen from
the action of Pope Sixtus IV., in 1 482. In that year, in a letter
addressed to the rector, doctors, and students of the University of
Tubingen, Sixtus granted a special and full dispensation to those
who should receive the cadavera of certain malefactors executed
for capital crimes in accordance with the civil law: u per justitiam
secularem" is the phrase in the original. They were given per-
mission to dissect and dismember these dead bodies, inasmuch as

THE STUDY OF HUMAN ANATOMY. 69

they desired thereby to render themselves learned and skilful in
the art of medicine, provided they would bury in the customary
manner such condemned men after they should be dissected and
dismembered.

The Grand Council of Venice, in 13Qg, passed a decree ordering
the medical college of that city to undertake a dissection once a
year. It is claimed that in Prague, as early as the foundation of
the University in 1348, the executioners were enjoined to deliver
the cadavera of malefactors to the school of medicine. Duke
Albrecht IV. imported an Italian anatomist, named Galeazzo, to
introduce the art of dissection into Vienna ; where the first anatom-
, ical demonstrations before the medical faculty were made in 1404.

In France, as early as 1376, Louis of Anjou permitted the sur-
geons of Montpellier to take the body of an executed criminal
annually for dissection. Charles the Bad. King of Navarre and
Lord of Montpellier, ratified this grant in 1377; as did King
Charles VI. in 1396 ; and King Charles VIII. in 1484, and again
in 1496. A similar grant was made in Scotland in 1505, as we
learn from the following extract taken from the Charter given by
the Town Council of Edinburgh to the Surgeons' Company, July
1, 1505, and ratified by King James IV. in the following year:
" And als that everie man that is to be maid frieman and maister
amangis ws be examit and p^evit in thir poyntis following Thatt
18 to say That he knaw anatomea nature and complexioun of
every member In manis bodie. And in lyke wayes he knaw all
the vaynis of the samyn thatt he may mak flewbothomea in dew
tyme. And als thatt he knaw in quhilk member the signe hes
domination for the tyme for every man aucht to knaw the nature
and substance of every thing thatt he wirkis or ellis he is negligent.
And that we may have anis in the yeir ane condampnit man after
he bedeid to mak anatomea of quhairthrow we may haf experience
Ilk ane to instruct others And we sail do suffrage for the soule."
By Act of Parliament, 32 Henry VIII., cap. 42, in 1540, it was
granted to the Barber-Surgeons of London to take " yearly forever

four persons condemned, adjudged and put to Death for

Felony by the due Order of the King's Highness, and to

make Incision of the same dead Bodies, or otherwise to order the
aarae after their said Discretions at their pleasures, for their further
and better Knowledge, Instruction, Insight, Learning, and Expe-
rience in the said Science or Faculty of Surgery."

70 E. M. HABTWELL.

Vesalius, a Fleming, born in 1514, did more than all bis prede-
cessors to overthrow Galenism and place medicine upon a rational
basis, and well deserves his title of the Father of Modern Anatomv.
Yet, despite the concessions we have noticed made by prelates,
kings and parliaments to, the early anatomists, Vesalius and his
students were obliged, in the words of Hallam, " to prowl by night
in charnel-houses, to dig up the dead from the grave ; they climbed
the gibbet in fear and silence to steal the mouldering carcass of the
murderer at the risk of ignominious punishment and the secret
stings of superstitious remorse." Vesalius began to dissect while
a youth in his teens. For a time he studied under the famous
French anatomist, Jacques Du Bois, who demonstrated the anatomy
of Galen on the carcasses of dogs. But Vesalius forsook Paris
for Italy, drawn thither by the reputation of the schools whence
Leonardo da Vinci and Michael Angelo derived their knowledge
of human anatomy. Before he was twenty -eight, as has been well
said, " Vesalius discovered. a new world," and held at one time the
professorship of anatomy in the universities of Pisa, Padua and
Bologna. He died the victim of the Spanish Inquisition. His
inspection, with the consent of the relatives, of the body of a
Spanish grandee, whose heart feebly contracted under the knife,
brought him before the Inquisition, and would have led him to the
stake but for the intercession of the King. Compelled to journey
to Jerusalem by way of penance, Vesalius was shipwrecked, in
1564, on the island of Zante. It is said that' he there starved to
death, and that unless a liberal goldsmith had defrayed the funeral
charges, the remains of the greatest anatomist the world had seen
would have been devoured by birds of prey.

The Italian schools under Vesalius and his successors, Fa 1 lop i us,
Columbus and Fabricius, exerted a wide and potent influence
upon European medicine. This influence was sooner felt and
more marked in France, Germany and Holland than in Eng-
land and Scotland. The following statements, made by Billroth,
may serve to indicate the favor in which anatomy was held in
Germany :

In the Privilegia granted by the Landgrave Wilhelm von
Hessen to the University of Marburg, in 1653, it is provided that
" in the medical faculty at the start there shall be two doctors in
pay, who, in addition to the theory, shall conduct the practice of
anatomy and of botany with the youth." The statutes of the

THE STUDY OF HUMAN ANATOMY. 71

medical faculty at Marburg for the same year, Title IV, read as
follows : —

"(1.) It is clear that anatomy, next after psychology, forms the
chief part of universal physiology. Since there is a twofold
method of teaching it, one that is ordinarily practiced in anatomical
theatres in the presence of many spectators, and the other which is
employed by the holders of scholastic chairs, let neither of them be
intermitted. Let both of them, as well publicly as privately, be
practiced.

"(2.) Let also' the art of dissection and of skillfully handling
and applying the knife in individual parts be shown, in order that
a difference may be noted between physical and medical or practical
anatomy. The various skeletons, also, both male and female, of
common and exotic animals shall be prepared, in order that not
only the structure of the skeleton, but also the whole of oste-
ology, may become known to students of medicine as well as of
surgery.

"(3.) Let pregnant women be dissected as well as others. Let
mid-wives as well as others be admitted.

"(4.) Let not those who are condemned to death be opened
alive, but let living things of every kind, as insects, serpents,
aquatic animals, birds, and quadrupeds, be dissected. Especially
let those studying anatomy observe, more precisely than butchers
would, domestic quadrupeds while they are being slaughtered.

"(5.) Moreover, let the bodies of atrocious criminals, whether
they have been beheaded or hanged, be designated for dissection.
Let them not be kept back by the magistracy when they are
sought for this purpose, in order that those who have done as
much evil as they could when alive, may, after death, on the other
hand, be of as much service and use as possible/'

We shall confine our attention chiefly to the history of anatomy
in Great Britain; inasmuch as in the development of anatomy in
America, the influence of Edinburgh and London is more readily
traced than that of Paris and Leyden.

Twenty-five years after the passage of the Act of 32 Henry
VIII., Queen Elizabeth granted to the College of Physicians, of
London, the bodies of four felons executed in Middlesex, " that the
president or other persons appointed by the college might, observ-
ing all decent respect for human flesh, dissect the same." In
1663, Charles II. increased the number of felons' bodies, annually
8

72 E. M. HABTWELL.

granted to the physicians, to six. The Act of 22 George II., c.
37, 1752, required the dissection or hanging in chains of the
bodies of all executed murderers in order that " some further Terror
and peculiar Mark of Infamy might be added to the Punishment
of Death." The provision of this Act regarding the dissection of
murderers remained unrepealed till the passage of the so-called
Warburton Anatomy Act, in 1832, while the provision regarding
the hanging of a murderer's body in chains remained in force till
1861, when it was repealed.

These were the only legalized sources for the supply of anatomi-
cal material in England prior to 1832. Such provisions might, at
first sight, seem generous and ample, yet they were not. We find
Dr. William Hunter, in 1763, in vain asking of the King a grant
of land sufficient for the site of an anatomical school in London,
which he proposed to endow with something like £7,000, and one
of the finest anatomical collections in Europe. In his memorial
to the Earl of Bute, Hunter writes: "Of the very few who profess
or teach this art iu any part of Great Britain, London excepted,
there are none who can be supplied with dead bodies for the
private use of students. They can with difficulty procure only so
many as are absolutely necessary for the public demonstrations of
the principal and well-known parts of the body. Hence it is that
the students never learn the practical part, and the teachers them-
selves can hardly make improvements, because they cannot have
subjects for private experiments and enquiries. Anatomy was not
upon a much better footing, even in London, till the year 1746."

In 1832, Parliament passed the Warburton Anatomy Act,
which is still in force throughout Great Britain and Ireland — in
all its essential features. To understand its significance and that
of "Burking," which really caused Parliament to enact it; we
must glance at the Edinburgh School of Anatomy.

We have already noticed the grant of anatomical material con-
tained in the charter of the Surgeons' Company, made in 1505.
The beginning of the Edinburgh Anatomical School was in 1694;
when the Town Council, on the 24th of October, in response to
the petition of Alexander Monteith, granted him "any vacant,
waste room in the correction house, or any other thereabouts
belonging to the Town." Monteith also obtained a grant of
"those that dye in the correction house; and the bodies of fund-
lings that dye upon the breast." The Surgeons' Company were

THE STUDY OF HUMAN ANATOMY. 73

granted, nine days later, "the bodies of fondlings who dye betwix
the tyme that they are weaned and their being put to schools or
trades; also the dead bodies of such as are stiflet in the birth,
which are exposed and have none to owne them; as also the dead
bodies of such as are/elo de se and have none to owne them; like-
waves the bodies of such as are put to death by sentence of the
magistrat, and have none to owne them." Certain interesting
conditions were attached to the grants to Monteith and the Sur-
geons. The dissection was to be during the winter, from one
equinox to the other; all the "gross intestines" were to be buried
within forty-eight hours, and the rest of the body within ten days
at the grantees' expense. The regular apprentices of the Surgeons
were to be admitted at half price, and any magistrate who thought
fit might attend the dissection. In the grant to the Surgeons, no
mention is made of the gross intestines, according to Dr. J.
Struthers, from whose sketch of the Edinburgh Anatomical
School these facts are taken ; but it is provided " that the petitioners
shall, before the terme of Michaelmas 1697 years, build, repair, and
have in readiness, ane anatomicall theatre where they shall once
a year have ane public anatomicall dissection, as much as can be
showen upon ane body, and if the failzie thir presents to be void
and null." The Anatomical Theatre of the Surgeons was reported
finished to the Town Council, December 17, 1697. The Council
ratified its grant of 1694, and, the same day, the Surgeons chose
a committee "to appoint the method of public dissections, and the
operators." In 1705, the Council gave <£15 salary to Robert
Elliott, the first Professor of Anatomy in Edinburgh. In 1720,
the Town Council elected Alexander Monro, primus, Professor of
Anatomy. In 1725, he removed from Surgeons' Hall to the
University buildings, because of the violence of a mob which had
attempted to demolish the Surgeons' Theatre, on account of the
supposed violation of graves. In 1722, the apprentices of the
Surgeons' Company were obliged, in their indentures, to subscribe
to "an obligation that they would altogether avoid raising the
dead."

Under the Monros, father, son and grandson, who held between
them the University Chair of Anatomy from 1720 till 1846, the
school became widely famous. Many of the early American phy-
sicians and anatomists studied at Edinburgh; where, early in this
century, there were several extramural private schools of anatomy.

74 E. M. HARTWELL.

Of these, ' that of Dr. Robert Knox was the most famous and
frequented. In the winter of 1828-29, he bad a class of 505: the
largest in Europe.

For years the demand for anatomical material had exceeded the
legal supply in Great Britain. As early as 1826, Parliament was
petitioned, but in vain, to give aid and protection to the anato-
mists,— who were forced to depend on the resurrection men for
subjects. Bodies often brought £10 each, in Edinburgh and
London; in one instance a subject was sold for £30. When the
home supply ran short, the Scotch anatomists were furnished with
stolen bodies from England, Ireland, and even France. "The
increased demand and higher pay for material/' says Lonsdale,
(Knox's biographer), "generated sad recklessness and brutality.
Quarrels arose over the spoils; the jealousy of rival factions of the
different schools, and the frequent attempts to outwit each other,
led to personal denunciations and a fearful publicity." In response
to numerous petitions from the medical profession, a "Select Com-
mittee of the Commons/' to inquire into the hindrances to the
study of anatomy, was appointed April 22, 1828. Its report was
rendered on the twenty-second of July, following. In 1788, the
Court of King's Bench decided, in the first reported case of the
sort, that it was a misdemeanor at common law to carry away a
dead body from a church-yard, although for the purpose of dissec-
tion, as being an offence contra bonos mores and common decency.
The Select Committee stated in its report, which was favorable to
the petitioners, that, under the law as then interpreted, there was
scarcely a student or teacher of anatomy in England who was not
indictable for a misdemeanor ; and also that medical men " were
liable in a civil action to damages for errors in practice, due to
professional ignorance; though at the same time they might be
visited with penalties as criminals for endeavoring to take the only
means of obtaining professional knowledge." It was not until the
following year, when the complaints of the anatomists and the
report of the committee had been emphatically endorsed by the
"Burking" horrors of Edinburgh, that leave was obtained, on the
fourth of May, to bring in a " Bill, to Prevent the Disinterment of
Dead Bodies, and for the Better Regulation of Our Schools of
Anatomy."

On the second of November, 1828, it was noised about in Edin-
burgh that a woman had been murdered on All Hallow Eve for

THE STUDY OF HUMAN ANATOMY. 75

the sake of her body, which was found in the dissecting room of
Dr. Knox. In the investigation which followed it was discovered
that William Hare, the keeper of a low lodging house in the West
Port, and one of his lodgers, William Burke, had, within less than
a year, committed sixteen murders, and disposed of the bodies of
their victims to the teachers of anatomy. The " Burke" method
was to suffocate the victim, already dead drunk. Throttling was
not resorted to: the nose and mouth were kept tightly closed, and
the smothering was soon effected. It was impossible to connect
Knox with these villians in any way, except as a receiver of stolen
goods for the benefit of the public. Hare turned State's evidence,
but Burke was found guilty, hanged and dissected. His skeleton
adorns the Anatomical Museum of the University of Edinburgh.

The Bill alluded to above was brought into Parliament May 5,
1829, but was thrown out in the House of Lords a month later.
It was not until August 1, 1832, after a long discussion in which
Sir James Mackintosh and Mr. Macau lay took part, that the
" Warburton Bill for Regulating Schools of Anatomy " was enacted.
At this distance in space and time the deliberateness of Parliament
seems a trifle strained in the face of such facts as we have stated ;
but one of the chief glories of the British Constitution is its slow
growth, we believe.

The Warburton Act is, with some trifling amendments, still in
force. Its effect has been to protect the sepulchres of the dead
and, in the long run, to furnish an adequate supply of subjects.
As, however, Massachusetts anticipated Great Britain by more
than a year in legalizing anatomy, in a law based upon the same
principles as those embodied in the English Act, we forego any
special consideration of the terms and provisions of the latter.

Part Second.

ANATOMY IN AMERICA.

European and American anatomy have both developed along
the same lines, but the European type is more highly specialized.
Nearly all the developmental stages through which European
anatomical science has passed are to-day represented in various
States of the American Union. In some States it is a secret and
perilous pursuit; in others it has gained legal protection ; in a few
it has attained, perhaps, to the dignity of an ungenerously fostered
science.

76 E. M. HARTWELL.

The earliest utterance in America, in recognition of the import-
ance of anatomical studies, seems to have been made in Massachu-
setts. In " The Cleare Sun-Shine of the Gospel Breaking upon
the Indians in New England " is found a letter dated " Roxbury,
24 September 1647," from John Eliot to the Rev. Thomas Shep-
hard of " Cambridge in New England." The Apostle declares of
the Indians that "all the refuge they have and relie upon in time
of sickness is their Powwaws, who, by an tick, foolish and notional
conceits delude the poor people, so that it is a very needfull thing
to informe them in the use of Physick, and a most effectual 1 meanes
to take them off from their Powwawing. Some of the wiser sort
I have stirred up to get this skill ; I have showed them the Anat-
omy of man's body, and some generall principles of Physick. I
have had many thoughts in my heart that it were a singular good
work, if the Lord would stirre up the hearts of some or other of
his people in England to give some maintenance toward some
Schoole or Collegiate exercise this way, wherein there should be
Anatomies and other instructions that way." It is unlikely that
the Apostle Eliot added dissections to his lectures on "the Anat-
omy of man's body ; " for later in the same letter he deplores the
fact that " our young students in Physick have onely theoretical 1
knowledge, and are forced to fall to practice before ever they saw
an Anatomy made," and says, " We never had but one Anatomy
in the Countrey, which Mr. Giles Firman (now in England) did
make and read upon very well."

The "first Anatomy in the Countrey" was doubtless made
without the warrant of legal enactment; certainly the majority of
dissections since then have been so made. The first statutory
provision regarding anatomy in America seems to be the Massa-
chusetts Act of 1784, by the terms of which the bodies of those
killed in duels and of those executed for killing another in a duel
might be given up to the surgeons "to be dissected and anato-
mized." In 1831 Massachusetts anticipated all her sister States,
and England as well, by legalizing the study of "anatomy in
certain cases."

In the Diary of Samuel Sewall, of Boston, recently published
by the Massachusetts Historical Society, is found under date of
September 22, 1676, the following entry : " Spent the day from 9
in the M. with Mr. [Dr.] Brakenbury, Mr. Thomson, Butler,
Hooper, Cragg, Pemberton, dissecting the middle-most of the

THE STUDY OF HUMAN ANATOMY. 77

Indians executed the day before. X, who taking the fp in hand,
affirmed it to be the stomach."

The earliest reference that I have found to a post-mortem ex-
amination in America is contained in a manuscript order of the
Council of Lord Baltimore, dated St. Mary's, in Maryland, July
20, 1670. In it John Stansley and John Peerce, Chyrurgeons,
are ordered to view, on Monday, August 8, 1670, the head of one
Benjamin Price, supposed to have been killed by the Indians. It
was brought out in connection with the Salem witchcraft trials, in
1092, that "about seventeen years before," a jury had been impan-
elled upon the body of a man that had died suddenly in the house
of Giles Corey, and that the jury, among whom was Dr. Zerub-
babel Endicott, found the man " bruised to death, and having
dodders of blood about the Heart/' This would indicate that a
post-mortem examination was made in Massachusetts as early as
1675, fifteen years prior to that made on the body of Governor
Slaughter, of New York, which is usually cited as the first recorded
autopsy in America. In 1690, Governor Slaughter died suddenly,
under circumstances which excited suspicions of poisoning. Dr.
Johannes Kerf by le, assisted by five physicians, examined the body.
The Council ordered £8 8 s. to be paid the surgeons for their ex-
amination.

It is recorded that Dr. John Bard and Dr. Peter Middleton, of
New York city, in 1750 injected and dissected the body of Her-
manus Carroll, an executed criminal, " for the instruction of the
young men then engaged in the study of medicine." This was
thirty-nine years before the State of New York legalized the dis-
section of the bodies of malefactors executed for arson, burglary,
or murder. Though Pennsylvania passed no anatomy Act until
1867, the first American medical school was organized in Phila-
delphia in 1765, by Drs. Morgan and Shippen, natives of that city.
Dr. William Shippen, Jr., a pupil of John and William Hunter,
gave, in 1762, a systematic course of lectures on anatomy. This
first course of lectures by Dr. Shippen, is usually termed the first
full and scientific course of anatomical lectures given in America ;
although Dr. Cadwallader, as early as 1751, made dissections for
the benefit of the physicians of Philadelphia, and Thomas Wood,
surgeon, in 1752 advertised in the New York papers "a course on
osteology and myology in the city of New Brunswick, N. J.," to
be followed, in case of proper encouragement, by a course in angi-

78 E. M. HARTWELL.

ology and neurology, and a course of operations on the dead body.
It should also be noted that Dr. William Hunter, educated at
Edinburgh under the elder Monro, who came to America in 1752,
gave lectures on anatomy and surgery in Newport, R. I., in the
years 1754, 1755, and 1756.

Shippen's courses were so successful that in 1765 the Medical
College of Philadelphia was organized with two professorships.
Dr. Shippen held the chair of " anatomy and surgery ; " that of
the "theory and practice of physic" was filled by Dr. John
Morgan.

A brief consideration of the character and career of Dr. William
Shippen, Jr., the first Professor of Anatomy and Surgery in
America, may well detain us for a few moments. His father, Dr.
William Shippen, was an eminent physician in Philadelphia, in
which city the son was born, in 1786. Young Shippen graduated
in 1754, at the College of New Jersey, of which institution his
father was one of the founders. After studying medicine for three
years with his father, he repaired to Europe, where he studied at
Edinburgh and London. He returned to Philadelphia in 1762,
in which year, at the age of twenty-six, he gave his first course of
lectures on anatomy. One of his successors in the chair of anatomy
— Dr. W. E. Horner — says: "Dr. Shippen seems to have been
intended by nature to lay the corner-stone of the immense edifice
of medicine, which has since been erected in this country. Aged
twenty-six, at the period alluded to, uncommonly perfect in his
form and engaging in his aspect ; his manners were those of a
finished gentleman; his enunciation was fine; his temper invari-
ably sprightly and good, could neither be excited by rancor, nor
rendered sullen and morose by opposition. To the personal ad-
vantages stated, and those of extensive hereditary friendship and
family alliance, Dr. Shippen added foreign study — at that day all
important in public estimation, from the want of opportunities of
instruction here. While in London he lived in the family of Mr.
John Hunter, the celebrated surgeon, and followed the lectures of
Dr. William Hunter on anatomy and mid-wifery. He enjoyed
the advantages of great intimacy with Sir John Pringle and Dr.
Fothergill. To the incentive of such illustrious associations we
may attribute much of the energy and determination which marked
his subsequent career. Dr. Shippen arrived in Philadelphia in the
Spring of 1 762, having completed his studies and gained from his
preceptors the reputation of great natural talents."

THE STUDY OF HUMAN ANATOMY. 79

In the Pennsylvania Gazette, published by B. Franklin, Poet-
master, and D. Hall, November 11, 1762, I find a card from Dr.
Shippen which, inasmuch as I cannot find that it has been repub-
lished, I venture to quote as a whole :

Philadelphia, November 11.
Mb. Hall. Sir:

Please to inform the Public that a Course of Anatomical Lectures
will be opened this Winter in Philadelphia for the Advantage of the
young Gentlemen now engaged in the Study of Physic in this and the
neighboring Provinces, whose Circumstances and Connections will not
admit of their going abroad for Improvement to the Anatomical Schools
of Europe ; and also for the entertainment of any Gentlemen who may
have the Curiosity to understand the Anatomy of the Human Frame.

In these Lectures the Situation, Figure and Structure of all the
parts of the Human Body will be demonstrated; their respective uses
explained, and, as far as a Course of Anatomy will permit, their
Diseases, with the Indications and Method of Cure, briefly treated of;
all the necessary Operations in Surgery will be performed, a Course of
Bandages exhibited, and the whole conclude with an Explanation of
some of the curious Phenomena that arise from an examination of the
Gravid Uterus, and a few plain general Directions in the Study and
Practice of Midwifery.

The Necessity and public Utility of such a Course in this growing
Country, and the Method to be pursued therein, will be more particu-
larly explained in an Introductory Lecture, to be delivered the 16th
Instant, at six o'clock in the Evening, at the State House, by William
Shippen, jun., M. D.

N. B. The Managers and Physicians of the Pennsylvania Hospital,
at a Special Meeting, have generously consented to countenance and
encourage this undertaking; and to make it more entertaining and
profitable, have granted him the use of some curious Anatomical Casts
and Drawings (just arrived in the Carolina, Capt. Friend) presented
by the judicious and benevolent Doctor Fothergill, who has improved
every Opportunity of promoting the Interest and Usefulness of that
noble and flourishing Institution.

The Pennsylvania Gazette, of November 25, 1762, contains the
following announcement :

Dr. Shippen's anatomical lectures will begin to-morrow evening at
six o'clock, at his father's house, in Fourth street. Tickets for the
9

80 E. M. HART WELL.

course to be had of the doctor at five pistoles each, and any gentlemen
who incline to see the subject prepared for the lectures, and learn the
art of dissecting, injections, etc., are to pay five pistoles more.

It is stated that his first class numbered twelve. "Having thus
started, it is not to be understood," says Dr. Horner, "that the
lectures proceeded without occasional interruptions from popular
indignation; for .the city being small, almost everyone knew what
was going on in it. The house was frequently stoned, and the
windows broken; and on one occasion, Dr. Shippen's life was put
into imminent danger. While engaged within, the populace
assembled tumultously around the house. His carriage fortu-
nately was at the door, and the people supposing that he was in it
made their first attack there. The windows of the carriage being
up, they were speedily demolished with stones, and a musket ball
was shot through the body of the carriage; the coachman applied
the whip to his horses and only saved himself and his vehicle by a
rapid retreat under a shower of missiles. The Doctor hearing the
uproar, ascertained its cause, and extricated himself through a
private alley."

Possibly the riot above described by Dr. Horner, may have
elicited the following utterance from Dr. Shippen, which is printed
in the Pennsylvania Qazette, December 26, 1765:

It has given Dr. Shippen much Pain to hear that notwithstanding
all the Caution and Care he has taken to preserve the utmost Decency
in opening and dissecting dead Bodies, which he has persevered in
chiefly from the Motive of being useful to Mankind, some evil-minded
Persons, either wantonly or maliciously, have reported to his Dis-
advantage that he has taken up some Persons who were buried in the
Church Bnrying Ground, which has disturbed the Minds of some of
his worthy Fellow Citizens. The Doctor with much Pleasure, improves
this Opportunity to declare that the Report is absolutely false ; and to
assure them that the Bodies he dissected were either of Persons who
had wilfully murdered themselves or were publicly executed, except
now and then one from the Potter's Field, whose Death was owing to
some particular Disease ; and that he never had one Body from the
Church.

In Chapter CCXI of the "History of the City of Philadelphia,"
written by Westcott Thompson, but not yet published in book
form, are found the following statements regarding Dr. Shippen:

THE STUDY OF HUMAN ANATOMY. 81

"Late in November 1762, Dr. Shippen received the first subject for
dissection of which there is any record. A negro man having cut his
throat with a glass bottle, from the effect of which he died, the action
upon his case is thus recorded by the Gazette of December 2. ' After
the coroner's jury had pronounced him guilty of self murder, his body
was immediately ordered by authority to Dr. Shippen's anatomical
theatre/ this accession to the stock of the dissecting room must have
been received a day or two after the opening lecture.

14 In September 1765, Dr. Shippen was compelled to deny publicly
that he had taken dead bodies for the purposes of dissection from the
church burying grounds. In September 1768, he was again obliged to
contradict the rumor that he had taken dead bodies from the city
burying grounds for purposes of dissection. In 1770, considerable
excitement existed in the city in relation to the supposed removal of
dead bodies from the city burying grounds for dissection in the ana*
tomical department of the college. It was probably about that time
that the circumstances happened described by Dr. Carson. [History
of the Medical Department of the University of Pennsylvania, pp. 81
and 217.] 'On one occasion Dr. Shippen's house was mobbed and
only by exercising great tact and by the judicious interference of his
friends, and of the authorities, was he saved from the entire destruction
of his accumulated materials for teaching. This event was known for
years as the sailors' mob.'

"Dr. Shippen in Bradford's Journal of January 11, 1770, published
an address to the public in which he said there were wicked and mali-
cious reports of his taking up bodies from several burying grounds.
He said ' I declare that I never have had, and that I never will have
directly or indirectly, any subject from any burying ground of any
Christian denomination whatever.' He said that upon two of the
families terrified by this report, he had waited, in order to vindicate
himself. He had tried to trace out the authors of the reports but had
failed. It was generally believed that he had taken up the body of a
young lady from Christ Church burying ground, ' but within a few days
the grave had been opened, and the body found there.' Another body
was that of a woman, whose name is given by Dr. Shippen. He says
that 'she died in the middle of the summer of 1769, of a putrid fever,
and yet I am charged with dissecting her body in the middle of winter.'
In corroboration of this address he appended an affidavit by Joseph
Harrison who stated that he was a student of medicine and had lived
with Dr. Shippen, Sr., as an apprentice, ' for the last eight years ; ' that
he had regularly attended the courses of Dr. Shippen, Jr., and knew
where the subjects employed in his lectures were from. He said, ' none

82 E. M. HARTWELL.

were ever taken oat of any burying ground of a Religions Society in
this city.'"

When Dr. Ship pen's lectures were interrupted, in 1775, by the
breaking out of the Revolution, his class numbered between thirty
and forty students. Early in 1777, he was appointed Medical
Director General of the Continental army. In 1778, he resumed
his lectures in Philadelphia. In 1781, he resigned his position in
the army to devote himself to the medical school. Dr. Caspar
Wistar became Shippen's associate in 1792. Dr. Shippen died in
1808.

In New York and Massachusetts, as in Pennsylvania, the
anatomists were the founders of the first medical schools. The
medical department of King's, now Columbia College, was organ-
ized in New York, in 1767. Dr. Samuel Clossy, an Irishman,
who began his course of lectures on anatomy in New York in
1763, was chosen the first professor of anatomy in King's. Dr.
John Warren, who from 1777, till the close of the Revolution,
had served as surgeon-in-chief of the military hospitals at Boston,
gave a private course of dissections to a class of medical students
in that city in 1780. In the following year he gave a public
course of anatomical lectures, the success of which led to the
organization of the Harvard Medical School in 1782. Dr. Warren
was the first professor in the new school. He was for many years
its presiding genius, and held the professorship of anatomy and
surgery till his death in 1815. It was chiefly through the efforts
of Dr. Nathan Smith, that the Dartmouth Medical School was
founded, in 1797. Dr. Smith was appointed "to deliver public
lectures upon Anatomy, Surgery, Chemistry/ Materia Medica, and
the Theory and Practice of Physic." To the Dartmouth School
is usually assigned the fourth and final place on the list of Ameri-
can schools of medicine founded before 1800.

Thanks to the efforts of Thomas Jefferson, in 1779, Virginia can
claim a place on that list for the medical department of William
and Mary College. "I effected in that year, 1779," he says in
his autobiography, "a change in the organization of that institu-
tion by abolishing the Grammar school and the two professorships
of Divinity and Oriental languages, and substituting a professor-
ship of Law and Police, one of Anatomy, Medicine, and Chemis-
try, and one of Modern Languages." In 1778, Mr. Jefferson drew

TEE STUDY OF HUMAN ANATOMY. 88

up a "Bill proportioning Crimes and punishments in Cases hereto-
fore capital." Among its provisipns was the following: "If any
person commit petty treason, or a husband murder his wife, a
parent his child, or a child his parent r he shall suffer death by
hanging, and his body be delivered to Anatomists to be dissected."
This bill was lost by the majority of a single vote, and Virginia
lost the opportunity of passing the first American Act to legalize
anatomy in even a small way. Virginia as yet has no anatomy
act.

In December, 1692, the province of Massachusetts Bay incor-
porated the major portion of the English Act of 1604 against
witchcraft among its statutes. The history and provisions of this
Act are worthy of more than passing mention, because it contains
not only the first American, but also the first English, statutory
prohibition of the desecration of graves, and indicates full well
that the belief in sorcery was a potent factor in popular prejudice
against human dissections. In the preamble to an Act for "the
appointing of Physicians and Surgeons," passed in 3 Henry VIII.,
1511, it is recited that "so far forth were the Science and Cunning
of Physick and Surgery practised by ignorant persons, that com-
mon Artificers, as Smiths, Weavers, and Women, boldly and
accustom ably took upon themselves great cures, and things of
great Difficulty, in the which they partly use Sorcery and Witch-
craft, partly apply such medicines unto the Disease as be very
noious and nothing meet therefor." The practice of witchcraft
was first made a felony, punishable witn death and the forfeiture
of estate to the King, in 1541. This Act of the Parliament of 33
Henry VIII. was repealed six years later, in the first year of
Edward VI.; but in 1565, the fifth year of Queen Elizabeth, it
was reenacted with a saving clause, whereby dower was secured to
the widow and inheritance to the heir of the felon. In 1604, the
first year of James I., the Act of 5 Elizabeth, as well as that of the
9th Parliament of Mary of Scotland, was repealed, and an Act for
"the better restraining and more severe punishing of witchcraft
and dealing with evil and wicked spirits," was passed. It con-
tained the following provision, new to the English law: "If any
person shall take up any dead man, woman, or child out of his,
her, or their grave, or any other place where the dead body resteth,
or the skin, bone, or any other part of any dead person, to be
employed in any manner of witchcraft, inchantment, charm, or

84 E. M. EAR TWELL.

sorcery, whereby any person shall be killed, destroyed, wasted,
consumed, pined, or lamed in his^or her body, or any part thereof/'
every such offender "shall suffer pains of death as a felon, and shall
lose the benefit of clergy and sanctuary."

This Act was cited formally in indictments drawn in Maryland
in 1674, and in Massachusetts in the spring of 1692, and was
acknowledged to be in full force in Pennsylvania in 1684. Massa-
chusetts seems to have been the only colony to embody it in its
laws. The Privy Council repealed the Act in 1695, because it
was " not found to agree with ye Statute of King James the First
whereby ye Dower is saved to ye Widow and ye Inheritance to ye
heir of ye party convicted." The English Act remained un-
repealed till 1736; and, so late as 1712, was declared to be in
force in South Carolina. It does not appear that any "resurrec-
tionist" was ever convicted under it in America. The first
American Act to prevent the digging up of bodies for dissection,
was the New York Act of 1789.

As we have already seen, Pennsylvania, New York, Virginia,
Massachusetts and New Hampshire all had medical schools pre-
viously to 1800. As late as 1782, when the Harvard Medical
School was organized, no one of the above-mentioned States had a
law in its statute books touching the dissection of the dead or the
desecration of their graves. The utmost help given to anatomists
was the occasional allowance of the body of a suicide or executed
criminal.

Possibly, Governor John Winthrop, who was read in physic,
may have authorized his kinsman, Giles Firmin, to make the
anatomy mentioned by Eliot. Prior to the Revolution, the royal
governors could order the dissection of a murderer's body. In
1778 the State of Virginia refused to sanction the dissection of
executed murderers; and has apparently remained iu a state of
arrested development ever since, so far as any appreciation of the
claims of anatomy is concerned. Massachusetts, in 1784, passed a
law allowing the dissection of dead duelists, thereby unwittingly
reproducing in spirit, though riot in letter, a canon of the mediaeval
church, which denied Christian burial to men slaiu in tournaments.
New York, in 1789, in order that science might not be injured by
its law of that year regarding disinterment, made it lawful for the
courts to add dissection to the death penalty in cases of murder,
arson and burglary. The First Congress of the United States, by

TEE STUDY OF HUMAN ANATOMY. 85

the act of April 30, 1790, gave federal judges the discretion of
adding dissection to the sentence of convicted murderers. A simi-
lar act was passed by New Jersey in 1796. No trace of progress,
worth mentioning, in this class of legislation, since the enactments
noted, is to be found in the most recent revisions of statutes, either
of the United States or of New Jersey.

The Act of Massachusetts, passed in 1784, against duelling, is a
noteworthy one, by reason of the fact that it contains the first

• authorization on the part of an American legislature of the dissec-
tion of the dead bodies of malefactors. The province had enacted
laws for the prevention of duelling, in 1719 and 1729. That of
1719 provided penalties in the way of fine, imprisonment, and
corporal punishment — any or all of them, at the court's discretion
— for those convicted of engaging in, or challenging another to
engage in, a duel. Under the Act of 1729, duellists and their
accomplices were carried in a cart to the gallows with a rope about
the neck, "and after sitting for the space of one hour on the
gallows, with the rope about his neck as aforesaid," the offender
was confined in the common jail for one year, and at the expira-
tion of his sentence was required to find sureties for his good
behavior for the succeeding twelvemonth. The Acts of 1729 and
1784, both denied Christian burial to the bodies of men killed in
a duel. Moreover, it was provided in section 3 of the Act of

' 1784, "that when it shall appear by the coroner's inquest that any
person hath been killed in fighting a duel, the coroner of the
county where the fact was committed shall be directed and
empowered to take effectual care that the body of such person so
killed be immediately secured and buried without a coffin, with a
stake drove through the body, at or near the usual place of execu-
tion, or shall deliver the body to any surgeon or surgeons, to be
dissected or anatomized, that shall request the same and engage
to apply the body to that use." Section 4 ordains "that any
person who shall slay or kill another in a duel, and shall, upon
conviction thereof on an indictment for murder, receive sentence
of death, part of the judgment of the court upon such conviction
shall be that the body be delivered to any surgeon or surgeons, to
be dissected and anatomized, that shall appear in a reasonable
time after execution to take the body and engage to apply it to
that purpose."

86 E. M. HABTWELL.

If the Massachusetts legislators in 1784 had any intention of
recognizing the needs of the anatomists, they failed to declare it,
so that New York was the first State, by section 2 of its Act of
1789, to express the desire that "science might not in this respect
be injured by preventing the dissection of proper subjects." It
was not till the passage of the Massachusetts Act of 1831 that any
State really undertook to " legalize the study of anatomy."

It is most likely that the provisions of the Act of 1784 touching
dissection were designed to make duelling a specially infamous
offence. This was quite in keeping, with the English law re-
garding dissection. In 1752, the Parliament of 22 George II., in
order that "some further Terror and peculiar Mark of Infamy
might be added to the Punishment of Death," legalized the
delivery of the bodies of executed murderers to the Surgeons for
dissection. This must have been the Act from which the royal
governors derived authority to dispose of murderers' bodies in
Massachusetts in the manner indicated in the following extract,
taken from the Life of Dr. John Warren, by Edward Warren,
M. D., page 230: "At this period [just prior to the Revolution]
the governor had the disposal of the body of the criminal after
execution. He might order its delivery to the man's friends, to
any one to whom he himself assigned it, or to a surgeon. The
prisoner, with the governor's assent, might make his own arrange-
ments even for the sale of his body, if he was so disposed, either
for the benefit of his family or his own brief enjoyment."

It is to be remarked that the Act of 1752 required the judges to
add either dissection or hanging in chains to the death sentence of
murderers, and that previously to 1832, when the Warburton
Anatomy Bill was passed, there seems to have been no warrant in
English law for any sort of bargain concerning a cadaver. The
only legal mode of disposing of a dead body, excepting in case of
malefactors, was to bury it. Once buried, it was an indictable
offence at common law for any person to exhume it, except by the
leave of the proper officers.

The New York Act of 1789 is of especial interest; both on
account of the circumstances which led to its enactment and because
it may fairly be considered the germ of all subsequent American
legislation concerning the cadaver. The Act of 1789 seems to
have owed its existence to the Doctors' Riot, in New York City, ,
in April, 1788. If you will turn to the issue of the New York

THE 8 TUD Y OF HUM A N ANA TO MY. 8T

Journal and Patriotic Daily Register, for Tuesday, April 15, 1788,
you will find the following : " As a concise statement of the sad
confusion of the city since Sunday last could not be ascertained for
this day's paper, it was thought proper to postpone it till such an
one could be had. It is devoutly to be hoped, in the meantime,
that those who feel themselves injured by the DOCTORS will
seriously replect upon the fatal ERROR of revenging their
cause upon the public at large." Imagine the New York Herald
apologizing for its inability to give a concise statement concerning
a riot two days old ! On Wednesday the Register avows a peculiar
satisfaction in announcing " that the unhappy convulsions of this
city have very considerably subsided," and promises " some par-
ticulars respecting this melancholy transaction from peace to horrid
war in the Weekly Register to-morrow." The charge of "his
Honor Chief-Justice Morris," delivered the day previous to the
grand jury at the City Hall, is contained in the Weekly Register of
Thursday, April 17, 1788 ; but one looks in vain for the promised
particulars in that or the succeeding issues of the Daily Patriotic
Register. The affidavit of Dr. Richard Bayley, executed April 14,
is found in the Register of the following day. In it he denies any
agency or concern " in removing the bodies of any person or per-
sons, interred in any church-yard or cemetery, belonging to any
place of public worship, and that he hath not offered any sum of
money to procure any human body so interred, for the purposes of
dissection," and further saith, " that no person or persons under
his tuition have had any agency or concern in digging up or
removing any dead body interred in any of the church-yards or
cemeteries, to his knowledge or belief, and further this deponent
saith not." Similar affidavits on the part of Efyenezer Graham,
John Parker, and George Gillaspy, pupils of Dr. Charles
McKnight, professor of anatomy ; also of Dr. McKnight himself,
and of John Hicks, Sen., are to be found in the Weekly Register,
which contains not only the Chief-Justice's charge, already men-
tioned, but also the card of William Neilson, Foreman of the
Grand Jury. The grand jury " do request that those persons who
can give any information that may lead to a discovery will ac-
quaint them therewith during their present sitting, at Simmons1
Tavern, in Wall Street." We find the local news of the New
York of a hundred years since best reported in the New York
letters of the Boston and Philadelphia papers. The Boston Gazette
10

88 E. M. EARTWELL.

and the Country Journal, in its issues of April 28 and May 5, 1788,
contains full accounts of the New York mob. The first account
is from a letter written April 16, by one who had borne arms
against the rioters. I give the second, both because it is shorter,
and because its writer seems to have taken especial pains to be
accurate.

New York, April 25th.

As exaggerated accounts of the late riots in this city have been cir-
culated through different parts of the country, we have obtained the
following particulars of that unhappy event: During the last Winter,
some students of physic and other persons had dug up from several of
the cemeteries in this city a number of dead bodies for dissection.
This practice had been conducted in so indecent a manner that it raised
a considerable clamor among the people. The interments not only of
strangers and the blacks had been disturbed, but the corpses of some
respectable persons had been removed. These circumstances most
sensibly agitated the feelings of the friends of the deceased, and
wrought up the passions of the populace to a ferment.

On Sunday, the 13th instant,' a number of boys, we are informed,
who were playing in the rear of the hospital, perceived a limb which
was imprudently hung out of the window to dry ; they immediately
informed some persons; a multitude soon collected, entered the hospital,
and, in their fury, destroyed a number of anatomical preparations, some
of which, we are told, were imported from foreign countries; one or
two fresh subjects were found, all of which were interred the same
evening. Several young doctors narrowly escaped the fury of the
people, and would inevitably have suffered very seriously had not His
Honor the Mayor, the Sheriff and some other persons interfered and
rescued them by lodging them in a gaol. The friends of good order
hoped that the affair would have ended here ; but they were unhappily
mistaken. On Monday morning a number of people collected, and
were determined to search the houses of the suspected physicians. His
Excellency the Governor, His Honor the Chancellor, and His Worship
the Mayor, finding that the passions of the people were irritated, went
among them and endeavored to dissuade them from committing un-
necessary depredations. They addressed the people pathetically and
promised them every satisfaction which the laws of the country can
give. This had considerable effect upon many, who, after examining
the houses of the suspected doctors, retired to their homes. But in the
afternoon the affair assumed a different aspect. A mob, more fond of
riot and confusion than a reliance upon the promises of the magistrates
and obedience to the laws, went to the gaol and demanded the doctors

THE STUDY OF HUMAN ANATOMY. 89

who were there imprisoned. The magistrates finding that the mild
language of persuasion was of no avail, were obliged to order out the
militia to suppress the riot, to maintain the dignity of the government,
and protect the gaol. A small party of about eighteen armed men as-
sembled at 3 o'clock and marched thither. The mob permitted them
to pass through with no other insult than a few volleys of stones, dirt,
etc. Another party of about twelve men, about an hour afterward,
made a similar attempt, but having no order to resist, the mob sur-
rounded them, seized and destroyed their arms. This gave the mobility
fresh courage; they then endeavored to force the gaol; but were
repulsed by a handful of men, who barely sustained an attack of several
hours. They then destroyed the windows of that building with stones,
and tore down part of the fence. At dusk another party of armed
citizens marched to the relief of the gaol, and, as they approached it,
the mob huzzaing began a heavy fire with brickbats, etc. Several of
this party were much hurt, and in their own defence were obliged to
fire; upon which three or four persons were killed and a number
wounded. The mob shortly after dispersed. On Tuesday morning
the militia of General Malcom's brigade and Col. Bau man's regiment
of artillery were ordered out, and a detachment from each were under
arms during that day and the subsequent night. But happily the mob
did not again collect, and the peace of the city is once more restored.

Dr. J. M. Toner, of Washington, in his useful Annals of Medical
Progress, states, p. 97, that "the doctors' mob in 1788, marked
the last serious resistance of the populace to the teaching of practi-
cal anatomy in America/' The very next year, however, as we
learn from Griffiths' "Annals of Baltimore," the body of "one
Cassidy, lately executed, was obtained for dissection, but was
discovered by the populace and taken from the gentlemen who
were then studying anatomy and surgery." Dr. Nathaniel Potter,
in a pamphlet published in 18*38, entitled "Some Account of the
Rise and Progress of the University of Maryland," alludes to the
destruction in 1807 of the Anatomical Theatre of Dr. Jno. B.
Davidge, then a private teacher of anatomy and surgery. Dr.
Davidge had erected a small anatomical theatre, at his own
expense and on his own ground. " It was discovered by the
populace that he had introduced a, subject for dissection; the
assemblage of a few boys before the door was soon accumulated
into a thickly embodied mob, which demolished the house and
put a period to all further proceedings for that season." "Such
were the vulgar prejudices against dissections," he adds, "that

90 & M. HARTWELL.

little sympathy was felt for the doctor's loss." I have been told
that a somewhat similar riot occurred in New Haven about the
year 1820; I have not been able to verify the statement, however.

The name of Warren is most intimately associated with the rise
and progress of anatomical science in Massachusetts. Dr. John
Warren while a student in Harvard College, where he graduated
in 1771, was the leading spirit in forming a private anatomical
society, composed of students. He says of it that " brutes were
dissected and demonstrations on the bones of the human skeleton
were delivered by the members." The Anatomical Society and
the Spunker Club, to which there are frequent allusions in the Life
of Dr. John Warren, seem to have been identical. Dr. Warren
was the principal lecturer of the club. His most zealous asso-
ciates were his classmates, Jonathan Norwood, William Eustis, class
of 1772, and David Townsend and Samuel Adams, of the class of
1770. Adams was a son of Samuel Adams the patriot. Eustis,
Adams, and Warren all studied medicine with an elder brother of
the latter, Dr. and General Joseph Warren. Eustis, Warren,
Townsend, and Adams became surgeons in the Continental array.
Adams died in 1778. Eustis, lived to become governor of Massa-
chusetts in 1823. Warren was surgeon-general of the military
hospital at Boston, from June, 1777, till the close of the Revolu-
tion, and was the first professor of anatomy and surgery of the
Harvard Medical School, of which he was practically the founder.

Some notion of the methods of study of the Spunker Club may
be gained from the following extracts from letters written by
Eustis to Warren, prior to 1775: "This may serve to inform you
that as soon as the body of Levi Ames was pronounced dead, by
Dr. Jeffries, it was delivered by the sheriff to a person who carried
it in a cart to the water side, where it was received into a boat
filled with about twelve of Stillman's crew, who rowed it over to

Dorchester Point When we saw the boat at Dorchester

Point, we had a consultation, and Norwood, David, One Allen and
myself took chaise and rode round to the Point, Spunkers likfe;
but the many obstacles we had to encounter, made it eleven o'clock
before we reached the Point, where we searched and searched, and
rid, hunted, and waded, but, alas, in vain ! There was no corpse

to be found We have a from another place, so

Church shan't be disappointed. P. S. By the way, we have since
heard that Stillman's gang rowed him back from the Point up to

TEE STUDY OF HUMAN ANATOMY. 91

the town, and after laying him out in mode and figure buried him,
God knows where ! Clark & Co, went to the Point to look for
him, but were disappointed, as well as we." No wonder that the
same writer, in another letter, says, " Good heavens I to reflect on
the continued bars we are meeting in our pursuits! It seems as
if fate had placed medical knowledge profunda in puteo, saxis et
trix tnobilibus submersa"

It is not yet one hundred years since Dr. John Warren delivered
the first course of public anatomical lectures ever given in Massa-
chusetts, in compliance with a vote of invitation passed by the
Boston Medical Society, November 3, 1781. It is scarcely fifty
years since the Massachusetts Medical Society began to agitate
the question of legalizing the study of anatomy. The Harvard
Medical School, in the ninety-eight years of its history, has had
but three professors of anatomy, namely, Dr. John Warren, pro-
fessor of anatomy and surgery from 1782 till 1815, when he died ;
Dr. John C. Warren, professor of anatomy and surgery from 1815
to 1847, when he resigned; and Dr. Oliver Wendell Holmes, pro-
fessor of anatomy, who, like the elder Warren, has held his chair
thirty-three years.

Dr. John Warren's son and successor, Dr. John C. Warren, was
three years old in 1781, the year the Massachusetts Medical Soci-
ety was incorporated. Fifty years later, as one of the most prom-
inent members of that society, February 2, 1831, he lectured before
the members of the Massachusetts Legislature, in the representa-
tives' chamber, on the Study of Anatomy, in accordance with a
vote of the house of representatives, passed January 29, 1831. At
the time of this lecture the anatomy bill, which becafce a law ou
the 28th of that month, was still pending.

No better testimony concerning the obstacles which beset the
pursuit of anatomical science during those fifty years can be given
than is found in the Biographical Notes of Dr. John C. Warren,
from which we quote : " No occurrences in the course of my life
have given me more trouble and anxiety than the procuring of
subjects for dissection. My father begau to dissect early in the
Revolutionary War. He obtained the office of army-surgeon when
the Revolution broke out, and was able to procure a multitude of
subjects from having access to the bodies of soldiers who had died
without relations. In consequence of these opportunities he began
to lecture on anatomy in 1781. After the peace there was great

92 E. M. HABTWELL.

difficulty in getting subjects. Bodies of executed criminals were
occasionally procured, and sometimes a pauper subject was ob-
tained, averaging not more than two a year. While in college I
began the business of getting subjects in 1796. Having under-
stood that a man without relations was to be buried in the North

Burying-Ground, I formed a party When my father

came up in the morning to lecture, and found that I had been
engaged in this scrape, he was very much alarmed, but when the
body was uncovered, and he saw what a fine, healthy subject it
was, he seemed to be as much pleased as I ever saw him. This
body lasted the course through. Things went on this way till
1807, when, with the cooperation of my father, I opened a dis-
secting-room at 49 Marlborough Street. Here, by the aid of
students, a large supply of bodies was obtained for some years,
affording abundant means of dissection to physicians and students.
In the meantime, however, schools began to be formed in other
parts of New England, and students were sent to Boston to pro-
cure subjects. The exhumations were conducted in a careless
way. Thus the suspicion of the police was excited ; they were
directed to employ all the preventive measures possible, and
watches were set in the burying-grounds. Thus the procuring of
bodies was very much diminished, and we were obliged to resort
to the most dangerous expedients, and, finally, to the city of New
York, at a great expense of money and great hazard of being
discovered. Two or three times our agents were actually seized
by the police, and recognized to appear in court. One or two
were brought in guilty, and punished by fine, but the law officers,
being more liberal in their views than the city officers, made the
penalty as small as possible. Constant efforts were necessary to
carry on this business, and every species of danger was involved

in its prosecution At that time scarcely any exhumation

occurred without accidents of the most disagreeable and sometimes
painful character. The record of them would make a black-book,
which, though the odium of it should belong to few individuals,
would do no credit to the enlightenment of Boston in the nine-
teenth century, and convey an idea of the state of feeling of a
professor of anatomy on the approach and during the course of
his anatomical pursuits.

"Sometimes popular excitement was got up, and the medical
college threatened. I had reasons, at some periods, even to

THE STUDY OF HUMAN ANATOMY, 93

apprehend attacks on my dwelling-house. Whenever the lectures
approached, a state of incessant anxiety came with them. At
length the pressure was so great that it was resolved to make an
effort in the legislature, though with little hope of success."

If it were necessary, evidence to corroborate that of Dr. Warren
might be indefinitely multiplied from the published, and unpub-
lished traditions of the elders. We content ourselves with the
mention of one episode. About 1820 a highly respectable phy-
sician of Eastern Massachusetts, being detected in anatomical
pursuits, was obliged to flee the State. In a distant community,
which to this day has no anatomy Act, he won eminence as a
teacher of anatomy and practitioner of medicine.

Dr. H. I. Bowditch, in his Life of Amos Twitchell, M. D., treats
fully of the condition of affairs in New England, when the law
said, as he puts it, " A man who is found with a body in his pos-
session for the purpose of dissection shall be considered guilty of
a felony."

It was chiefly due to the efforts of the Massachusetts Medical
Society that Massachusetts, in 1831, was induced to anticipate all
English-speaking States in the enactment of a liberal law regarding
anatomical science. The first definite action of the society seems
to have been taken by the councillors February 4, 1829, when, on
the motion of Dr. A. L. Peirson, of Salem, a committee, consist-
ing of Drs. John C. Warren, E. Alden and A. L. Peirson, was
appointed "to prepare a petition to the legislature to modify the
existing laws which now operate to prohibit the procuring of sub-
jects for anatomical dissections." Previous attempts, however,
seem to have been made to weaken popular and legislative preju-
dices. Public attention had been forcibly called as early as 1820,
in the case of the physician above alluded to, to the unsatisfactory
working of the law of 1815, "to protect the sepulchres of the
dead." It is said that a year or two later a private teacher of
anatomy, in Boston, found one morning on his dissecting-table
the body of a promineut actor, then recently deceased. The
anatomist, who had been a particular admirer and friend of the
actor's, caused the body to be returned to the tomb, under Trinity
Church, from which it had been stolen, and acquainted the authori-
ties with the circumstance. This occurrence seems never to have
been made public, but the physicians and authorities agreed that
the laws must be amended. Doubtless they concluded that the

94 E. M. HART WELL.

public must be enlightened before anything could be gained from
the legislature, for, in 1825, Wells and Lilly reprinted in pamphlet
form an article on "The Importance of the Study of Anatomy,
with some Additional Remarks," from the Westminster Review of
1824. Some writers allude to efforts before the legislature in 1828,
but we have found no documentary proof of any legislative action
previous to that in the house of representatives, February 3, 1829,
when the Committee on the Judiciary was instructed, on motion of
Mr. F, A. Packard of Springfield, " to inquire into the expediency
of making any farther legal provisions to protect the sepulchres of
the dead from violation." In accordance with these instructions,
on February 14th the Committee reported a bill, which, on being
read a second time, February 24th, was indefinitely postponed on
the motion of Mr. Thomas B. Strong, of Pittsfield. The secretary
of the Massachusetts Medical Society at this time was Dr. George
Hayward. In the North American Review for January, 1831, he
says that this proposition, above noted, to mitigate the severity of
the law " was hardly listened to with decency ; members seemed
anxious to outdo each other in expressions of abhorrence ; and the
bill was not even allowed a second reading."

History repeats itself in the case of anatomy Acts no less than
in other departments. In 1866, an anatomy bill, after passing the
Pennsylvania house of representatives, was withdrawn from the
senate of that State, because a too influential member of that body
objected to it as being " unworthy of the age in which we live."
The next year, however, when it was made manifest that "the
bodies of distinguished legislators themselves, after a life full of
good works, were no longer safe in their graves," both senate and
house passed "An Act for the promotion of medical science, and
to prevent the traffic in human bodies, in the city of Philadelphia
and the county of Allegheny."

At the annual meeting of the Fellows of the Massachusetts
Medical Society, June 3, 1829, the committee of three, appointed
by the councillors in February, reported that it was inexpedient to
act upon the petition prepared by them to be presented to the
legislature. After a full discussion of the report it was agreed to
refer the whole subject to a committee of nine. The committee
was requested to report at the October meeting of the councillors;
and the councillors were authorized to take such measures as they
might deem necessary in behalf of the society. The following

THE STUDY OF HUMAN ANATOMY. 95

named gentlemen were chosen to serve on this committee: Drs.
A. L, Peirson, of Salem ; John C. Warren, John D. Wells, John
Ware, William Ingalls, and George C. Shattuck, of Boston;
Nathaniel Miller, of Franklin ; Nehemiah Cutter, of Pepperell,
and John Brooks of Bernardston. When the councillors of the
society met, October 7th, the committee reported that on September
1st a circular letter to the Fellows of the society had been issued,
" with a view of advancing the objects proposed by their appoint-
ment," and they recommended to the councillors to cause a petition
to be prepared and presented at the winter session of the general
court. It was voted to continue the committee, and to authorize
it to incur an expense not exceeding one hundred and fifty dollars.

A circular letter, dated Salem, September 1, 1829, and signed
by all of the committee excepting Dr. Miller and Dr. Cutter,
solicits the aid of every influential member of the society in remov-
ing the popular prejudice against dissection, "especially as it
exists in the minds of members of the legislature." The points
upon which it was intended to rely in the proposed petition to the
legislature, are as follows: "(1.) Anatomical knowledge is abso-
lutely necessary in all branches of our profession. (2.) This
knowledge can only be acquired by dissection. (3.) So far as the
poor are concerned, it is for their especial benefit that all physicians
should learn anatomy thoroughly. (4.) It is believed that the
diseases and lameness of many paupers have passed from a curable
to an incurable condition for the lack of surgical skill, which could
only have been derived from a knowledge of practical anatomy.
(5.) All lovers of good morals must feel desirious to prevent the
growth of a body of people who make it aJbusiness to violate the
sepulchres of th$ dead. (6.) The public, as a body, have a greater
degree of interest in this matter than even physicians." The
Fellows are urged to lay the subject before the members of the
legislature, with whom they may be acquainted, and to inform the
committee, before October 1st, concerning their own views and the
course of public opinion in their vicinity.

The petition authorized by the councillors, and alluded to by
the committee in the circular, which was probably written by
Dr. Peirson, seems to have taken the shape of an "Address to the
Community on the necessity of legalizing the Study of Anatomy :
By order of the Massachusetts Medical Society." In the address,
which covers twenty-seven pages, and bears the imprint of Per-
il

96 E. M. HARTWELL.

kins & Marvin, Boston, 1829, the points of the Salem circular are
amplified and enforced. The address is noticed in the American
Journal of Medical Sciences, vol. vi, p. 210, by Dr. W. E. Horner,
of Philadelphia, who characterizes it as " a candid and open expo-
sition of difficulties, and of the means of relieving them." " It
'is," he says, "a statement directly to the point, and must have
weight, if common sense and common philanthrophy are to be
arbiters. It proposes that the legal restrictions upon dissections
shall not apply in the case of individuals who have no living
relatives, and who have been kept at the public expense/' Dr.
George Hay ward declares that " this address made a deep impres-
sion on the thinking part of society, and wrought a marvellous
change in public opinion." At their meeting, on February 3,
1830, the councillors of the Medical Society authorized the com-
mittee of nine to print a new edition of not more than ten thousand
copies of the Address to the Community.

Meanwhile, on January 22d, in accordance with a motion made
by Mr. Mason, of Boston, in the house of representatives, the
Committee on the Judiciary had been instructed to inquire into the
expediency of farther legislation for the protection of sepulchres.
The Judiciary Committee consisted of Messrs. L. Saltonstall, of
Salem ; L. Shaw, of Boston ; Newton, of Worcester ; Mann, of
Dedham; and Whitman, of Pembroke. Mr. Saltonstall, the
chairman, made a detailed report FebruSry 25, 1830, in which it
was recommended that the farther consideration of the matter be
referred to the first session of the next legislature. The report
lay upon the table till March 11th, when it was taken up, accepted
and ordered to be published in the " newspapers which print the
laws of the commonwealth." This report is printed as " No. 51,
House Documents, pp. 756-764, Documents of Massachusetts,
Political Year 1829, and January Session 1830." The report is
eminently liberal in spirit, and judicial in tone, and is written
clearly and concisely. Although the committee reach the conclu-
sion that the existing law, that of 1815, is unfair to the medical
profession and inconsistent with the best interests of the commu-
nity, they refrain from proposing any alteration of it, believing
that public opinion has not become sufficiently enlightened to
warrant such action.

Governor Levi Lincoln, in his address to the legislature, deliv-
ered May 29, 1830, at the opening of the summer session, declares

THE STUDY OF HUMAN ANATOMY. 97

that the frank and manly representation by the medical faculty of
the embarrassments and difficulties of acquiring a knowledge of
anatomy deserves the most respectful regard. "It may be," he
says, "that this subject is of a nature too delicate for direct legis-
lation; but the public mind should be instructed in its interesting
importance. Let it be shown that the knowledge which is sought
in the science of anatomy concerns all the living, and that without
it the accidents and ills of life which art might remedy are beyond
relief. Let the reason of men be addressed, and prejudice be
dispelled by information and the force of argument. It may then
come to be understood that a community which demands the
exercise of skill and denies the means to acquire it, which punishes
ignorance and precludes the possibility of removing it, is scarcely
more compassionate than that Egyptian harshness which imposed
the impracticable task in cruel oppression of the inability to per-
form it It is not my purpose to propose any definite act

for your adoption. I would commend the subject only to the
discreetness of your counsels."

On May 31st, Mr. John Brazer Davis, of Boston, moved in the
house of representatives, and it was ordered, "That so much of his
Excellency the Governor's speech, as relates to a modification of
the laws in relation to the study of anatomy, be referred to a select
oommittee.,, The gentlemen chosen to act as such committee, were
Messrs. J. B. Davis, of Boston ; G. Willard, of Uxbridge ; A.
Hutchinson, of Pepperell ; L. W. Humphreys, of Southwick, and
J. B. Flint, of Boston. The day after their appointment, the
oommittee reported through Mr. Davis, that the subject be referred
to the next session of the legislature, and the report was accepted.

On the 1st of January, 1831, the select committee made its

import, and brought in a bill "more effectually to protect the

sepulchres of the dead and to legalize the study of anatomy in

oertain cases." The report was written by the chairman of the

committee, Mr. Davis. The report constitutes No. 4 of the House

I)ocument8 for 1831, and in the printed copy is dated January 6.

I*age9 3-82 inclusive are devoted to the report proper; the bill is

found on pages 83-86; the list of documents accompanying the

import is found on page 87; and the documents themselves fill

twenty-nine pages more.

This is altogether the most exhaustive document on the subject
that we have seen; inasmuch as the committee undertakes to con-

98 & M. HARTWELL.

aider, in "all its aspects, the subject committed to them, and to
present not only the results, but the details, of their researches
and reasonings on it." We shall not undertake to outline it within
the limits of a latter-day paper, in face of the fact that twenty
pages octavo are taken up in tracing " the progress of anatomical
science from the first rude attempts of the Greeks, through a slow
progress of near two thousand years," before it is attempted to
show, in nearly thirty-six pages more, that " the study and knowl-
edge of anatomy are essential to the safe and successful practice
of medicine." We unhesitatingly recommend this "faithful com-
pilation of the facts and reasonings of distinguished men, who have
devoted their attention to this subject," to the consideration of
those who have to snatch time from the practice of medicine to get
up "inaugural addresses" for medical colleges in States still fifty
years behind the times. They will find Dr. Southwood Smith's
"Uses of the Dead to the Living," and the "Report of a Select
Committee of Parliament on the Hindrances to the Study of
Anatomy, London, 1828," poor beside, and because of, the riches
of this report of the Davis committee.

The legal status of dissection is noticed in the report as follows:
"In 1815 a law was passed for the protection of the sepulchres of
the dead, which punished the exhumation of any dead body or the
knowingly and wilfully receiving, concealing, or disposing of any
such dead body, by a fine of not more than one thousand dollars,
or imprisonment for not more than one year. Before the passing
of this Act, several cases at common law were brought before the
Supreme Judicial Court, in all of which, where there was a con-
viction, the party was punished. Where it appeared that the
exhumation was for subjects for dissection, « small fine was im-
posed. The last case of this kind was against a now eminent
physician, then of Essex county, in which several important law
points were raised; but the case does not appear to have been
reported. Under the statute there have been several prosecutions,
convictions, and punishments. With truth it may be said that in
Massachusetts a student or teacher of anatomy cannot be found
who is not indictable under the statute of 1815."

"While the law of this Commonwealth is thus severe against
the exhumation of dead bodies, another law has been passed, by
which every practitioner of medicine is required to obtain a degree
at Harvard University, or license from the Medical Society, before

THE STUDY OF HUMAN ANATOMY. 99

he can maintain an action for debt for his professional services.
The license or degree is given on examination, and one of the
prerequisites for this examination is that the applicant shall have
gone through such a course of dissection as shall give him a
minute knowledge of anatomy.

"The only legalized mode of supplying subjects for dissection is
the sentence or order of the Supreme Judicial Court of this State
and of the Circuit Court of the United States in capital convic-
tions within their respective jurisdictions. The insufficiency of
this supply may be inferred from the statements of the secretary of
the Commonwealth and of the clerk of the United States District
Court. The former states, in answer to inquiries addressed him
by the chairman of this committee, that the whole number of exe-
cutions or suicides of convicts from January 1, 1800, to December
31, 1830, is but twenty-six — less than one a year. The clerk of
the United States District Court, in reply to like inquiries, states
that from the adoption of the federal constitution and the first
organization of the federal courts down to the present time the
whole number of executions and of suicides of convicts sentenced
by that court in this district is but fourteen, — being about one in
three years."

February 26, the clerks of the two houses caused the enacted
bill to be laid before Governor Lincoln, by whom it was approved
and signed February 28, 1831.

The wisdom of the Medical Society and the select committee in
acting on Governor Lincoln's recommendation that " the reason of
men be addressed, and prejudice be dispelled by information and
the force of argument, is justified by the lack of opposition to the
enactment of the Davis bill. The Boston Advertiser for February
11, 1831, notes the fact that on the day previous the Davis bill had
passed to a third reading in the house by a vote almost unanimous.
It adds : "No discussion took place touching the general provisions
or tendency of the bill. Several amendments were offered relating
to the details only. No one expressed any sentiments or opinions
in opposition to the general features of the bill; but it received the
approbation of all as a necessary step in the progress of improve-
ment." This shows a marked change in public opinion since 1829,
" when," to use the words of Dr. G. Hay ward, " the proposition
to mitigate the severity of the law against those engaged in dissec-
tion, was driven almost by acclamation from the legislature."

100 E. M. HART WELL.

Subsequent legislation has considerably modified the act of Feb-
ruary 28, 1831, as may be seen on consulting the Acts of April 1,
1834, March 26, 1845, May 10, 1855, and March 28, 1867. By
the Act of 1845, chapter 242, former Acts are simplified, amended
and improved. Section 1 provides that the overseers of the poor
of any town, and the mayor and aldermen of any city, in the com-
monwealth, " shall, upon request, give permission to any regular
physician, duly qualified according to law, to take the dead bodies
of such persons as are required to be buried at the public expense
within their respective towns or cities ; " and also makes it " the
duty of all persons having charge of any poorhouse, work-house,
or house of industry, in which any person required to be buried at
the public expense shall die, immediately to give notice thereof to
the overseers of the poor of the town, or the mayor and aldermen

of the city, and the dead body of such person shall not,

except in cases of necessity, be buried, nor shall the same be dis-
sected or mutilated until such notice shall have been given and the
permission therefor granted." According to section 2, " no such
body shall in any case be surrendered if the deceased person, dur-
ing his last sickness, of his own accord, requested to be buried."
Excepting the repeal of sections 10 and 11 of the Revised Statutes
of 1835, the Act of 1845 contains .no other noteworthy new pro-
vision.

Chapter 323, Laws of Massachusetts, 1855, section 1, confers
the powers and duties of overseers of the poor, as defined in chap-
ter 242, Laws of 1845, upon " overseers and superintendents of
State almshouses." Section 2 contains provisions new to the
statute book. It reads : " Whoever buys, sells, or has in his pos-
session for the purpose of buying, or selling, or trafficking in, the
dead body of any human being shall be punished by fine of not less
than fifty, nor exceeding five hundred dollars, or by imprisonment
in the jail not less than three months, nor exceeding three years."
The duty of giving immediate notice to the proper authorities of
the death of friendless persons in the institutions under their con-
trol, devolved by the Act of 1845 upon the directors of houses of
industry, etc., etc., is also, by the Act of March 28, 1857, laid
upon the board of directors of public institutions of Boston.

So far as the writer has been able to learn, the Massachusetts
legislature has enacted nothing of interest concerning anatomical
science since 1857.

THE STUDY OF HUMAN ANATOMY. 101

We have already noticed the provisions of the Act of 1784,
concerning the dissection of (^ead duellists. The Act of 1784 was

repealed March 15, 1805, when the following was enacted :

"Justices of said court, before whom the conviction shall be, shall
in cases of murder committed in a duel, and in other cases, may,
at their discretion, further sentence and order the body of such
convict to be dissected and anatomized."

In chapter 125, section 2,\>age 716, Revised Statutes 1835, we
find no mention of "murder committed in a duel;" but we do find
that "in every case of a conviction of the crime of murder, the
court may, in their discretion, order the convict to be dissected,
and the sheriff shall deliver the dead body of such convict to a
professor of anatomy and surgery in some college or public semi-
nary, if requested; otherwise it shall be delivered to any surgeon
who may be attending to receive it, and who will engage for the
dissection thereof." The last revision of the Massachusetts
statutes contains the above provision for the dissection of a dead
murderer's body, practically unchanged, excepting this saving
clause: "unless his friends desire it for interment."

The Massachusetts Anatomy Act of 1831, was productive of
results in two directions; it lightened the burdens of the teachers
of anatomy in that State, and it led to the enactment of similar
laws in other States. Connecticut passed a liberal Act, modelled
on that of Massachusetts, June 5, 1833, but rej>ealed the same
June 5, 1834. New Hampshire legalized anatomy in 1834, but
rescinded its action in 1842. Michigan passed "an Act to facili-
tate the study of anatomy," March 9, 1844, but repealed it April
"7, 1851. New York is entitled to the place of honor next to
^Massachusetts, on the list of States which have consistently
endeavored to promote anatomical science. The New York law
*>f April 1, 1854, has never been repealed; on the contrary, it has
lbeen improved, notably by the amendatory Act of June 3, 1879.
Referring to the Massachusetts law of 1831, as amended in
1845, Dr. John C. Warren, says: "The Superintendent of the
House of Industry opposed great difficulties to the execution of
this law; but he dying in 1847, an ample supply was obtained for
the medical school afterwards, particularly in consequence of the
influx of Irish paupers, and the great mortality among them."
Concerning the working of the same law; Dr. George Hay ward,
writing in 1855: "The supply has not been, perhaps, as great as

102 E. M. HARTWELL.

could be wished; but, with the increase of population and pauper-
ism, this objection will pass away." We doubt, if in the judgment
of the anatomists of the Harvard Medical School, "this objection"
has " passed away." We incline to the belief that "with the in-
crease of population and pauperism," there has been, at least, an
equal increase of demagogues, and that no class of men in Massa-
chusetts have a more realizing sense than have its anatomists of the
relation existing between eternal vigilance and the price of liberty.

The city government of Boston, November 3, 1869, ordered
"that permits be issued by the city clerk, until otherwise ordered,
to the surgeons of the Harvard Medical School to take the dead
bodies of such persons dying at Deer Island, or the House of
Correction, the County Jail, or City Hospital, as may be required
to be buried at the public expense." The statutory restrictions
concerning the delivery of unclaimed bodies are embodied in the
remainder of the ordinance. The anatomists of Baltimore, Wash-
ington, and New Orleans, might fairly consider this Boston ordi-
nance a liberal one, for they are still obliged to dissect without
legal warrant, or not at all. On the other hand, in Germany or
France, where for years the dissecting rooms have been furnished
with the unclaimed dead by the police, this ordinance would, un-
questionably, be considered imperfect and illiberal.

It is unfortunate that American anatomists are forced to dance
attendance upon public functionaries for "permits;" as they are
thereby put in the false position of seeking as a personal favor
what ought to be furnished them for the furtherance of the public
welfare. Possibly, the time is not yet ripe for the Massachusetts
auatomists to demand that the unclaimed dead of Spriugfield, Fall
River, Worcester, Lowell, in short, the entire State, as well as of
Bostou, should be delivered to them at their dissecting rooms; but
such a consummation is none the less devoutly to be wished.
Massachusetts led off in legalizing the dissection of bodies required
to be buried at the public expense. Would that she might in-
augurate an administrative reform which should prevent the
present wasteful decomposition of valuable material at the bot-
tom of graves, and preclude the necessity which requires one who
is bent on thoroughly learning practical anatomy in all its
branches, to seek the anatomical institutes of Europe.

The legal status of anatomy in America, at the beginning of
the century, is well illustrated by the Connecticut Acts of 1810.

THE STUDY OF HUMAN ANATOMY. 103

At the May session of that year, it was made punishable by a fine
of at least one hundred dollars and imprisonment in the county
jail for at least three months, for any one secretly to disinter the
body of any deceased person for the purpose of dissection, or in
any way to aid in so doing, or knowingly " to assist in any surgical
or anatomical experiments therewith or dissections thereof." At
the October session it was enacted that there should be a " medical
institution of Yale College," one of whose four professors should
teach anatomy, surgery, and midwifery j and that, as speedily as
the college funds would allow, a collection of anatomical prepara-
tions should be provided.

The Massachusetts Act of 1784 only authorized dissection of
dead duellists as a mark of infamy ; therefore, the New York Act
of 1789 must be considered as the first American anatomy law.
This Act was passed the year after the famous "Doctors' Mob" in
New York city, and is entitled, "An Act to prevent the Odious
Practice of Digging up and removing, for the purpose of Dissec-
tion, Dead Bodies interred in Cemeteries or Burial Places." It
comprises two sections. Section I. provides that any person
convicted of removing any dead body from its place of sepulture,
for the purpose of dissection or with intent to dissect, or of dis-
secting or assisting to dissect, such body, " shall be adjudged to
stand in the pillory or to suffer other corporal punishment, not
extending to life or limb, and shall also pay such fine and suffer
such imprisonment as the court shall in their discretion think
Proper to direct." In Section II. it is further enacted, " In order
that science may not in this respect be injured by preventing the
dissection of proper subjects, that when any offender shall be
conV'icted of murder, arson, or burglary, for which he shall be
^ntenced to suffer death, the court may, at their discretion, add
*° the judgment that the body of such offender shall be delivered
*° * surgeon for dissection." Massachusetts made the first con-
siderable improvement on this New York Act when in 1831, it
Passecl a statute authorizing, under certain restrictions, the deliv-
^y to the anatomists of the unclaimed bodies "of deceased persons
Quired to be buried at the public expense."

Enactments similar to the New York Act of 1789, Section I.,

**ave since been passed by the following States: Alabama,

Arkansas, California, Connecticut, Georgia, Illinois, Indiana,

Iowa, Kansas, Kentucky, Maine, Massachusetts, Michigan,

104 , E. M EAETWELL.

Minnesota, Mississippi, Missouri, Nebraska, New Hampshire,
Ohio, Oregon, Pennsylvania, Rhode Island, Tennessee, Texas,
Vermont, Virginia, West Virginia, and Wisconsin. Of the
above-mentioned States, Kentucky, Oregon, Rhode Island, Texas
and West Virginia have no anatomy Acts ; while Rhode Island,
Texas, and West Virginia have no medical schools. The laws of
nine States, namely, Colorado, Delaware, Florida, Louisiana, Mary-
land, Nevada, New Jersey, North Carolina, and South Carolina,
are, so far as the writer has been able to learn, silent regarding
grave-robbery. While the Territories of Dakota, Utah, Washing-
ton, and Wyoming have laws for the protection of sepulchres, the
District of Columbia has no such law, although one was inserted
into the proposed code of 1857, which failed of adoption.

The second section of the New York Act of 1789 has developed
into the Acts of twenty-four States. The following named States
have legalized dissection: Alabama, Arkansas, California, Colo-
rado, Connecticut, Georgia, Illinois, Indiana, Iowa, Kansas,
Maine, Massachusetts, Michigan, Minnesota, Missouri, Nebraska,
New Hampshire, New Jersey, New York, Ohio, Pennsylvania,
Tennessee, Vermont, and Wisconsin.

The dissection of executed criminals, as such, is still lawful
within the special jurisdiction of the United States Government
and in the following States : Alabama, Arkansas, Colorado, Con-
necticut, Georgia, Illinois, Indiana, Kansas, Massachusetts, Mis-
souri, Nebraska, and New Jersey. Nebraska, like the United
States and New Jersey, makes no provision other than this for its
anatomists. Unlike them, however, it has a penal statute regard-
ing grave robbery. Alabama, Georgia, Missouri, and Tennessee
allow the dissection not only of executed criminals, but also of
"other persons with the consent of their friends."

Kentucky, Mississippi, Oregon, Rhode Island, Texas, Virginia
and West Virginia are without laws of any kind regarding dissec-
tion, though they all forbid violation of sepulture. The most
backward of the United States are those which have no statute
touching either dissection or grave robbery. In this class we find
Delaware, Florida, Louisiana, Maryland, Nevada, North Carolina,
South Carolina and the District of Columbia. The Territories of

(mods, Idaho, Montana, and New Mexico are similarly indiffer-
t to the science of anatomy and the sanctity of burial-places.
in of these States, like Maryland, Louisiana and South Caro-

THE STUDY OF HUMAN ANATOMY. 105

lina, contain medical schools. In order to punish body-snatching
Maryland is to-day obliged to fall back on the common law of
England, although the common law penalty was superseded nearly
fifty years since, by the passage of the Warburton anatomy act.
All things considered, the attitude of the Italian cities of the four-
teenth century and that of the empire of Japan of to-day must be
characterized as more liberal and enlightened regarding the alpha-
bet of medicine than that of the United States and of very many
of the individual States.

The Acts of the following States may be termed fairly liberal :
Arkansas, California, Connecticut, Illinois, Indiana, Iowa, Kansas,
Massachusetts, Michigan, Minnesota, New Hampshire, New York,
Ohio, Pennsylvania, and Wisconsin.

The Acts of Alabama, Colorado, Georgia, Maine, Missouri,
Nebraska, New Jersey, Tennessee, and Vermont are illiberal.

In 1869 Maine enacted "that when any person convicted of

crime dies or is executed in the State prison or any jail, the warden

or keepers shall, on request, deliver his body to instructors in

Medical schools established by law." In February, 1876, capital

punishment was abolished ; so that at present in Maine it is legal

*° dissect only the body of a person who " requests during his life

that his body may be delivered to a regular physician or surgeon

*°r the advancement of anatomical science, after his death, unless

some kindred or friend within three days" asks to have it buried;

or the body of a convict who has not at any time requested to be

buried, and whose friends and kindred fail for three days after his

de3.th to ask for his burial.

The statute of Tennessee, unless it has been repealed since 1871,

18 quite as liberal as that of Maine. It provides that no penalty

8uall "apply to regular physicians to whom the bodies of criminals

^"^ delivered pursuant to law, or to dissection of slaves by consent

°* th€jr masters, or of other persons by consent of their relatives."

Tfae New York Act of June 3, 1879, may be instanced as a type

°* the liberal class of American Acts. It reads: "It shall be

. ^ftal in cities whose population exceeds 30,000 inhabitants, and

111 bounties containing said cities, to deliver to the professors and

•"^^here in medical colleges and schools in this State, and for said

PT°fessors and teachers to receive, the remains or body of any

"fce^agecl person for the purposes of medical and surgical study : -

Provided that said remains shall not have been regularly interred,

106 E. M. HARTWELL.

and shall not have been desired for interment by any relative or
friend of said deceased person within twenty-four hours after death ;
provided, also, that the remains of no person who may be known
to have relatives or friends shall be so delivered or received with-
out the consent of said relatives or friends; and provided that
the remains of no one detained for debt, or as a witness, or on
suspicion of crime, or of any traveller, nor of any person who shall
have expressed a desire in his or her last sickness that his or her
body may be interred, shall be delivered or received as aforesaid,
but shall be buried in the usual manner; and provided, also, that
in case the remains of any person so delivered or received shall be
subsequently claimed by any surviving relative or friend th&y shall
be given up to said relative or friend for interment. And it shall
be the duty of the said professors and teachers decently to bury in
some public cemetery the remains of all bodies after they shall
have answered the purposes of study aforesaid ; and for any neglect
or violation of this provision of this Act the party so neglecting
shall forfeit and pay a penalty of not less than $25 nor more than
$50, to be sued for by the health officers of said cities, or of other
places, for the benefit of their department." An earlier law of
New York forbids all traffic in subjects, or any use of them, except
for anatomical purposes, under penalty of imprisonment in jail for
not more than a year.

To summarize the legislation from 1789 to 1881, we may say
that twenty-four States allow dissection; fifteen States have liberal
anatomy Acts, while nine have illiberal ones ; the laws of fourteen
States are silent regarding anatomy, excepting their laws on mal-
practice ; twenty-eight States forbid the desecration of graves, while
the laws of ten States are silent regarding it ; the laws of the six
States are silent touching both dissection and disinterment; Dakota
alone of the eight Territories allows dissection ; four Territories
forbid exhumation, and four have no enactment regarding it;
twelve States aud oue Territory require the burial of oadavera
disseciiu

The District of Columbia occupies a unique position among the
capitals of civilized States in that the studies of its anatomists and
the graves of its dead are alike unprotected by statutory enact-
ments. The United States government sends Washington resur-
rectionists to jail when it can: but it has recently utilized in the
examinations before the Navy Board, in the city of Washington,

TEE STUDY OF HUMAN ANATOMY. 107

as many as twelve subjects, which* could be procured by stealth
only.

The most elaborate, the most liberal, and also the most stringent
of the American anatomy Acts have been passed within the last
five years. Those of Indiana, Ohio, and New York, were passed
in 1879 ; and the amended Act of Iowa, March 26, 1880. So far
as I can learn the amended Michigan Act, approved March 2,
1881, is the latest American Anatomy Act.

The Michigan Act of 1844, which, as we have noticed, was
repealed in 1851, required the officers of the State prison ta sur-
render the bodies of all unfriended convicts dying in their prison
to any agent of the medical society of the State who should present
an order for the same signed by the president of the society. Simi-
larly, the unclaimed bodies of convicts dying in a county jail,
under sentence of six months imprisonment or more, were deliv-
erable to the agents of the medical society of the county in which
the jail was situated. In 1867 a new Act was passed, which has
8IQce been thrice amended,— once in 1871, again in 1875, April
27th, and again by the Act of March 2, 1881. The last men-
tioned Act contains provisions which render it the most advanced

and liberal of all American Anatomy Acts ; I therefore give it in
fall.

€€ Sec. 1. The People of the Slate of Michigan enact, That sections
1 and 2 of act number 138 of the session laws of 1875, approved
April 27, 1875, being sections 2110 and 2111 of chapter 65 of the
COri^piled laws of 1871, as amended, be amended so as to read as
follows:

"(2110). Sec. 1. Any member of either of the following boards,
ar*d any of the following named officers or persons, to wit: The
°°arcl of health of any city, village, or township, the common
COl^ticil of any city, the board of trustees of any village, the mayor
°* ariy city, president of any village, any board, or officer having
tne direction, management, charge, or control, in whole to in part,
°* any prison, house of correction, work-house, jail, or lock-up,
c°vinty superintendents of the poor, keepers of poor-houses and
all**sbouses, any physician, or other person in charge of any poor-
"Ouse or almshouse, sheriff, coroners, the board of State commis-
sioners, the board of trustees, board of control, and all officers,
Physicians, and persons in charge, in whole or in part, of any
u*8titution for the deaf and dumb, blind, and insane, or other

[

108 E. M. SAB T WELL.

charitable institution founded or supported, in whole or in part,
at public expense, having in his or their possession or control the
dead body of any person not claimed by any relative, or legal
representative, as hereinafter provided, and which may be required
to be buried at public expense, or the expense of any one of such
public or charitable institutions, shall deliver such dead body or
bodies, within thirty-six hours after death, or after he or they
shall become possessed thereof, to the express or railway company
.at the nearest railway station, placed in a plain coffin and enclosed
in a strong box, securely fastened, and plainly directed to the
" Demonstrator of anatomy, of the University of Michigan, Ann
Arbor, Mich.," excepting only the dead bodies of such persons as
shall have died from some infectious disease. And such boards,
common councils, officers, or other persons making such shipment
shall take the usual shipping receipt for such package, and shall
notify the consignee of such shipment by letter, mailed on the day
the package is so delivered as aforesaid; and shall also inclose in
such letter a statement giving, as nearly as can be ascertained, the
name, age, residence, and cause of death of such deceased person ;
.and the name and postoffice address of the known relatives of such
deceased person, whose body has been shipped as aforesaid ; and
also a statement of the costs and expenses which have been incurred
in the procuring of the coffin, box, preparation of body for ship-
ment, and shipping the same. And, upon the receipt of such con-
signment, the said demonstrator of anatomy of the University of
Michigan shall immediately forward to such officers, board, coun-
cillor institution, or persons making such shipment, or incurring
such expenses, the amount thereof, not exceeding in any case the
sum of fifteen dollars : Provided, Such dead body shall not be so
shipped or delivered as aforesaid, if it shall be requested in good
faith for interment by any relative before the same shall be shipped
a? aforesaid, and in case the dead body of any person, so delivered
or shipped as aforesaid, be subsequently claimed or demanded of
said demonstrator of anatomy, or of any other person or institu-
tion, into whose possession or under whose control it may have
been placed, by virtue of the provisions of this law, by any rela-
tive or legal representative of such deceased person, for private
interment, it shall be given up to such claimant even after the
same shall have been interred, as hereinafter provided. Such
bodies shall be used only for the purposes hereinafter mentioned,

THE STUDY OF HUMAN ANATOMY. 10*

and shall then, in all cases, be interred in some suitable place, kept
for that purpose, and a correct record shall be kept of every such
body, and all matters by which such body may be identified com-
ing to the knowledge of the person or officer at any time in charge
of such bodies, shall be faithfully recorded at length in a book to.
be kept for such purposes, to the end that the same may be at any
time traced and recovered by the friends and relatives of such
deceased person : And provided further, That the institution,
board, council, officer, or person aforesaid in charge of any such
body as aforesaid shall, immediately after the death of such per-
son, notify, if possible, by telegraph, or otherwise by letter, one or
more of the nearest known relatives of such deceased person of the
death of such person ; and in no case shall the body of any such
deceased person be delivered or shipped as aforesaid until after the
expiration of twenty-four hours from death ; and every individual
officer or party violating any of the provisions of this section shall
be deemed guilty of a misdemeanor.

"(2111). 8bc. 2. The bodies so delivered, or shipped as aforesaid,.
shall be used for the advancement of anatomical science in this
State and in the following institutions of learning only, viz: The
University of Michigan, Detroit Medical College, and Michigan
College of Medicine. And said bodies shall be distributed to and
smong the same equitably, the number assigned to each by said
demonstrator of anatomy, shall be proportional to that of its stu-
dents in actual attendance. And each of said institutions shall
pay quarterly to said demonstrator its ratable proportion of the
expenses borne aud incurred under this act: Provided, hotvever,.
"That said demonstrator of anatomy, upon the receipt of every
body, under and by virtue of the provisions of this act, shall
cause the same to be embalmed or put in a state of preservation,.
»nd shall not permit the same to be delivered to either of said
institutions for the purpose of dissection, until the same shall have
been in his possession at least ten days. And it shall be the duty
of said demonstrator of anatomy, upon the receipt of every body,
to immediately notify the relatives of such deceased person, if
known, of the receipt of such body, either by mail or telegraphy
as he may deem best. And that said body will be preserved
intact, for the space of ten days, in which time such relative will
be entitled to said body for the purpose of private interment, upon
payment of the expenses already incurred. And if the relatives

110 KM. HARTWELL.

or legal representative of such deceased person shall request said
body for the purpose of interment, and shall pay said expenses, said
demonstrator shall deliver to such relative or legal representative
the said body, together with the said coffin and box enclosing the
same. But in case said body shall not be requested by such rela-
tives until after the same shall have been applied to the purposes
intended, the remains thereof, together with the coffin and box
aforesaid, shall be delivered without charge : Provided, That the
University of Michigan, Detroit Medical College and Michigan
College of Medicine aforesaid, and each and every other medical
institution shall not receive into their possession any bodies pro-
cured in this State other than those provided for by the provisions
of this act, and every individual or party violating this provision
shall be deemed guilty of a misdemeanor."

Indiana had not legalized dissection when, in the spring of
1878, the body of the Hon. J. Scott Harrison, a son of the late
William Henry Harrison, President of the United States, having
been stolen from its grave near to the Ohio line, was found by the
son of the deceased, the day after his burial, in a Cincinnati dis-
secting-room, whither he had gone in search of another body.
The only penalty for grave-robbery under the Indiana statutes was
a fine not exceeding one thousand dollars, provided by the Act of
June 14, 1852. This case of resurrecting led to the improvement
in 1879 of the laws of both Indiana and Ohio. Possibly the
stringent amendment to the Iowa law, passed March 26, 1880,
might be traced to the outrage of the Harrison tomb.

Chapter LXV. of the laws of the fifty-first session of the Gen-
eral Assembly of Indiana is "an Act in relation to the use of
human bodies for the purpose of dissection; to require a record
thereof to be kept, and to punish the unlawful possession or dis-
section of such bodies and the violation of graves."

Section 1 requires that all institutions or persons engaged in
dissection shall keep a record book containing full particulars
regarding every corpse received for dissection.

Section 2 makes it punishable by a fine of " not less than one
hundred nor more than five hundred dollars, to which may be
added imprisonment in the county jail for any period not less
than one month nor more than one year," if the person having
the custody of the record required by section 1 fail or refuse to
produce it. Section 3 declares it a felony, punishable by impris-

THE 8 TUD Y OF HUMAN A NA TO MY. 1 1 1

on men t for not less than one nor more than five years, for any
person to "receive, or have in possession, or dissect, or permit to
be dissected, .... any such body of which the record required by
section 1 shall not have been made." Making a false entry in
the record is made a felony by section 4, punishable by not less
than one nor more than three years' imprisonment in the State
prison.

Imprisonment in the State prison for not less than two nor
more than five years is the penalty provided by section 5 for the
felon " who shall dissect, or have in his possession for the purpose
of dissection, any human body, or any part thereof, other than
euch as are or may be given by law for such uses." Section 6
makes those who " have the supervision of the dissecting-room
and of the instruction given therein " responsible " for bodies
received or found therein." Section 7 relates to illegal exhuma-
tion which is made a felony, punishable by imprisonment in the
State prison " for not less than three nor more than ten years."
-According to section 9, one who knowingly aids in concealing an
unlawfully procured body is liable, as a felon, to imprisonment in
the State prison for from one to three years. Section 10 declares
that " any person who shall buy or receive, by gift or otherwise,
3ny dead human body, or any part thereof, knowing the same to

iave been disinterred in violation of this act, shall be

deemed an accessory to such offense, and, on conviction thereof,
l)e punished in like manner as is prescribed in the preceding
section."

Chapter LXVI of the Session Laws of Indiana for 1879, is an
-Act " to promote the science of medicine and surgery by providing
xnethods whereby human subjects for anatomical and scientific dis-
section and experiment may be lawfully obtained, and prescribing
penalties for violation thereof." The Act is a liberal one. The
penalties provided for its violation are severe. Its fifth section is
unusual in its provisions.

" Sect. 5. In case of any vagrant found dead, or in case of any
person killed while committing a felony, or if any prisoner is con-
victed of felony and justifiably killed in attempting to escape from
prison or from officers of* the law having him or her in lawful
custody, upon the body of which person an inquest may lawfully
be held, and shall be held by the coroner or other officer thereto
lawfully authorized, it shall be the duty of such inquest to inquire
13

112 E. M.HABTWELL.

as to the existence and residence of any next of kin of such de-
ceased person ; and if it shall be the verdict of such inquest that
the person so found dead or killed had no next of kin, the coroner
or other officer holding such inquest may at his discretion, and
with the approval of the sheriff of the county wherein such inquest
is held, upon the request in writing of the faculty or other author-
ities of any duly incorporated and organized medical college or
medical association within this State, in operation nearest to the
place of such inquest, deliver such dead body to such college for
the scientific purposes thereof, taking a proper descriptive receipt
therefor, which shall be filed with the clerk of the county."

Ohio, as early as 1831, enacted penalties for grave robbery, but
did not pass any "Act to encourage the study of anatomy," till
March 26, 1870, when an inadequate law with the above title was
passed. House bill No. 216, Ohio legislature, 1878, embodied an
attempt to repeal the Act of 1870, in the following remarkable
terms : —

" Whereas, by the laws of this State the bodies of criminals,
executed for heinous offences, unless said criminals are poor and
friendless, are entitled to decent burial ; and whereas, poverty is
no crime, and the poor, honest, friendless man, in life and in death,
should before the law be the equal, at least, of the depraved crim-
inal ; and whereas, by the laws of this State the bodies of deceased
and unclaimed poor are authorized to be given over to certain
colleges and schools for dissection ; therefore, —

"Sect. 1. Be it enacted, etc., That an Act entitled 'An Act to
encourage the study of anatomy/ passed March 25, 1870, be and
the same is hereby repealed.

"Sect. 2. This Act shall take effect and be in force from and
after its passage."

The person who introduced this bill, meeting with unexpected
opposition, finally withdrew it, saying that he had " only intro-
duced it for fun." The Harrison horror satisfied the Ohio legis-
lators that anatomy could not be regulated by jocose legislation ;
and an earnest attempt was made to protect alike the anatomists
and the dead, as may be seen on consulting section 3763 of the
Revised Statutes of Ohio, 1880.

From 1851 till 1880 it was provided, in the chapter of the code
of Iowa which relates to offenses against chastity and decency, that
every offender who should illegally disinter, or assist in disinter-

THE STUDY OF HUMAN ANATOMY. 113

ring or concealing any human body, should "be punished by
imprisonment in the county jail not exceeding one year, or by fine
not exceeding $1000, or by both fine and imprisonment." By
Act of March 26, 1880, embodied in section 4019J of Revised
Statutes of Iowa, every such offender is now liable to imprison-
ment " in the penitentiary not more than two years, or by fine not
exceeding $2500, or by both fine and imprisonment." By the Act
of April 22, 1872, it is allowed in Iowa, under the customary
restrictions, for any coroner or undertaker in any county or city in
which the population exceeds one thousand inhabitants to deliver
to any medical college or school, or any physician in the State, for
the purpose of medical or surgical study, the body of any deceased
person, except where such body had been interred or dressed for
interment.

I have endeavored to ascertain some facts as to the amount and
cost of the dissection done in our American schools of medicine.
I can find no statistics on the question. The following statement
is based on the figures of the forthcoming report for 1879 of
<xeneral John Eaton, United States Commissioner of Education,
and on such data as have been kindly furnished me by several
prominent teachers of anatomy. The total number of medical
etudents of "all sorts" in the United States, in 1879, was 13,321,
showing an increase of 1,484 over 1878, and of 7,378 over 1870.
Of these 9,603 were in attendance upon 988 instructors in 68
co-called regular schools, in 26 States and the District of Colum-
bia. The increase of regular students in 1879 over 1878 was
1,317. In 12 States with liberal anatomy laws there were 34
schools, with 599 instructors and 5,294 students.

Indiana and Ohio joined the column of liberal States in 1879,
with a total number of 1,219 students; whereas in 1878 the total
number in those States was 945. In 6 States with illiberal laws
there were 18 schools with 228 instructors and 1,672 students;
and in 8 States and the District of Columbia there were 1,652
students in 15 schools, with 122 instructors, unprotected by law
in the study of practical anatomy. Kentucky, with 4 schools and
603 students, had no anatomy law. The District of .Columbia
had 158 students in 3 schools; also 1 President of the United
States and 1 Congress, ditto, but no anatomy law. Maryland
with 2 schools and 468 students; Louisiana with 1 school and
193 students; South Carolina with 1 School and 71 students;

114 E. M. HARTWELL.

and North Carolina with 1 school and 7 students had no anatomy
Act and no statute forbidding disinterment of the dead. The city
of Baltimore buried 577 unclaimed dead bodies in 1880, while her
anatomists were obliged to use stolen subjects or none.

During the winter of 1879-80, in 11 medical schools in 6 States
and the District of Columbia, there were 1,944 students in attend-
ance, of whom 1,255 dissected, and 609 dissected more than one
" part." On the average the dissection of two parts is required
for a degree. The average cost of a part was $3.00, the extremes
being $9.00 and nothing. The demonstrator's ticket is not reckoned
in the cost per part. The average cost of subjects to the schools was
$18.72; the extremes of price being $3.00 and $50.00. Usually
5 students dissect on a single subject, but in one school 8 and in
another 10 students work on the same subject, alternately reading
and dissecting. Of 445 subjects used, not more than 39 were used
by students in making surgical operations on the cadaver. Three
only of the eleven schools claim to prescribe such a course of opera-
tions ; but judging from the number of students who took it, it is
a medical rather than a legal prescription. Of the 1,255 students
who dissected, 465 using 133 subjects were unprotected by law in
so doing. On the basis indicated above, it is computed that
between 3,400 and 3,500 subjects should have been used by the
students in the regular medical schools of the United States. The
official returns show that in France in 1876, 3,463 subjects were
delivered in accordance with law, at the anatomical theatres of
schools having an aggregate of 5,624 students.

We have traced, thus far, the course of practical anatomy in
America from the time of Giles Firmin till the close of the last
century ; and have considered in a more detailed way the develop-
ment of what may be characterized as the most typical of the
American Anatomy Acts, namely, the Massachusetts law. The
same obstacles of prejudice and apathy which beset the anatomists
of our younger States, have been operative in every land where anat-
omy has gained a foothold, since the days of Ptolemy. It would
be interesting to attempt to analyze the popular prejudice against
human dissection, which prejudice is a strange compound of pagan
superstition, Christian materialism, and an innate aversion to the
morals, aims, and manners of the average American medical
student. Such an attempt would take us too far afield. It is
note- worthy, however, that anatomy has flourished chiefly under

THE STUDY OF HUMAN ANATOMY. 115

the rule of princes and prelates. Anatomists have usually found
republics, to say the least, ungrateful. We ought not to be sur-
prised, therefore, when we consider American Anatomy Acts as
a class, to find certain of our States no more enlightened in this
regard than was France when Vesalius had to contend by night
with vultures and prowling dogs for the carcase of the murderer
or the suicide. The utmost help that several of our States give to
anatomists is the occasional gift of the body of an executed male-
factor; while others of them have not attained even to that
mediaeval stage of generosity.

The guild spirit which led to the incorporation of the Edin-
burgh Surgeons as a "Company," in 1505, and the incorporation
of the "Mystery and Commonalty of Barbers and Surgeons of
London," in 1540, may be said to characterize the majority of our
American medical colleges which are, as has been well said by
President Eliot of Harvard University, managed as commercial
ventures. This trading monopolizing spirit is more marked in
British than in Continental schools of medicine. The radical
difference between European and American medical education
results from the woeful lack, on this side of the Atlantic, of the
well-considered, consistent, and responsible State supervision exer-
cised over the teachers, students, and practitioners of medicine in
most European countries. In no department of medical education
is this difference more strongly marked than in that of anatomy.
It is equally clear whether we consider the training and attain-
ments of the teachers, the amount of practical knowledge required
of the students, or the laws regulating the supply of material in
this department.

It is no less certain that the German and French schools of
anatomy outrank the British, than that the latter outrank the
American. While one might, from sources to be found in the
libraries of Washington, Boston, and Baltimore, trace the develop-
ment of the French laws concerning the cadaver, I find it impos-
ble to make any detailed statement, based on authentic documents,
regarding the laws which regulate the organization and mainte-
nance of the German institutes of anatomy. It may be stated,
however, that an Act which should embody the best features of
the best American Anatomy Acts, while it would compare favora-
bly with the British laws, would fall far short of the French, in
point of comprehensiveness and liberality; and it is safe to say

116 E. M. HARTWELL.

that do medical school in the United States combines the rigid
requirements of Vienna and Prague, of seventy years ago, with
anything like the wealth of opportunity offered to-day at Paris
and Bonn. One who should desire to become a thoroughly facpert
anatomist through the dissection of the dead rather than by mang-
ling the living, would be justified in going from America to Ger-
many or France simply on grounds of economy. The depopu-
lation of American medical colleges, owing to such a cause, need,
however, not be feared, so long as the present public and profes-
sional indifference to ignorance of the fundamental facts of medical
science obtains.

ALTERNATION OP PERIODS OP REST WITH
PERIODS OP ACTIVITY IN THE SEGMENT-
ING EGGS OP VERTEBRATES. By W. K. BROOKS,
Ph. D., Associate in Biology. With Plate VIII.

In the first volume of this Journal I have called attention to
theJacLihaLthe well-known contraption nf t.hft mollnaran shtct after.

Notb. — Od pages 28 and 29 of "A Centennial Address," delivered before
the Massachusetts Medical Society, June 7, 1881, by Samuel Abbott Green,
M. D , received while this paper was going through the press, I find interesting
allusions made to Giles Firmin, concerning whom the apostle Eliot speaks in
the letter quoted on page 76 of these Studies. "An anatomy is the old name,"
says Dr. Green, " for a skeleton, and Mr. Firman may be considered, in point of
time, the first medical lecturer in the country. His instruction must have been
crude, and comprised little more than informal talks about the dry bones before
him ; but even this would be a great help to the learners. At any rate it seems
to have excited an interest in the subject, for the recommendation is made at
the session of the General Court, beginning October 27, 1647 — a few weeks
later than the date of Eliot's letter, — that " we conceive it very necessary
y1 such as studies phisick, or chirurgery may have liberty to reade anotomy &
to anotomize once in foure yeares some malefacto7 in case there be such as the
Courte shall alow of." — General Court Records^ ii, 175.

egg under constant observation until I saw it undergo segmenta-
tion, or satisfied myself that it was dead, and the result was quite
interesting, since it showed that the periods of change, which are
rather short, are separated from each other by extremely long
periods of rest.

The blastoderm of the egg which was selected is shown in
Figure 1, Plate VIII, as seen from above, magnified eighty
diameters. It is divided into eight spherules, which are sym-

117

116 E. M. HARTWELL.

that do medical school in the United States combines the rigid
requirements of Vienna and Prague, of seventy years ago, with
anything like the wealth of opportunity offered to-day at Paris
and Bonn. One who should desire to become a thoroughly Expert
anatomist through the dissection of the dead rather than by mang-
ling the living, would be justified in going from America to Ger-
many or France simply on grounds of economy. The depopu-
lation of American medical colleges, owing to such a cause, need,
however, not be feared, so long as the present public and profes-

ALTERNATION OP PERIODS OP REST WITH
PERIODS OP ACTIVITY IN THE SEGMENT-
ING EGGS OP VERTEBRATES. By W. K. BROOKS,
Ph. D., Associate in Biology. With Plate VIII.

In the first volume of this Journal I have called attention to
the fact that the well-known contraction of the molluscan egg after
each division is the external indication, at least in the Fresh-water
Pulmonates and the Oyster, of an alternation of periods of rest
with periods of activity.

I have suggested, (Vol. I, No. 2, page 78), that this alternation
may be due to the need for an accumulation of energy, by the
assimilation of the food contained in the egg, in order to overcome
the physical resistance of the protoplasm.

According to this view the separation of the periods of activity
by intervening periods of rest is the essential feature, and the con-
traction after each division a secondary phenomenon; and it is
therefore interesting to find/ in eggs where the blastoderm is small
and the food yolk large and inelastic, that while there is no con-
traction after each division there is, during the early stages at
least, a well marked period of rest after each period of activity.

During the summer of 1880 I obtained, at the marine labora-
tory of the Johns Hopkins University, a number of large fish-eggs,
which are probably those of Batrachus tau (Linn.) While many
of the eggs appeared to be perfectly healthy, and while I found
them in various stages of segmentation, I at first failed to observe
any change whatever in a single egg, even after several hours
observation.

As it seemed possible that this might be due to the rapidity of
the change when it did take place, I determined to keep a single
egg under constant observation until I saw it undergo segmenta-
tion, or satisfied myself that it was dead, and the result was quite
interesting, since it showed that the periods of change, which are
rather short, are separated from each other by extremely long
periods of rest.

The blastoderm of the egg which was selected is shown in
Figure 1, Plate VIII, as seen from above, magnified eighty
diameters. It is divided into eight spherules, which are sym-

117

118 W. K. BROOKS.

metrically placed on the sides of a longitudinal axis. At one end
of this axis there are two large spherules 1, and, following these,
a second, somewhat larger pair 2 ; then a very small pair 3, and
at the opposite end of the axis a fourth pair 4, nearly, but not
quite, as large as the first pair. This egg was so perfectly sym-
metrical, and its spherules so well defined, that I felt sure that it
was alive, and therefore determined to keep constant watch of it
until some change took place. I do not know how long it had
been in this condition before I placed it under the microscope, but,
for two hours after, it exhibited no visible change whatever. At
the end of this time nuclei became visible in the cells 4 and 3 and
soon divided, and at the end of five minutes each of the spherules 3
had divided into two, as shown in Figure 2 ; each of the spherules 4
had a double nucleus, one of the cells 2 a double nucleus, and the
other a single one.

In five minutes more, Figure 3, all the spherules were in some
stage of division, but this was more advanced on one side of the
axis than on the other. In five minutes more, Figure 4, all the
spherules except 2 and 4 on one side were perfectly divided. In
ten minutes more, Figure 5, the division was completed; the
blastoderm was divided into sixteen spherules, and these were
symmetrically arranged in pairs, on the two sides of a long axis,
which was identical with that of Figure 1.

The perfect bilateral symmetry of this stage formed such a
marked contrast to all the stages between it and Figure 1, that I
felt confident that it marked the end of a period of segmenting
activity, and that a period of rest would now follow.

The result fully justified this supposition, for, although I
watched it for more than three hours, no more change was visi-
ble, and when I retired at night it was as shown in the figure.

It was not dead, for the next morning the blastoderm was found
to be divided up into a great number of small cells, as shown in
Figure 6, which is a little more magnified than the other figures.

During the summer I observed the same phenomenon in the
segmenting egg of an Arthropod, and it was observed by Mr.
Wilson in Annelid eggs. Dr. Clarke has also observed it in
Amblystoma, and I think we may conclude that it is charac-
teristic of segmentation in general ; that wherever circumstances
admit of a careful time-record, the active changes will be found to
be separated from each other by periods during which there is no
visible external change.

A NEW METHOD OP STUDYING THE MAMMA-
LIAN HEART. By H. NEWELL MARTIN, M. A., D. Sc.
M. D. With Plate TX.

In the course of some experiments made by me in conjunction
with Dr. W. T. Sedgwick, on blood pressure in the coronary
arteries of the heart, the fact was impressed upon me that the
mammalian heart is no such fragile organ as one is usually
inclined to assume, but possesses a very considerable power of
bearing manipulation. On the other hand, I knew of various
unsuccessful attempts to isolate the mammalian heart and study
its physiology apart from the influence of extrinsic nerve centres,
in a manner more or less similar to the methods so frequently
used for physiological investigations on the heart of a cold-blooded
animal ; the mammalian heart, however, always died before any
observations could be made on it Thinking over the apparent
contradiction, it occurred to me that the essential difference proba-
bly lay in the coronary circulation ; in the frog, as is well known,
there are no coronary arteries or veins, the thin auricles and
spongy ventricle being nourished by the blood flowing through
the cardiac chambers, but in the mammal the thick-walled heart
has a special circulatory system of its own and needs a steady
flow through its vessels, and cannot be nourished (as appears to
have been forgotten) by merely keeping up a stream through
auricles and ventricles. The greater respiratory needs of the heart
of the warm-blooded animal also needed consideration; the lungs
ought either to be left connected with it, or replaced by some other
efficient aerating apparatus; if entirely separated from the central
nervous system there seemed no need to replace the natural lung
by an artificial one, and, though I hope ultimately to do this, my
work hitherto has been confined to the study of heart and lungs
living together, when all the rest of the body of the animal was
dead. Under such circumstances, with uniform artificial respira-
tion, the lungs may be regarded as purely physical organs adapted
for gaseous diffusion ; and probably better for this purpose than
any substitute which could be constructed.

My first experiments were made with cats. The animal was
narcotised with morphia, tracheotomised, and a cannula put in the
14 119

120 H. NEWELL MARTIN.

left carotid. Then the thorax was opened, (artificial respiration
being started), the innominate artery tied beyond the origin of the
left carotid but proximal to the point where the right subclavian
and right carotid separate ; the left subclavian was ligatured near
its origin ; and the aortic arch tied immediately beyond the
organ of the left subclavian. Finally, the superior and inferior
cavffi and the root of one lung were tied ; the cannula in the left
carotid was connected with the manometer of the kymographion,
and tracings taken in the usual manner. Under these circum-
stances the course of the blood was — left auricle, left ventricle,
aortic arch and the ligatured arterial stumps connected with it, the
coronary vessels, the right auricle, the right ventricle, the pulmon-
ary circulation through one lung, and back to the left auricle. All
circulation was cut off from every organ in the body except heart
and lungs; the brain and spinal cord soon died, the muscles
became rigid, and kidneys and liver had no longer any physiological
connection, either through the nervous system or the blood, with
the heart; which, though still in the body, was physiologically
isolated from everything but the lungs; yet as my preliminary
experiments shewed (Johns Hopkins University Circular, No. 10,
p. 127, April, 1881,) the heart went on beating with considerable
force and regularity for more than an hour.

The method, however, still left much to be desired ; I wanted
the heart alive much longer ; a means of keeping it at a uniform
temperature ; a method of renewing the blood which, either be-
cause clogged with waste products usually removed by the kidneys
or other organs, or because certain nutritive materials in it were
used up, ceased to be efficient in keeping the heart alive after a
certain time ; and opportunity to run blood, to which various sub-
stances had been added, through the heart from time to time in
order to study their action upon it.

After several attempts the apparatus represented in Plate IX
was devised, and has been found to answer admirably ; with it I
have kept a heart, isolated physiologically from everything but the
lungs, beating with beautiful regularity for more than five hours,
and have no doubt I could keep it considerably longer were that
necessary.

In the plate the heart is represented very diagrammatically and
of hugely disproportionate size; the pulmonary vessels also are
entirely omitted, as they are not interfered with in the experiment.

THE MAMMALIAN HEART. 121

At first I thought the immense disproportion in capacity between
the complete pulmonary system of vessels and the systemic circu-
lation reduced to only its coronary portion would injure the work-
ing of the heart, and I tied up, as above stated, the root of one
lung and sometimes one or two lobes of the other ; but I have
since found that this is quite unnecessary ; the left auricle takes
only what it wants, no matter how much blood is accumulated in
the lungs, and the circulation is thus confined to the quantity of
blood which under a given aortic pressure is sent through the
coronary system in a given time.

The course of an experiment is as follows : Tracheotomy having
been performed, each pneumogastric nerve is divided in the neck ;
this is, I find, of importance as saving the heart from the effects of
powerful dyspnoeic inhibition when subsequently all the cerebral
circulation is cut off. A cannula, p, is then placed in the left
carotid, o; and another, «, in the right carotid, r; the purpose of
these will be mentioned presently. Next the first pair of costal
cartilages and the piece of sternum between them are resected,
artificial respiration started, and the internal mammary arteries
found and ligatured where they pass forwards between the apices
of the lungs. The sternum and the sternal ends of the ribs are
then cut away down to the diaphragm, and if the day is cold a
cloth soaked in moderately hot water laid over the posterior half
of the chest so as to keep lungs and heart warm, care being taken
that it does not touch the pericardium ; this hot damp cloth is
renewed from time to time as necessary ; on a warm day it may be
omitted.

Next the superior cava is pushed aside and the right subclavian
artery, w} clamped and opened. The bulb of a slender ther-
mometer, a, is then placed in the vessel and, the clamp being
removed, is pushed down into the innominate trunk and tied so
as to keep it there. This gives the temperature of the blood flow-
ing through the heart, which cannot be deduced accurately from
the temperature of the chamber in which the apparatus is placed ;
partly because the blood warms and cools more slowly than the
air in the box, and partly because in its circuit through the lungs
it is cooled. A very small twig given off from the innominate trunk
to the anterior mediastinum is also tied. Next the left subclavian,
m, is isolated and a cannula, x} placed in it ; and the aortic arch, /,
tied just beyond the origin of the left subclavian. When the sub-

22 H. NEWELL MABTI*.

clavians and aorta are tied (the carotid flow being already stopped)
anaemic or dyspnceic convulsions occur, and arterial pressure rises
very high, as evidenced by the great size to which the stamps
connected with the aortic arch become distended ; to obviate this
strain on the heart, the aortic arch is tied as quickly as possible
after putting the cannula in the left subclavian, and before the
dyspnoea is extreme a large quantity of blood drawn off through
the cannula, *, in the right carotid ; when what appears sufficient
is drawn the screw-clamp u is tightened up again. Finally the
inferior cava, e, is ligatured, and the azygos vein, /; and a can-
nula, A, put in the superior cava, g. This finishes the operative
procedure.

To get rid of the blood now present in the heart and lungs,
which would be apt to clot in the cannula during a subsequent
prolonged observation, and to replace it by defibrinated blood, of
which about two litres are obtained from other dogs before the ex-
periment, is the next step. The cannula h is filled with whipped
blood and connected with a funnel containing the same warmed to
35° C; the clamp t on the right carotid is then again opened and
from 300 to 400 c. c. of defibrinated blood run through the heart
and lungs — in by the superior cava and out by the carotid — wash-
ing out and replacing the blood previously present; the blood
drawn is whipped and strained and added to the stock on hand.
The supply should be slow and sent in under a pressure equal
to that exerted by a column of blood about 20 centimetres in
height. The carotid is then again clamped and the vena cava a
second or two later, after the heart and lungs have filled up with
blood. The funnel is now removed and the heart, still lying in
the chest, is ready for transference to the chamber in which it
is to be kept warm and moist arid fed with fresh defibrinated
blood.

This chamber consists of a box five feet long, three high, and
two and a half wide. It has no floor; has one wooden end, I; a
wooden back ; a glass front ; a glass roof, K ; and a glass end, L.
The front can be entirely removed and has also a door in it
through which matters can from time to time be arranged inside
and temperatures read off without removing the whole front. The
chamber rests on a galvanized iron trough, DD, which contains
about an inch and a half of water. In it is a Bunsen's regulator
connected with the burners CC} and serving to maintain a uniform

THE MAMMALIAN HEART. 123

temperature in the interior. In the chamber about an hour before
the experiment are placed the glass cylinders 27 and 28, each con-
taining about 800 c. c. of fresh whipped and strained dog's blood,
which has thus time to attain the temperature of the interior of
the box.

All being ready the front of the chamber is removed and the
dog stretcher OGf having on it the dead body of the dog with the
living heart and and lungs, is put in. The heart alone is indicated
in the diagram to make description of its connections easier. The
cylinders 27 and 28 are elevated on a block at the anterior end of
the stretcher, so that their lower ends are ten or twelve centimetres
above the auricular end of the heart. These cylinders are Mar-
riott's flasks. Each is closed air-tight at the top by a cork through
which four tubes pass; one tube in each case (9 and 12 respec-
tively) allows air to enter from the interior of the chamber and
reaches to near the bottom ; another (5, 6) dips a little deeper into
the blood and acts as a syphon to draw it off. The remaining tubes
(7 and 10, 8 and 11, respectively,) only reach a short way through
the cork. Each has on its upper end a bit of rubber tubing which
can be closed air-tight by a clamp, and is so when the cylinder is
in use. These short tubes are for filling the reservoirs ; when one
cylinder is nearly empty, as for instance 27 in the diagram, the
clamp, 2, on the tube leading from it to the heart is screwed up,
and the communication between the heart and the other reservoir
opened; while this second one is feeding the heart the first is
refilled by opening the clamps 18 and 17, putting the funnel 19
on the rubber tubing of 11, and refilling the reservoir through it;
as the blood enters the air escapes through 10 ; when the cylinder
is filled the clamps 17 and 18 are again screwed tight and the
cylinder is again ready for use long before its fellow has emptied.

The syphons leading from each Marriott's flask meet in the
Y-piece z from which passes the rubber tubing i. As soon as the
animal is placed in the chamber this bit of tubing is filled with
blood by opening its connection with one of the reservoirs and is
immediately slipped over the end of the cannula, A, in the superior
cava, from which the clamp is removed : the heart is thus steadily
supplied with blood from each reservoir in turn. The outflow
tube, 9, passes from the left carotid, o, which is not used for the
preliminary bleeding and washing out which, with the object of
avoiding any clotting in the left, arc done through the right carotid

124 H. NEWELL MARTIN.

as above described ; now that there is only defibrinated blood to
deal with there is no longer any danger of such clotting. Over the
cannula, p, is slipped one end of the rubber tube, q, which leads to
the glass tube 21, which passes through the wooden end of the box
and has on it a stopcock, 22, beyond which the tube curves round
and reenters the box. By means of the stopcock the rate of irri-
gation can be regulated without opening the chamber ; the blood
which flows through is received in the vessel 24, which is set aside
within the box and replaced by another from time to time as
necessary, until one of the Marriott's flasks needs refilling. In
this way the blood being nearly always inside the chamber does
not get a chance to cool more than a degree or two, and so has
ample time to heat up again to the proper point while the other
Marriott's flask is emptying. The rate of flow permitted is usually
a pretty rapid dropping ; but if a low arterial pressure is desired the
stop-cock, 22, is opened wider ; if a higher it is more closed. Even
a slow dropping keeps the heart well alive for a long time; if
signs of feebleness come on, all that is needed is to open the stop-
cock wide for a few seconds and thoroughly renew the blood in the
heart.

Arterial pressure and the pulse curves are obtained from the
mercurial manometer 26. This, by means of connecting tubes,
filled with sodic carbonate solution in the usual manner, is attached
to the cannula x in the left subclavian.

All the connections having been made the front is replaced on
the chamber and henceforth the heart beats on in it without dis-
turbance, except as from time to time a small door is opened to
change the receptacle 24, or take out blood to refill one of the
Marriott's flasks and change the one connected with the heart by
opening or closing the clamps 1 or 2, or note the temperature of
the thermometer a.

The description of the various connections to be made after the
animal is placed in the chamber takes some time, but the whole
thing is done in two or three minutes. While the front of the
chamber is out the air in it cools considerably, but the blood of
course much less on account of its high specific heat, and in a very
few minutes, while one waits for the heart to get uniform and to
be sure that brain and spinal cord are dead, all inside is again at a
uniform temperature and a series of observations can be com-
menced. Before commencing these I always wait until all signs of

THE MAMMALIAN HEART. 125

reflex excitability are lost and the muscles begin to exhibit rigor ;
this occurs at latest iu half an hour after ligaturing the various
arteries. Sometimes Traube's curves are seen for a few minutes
after the animal is placed in position, shewing that the medulla is
not quite dead; but they very soon pass off never to return,
though when the heart begins to die something simulating them
(to which I will return later) usually occurs..

It is, I think, clear that by this plan of work the study of the
physiology of the mammalian heart is made possible to an extent
never before attainable ; I have now made a considerable number
of observations which shew that for at least four hours and often
for considerably longer, great regularity and power in the heart's
beat can be maintained. I give below in tabular form the succes-
sive observations as to pressure in the subclavian and pulse rate
made in two experiments, which shew the perfect availability of the
method. To investigate the direct action of any drug on the heart
one would have only to inject it by a hypodermic syringe into the
cardiac end of the tube t, as in the usual manner of injecting curari
into a vein. By altering the temperature of the chamber one can
readily study the effect of various temperatures on the pulse rate,
arterial pressure being kept at a given level while the tracings (at
intervals of five or ten minutes) are being taken, by altering the
outflow through the stopcock, if necessary ; between the readings
a uniform flow is kept up irrespective of arterial pressure. By
keeping the temperature constant and altering the stopcock the
direct influence of various arterial pressures on the pulse rate can
be readily studied. On these two latter points I have already
made a number of interesting observations, which are not, how-
ever, yet quite ready for publication. The chemical products of
muscular work apart from those eliminated by the lungs must also
accumulate in the blood which has flowed round and round the
beating heart for hours, and probably can there be examined better
than in any other organ at present at our disposal. It seems
also to me practicable to unite a given organ, say kidney or liver,
with the heart and keep it alive for study, but this I have not yet
tried. At any rate it is clear that a large field for investigation of
various points of great interest is made available for study under
much more favorable circumstances than hitherto.

When the heart begins to die the first symptom is an irregular
rhythm which cannot be removed by free irrigation with the blood

126 H. NEWELL MARTIN.

in the reservoirs. Whether this is immediately due to changes in
the heart itself, or to the consumption of food materials in the
stock of blood, or to the accumulation in it of wastes usually
removed by the kidneys or other organs I cannot at present state.
Whether it be due to the first of the above causes could readily be
decided by taking an entirely fresh stock of defibrinated blood.
The irregularity manifests itself by a large beat followed by three
or four smaller ones, and so on for more than an hour. Then the
small beats become feebler and feebler, and, arterial pressure being
consequently very low, the pulse due to the more powerful beat
very conspicuous. Finally the large beats alone remain, and they
gradually become less and less until they disappear. In its earlier
stages the phenomenon has an interesting resemblance to the sec-
ondary rhythm observed in the frog's heart under certain circum-
stances ; it is what I referred to above in stating that late in the
experiment something simulating Traube's curves is often seen.

For the guidance of those who may repeat the experiment, I
may add that the thing most to be avoided is sending blood into
the superior cava too fast or under too high a pressure; this is far
more fatal than considerable cooling or delay.

The following tables give the results of two experiments. In
each case the number indicated in the column headed " pressure "
is the pressure in millimetres of mercury indicated by the mano-
meter connected with the left subclavian artery. The numbers in
the column headed " pulse " give the number of heart beats per
minute. Temperatures, when given, (Table II.) are not accurately
those of the heart or blood, but those of the chamber in which the
heart lay. The introduction of a thermometer into the innomi-
nate trunk is one which I have only used in later experiments on
the influence of temperature changes on the pulse rate, when an
accurate knowledge of temperature was essential ; in the experi-
' ments given here the point I had in view was merely to determine
whether an isolated heart could be kept alive long enongh for
study ; and accuracy as regards temperature readings within a de-
gree or two was not essential.

Table I records the first experiment, which showed me that the
end I had in view was really attainable, and is given partly, per-
haps, because I have a special interest in it on that account, but
chiefly because it illustrates how well the heart will live under
very rough experimental conditions. At the time when it was

THE MAMMALIAN HEART.

127

made I had not arranged any warm chamber, and the heart was
simply warmed in the roughest manner by inverting a tin pan
over the body of the dog and putting a Bunsen's burner under
this; with some wet cloths to keep the atmosphere moist. From
time to time, the gas was turned down or up as I thought the
temperature round the heart was too high or too low, but no ther-
mometer readings were taken, and the temperature no doubt varied
very much in the course of the experiment. At this time also the
use of the Marriott's flasks had not been thought of: from time to
time, as the heart seemed weakening, fifty cubic centimetres of
whipped blood were run in by the vena cava and an approximately
equal bulk removed through the carotid. The numbers given
therefore as to pulse rate and arterial pressure have little or no
value; and the whole experiment simply serves to show with what
rude appliances the isolated heart can be kept at work for a long
time when the coronary circulation is maintained.

Table I.

Experiment of April 1, 1881.

Time. P.M.

1 h. 35'.

Pressure.

1 h. 40'.

2 h. 20'.

2 h. 22'.

2 h. 23'.

104

2 h. 30'.

102

2 h. 37'.

118

2 h. 40'.

2 h. 50'.

100

3 h. 04'.

100

3 h. 21'.

3 h. 28'.

104

3 h. 50'.

3 h. 51'.

3 h. 52'.

112

4 h. 06'.

4 h. 13'.

4 h. 15'.

4 h. 16'.

[Remarks.

Finished tying up all the vessels
but those of the pulmonary
and coronary circuits.

Fresh blood run in.

Fresh blood run in at 3 h. 3'.
Cold blood run in at 3 h. 27'.
Fresh warm blood run through.

Fresh warm blood run through.

128

H. NEWELL MARTIN.

Table I. — Continued.

Time. P. M.

Pressure.

Pulse.

Remarks.

4 h. 29'.

Fresh warm blood ran through.

4 h. 30'.

4 h. 39'.

Fresh warm blood run through.

4 h. 40'.

4 h. 47'.

4 h. 59'.

Fresh warm blood run through.

5 h. 00'.

5 h. 09'.

5 h. 10'.

Fresh bldod taken from another
dog and not used before in
the course of this experiment,
run through.

5 h. 1 8'.

140

5 h. 23'.

5 h. 26'.

Fresh blood.

5 h. 29'.

116

5 h. 33'.

5 h. 35'.

Fresh blood.

5 h. 40'.

5 h. 44'.

Fresh blood.

5h. 45'.

102

Chronograph pen out of order,
so the pulse rate cannot be
given.

5 h. 48'.

5 h. 53'.

Fresh blood.

5 h. 55'.

6 h. 00'.

6 h. 02'.

Fresh blood.

6 h. 03'.

6 h. 11'.

Fresh blood.

6 h. 14'.

6 h. 20'.

6 h. 22'.

Fresh blood run in ; none drawn
off.

6 h. 24'.

118

6 h. 30'.

6 h. 35'.

6 h. 36'.

Fresh blood run in; none drawn
off.

6 h. 38'.

118

100

6 h. 41'.

The beat immediately after-
wards became very irregular,
and ceased finally at 7 h. 10'.

THE MAMMALIAN HEART.

129

The above experiment, as already stated, justifies no conclusions
except that an isolated mammalian heart can be kept beating for
several hours. It, however, suggests (and subsequent experiments,
which I hope shortly to publish, confirm) that the pulse rate of
the isolated heart is very independent of arterial pressure, though,
as no accurate temperature observations were made in this case,
the experiment by itself is not worth much in that respect.

Table II.

Experiment op May 26, 1881.

Time. P. M.

1 h. 50'.

2 h. 05'.

2h. 15'.

2 h. 45'.

3 h. 00'.
3 h. 15'.

3 h. 55'.
4h. 15'.

4 h. 35'.
4 h. 50'.
5h. 10'.
5h. 45'.
6 h. 00'.
6 h. 15'.

Temp, in
degrees C

95°
99°

98°

99°

100°

99°

86
87
90
91
87
86
68
64
60
56

118
118
120
120
118
120
117
117
118
117

Notes.

All vessels tied bat those of
the coronary and pulmonary
circuits. Then 150 c. c. of
warm whipped blood sent
through the heart in order
to wash out the blood .al-
ready in it and in the lungs.
Animal removed to warm cham-
ber and the irrigation started
from the Marriott's flasks and
maintained thenceforth.
Poise rate not known, as the
chronograph was not work-
ing.

Arterial pressure now began
to fall markedly, and while
a fresh supply of blood was
being obtained from another
dog (that in use having al-
ready circulated round the
heart many times, and being
presumably full of wastes)
the orpan ceased to beat at
6 h. 45'.

130 H. NEWELL MARTIN.

The experiment described in Table II was made in the warm
chamber described in the preceding pages and with the Marriott's
flasks, giving a uniform instead of the intermittent supply of fresh
blood used in the experiment of Table I. It is one of a number
which all shew the great regularity which can be obtained for
some hours in the heart's work under such circumstances; and
hence the possibility of readily observing the influence on its
activity of various conditions and of drugs: in other words, it
indicates that the separated organ is in a fit condition for physio-
logical or therapeutical experiment.

During the earlier part of the above experiment (from 2.15 to
3.00 P. M.) the chamber and its contents were considerably cooled
in consequence of one of the Marriott's flasks being out of order
and necessitating the keeping open of the doors, for its repair.
When this was accomplished, we find for the subsequent two hours
(3 h. 00' to 4 h. 50') a very remarkable uniformity in the heart's
work. Arterial pressure only varies between 86 and 91 mm. of
mercury, and the pulse rate between 118 and 120 per minute.
Probably under no conditions would a heart still connected physi-
ologically with the rest of the body display so great a uniformity
in its activity for so long a time. The pulse, it will be seen, still
remained very regular to the end of the experiment, although
arterial pressure fell ; this again illustrates the slight influence
exerted by aortic pressure upon the rhythm of the isolated heart.

A NOTE ON THE PROCESSES CONCERNED IN
THE SECRETION OF THE PEPSIN-FORMING
GLANDS OF THE FROG. By HENRY SEWALL,
Ph. D., Associate in Biology, Johns Hopkins University.

It has been shown, chiefly through the labors of Langley, that
the oesophageal glands of the frog undergo in digestion marked
histological changes. When the animal in healthy condition has
fasted several days, the oesophageal glands are full of fine granules
throughout, and no boundary lines between the cells can be made
out in the fresh gland. Examined two or three hours after feed-
ing the glands are found to be void of granules on their outer
borders, the hyaline matrix alone remaining; this process may
extend until all the granules have disappeared except a larger or
smaller group collected round the gland lumen. The return to
the normal resting appearance occurs usually one to two days after
light feeding. These changes are in certain species, as in R. tem-
poraria, so well marked that I thought it advantageous to use
them as a sign of the secretory activity of the oesophageal or pep-
sin-forming glands, when these were excited by the absorption of
different food materials from other than the usual surfaces. It
was thought that one might thus get a better idea of the conditions
and mechanism of secretion. The experiments were conducted in
the cold months, from February to May, and the frogs were in
large part taken fresh from the mud in which they had buried
themselves for the winter. The great changes necessarily taking
place in the animals while coming into a more active condition, no
doubt account for much of the want of uniformity which was
observed in the behavior of many of the specimens examined.
The animals chiefly used were small specimens of the bull-frog,
jR. mugiens. Of two frogs apparently alike it sometimes happened
that one would show distinct secretory changes in the oesophageal
glands three hours after feeding on meat or beef fibrin, while the
other examined at the same time preserved the hungering appear-
ance of its glands unchanged, but this was exceptional.

In a small spotted frog, R. halecina, the diminution of granules
in digestion, if it occurs at all, goes on very slowly. I was not

131

132 BENE Y SE WA LL.

able to detect a disappearance of the granules in this frog in less
than twenty-four hours after feediug, though digestion had evi-
dently been active in the interval. A conclusion may be stated
in advance that secretion apparently involves the glands simulta-
neously in two opposite activities, a breaking down and a building
up, and it is the ratio of the vigor of these changes which deter-
mines the histological appearance of the gland at any time. Nearly
all the microscopic examinations were made upon the fresh gland
by snipping off a piece of the mucous membrane of the oesophagus
and mounting quickly in iodized amniotic fluid of the sheep.

For convenience in description I will indicate by letters the
various solutions used as food stimuli. A was a solution obtained
by extracting beef muscle with 0.5 per cent. NaCl. B was a con-
centrated commercial " peptone,"% said to be the product of the
peptic digestion of beef muscle; this was diluted five to ten times
before use; it contained probably the flesh " extractives " as well
as peptone. C was a concentrated solution obtained by boiling
dog's muscle in water. D was a strong watery solution of pep-
tone, made in the laboratory by the peptic digestion of fibrin.
E was 0.5 per cent. NaCl.

In all cases comparative examinations were made of similar
frogs unmolested and experimented upon.

The fluids mentioned were injected in quantities usually of 1 to
2 c. c, either into the rectum or under the skin.

When the injection was into the rectum the fluid was usually
allowed to flow from a pipette inserted into the anus ; a safer way
was subsequently found in gently tying the frog in a prone posi-
tion and allowing the solution to run from a burette through a
cannula into the rectum. There was no evidence in any case of
fluid having reached the stomach. The bladder and rectum only
appeared to be filled.

1 . Injection into the rectum.

Injection of both A and B into the rectum caused in the hungry
frog marked disappearance of granules from the oesophageal glands.
The evidence as to C was not satisfactory. E appeared also to
have a distinct effect. It is to be remarked that the disappearance
of granules begins very quickly after the injection, the process is
rapid and recovery to the original condition of full granulation is
speedy, except in the case of injection with NaCl. The diminu-
tion of granules is marked twenty minutes after injection, and in

PEPSIN-FORMING GLANDS OF THE FROG. 133

fifty to eighty minutes the glands have again become granular
throughout.

The stomach of a frog which has hungered several days con-
tains generally very little mucous fluid, which is usually acid.
The stomach wall itself appears to be always acid. In the expe-
riments described above there was no considerable increase in
the stomach contents accompanying the histological change of
the oesophagus. And it may be said here that I have' been able
to discover no relation between the amount of fluid secreted into
the stomach during any period and the histological appearance of
the oesophageal glands at that time.

2. Injection into the dorsal lymph sac.

There was found no distinct evidence of a diminution of gran-
ules following a hypodermic injection of the fluids enumerated
above. On the contrary, the injection of B was almost always
succeeded by an accumulation of granules in the glands, even
under conditions, as in active digestion, in which a diminution of
granules was to have been expected. Such an injection apparently
accelerates and intensifies the normal digestion.

3. The relation between the injection into the lymph sac and the
amount of secretion found in the stomach.

When C was injected under the skin of a fasting frog, the
stomach was found at the end of one to two hours very much
distended, with a rather thin, neutral or slightly acid, raucous fluid.
The same result to somewhat less extent followed the use of D.
This effect from B was particularly marked where the frog had
been previously fed. The results from A were in the same direc-
tion but less noticeable, and no such effect followed the use of E.
No experiments were made to determine the digestive value of
this secretion. It appeared to increase in quantity for a period
considerably longer than that required by the artificially excited
glands to recover their resting appearance. The causes that pro-
duce the secretion seem rather to increase than diminish the
granules of the glands.

As to what are the steps in the recovery of granules by the
glands after their disappearance in digestion, nothing decisive can
be said. There was one well marked histological character which
distinguished the glands of certain frogs, the meaning of which
seems well worth investigation.

i34 HENRY SEW ALL.

This peculiarity was the presence of very large well defined
masses in the gland cells, usually in their outer part. These
masses were strongly suggestive of some of the forms of lymph
corpuscles which are numerous in the wall of the oesophagus.
They have much the refractive characters of the fat or of fresh
fat cells. Sometimes they exist as clumps of highly refractile
granules, and it is clear that their substance exists in very different
states of division in the glands.

These bodies stain black with osmic acid. They are dissolved
by ether. This reagent dissolves out also all of the granules from
the oesophageal glands, leaving behind only a clear gland substance
with the cell nuclei imbedded therein. It may be observed that it
is unsafe to draw conclusions from the appearance of specimens
preserved in balsam, for the preliminary treatment with the
clearing fluid dissolves out many of the finer granules previously
present in the cells.

The granule masses referred to were by far most numerous in
glands which, there was reason to think, were being actively
regenerated as to their granules; that is after a long period of
normal digestion or in the case of the injection of a food solution,
as By during normal digestion. The reactions indicate that they
are of fatty nature.

General concluiions :

The general conclusions toward which these results lead are that
the secretory changes in the glands of the oesophagus are started
by the mere absorption of matter from the alimentary canal but
that the regeneration of the glands depends upon the presence of
new matter in the blood itself.

The presence of foreign matter in the blood may cause an ex-
tensive secretion into the stomach and if in this case the secretion
comes from the oesophageal glands these are rebuilt quite as fast
as they are broken down. It is probable that the secretory process
is initiated by stimuli in the alimentary canal but is chiefly carried
out by the influence of substances newly absorbed into the blood.

The secretion of the acid of the gastric juice seems to be due to
the presence of food matter resting in and absorbed by the stomach
itself. The results obtained from these experiments give general
support to Setoff's views concerning peptogenic substances.

LIST OP MEDUSA POUND AT BEAUFORT, N. C.„
DURING THE SUMMERS OF 1880 AND 1881. By
W. K. BROOKS.

During the two seasons which we have spent at Beaufort the
members of our party have derived great benefit from the lists of
the Vertebrate and Invertebrate fauna of Fort Macon, by Drs*
Yarrow and Coues, which were published by the Philadelphia
Academy of Sciences in 1871. We have been able to make
many additions to these lists, especially in the various groups-
of invertebrates, and as the authors made no attempt to collect or
identify the medusae of these waters I have drawn up the following
list of Acalephs and Ctenophorae, from the notes which I made
during the summers of 1880 and 1881.

McCrady aud L. Agassiz have studied the medusae of South
Carolina, and I give, for convenience of reference, a list, compiled
from these authors by A. Agassiz (N. A. Acalephs), of the forma
which occur at Charleston, for comparison with my own list of
Beaufort species.

Acalephs of Charleston, 8. O.

(From A. Agassiz' N. A. Acalephs, pp. 223-4.)

Bolena Uttoralis, McCr.

Mnemiopsis Oardeni, Ag.

Beroe punctata, Esch.

Idyiopsis Clorkii, Ag.

Stomolophus meleagris, Ag.

(10

(«■)

Oyanea versicolor, Ag.

Foveola dctonaria, A. Ag.

(3.)

(*•)

(5.)

(6.)

Persa incolorata, McCr.

(7.)

Liriope scutigera, McCr.

(8.)

Oceania folliata, Ag.

(9.)

EuehtUota ventricularis, McCr.

(10.)

(11.)

Acalephs CoUected at Beaufort during:
the Summers of 1880 and 1881.

Mnemiopsis Oardeni, Ag.
Mnemiopsis Leidyi, A. Ag.

Idyiopsia Clarkii, Ag.
Stomolophus meleagris.
Dactylomctra quinquecirra, Ag.

Foveola octonaria, A. Ag.
Cimina discoides, Fewkcs.
Cheiropaalamus quadrumanus, P.

Muller.
Tamoya haplonema, F. Muller.
Per sa incolorata, McCr.
Liriope scutigera, McCr.
Liriope scutigera, A. Ag.
Oceania folliata, Ag.
Eucheilota ventricular is, McCr.
Dipleuron parvum, sp. nv.

135

186

W. K. BROOKS.

Clytia bicophora, Ag.
Platypyxis cylindrical Ag.
Eucopc divaricate, A. Ag.

Obclia commisuralis, McCr.
Eirene gibbosa, Ag.
Eutima mira, McCr.

Eutima variabilis, McOr.
Aglaophenia tricuspid, Ag.
Aglaophenia trifida.

Plumularia quadridens, McCr.
Plumularia (Catharina-like), McCr.
Dynamena comicina, McCr,

Diphasia (nigra-like)t Ag.
Margelis Carolinensis, Ag.
Nemopsis Bachei, Ag.
Eudcndrium ramosum, McCr.
Turritopsis nutricula, McCr.
Stomatoca apacata, McCr.
WiUia ornata, McCr.
Dipurina cervicata, McCr.
Dipurina strangulata, McCr.
Corynetis Agassizii, McCr.
Oemmaria gemmoaa, McCr.
Pennaria tiarella, McCr.

Ectopleura turricula, Ag.

Parypha cristata, Ag.

Hydractinia polyclina, Ag.
Eudoxia alata, McCr.
Diphyes pusilla, McCr.
Physalia arethusa, Til.
Velella mutica, Bose.
Porpita linniana, Less.

(9.) Campanularia noliformis, McCr.

(12.) Eucopa obligua, sp. nv.

Obelia commisuralis, McCr.

Eirene gibbosa, Ag.

Eutima mira, McCr.
(18.) Eutima cuculata, sp. nv.
(14.) Eutima emarginata, sp. nv.

(15.) Nematophorus, sp. nv.

(16.) Dynamena bilatteralis, sp. nv.

(17.) Margelis Carolinensis, Ag.
(18.) Nemopsis Bachei, Ag.

Eudendrium ramosum, McCr.
(19.) Turritopsis nutricula, McCr.
(20.) Stomatoca apacata, McCr.
(21.) WiUia ornata, McCr.

Dipurina strangulata, McCr.
Corynetis Agassizii, McCr.

Pennaria tiarella, McCr.
(22.) Pennaria inomata, sp. nv.

Ectopleura ochracca, A. Ag.

Parypha cristata, Ag.
(23.) Steenstrupia gracilis, sp. nv.
(24.) Hydractinia polyclina, Ag.

Eudoxia alata, McCr.

Diphyes pusilla, McCr.

Physalia arethusa, Til.

Porpita linniana, Less.
Nanomia car a, A. Ag.

(1.) Stomolophu8 mdeagriBy Ag.

We found no living specimens of this species in 1880, although
the remains of two or three were found on the sand bars at low
tide, early in June.

In June, 1881, living specimens were extremely abundant both
outside the bar and in the sounds.

BEAUFORT MEDU82E. 13T

They could be seen floating or swimming at the surface on all
sides of the boat, and although they were so shy that they sunk
when approached, they were so abundant that we easily captured
all we could carry home. Those which we secured were from
four inches to twelve inches across the opening of the umbrella,
although larger specimens were seen.

Later in the season they were less abundant, but we found
specimens occasionally through June, July and August.

The fact that such a large and conspicuous species should be so-
abundant one year and almost absent another year shows the im-
possibility of thoroughly studying the fauna of our coast without
permanent marine stations.

(2.) Dadylometra quinquecirra, Ag.

This medusa is found in abundance all through the summer in
the lower part of the Chesapeake Bay. We never found it inside
the inlet at Beaufort, although we occasionally found it just outside
the bar, and early in September, 1880, it was common.

The southern form swims at the surface at all hours of the day
and night, and as it differs from A. Agassiz' description in several
slight particulars, it is probably a well marked southern variety.

(3.) Foveola octonaria, A. Ag.

Rather abundant in June and early July. Although TarritopsU
%uiricula is our most common medusa, we never found the young
Cunina in its bell.

(4.) Cunina discoides, Fewkes.

In August, 1880, I procured a single mutilated specimen which
i very similar in general form to Fewkes figure of Cunina dis-
<xndes1 although it has but twelve tentacles, and eight sense organs,

(5.) Cheiropsalamu8 quadrumanus, F. Muller.

This interesting medusa will probably be found to be by no
means rare along our coast, although it is seldom found at the
surface.

McCrady has found one specimen at Charleston, and one at
Port Royal.

158 W. K. BROOKS.

In July, 1880, we found a few specimens on the sand bars at
low tide, and throughout July, August and September we got
specimens in from three to eight fathoms outside the bar, on sandy
bottom. The specimens were taken from the bottom with the
trawl, and we found none at the surface, although those which we
kept in aquaria in the house swam near the surface. They were
from one inch to five inches across the umbrella.

(6.) Tamoya haplonema, Fr. Muller.

In July, 1880, a fisherman brought me a single living female of
this species. We found no others.

(7.) Perm incolorata, McCr.

Found occasionally at night, swimming at the surface, from
June 24th to August 8th. It is a very delicate species but many
of our specimens were perfect and healthy. We found twenty or
thirty in all. It is one of our most rare medusae. McCrady
found four specimens, and Haeckel has found other species of the
genus, but it seems to have entirely escaped other observers.

(8.) Liriope scutigera, McCr.

McCrady's Liriope scutigera is one of the most common medusas
at Beaufort, and as we found specimens at all stages of growth, we
were able to trace the whole of the interesting metamorphosis, and
to decide that it is not the same as L. scutigera, A. Ag. A single
specimen Nvhich seemed to belong to the latter species, was found
in July, 1880.

(9.) Oceania folliata, Ag.

We were able to trace the whole life-history of this abundant
species, and to settle a number of doubtful points concerning it.

The hydra — Campanularia noliformis, McCr. — is very like
Agassiz' Platypyxis cylindrica, but may be distinguished from it
by several constant features.

The upper or distal end of the reproductive calyx, is truncated
squarely instead of flaring, and the outline of the calyx is alike in
side and front view.

BE A UFOR T MED USJS. 139

The four or five medusae which it contains are nearly equal in
size, and they are discharged in quick succession, the last escaping
within a few minutes after the first.

The medusa, Epenthesis folliata, McCr. is very similar to Oceania
languida, A. Ag., but the tentacles and otocysts develop as A.
Agassiz describes them in Clyiia bicophora, Ag.

The difference between the hydra and Platypyxis cylindrica, is
so slight that a thorough knowledge of the life history of the latter
may show that it is only a northern variety; but there can be no
question as to the specific distinctness of the medusa from Oceania
languida.

(10.) Eucheilota ventricularis, McCr.

Mature and nearly mature medusaa are common at Beaufort,
from July 15th to the end of August, but the young ones were
more rare, although I was able to get a sufficiently complete series
to show that the young medusa found at Naushon, by Alex.
Agassiz, undoubtedly belongs to this species.

(11.) Dipleuron, novum genus.

Medusa with four radiating chymiferous tubes, four radial ten-
tacles with basal cirri, and twelve otocysts, four interradial and
eight on the sides of bases of radial tentacles. Reproductive organs
two, nearly spherical, on two opposite chymiferous tubes, near bell
margin. Stomach short, with simple mouth, without oral tentacles.

Dipleuron parvum, sp. nv.

Umbrella nearly as high as wide in profile view, with greatest
transverse diameter about half-way up, where there is a distinct
angle in the outline. Umbrella of uniform thickness from top to
free edge ; elliptical when seen from above or below, with major
axis nearly twice as long as minor axis. Proboscis a little enlarged
at the circular mouth, which has a simple edge. Reproductive
organs spherical, two in number, on two opposite radiating tubes
near bell margin, with a large central chamber, opening into radi-
ating tube by a long narrow vertical slit.

The four radial tentacles are usually carried with their tips
tamed upwards. Each tentacle carries at its base two small

140 W. K. BROOKS.

twisted cirri, and consists of a swollen pigmented bulb which
passes gradually into a long slender filament, which is usually
coiled in a loose spiral.

Otocysts twelve in number, of two kinds ; four large ones half-
way between the tentacles, and eight smaller ones, two at the
base of each tentacle. Each otocyst has a single otolith, and the
small otocysts are sometimes absent.

The largest specimens are about T$for inch in longest diameter.

This species is common at Beaufort, from June 5th to August
8th, and sexually mature specimens of both sexes are frequently
found.

It somewhat resembles A. Agassiz* JEucheilota duodecimals,
(Phialium dodecasemum, Haeckel,) except that the reproductive
organs are always two, and spherical.

(12.) Eucope obliqua, sp. nv.

Communities from half to two-thirds of an inch high. Hy-
drotheca slightly faring at edge. Knee oblique, lowest on side
nearest main stem, and highest on outside. Stem with from five
to seven annulations above each fork. Hydranths colorless, with
about thirty tentacles so placed that their tips form two circlets.

Reproductive calycles long, nearly cylindrical, abruptly trun-
cated at tip.

Medusae arranged in two rows; seven or eight maturing together.

When discharged the medusa is about tA^ inch across disc, with
two otocysts and six or seven tentacles in each quadrant.

The hydrae were frequently found on floating pieces of Sargas-
sum and on drift wood. The number of tentacles at the time the
medusa escapes from the calycle is quite variable, and although
twenty-four seems to be the normal number, I did not find a single
specimen with exactly twenty-four. Usually three of the quadrants
had six each, and the fourth seven or sometimes five.

After the escape of the medusae the distal half of the calycle
falls off, and its proximal end becomes converted into an ordinary
hydrotheca.

(13.) JEulima cucvlata, sp. nv.

Umbrella flat : height about one-fourth diameter. Gelatinous
substance very thick in centre, bo that the cavity of the sub- urn-

BEAUFORT MEDUSA. 141

brella is very shallow, and makes less than half the total height of
bell.

The umbrella diminishes in thickness gradually towards the
bell margin, where it forms a thin edge. Prolongation into pro-
boscis conical above, prismatic below, more than twice as long as
height of umbrella. Stomach a little enlarged, forming about
one-fifth of total length of proboscis, with four simple lips. Four
radial tentacles, very long, slender, imperfectly retractile, with very
slight basal enlargements, without accessory cirri. Nine or ten
slight enlargements of circular tube in each quadrant, and a few
of the enlargements have accessory cirri. Two otocysts, with from
three to eight otoliths, in each quadrant.

Reproductive organs run along radiating tubes from circular
tube to conical part of proboscis, but they do not run down onto
prismatic portion.

About one third of an inch in diameter. Stomach and tentacular
bulbs intense green by reflected light; ectoderm of tentacular bulbs
sky-blue, and endoderm bright pink by transmitted light.

A few specimens were found August 7, 1880. The bases of the
tentacles are covered by small semicircular flaps or hoods, from the
gelatinous substance of the bell, and I have named the species from
these, although similar hoods are found in Eutima mira, McCr.

The species may readily be distinguished by its very flat disc-
like umbrella, and by the great length of the tentacles. When
these were thrown out to three times the diameter of the bell they
were fer from straight, but were thrown into a number of sharp
angular zig zag folds. At first sight this species might seem
to belong to Haeckel's genus Eutimium. Although the basal cirri
are entirely absent, careful examination shows that the marginal
enlargements and cirri are present, but very small.

(14.) Eutima emarginata, sp. nv.

During the summer we occasionally found specimens of what
seems to be another new species of Eutima, but a more complete
knowledge of its life history may possibly show that it is the young
of a described species. If so it must undergo considerable meta-
morphosis.

It may be described as follows :

Medusa with a rather low bell, one third as high as wide, with
a strongly emarginated rim. Gastrostyle about three times as long

142 W. K BROOKS.

as height of bells, prismatic, with four prominent ridges along the
radiating tubes. Stomach no wider than, and about one third as
long as style, with four simple lips. Radiating tubes enlarged to
form four fusiform chambers on lower end of style, just before they
join stomach. Four radiating tentacles, tapering gradually from
base to tip, and capable of almost perfect retraction, although they
are never extended much further than the length of the proboscis.

Two otocysts with three ossicles each, in each quadrant. From
ten to twelve enlargements and three or four cirri in each inter-
radius, and a cirrus on each side of base of each radial tentacle.

The reproductive organs were not observed.

No hoods over radial tentacles.

The largest specimens were about one third of an inch in
diameter.

(15.) Nematophorus, sp. nv.

On August 18th, 1880, we took with the trawl off Fort Macon,
in three fathoms of water, great quantities of a beautiful feather-
like hydroid community, the stems being a foot or more in height.
They were all torn away from their attachment, but there was no
way to decide whether they had been pulled up by the trawl or
were drifting specimens from a distance.

The hydranths were alive, but they soon died in confinement,
and I did not see any in an expanded state.

At the base of each pinna there is one of the rounded perforated
bodies upon which Clarke has founded the genus Nematophorus,
but our species is much more like a typical Aglaophenia than
Clarke's Nematophorus grandisy and I cannot, without specimens
for comparison, state positively that it is not one of the described
species of Aglaophenia.

(16.) Dynamena bUatteralis, sp. nv.

Stems simple, unbranded, slightly curved; from one-fourth of an
inch to one inch high; springing from a creeping hydrocaulus.
From five to twenty pairs of hydranths on each stem. Hydrothecas
long, in contact with each other along middle line of convex side of
stem for about two-thirds of their total length. The distal third
bends outwards almost at right angles, and the bilobed openings
are almost parallel to the stem.

BE A UFOR T MED USJE. 148

The tentacles of the hydranth are arranged in an ellipse, with
its long axis at right angles to the long axis of the stem. The
tentacles at the ends of this axis are the shortest, and those at the
ends of the minor axis, or the top and bottom tentacles, are the
shortest Tentacles about twenty-two. Reproductive calycles at
base of stem, nearly spherical, with two or three obscure annula-
tions, a short constricted stalk, and a small circular mouth.

This form bears a general resemblance to Dynamena cornidnay
but I have never seen anything like the horn-shaped reproductive
calycles which he describes.

It is very abundant at Beaufort all through the summer. When
kept in confinement in a small quantity of water, the tips of the
stems grew to a length of several inches, forming a slender trans-
parent spiral thread. When the tips of these threads come into
contact with the sides of the glass, they become attached, and
throwing out branches, become the hydrorhizae of new commu-
nities, which flourish after the parent stock has died.

(17.) Margelis Oarolinensis, Ag.

Very common all through the summer, but we did not find the
hydra.

(18.) Nemopsi8 Baohei.

A few specimens were found in the early spring of each season.
The Beaufort form seems to be a southern variety, for all the
specimens found differ from A. Agassiz' figure, and from sketches
which I made in 1874 in his laboratory at Newport, in the outline
of the bell, and in the form of the median radial tentacles. The
bell is more flattened and its diameter exceeds its height, and the
median tentacles have rather slender shafts, with abrupt enlarge-
ments at their tips.

(19.) Turritopsis ntUricula} McCr.

.This medusa is found all through the season, and is the most
common species at Beaufort.

The young stages figured by A. Agassiz do not belong to this
species.
2

144 W. K BROOKS.

Notwithstanding McCrady's excellent description and figures,
Fewkes has figured and described it as a new genus and species
Modeeria multiienticulata.

(20.) Stomatoca apacta, McCr.

Rather common at Beaufort all through the summer.

(21.) WiUia ornata, McCr.

This is a rare species at Beaufort, and I have not met with any
sexually matured specimens. Those I found were obtained on
July 12th and 13th and August 18th, 1880.

The largest specimens had four stolons running off from the
four corners of the stomach just below the inner ends of the
radiating tubes. Each stolon soon branched dichotomously, and
ended in a medusa bud.

(22.) Pennaria inornata, novum species.

Stems wiry, horn-colored, branching irregularly so as to build
up a loose arborescent tuft five or six inches high. Hydranths
irregularly placed, usually on short lateral branches from secondary
stems, sometimes on tips of secondary stems, and occasionally on
short branches which spring directly from sides of large trunks.

Stem has from five to seven annulations distal to each fork, and
an equal number proximal to each hydranth.

Hydranths nearly colorless, with a circlet of from ten to twenty
short tentacles — only one-third as loug as hydranth — near the base,
and three, or sometimes only two, circlets of short clavate tentacles
around the long slender manubrium. There are usually five of the
clavate tentacles in the distal set, more in the second set, while the
proximal set varies greatly and may be absent.

Taken with the trawl outside Fort Macon, August 18th, 1880.

(23.) Steenstrupia gracilis, novum species.

Umbrella bell shaped, circular in cross section, with a long,
conical, sharply pointed apex, which makes half the total length,
and contains a still longer undulating prolongation from the
stomach. One long tentacle and three rudimentary ones, one
longer than the other two, and all four without ocelli. The long

BE A UFOB T MED USJE. 1 45

tentacle — the dorsal tentacle of Haeckel — may be extended to
nearly twice the length of the umbrella including the apex ; it is
ringed, and ends in a spherical enlargement. The bulb at its base
is no larger than those of the other three tentacles, and it has no
ocellus.

The tentacle opposite it — the ventral tentacle of Haeckel, — is
about three times as long as the other two rudimentary tentacles,
and the length of these latter is about equal to their width.

Radiating tubes arch upwards a little from the stomach, and
then pass outwards and downwards in graceful curves to the
circular tube.

Stomach usually about three-fourths as long as the cavity of the
sub-umbrella, although it may be protruded from the opening.
It is a little swollen in the middle, and tapers gently towards each
end.

The sides of the umbrella are nearly uniformly thick from top
to bottom, and in profile view their outline passes into that of the
apex by a very gentle curve, which is first convex and then concave.

Length of apex A inch, height of umbrella A inch, ordinary
length of long tentacle about § inch.

Found only on June 20th, 1880, in Newport River.

This graceful medusa may readily be distinguished from Cory-
morpha pendule, Ag., by the elongated apex, as well as by the fact
that the longest of the rudimentary tentacles is opposite the long
tentacle.

It may be distinguished from Hybocodon by its circular outline
in cross section.

On August 2d, 1880, and on two or three days of the same
week, I took from muddy bottom in three or four feet of water, a
number of specimens of a solitary hydroid, which may be the
young hydra of this species.

The specimens were naked, about £ inch long, and they had at
the upper end of the long slender body, a circlet of about twelve
long slender tentacles, with pigment spots at their tips; and some
distance above these, a circlet of six short clavate tentacles, also
pigmented at their tips.

The animals fastened themselves to the bottom of the tumbler
in which they were kept, and I was able to change the water
without disturbing them. The lower end of the body soon became
encased in a sheath of grains of sand and other small particles.

146

W. K. BROOKS.

They multiplied by transverse fission, the upper end separating off,
and fastening itself to the glass near the old trunk, which soon
developed a new head.

(24.)

Hydractinia polyclina, Ag.

The Beaufort Hydractinia is quite different from the descriptions
of the northern form, but I made no careful examination of it.

Summary.

Alex. Agassiz includes Velella mutica among the Charleston
species on the authority of McCrady, but as McCrady only says,
"I have never seen a VeleUay" we may omit it, and we shall then
have 42 species recorded as occurring at Charleston, aud 43 found
in two summers at Beaufort. Of these 43. 27 or more than half
occur at Charleston, and eight of the remaining 16 are new.

This list is not complete as there are three or four other forms
which are not described at present, as the data are insufficient, and
as our stay at Beaufort included the summer months there are, no
doubt, a number of winter species which we have not obtained.

Our open boat was so ill adapted for facing the line of breakers
on the bar that it was seldom safe to venture outside for a pro-
longed excursion, so we did very little with the deep water forms,
which our short excursions lead us to believe are very numerous
and interesting.

November 9th, 1881.

ON THE ORIGIN OP THE SO-CALLED "TEST-
CELLS" IN THE ASCIDIAN OVUM. By J. PLAY-
FAIR MoMURRICH, B. A., Assistant in the Biological Labora-
tory, University of Toronto. With Plate X.

The following observations have been made in the hope of
elucidating to some extent the nature of the so-called "test-cells,"
so characteristic of the ova of Tunicates. These bodies have been
described by various authors as occurring in the eggs of most of
the commoner forms, and under normal circumstances probably
do not make their appearance until after fertilization. Lacaze-
Duthiers(l) states that in Molgula a true layer of "test-cells" is
wanting, and only the follicle-epithelium surrounds the newly
deposited ovum. Under abnormal circumstances, however, they
are formed at a much earlier period, and thus in most eggs that
have been observed, "test-cells" were present or soon made their
appearance.

My observations have been carried on for the most part on ova
of A8cidia amphora, but I have also confirmed most of them by
similar experiments on eggs of Cynthia ocellaia. I made use only
of mature or almost mature eggs, so that I am unable to give as
complete an account of some points as could be desired.

The mature eggs of A. amphora (PI. X, Fig. 1) have an average
diameter of about .255 mm. and present on optical section two dis-
tinctly marked zones, enclosing a semi-transparent granular mass,
the yolk. The outer of the two zones is formed by the follicle-
epithelium, consisting of a single layer of cells surrounding the
whole egg, and presenting on a surface view a polygonal appear-
ance. On examining a single cell which has become separated
from the egg, with a rather high magnifying power, its interior is
seen to be occupied almost entirely by a number of vacuoles, sepa-
rated from one another aud surrounded by very delicate bands of
protoplasm, which, in some of the angles formed by the meeting
of the polygonal vacuoles, appear as dark spots. (PI. X, Fig. 2.)
I have not been able to observe the development of these cells, but
Semper (2) has described it as it occurs in Molgula nana, where, in

147

148 J. PL AY FAIR McMUBBIGH.

the earliest observed stages, they appear as a layer of flat cells on
the surface of the egg, which, later on, become prismatic, and in
the interior of which 2-4 yellow granules make their appear-
ance. These afterward disappear and large vacuoles take their
place, pressing the protoplasm and nucleus to one side. In his
figures, the formation of the vacuoles has not advanced as far as in
the eggs I studied, but, on comparing his Figure 5, Plate I, with
my Figure 2, Plate X, it will at once be recognized that the ap-
pearance I have observed is to be accounted for in the same
manner, the vacuoles having become exceedingly abundant, and
pressed the original contents of the cell to the periphery, small
portions only being left in the intervals between the vacuoles.

Fol,(3) having succeeded in tracing the origin of these cells still
farther back in Phalhma intestinalisy states that they are not
formed from the ovary, but from the interior of the egg at the
boundary between the yolk and the nucleus, and wander thence to
the surface, where they form an epithelial layer round the egg.
One would fancy at first that the eminent observer had accidentally
confused " test-cells" with the follicle-epithelium, but that he has
not done so is evident from his also describing the " test-cells " as
formed later on. This discovery is of great interest, both from its
upsetting all former theories as to the formation of these cells,
which have hitherto always been considered as being formed from
the ovary, and also from the singular manner in which Kowa-
lewsky's theory in regard to the formation of the " test-cells " from
these cells has been turned upside down, these bodies (*. e. the
" test-cells") being formed independently from the yolk (as will be
seen hereafter) from which at an earlier period the follicle-cells
had also been derived. If M. Fol's observations are correct, it
is evident that the term " follicle-cell " is entirely a misnomer,
as is also indeed that of " test-cell," both being to a high degree
misleading to one who has not studied the history of the ap-
pellations.

Within this layer of so-called " follicle-cells" comes the second
zone of the egg, consisting of a transparent, apparently homoge-
neous structure, which, however, when acted on by acetic acid,
becomes markedly granular. This is the egg-shell or "chorion"
of some authors.

In the interior of the egg-shell and filling it almost completely
in the fresh ovum, is the yolk. On the average it measured

A8CIDIAN OVA. 149

.236 mm. and was of a yellowish gray color, due to the coloration

of the constituent granules. In the majority of the eggs of this

Ascidian I examined, no nuclei were visible either in the fresh

egg, or in those that had been subjected to the reaction of acetic

acid and glycerine, or osmic acid and carmine. In some, however,

a clear spot was noticeable, usually situated eccentrically (in one

instance at the periphery of the egg), and measuring .020-086

mm. One egg presented a rather peculiar abnormality, which I

deem worthy of being recorded. The peculiarity consisted in the

presence of two distinct nuclei, both situated eccentrically on the

same side of the egg and varying somewhat in size, the larger

measuring .06 mm. and the smaller .04 mm. I am certain that I

did not observe the male and female pronuclei, as the egg under

observation had just been removed from the ovary, so that it could

not have been impregnated any length of time, if at all, before

xny observation of it.

These were all the points to be observed in a perfectly fresh
ovum, but in one that had been removed from the ovary for a
tshort time, or which had been subjected to the action of various
Reagents, there was to be seen surrounding the yolk a layer of
fcodies, which have received the name of " test-cells " from the
idea that they were the cells of the developing ovum, from which,
ventually, the characteristic test of the adult Ascidian was formed.
owalewsky(4) in his paper on the development of Ascidia canina
tatee his belief that such is the fate of these cells, which, he also
^■maintains, have their origin from the epithelial cells of the egg-
bllicle. Later on, however, in his paper on the development of
yrosoma,(5) he withdraws the statement that the mantle is formed
rom the "test-cells," but still adheres to the opinion that these
re merely follicle-epithelial cells, which have migrated inwards
towards the yolk. Before the appearance of his first paper, how-
ever, Kupffer,(6) after investigating the subject, came to the conclu-
sion that the " test-cells " formed at the surface of the egg itself,
^rhich theory had been independently adopted by Metschnikoff.(7)
before the publication of Kowalewsky's second paper, Hertwig(8)
Shewed that the "test-cells" take no part in the formation of the
Xnantle, this being formed as a secretion of a homogeneous substance
from the epidermis, into which, later on, cells migrate from the
epidermis. Hertwig's observations were made on Phallusia mamil-
tata and P. virginea (?), and have been confirmed by Semper(2) by

150 J. PLAYFA1R McMURRIOH.

observations on Clavelina vitrea and Oynthia depres&a. In the
same year that Semper published his observations, a paper by
TJssow*9* appeared, in which the old theories first advanced by
Kowalewsky are revived and most emphatically insisted upon.
He says : " The outer mantle of the Tunicates is developed, not as
a secretion product of the epidermal cells of the inner mantle,
(Hertwig, Arsenjew,) but by the increase in number and growth of
the8o-called 'test-cells' (Kupffer, Kowalewski)," and again: "The
result of my observations on the formation of the so-called ' test-
cells ' is in complete accord with that of A. Kowalewski. The
yellow bodies are in fact nothing but cells of the Graafian
follicle . . . . "

Semper shews that in the several species in which he examined
the ova, the "test-cells" were formed in the interior of the egg,
and that by the* action of reagents, or even by exposure to sea-
water, these bodies might be produced in immature eggs. He
holds that they are devoid of a nucleus and of a cell-wall, and dis-
carding the term " test-cells/' substitutes instead that of " test-
drops."

My own observations having been confined to mature or almost
mature eggs, I cannot confirm Professor Semper's statement as to
the production of these peculiar bodies in immature eggs by means
of reagents, but these have the effect of producing them in most
cases almost immediately in mature eggs, even the exposure to sea-
water for a short time being sufficient for the purpose. Produced
in this manner these bodies are small and roundish in shape, and
in their interior numerous clear highly-refractive granules are to
be seen. I could detect no nucleus either in the fresh or in the
stained "drop," .and a limiting membrane was also apparently
wanting.

As regards their mode of origin I am in accord with the obser-
vations of Kupffer,(6) Metschinkoff/7) etc. When an egg has been
removed from the ovary for a few minutes, there appear in the
interior of the yolk, numerous clear spots situated nearer the
periphery than the centre. In no case did they niake their ap-
pearance at the centre of the yolk, and though in Figure 3, (PI. X,)
some appear to be very close to it, these in reality are peripheral
and appear indistinctly when an optical section of the egg is made
and accordingly have been represented. I accordingly conclude
that their origin is peripheral as stated by Metschinkoff.(7) They

ASOIDIAN OVA. 151

gradually migrate outwards, until they form a layer at the periph-
ery of the yolk (PI. X, Fig. 4), and then pass outside of it alto-
gether. Tne yolk at the same time contracts and leaves a space
between its circumference and the egg-membrane, in which the
€ test-cells" lie, forming at first a layer round the yolk (PL X,
-F^g*. 5), but as the contraction of the yolk proceeds, and the space
mes larger, they move away from the surface of the egg and

themselves irregularly. (PI. X, Fig. 6.)
should imagine that there is in a manner a separation of the
into two portions ; an outer, consisting of protoplasm with
jraratively few yolk granules, and an inner, containing most of
the ^olk granules and a small amount of protoplasm. The outer
zo**^ is of no further use in the process of development, and gradu-
a**y splits up into these "test-drops," their formation commencing
a** **fae inner part of the zone and proceeding outwards, until we
™^vcs numerous "test-drops" and nothing left of the egg but a
^^tt^e mass of food-granules, closely packed together in the re-
J**^ining protoplasm, from which the embryo is formed. Metschin-
^ describes this separation of the egg into two portions. He
: " In the greenish protoplasm of a young egg of Ascidia
%n-€^atinali8f fine yolk-granules collect round the nucleus; the
^^ttaber of these becomes continually greater, whereby only the
F^**i pheral portion of the protoplasm retains its greenish colora-
1o**- This layer now separates itself distinctly from the central
ar portion and splits up into a great number of round
ies which are the first * Tunic-cells,' " From this description
would imagine that the author implied that the "Tunic-cells"
« formed at the extreme periphery of the egg, which, however,
ot the case, for they make their appearance in its interior,
in treating a fresh ovum with a dilute solution of acetic acid
^~*" ^>r 2 drops of commercial acid to a watch-glassful of water) for
^^^^it half an hour, its appearance becomes considerably changed.
^"*^l-> X, Fig. 6.) The interglobular protoplasm of the " follicle-
^^*1^" becomes much more distinct, and, in consequence, the
S^c^^Qjgg themselves become more plainly marked off. The trans-
^^^^nt, apparently homogeneous egg- membrane becomes, ris men-
lc**^ed above, distinctly granular. The yolk contracts very much,
*^^suring on the average about .116 mm., half its original size,
is contraction leaves a clear space between the yolk and the
-membrane, which, however, is larger in one-half of its cir-
3

.158 J. PLAYFAIR McMURRIOH.

camferenoe than in the other, owing to the eccentric position
assumed by the contracted yolk. In this clear space are numerooi
" test-cells," not forming a layer round the yolk, as they usually
do in an egg that has been subjected for a short time only to tbi
action of acetic acid or sea- water, but scattered irregularly around
the yolk. The "test-cells" measure .008 mm. and present the
appearance described above. In eggs that have been left in aceti<
acid for a much longer period (6-20 hours) no further change
occur, showing that the acid has exerted its full influence on
them.

After exposure to sea-water for six hours, very much the same
appearance is presented as with dilute acetic acid. The " follicle-
oells," however, shew a tendency to separate from the egg-mem-
brane, which, on its part, does not present a granular appearance.
(PL X, Fig. 7.)

Upon running some strong picro-carmihe under a cover-glass,
below which were some ova in sea-water, very important changec
occurred. At first no "test-cells" were to be seen, but, as tin
picro-carmine gradually reached the egg, and the picric acid
exerted its action upon it, it gradually assumed a yellow hoe
while, at the same time, there appeared at its periphery mrnnj
small spherical bodies of a round or oval shape, the same suse ai
the "test-cells," and containing in their interior several highlj
refractive granules, which, in fact, render them apparent. N<
"test-cells" appear outside the yolk, which retains its original sise
The egg-membrane assumes a pink hue, and, after some time
becomes distinctly granular. The " follicle-cells " do not stain fc*
some time and show a tendency to separate from the egg-meon
brane. (PL X, Fig. 8.) The reaction produced by very dilufl
picro-carmine is also rather important. After being subjected M
this reagent for about half an hour, the eggs presented an appear
ance intermediate between that produced by the continued acticj
of dilute acetic acid and that following the employment of stroK
picro-carmine. (PL X, Fig. 9.) The yolk contracts to a slig^l
degree, and "test-cells" make their appearance, filling up fel
small space between the partly contracted yolk and the
membrane.

I also employed osmic acid in the following manner. The
were placed in a watch-glass containing sea-water, to which 1
2 drops of | per cent, osmic acid had previously been added, mi

ASCIDIAN OVA. 163

aiicrwed to remain there for from five to ten minutes, when they

were removed and stained with Beale's carmine. In most cases

flo ofcange occurred, the yolk remaining of its original size, and no

"tes*-cells" or clear spots made their appearance in the yolk, with

£be exception of one instance, in which I did perceive a number

of dear spots in the periphery of the yolk.

these results two questions are suggested: 1st. What are
" test-cells ? " 2d. How are the various phenomena caused
hy tlie various reagents to be explained ? I shall give the second
<lu«stion priority. The explanation that seems to me to be the
s**tt jplest, and that which bears the stamp of probability most dis-
^^c^tJy, is, that these phenomena are caused by the varying effects
°^ ^fce different reagents in producing a contraction of the proto*
of the yolk. Thus, osmic acid, which "fixes" the proto-
iramediately, allows of little or no contraction, and hence no
*^^t-cells" appear; with picric acid (which evidently is the con-
^^^ent of the picro-carmine that is active in producing the phe-
enon) a slight contraction takes place before the protoplasm
mes "fixed," whereby the "test-cells" are formed, but the
°°** fraction is not sufficient to cause them to pass outside the yolk;
*0(l, in the last place, with acetic acid and sea- water there is no
^^* lag of the protoplasm, and the contraction goes on to such an

nt that the "test-cells" are driven completely outside the
^° J 1$. Strong evidence in support of this theory is afforded by
~**^ variation in the action of picric acid, according to the strength
lr* ^*vhich it is used. For, as we have seen, in a dilute solution so
h contraction of the yolk is produced, that the "test-cells" do
%ly pass out.

ooordingly, then, the "test-cells" are formed by a contraction

^he protoplasm of the egg, and thus we can readily understand

**^ir formation in a developing egg, where the contraction pro-

^^^^*d by the process of cleavage would be quite sufficient to cause

**^ir extrusion from the yolk.

^&fe are now in a position to discuss the question as to the nature

* ^hese "test-cells." Semper (2) regards them as merely polar glob-

*^^, comparing them, in respect to their number, with those of

*•**« Mollusca. This theory is, however, untenable, for by the re-

*^^*tshes of Hertwig on the formation of the polar globules in the

^Sg* of H»mopis, Nephelis,(l0) Asteracanthion, Mytilus,(ll) and

°t**er forms, we know that the polar globules are formed by a true

154 J. PLAYFAIR McMURRIOH.

cell-division, and are themselves true cells, containing a nucleus,
whereas no such process has been observed during the formation
of the "test-cells," and I for my part am sure that it does not ob-
tain, and, as Semper himself insists, the "test-cells" are not true
cells, but merely "drops." Fol,(S) too, states that in Phallwna
intestinalis polar globules (two in number) are formed after the
disappearance of the original nucleus and after the formation of
"test-cells." Accordingly then, there is no morphological ho-
mology between the polar globules and the "test-cells." In the
eggs of certain forms, however, such as, in the Amphibia, Rana,(l0)
and in the Pisces, the Trout,(12) after the disappearance of the ger-
minal vesicle, peculiar bodies are extruded from the yolk without
any spindle-formation or cell-division, for which Hertwig proposes
the name of excreted bodies (Excretkorper) in contradistinction to
the polar globules formed by cell-division. These structures have
been supposed by the various authors to be the remains of the ger-
minal vesicle, and thus, as far as their mode of formation is con-
cerned, probably do not allow of comparison with the "test-cells,"
but since they resemble these latter in being bodies whose presence
in the egg is not necessary to its further development, and since
the cause of their appearance is evidently the same, viz : the con-
traction of the yolk induced by a stimulus, I think there can be
no objection to classifying the "test-cells" with them as Excret-
korper.

Wyville Thomson/13* however, has described bodies as occurring
in Antedon rosaceus which bear a closer homology to "test-cells"
than even these structures. He says: "Consequently on the con-
traction of the yolk, a number of minute spherical pale yellow
oil-globules are apparently pressed out into the space within the
Vitelline membrane." These bodies differ from "test-cells"
only in the fact that they are oil-globules, whereas "test-cells"
distinctly protoplasmic in their nature, and contain in their inter ion
several oil-globules usually. This distinction, however, is of com-
paratively little moment, and both in their mode of formation an
general appearance these Excretkorper — for so they also may
denominated— are evidently closely related to "test-cells" an-
perhaps identical with them.

I consider these "test-cells" to be simply masses of albumino
material containing two or three granules of the food-yolk, ai
presume that they are in reality only portions of the protopl

A8CIDIAN OVA. 155

of the egg, which have been forced out by the contraction. If
an egg, in which the "test-cells" have passed outside the yolk,
ke subjected to pressure sufficient to rupture the yolk-membrane,
Allowing the yolk to come into contact with the " test-cells," and
at the same time leaving the egg-shell intact, the "test-cells"
commingle completely with the yolk and cannot be distinguished
again. The granules to be observed in a "test-cell" have a perfect
resemblance, both in shape and appearance, to those remaining in
*he yolk as food, so that it may be presumed that they are in
jreality the same, and were originally situated in the yolk, in that
^portion of the protoplasm which formed the " test-cell," and were
extruded with it.

The reason why portions of the yolk, originally of use to the
embryo, have become useless and are extruded, must remain unde-
cided until the life-histories of more of the lower types of Ascidians
lave been fully worked out, but in all probability the explanation
5s to be sought for in a change in the life of an ancestral form,

"whereby the development became more rapid and less food-yolk

"was required, while, at the same time, little or no diminution in

the amount of yolk in the egg was produced.

TABLE OF REFERENCES.

1. Lacaze-Duthiers. Les ascidies simples des cdtes de la France.
Arch, de zool. exper. Vol. III. 1874.

2. Semper. Ueber die Entstehung der geschichteten cellulose-
epidermis der Ascidien. Arb. aus dem zool.-zoot. Inst, zu Wurz-
burg. Bd. II. 1875.

3. Fol. Sur la formation des ceufs chez les Ascidies. Journ,
de Micographie. T. I. 1877.

4. Kowalewsky. Weitere Studien uber die Entwickelung der
einfachen Ascidien. Archiv fur mikr. Anat. Bd. VII. 1870.

5. Kowalewsky. Ueber die Entwickelungsgeschichie von Pyro-
*oma. Archiv fur mikr. Anat. Bd. XI. 1875.

6. Knpffer. Die Stammverwandtschaft zwischen Ascidien und
^VirbeUhieren nach Untersuchungen uber die Enturickelung der

^^scidia canina. Archiv fur mikr. Anat. Bd. VI. 1870.

1 J. PLATFAIB McMUBRICH.

7. Metschinkoff. Zur Entwickelungsgeschichte der einfachei
Lscidien. Zeit. fur unseen. Zool. Bd. XXII. 1872.

8. Hertwig. Untersuchungen uber den Bau und die Entuncke
lung des CelUilosemantels der Tunicaten. Jen. Zeii. VII. 1872

9. Ussow. Zoologischrembryologische Untersuchungen: Du
ManteUhiere. Archivfur Naturg. Jahrg. XLI. 1875.

10. Hertwig. Beitrdge zur Kenntniss der Bildung, Befruchtun$
und Theilung des thierischen Eies. 2ter Theil. Morph. Jahrb
Bd. III. 1877.

1 1. Hertwig. Beitrdge zur Kenntniss der Bildung, Befruchiung
und Theilung des thierischen Eies. Ster Theil. Morph. Jahrb.
Bd. IV. 1878.

12. Oellacher. Beitrdge zur Entwickelungsgeschichte der Kn<h
ehenfische. Zeit. fur wissen. Zool. Bd. XXII. 1872.

18. Wyville Thomson. On the Embryogeny of Antedon rosacea^
Linclc (Gomatula rosacea of Lamarck). Phil. Trans. 1865.

EXPLANATION OF FIGURES.

Figure 2 is drawn with Hartnack obj. 9, oc. 2; all the rea%
drawn with Hartnack obj. 7, oc. 2.

Figure 1. — Fresh egg of Ascidia amphora.

Figure 2. — Follicle-cell.

Figure 3. — Egg after short exposure to sea-water.

Figure 4. — Egg after longer exposure to sea-water.

Figure 5. — Egg after still longer exposure to sea-water.

Figure 6. — Egg after exposure for half an hour to the action *

acetic acid.

Figure 7. — Egg after exposure to sea- water for six hours.

Figure 8. — Egg after the action of strong picro-carmine.

Figure 9. — Egg after the action of very dilute picro-carmine

-A CONTRIBUTION TO THE STUDY OP THE
BACTERIAL ORGANISMS COMMONLY FOUND
UPON EXPOSED MUCOUS SURFACES AND IN
THE ALIMENTARY CANAL OF HEALTHY
INDIVIDUALS. Illustrated bt Photo-Micrographs.1 By
GEO. M. STERNBERG, Surgeon V. S. Army, late "Fellow by
Courtesy " of the Johns Hopkins University. With Plates XI,
XII and XIII.

Introduction.

The observations recorded in the following paper and the
^photo-micrographs by which it is illustrated, were made in the
^Biological Laboratory of Johns Hopkins University, Baltimore,
Maryland, daring the months of June, July and August, 1881,
9t which time the writer was acting under the orders of the
^National Board of Health, and was engaged in special investiga-
tions which occupied a considerable portion of his time, and to
which this study was necessarily subsidiary.

Microscopists have long been familiar with the fact that a variety
of bacteria are constantly found in the alimentary canal of healthy
individuals, and that the examination with a sufficiently high
power of saliva or faeces never fails to demonstrate the presence of
a multitude of these micro-organisms of various forms. Some
microscopists to whom this fact is familiar, and whose studies have
shown them the widely extended distribution of the bacteria, both
within and without the human body, have shown a disposition to
ridicule the idea that these minute organisms, so universally pres-
ent, are capable under any circumstances of playing so important
a role in the etiology of infectious and epidemic diseases as has
been ascribed to them by believers in the "germ-theory." It must
be admitted that many extravagant and unfounded claims have
been made by over-enthusiastic supporters of this theory, and that
a scientific conservatism is very essential to him who would esti-
mate at their true value the facts developed by the numerous re-

1 Bead at the Cincinnati meeting of the A. A. A. S., Aug. 18th, 1881.

157

158 GEO. M. STERNBERG.

searches which have been made relating to the bacteria. The
literature of the subject is already enormous, and the yearly addi-
tions to it seem to grow almost in geometrical progression, showing
the rapidly increasing interest in the subject among physicians,
sanitarians and men of science generally, due to a more general
appreciation of the importance of the questions involved.1

It is evident that the time has passed when the spirit of investi-
gation can be arrested by the exhibition under the microscope of
the bacteria found in the saliva or feces of a healthy individual
and the magisterial dictum of an "expert microscopist" that these
minute organisms are entirely harmless.

That there are many widely distributed forms (species?) which
are ordinarily harmless, can not be questioned, but that pathogenic
bacteria exist, either as distinct species or as physiological varieties
(Pasteur) of common forms, is now definitely proven.

No apology, then, is needed for a study of this nature, the
object of which is to place upon record photographic representa-
tions of the common bacterial organisms found in the bodies of
healthy individuals and some observations relating to their physio-
logical properties and the best method of studying them.

It is evident that a precise knowledge of the morphology and
development — life-history— of these common forms is an essential
prerequisite to the recognition of unusual forms and to the en-
lightened study of the possible relation of such forms to any par-
ticular disease with which they may be found associated.

I call attention, however, fit passant, to the tact that recent re-
searches indicate that too much importance has heretofore been
attached to morphological distinctions, and that not only may the
same organism present distinct morphological peculiarities in dif-
ferent stages of its development, but that during the same stage
differences in size, if not in form, may result from conditions re-
lating to the environment — temperature, composition and reaction

1 Not*. — In the bibliography compiled by Magnin (ki The Bacteria," Little,
Brow a Jt Co.. Boston. l&H) and added to by myself, but which can br no
meaa* be considered complete, the references from iS3>)— 10 are seven: from
l!*40- 50, twelve: from I<30-o0. seventeen: from 18*0-?), sixty-three: from

lSTU-^U. above three hundred and drty. In the second volume of the %t Index
Catalogue :o Library of ihe Sury^jon-General's Ofice." just published, four
closely printed page* are required for the reference* relating to tk Charbon n
alone.

BACTERIA IN HEALTHY INDIVIDUALS. 159

of medium, presence or absence of oxygen, etc* On the other
handy organisms morphologically undistinguishable from each
other may possess different physiological properties.

The researches of some of the pioneers in this field of investi-
gation, and especially the discovery by Davaine of a bacillus in
the blood of Anthrax and of Obermeier of a spirillum in that of
neJa-psing fever, led many to anticipate that organisms morpho-
Jo^£*i*2ally distinct might eventually be discovered for each specific
d is^*»se.

his expectation has not been realized, and the germ-theory has
vigorously attacked by conservative opponents who have
jperly pointed out the morphological identity of Bacillus an-
and B. subtilis, and of Spirochaete Obermeieri and 8. pli-
which is not infrequently found in the mouth of healthy
viduals. This argument has, however, lost its force, and the
mon and usually harmless bacteria around us have acquired a
importance since it has been shown by Pasteur,1 Buchner,2
^""x~^«nfield,s Grawitz,4 and others, that, by special methods of culti-
on, pathogenic varieties may be developed from harmless or-
isms, and that, by certain treatment, deadly bacteria may so
^** lose their virulence as to produce only a mild, though protec-
* v*-=** form of disease. In a recent study5 of "A Fatal Form of
ticsemia in the Rabbit produced by the Sub-Cutaneous Injec-

**:**~fck of Human Saliva" I have obtained experimental evidence

anting in the same direction.

brief reference to these facts is all that I can permit myself

**^ "fcbe present paper, but I desire to call attention to certain possi-

1 ™ * ties which remain after the negative demonstration has been

T^^^e that no organisms are present in the blood of patients suffer-

-^^S' from a certain disease — that is, none demonstrable with the

lBt*"lie8t powers of the microscope as at present perfected. This

*4De l'attenuation du virus du cholera des poules." C. R. Ac. des Sc, XCI,
** - ^ "38-80.

*'Ueber die experimen telle Erzcugung des Milzbrand-Contagiums aus den
*^*-*pilzen." Munchen, 1880.

*' Further Investigations on Anthrax and Allied Diseases in Man and Ani-
tas." Brown Lectures, I-V; London Lancet, 1880, pp. 966-906; 1881, pp.
— *> 91-94, 163-164.

s See Bulletin National Board of Health, April 80, 1881, and succeeding
T^*de in the present number of this Journal.

160 GEO. M. STERNBERG.

negative demonstration by no means proves that the disease in
question is not a germ disease, for the habitat of the parasite may
be elsewhere than in the blood, which may not offer the proper
conditions for its development and from which it may be excluded
by vital or mechanical obstacles.

Bacteria are always present in the alimentary canal of healthy
men and animals, but that they do not find their way into the
blood-stream, or if so, are quickly disposed of, has been amply
proven by the negative results of microscopical examinations and
culture-experiments.

In the form of septicaemia in the rabbit which I have recently
studied, L c, I have invariably found an abundance of micrococci
in the effused serum in the sub-cutaneous cellular tissue of an ani-
mal recently dead, but these organisms are not always found in
the blood, and my observations indicate that they only invade the
circulating fluid during the last hours of life. Micro-organisms
have been found in many other localities without their presence
•being revealed by a microscopical examination of the blood;
e. g.} in effused liquids in the pleural and peritoneal cavities, in
pysemic abscesses, and in various tissues and organs of the body.
I have quite recently found an abundance of minute bacilli in the
substance of the heart of a rabbit, which died as the result of the
sub-cutaneous injection of a contaminated water (unpublished ex-
periment).

The possibility that pathogenic bacteria may become parasitic
upon the bronchial mucous membrane, or in the air-cells of the
lungs, should also be borne in mind. But, when we consider the
extent of the alimentary tract, the variety of substances taken as
food and drink, and the ready access which micro-organisms have
to this human culture-apparatus, kept as it is at a constant tem-
perature and supplied with pabulum suited to their development,
it seems probable that this is the locality where pathogenic organ-
isms may most frequently find the conditions favorable to their
multiplication. This view is supported by many facts connected
with the epidemic prevalence of pestilential diseases, and it is
generally admitted that patients suffering from typhoid fever and
cholera may sow the seeds (germs?) of these diseases in the dis-
charges from their bowels.

It is unnecessary to dwell further upon the possibilities in this
direction which make it important that the bacterial organisms

BACTERIA IN HEALTHY INDIVIDUALS. 161

present in the human body should be studied by modern scientific
methods — photography, isolation and cultivation in various media,
injection into animals, etc., etc., but I will refer for a moment to
another possibility which has occurred to me, which should, I
think, receive the attention of chemists and physiologists.

What is the rdle of those micro-organisms which are constantly
present in the alimentary canal of men and animals f

The fact that they are parasites does not exclude the possibility
of their playing an important physiological rdle in the animal
economy.

I am not speaking of accidental or occasional parasites, but of
t;]io6e which have probably been the commensals of man, and of
'fclie inferior animals frequented by them, from the earliest times.
t can hardly be possible that in the process of evolution the
nee of these parasites has had no influence upon the host, or
liat, to go no further back, in the gradual change from the mode
f life and habits of a nomadic savage to that of a civilized man,
he change in environment has had no modifying influence upon
hese micro-organ isms, which laboratory experiments show to be
susceptible to changes in temperature and in the composition
f the medium in which they are placed.
The question is frequently asked, " If bacteria are such terrible
things, how is it possible that we can exist upon the earth sur-
nded and infested as we are by them?" Certainly there would
an end to all animal life, or rather there would never have
n*a beginning, if living animals had no greater resisting power
to the attacks of these parasites, which by numbers and rapid
^fievelopmeut make up for their minute size, than has dead animal
^natter.

On the other hand, but for the power of these little giants to
JduII to pieces dead animal matter, we should have dead bodies
Spiled up on all sides of us in as perfect a state of preservation as
ned lobster or pickled tongue, and there being no return to the
il of the materials composing these bodies, our sequoias and oaks
'Xvould dwindle to lichens and mosses, and finally all vegetation
^vould disappear and the surface of the earth would be a barren
snd desolate wilderness, covered only with the inanimate forms of
successive generations of plants and animals.

162 GEO. M. STERNBERG.

Section 1.

A number of authors1 have given more or less extended accounts
of the micro-organisms found in the human mouth, and their
accounts agree so well with each other and with the results of
my own observations, that I should hardly think it necessary to
record these, but for the fact that I am able to present photographic
representations of the organisms described for comparison with
the illustrations drawn by other observers.

The special advantages which I claim for this method of illus-
tration are set forth in a paper contained in the last volume (1880)
of the Transactions of the American Association for the Advance-
ment of Science.

I would especially call attention to two recent papers, one by
Butlin, of England, and the other by Rappin, of France, both of
which are illustrated and show careful study.

1 Remak. " Diagnostische und pathologische Untersuchungen." Berlin,
1845, s. 221.

P/eufer. "Der Mundhohlen-Katarrb." flenle u. Pfeufer. Ztchft. f. Rat.
Med., Bd. 7, 1849, s. 180.

Miguel. " Untersuchungen uber den Zungenbeleg." Prager Viertel-Jahr-
schft., 1860, Bd. 28, s. 44.

Robin. «« Vegetaux Parasites." Paris, 1853, p. 845.

Niedhardt. " Mittheilungen uber die Veranderungen der Zunge in Krank-
heiten." Arch, der wissensch. Heilkunde, Bd. V, 1861, s. 294.

Hyde Salter. Todd's "Cyclopaedia of Anatomy and Physiology. " Art.
«• Tongue." Vol. IV, pt. 2, p. 1161.

Hallier. " Die pflanzlichcn Parasiten." Leipzig, 1866.

Kolliker. «' Handbuch der Gewebelehre." 6te Auflage, 1867, ss. 348-849.

Farlie Clarke, "Diseases of the Tongue." London, 1878, p. 98.

Billroth. " Coccobacteria septica." Berlin, 1874, s. 94.

Robin. " Lecons sur les Humeurs." Paris, 1874, p. 550.

Koch. "Untersuchungen uber Bacteria." Cohn's Beitrage zur Pflanzen,
Bd. II, Hft 3, s. 399.

Butlin. "On the Nature of the Fur on the Tongue." Proc. Boyal 8oc.,
London, Vol. XXVIII, p. 484.

Rappin. " Des Bactenes de la Bouche." These de Paris, No. 144, April,
1881.

BACTERIA IN HEALTHY INDIVIDUALS. 163

Methods of Research.

Collecting. — I have found the following to be the most satis-
factory method of collecting bacteria for examination with high
powers and for photography.

The slightest possible smear of the material to be examined is
allowed to dry upon a thin glass cover, and to secure a sufficiently,
uniform layer, it is usually best to spread it while moist with the
end of a glass slide.

Material is obtained from the mouth by scraping the surface of
"the tongue, or of the teeth, with a clean instrument; from the
female vagina by a speculum or digital examination ; and from
"the mouth of the male uretha by applying a thin glass cover di-
rectly to the moist mucous membrane at the extremity of the canal.
Staining. — A five-cent bottle of aniline violet ink furnishes an
mple supply of staining fluid of the best quality. Two or three
rops of this placed upon the thin cover will very quickly— one
three minutes — give to the bacterial organisms attached to its
urface a deep violet color. The cover is then to be washed by a
entle stream of pure water and is ready for immediate examina-
ion, or may be mounted for permanent preservation over a shallow
containing a solution of potassium acetate (Koch's method),
^^arbolic acid water (2-5 per cent.), camphor water, or simply dis-
tilled water.

Photographing. — To make satisfactory photographs of the
Smallest bacteria it is necessary to use a staining fluid which will
ive stronger photographic contrast, as the violet is transparent
w the actinic rays. I have employed for this purpose aniline
V>rown (recommended by Koch), or iodine solution (iodine 2-5
ins, potassium iodide q.s. to dissolve, distilled water 100 grains).
A recent writer (Soubbotine1) advises the use of osmic acid as a
xing solution to be used in advance of staining. This is doubt-
X«s8 desirable when specimens of blood or thin sections of tissue
*^ontaining bacteria are to be examined, as the normal histological
lements are better shown, but the method posesses no special
dvantages so far as the demonstration of vegetable organisms is
Concerned. It must be remembered that aniline solutions often

1 Arch, de Phys., 2e serie, VIII, p. 479.

164 GEO. M. STERNBERG.

contain a granular precipitate which might be mistaken by a
novice for deeply stained micrococci.

I cannot here give a detailed account of the technique of the art
of photo-micography, but will simply say that there are many
difficulties to be overcome, and that the best results can only be
obtained by the use of first-class objectives of high power, and by
skilful manipulation in the preparation of slides and projection of
a well-defined image, supplemented by a sufficient knowledge of
the technique of photography to ensure the making of well-timed,
well^developed, and properly intensified negatives. For, one who
has not the services of a practical photographer at his command,
the dry-plate process offers many advantages.

OaUure-experimenU. — A knowledge of the life-histories and
physiological properties of the various vegetable parasites which
infest the human body can only be obtained by well-devised and
carefully conducted culture-experiments. This method of research
is still in its infancy, but it has already given valuable results and
must doubtless be our main reliance for the advancement of science
in this direction. My own experiments have been made chiefly
with a view to testing methods and are preliminary to more ex-
tended studies which I hope to make in the future.

Culture-cells in which a drop of fluid — aqueous humour, etc. —
containing the organisms to be observed, is in contact with a thin
glass cover and surrounded by a limited quantity of air, are useful
and convenient for certain purposes, especially for the continuous
study of successive stages in the development — life-history — of
bacterial organisms. But the method of Pasteur — cultivation in
gross in sterilized fluids contained in glass flasks — offers decided
advantages so far as the isolation, preservation, and cultivation
of special forms, and the exclusion of atmospheric germs is
concerned ; and, also, because the considerable quantity of fluid
used gives material for physiological experiments — injections into
animals, etc.

The method which I have found most satisfactory, after a con-
siderable number of experiments with various forms of apparatus,
is a modification of that of Pasteur which I shall proceed to
describe in detail.

The culture-flasks which I employ are shown in Figure 1, Plate
XI, supported in small bottles in the position in which they are
introduced into the culture-oven.

BA C TERIA IN HEAL THY INDI VID UAL8. 165

The larger one, in the centre, is made from a Florence flask,
the neck of which has been drawn out into a capillary tube in the
flame of a Bunsen burner. The smaller flasks are of my own
manufacture, and are made from glass tubing of about £ inch
diameter. Bellows operated by the foot and a flame of con-
siderable size — gas or alcohol — will be required by one who pro-
poses to construct these little flasks for himself, but they could
doubtless be obtained at small expense from any thermometer-
maker. A little practice has enabled me to turn out twenty or
thirty in an hour, and I have found it much easier to make new
tubes than to clean old ones. I therefore throw them away when
they have been once used.

After blowing the bulb the lower end is drawn out in a capil-
lary tube and hermetically sealed in the flame. In this condition
t:tie flask, which is already sterilized by heat, may of course be
preserved indefinitely, free from contamination by atmospheric
rms.

To introduce a liquid into the flask, heat the bulb slightly,

off the sealed extremity of the tube and plunge it beneath

le surface of the liquid. If the liquid has already been sterilized,

mporary exposure to the air while several of the little flasks are

ing filled is not likely to result in the introduction of atnios-

ic germs — for any organisms which fall upon the surface of

liquid will be arrested there for a time, unless they are suh-

rged by mechanical means — stirring.

I have found it best, however, not to trust to the sterilization of
culture-liquid previously to its introduction into the flasks,
am in the habit of filling a considerable number of them at
ne time with filtered chicken-JauiV/on, Cohn's fluid, hay-infusion,
whatever culture-fluid I may desire to use; and, after again
I lermetically sealing the capillary extremity of the tubes, steriliza-
tion of the contents is effected by heat.

This is accomplished by placing the flasks in a bath of oil,
^melted parafine or concentrated salt-solution, and maintaining
them at a temperature of about 105° C. for an hour or more.
ionally a flask which has an exceptionally thin bulb will
x pi ode, and care must be taken by the operator that the hot oil is
%)ot thrown into his face by such an accident. This possibility
Snakes it desirable that a bath should be used having a fixed boil-
ing-point not exceeding 105°, and which consequently does not

166 GEO. M. STERNBERG

require watching. I have found a concentrated salt-solution to
fulfil this requirement.

After sterilizing, the flasks are washed to remove the salt-solu-
tion from their surface. They are then placed in a culture-oven
kept at a temperature of 95-100° Fah. (36-38° C.) for three or
four days to test the success of the previous operation — steriliza-
tion.

If the liquid contents remain transparent and no mycoderma
has formed upon the surface during this time, the flasks may be
put aside for future use and can be preserved indefinitely.

The process of sterilization sometimes causes a floculent pre-
cipitate to form when albuminous fluids are employed, although
they may have been previously boiled and filtered. This might
lead to the suspicion that they had broken down, but for the fact
that this precipitate is already present when the flasks are intro-
duced into the culture-oven, and no subsequent change takes place.

To inoculate the liquid contained in one of these flasks with
organisms from any source, the extremity of the tube is broken
off with forceps, the bulb being dependent, and by the application
of gentle heat — the heat of the hand is usually sufficient — enough
air is forced out to cause a little fluid to be drawn into the tube
upon immersing its extremity in the liquid and allowing the air
in the bulb to again contract by cooling.

A little experience will enable the operator to inoculate one
tube from another, to introduce a minute quantity of blood con-
taining organisms directly from the veins of a living animal, etc.,
without any danger of contamination by atmospheric germs. No
other method with which I am acquainted offers such security as
to sterilization of the culture-fluid and exclusion of foreign germs;
and a somewhat extended experience in a recent experimental
study, "A Fatal Form of Septicaemia," etc., /. c, has convinced
me that it has also decided advantages on the score of convenience.

The bottle which supports the inverted flask protects the capil-
lary extremity from dust, and labels are conveniently attached to
it. The formation of a mycoderma upon the surface of the fluid
is readily recognized, and contained organisms soon settle to the
bottom of the tube. Small quantities of fluid are conveniently
obtained for microscopical examination by breaking off the end of
the tube, forcing out a little of the contents on a clean slide and
immediately sealing the extremity again in the flame of a lamp.

BACTERIA IN HEALTHY INDIVIDUALS. 167

Another form of apparatus which I have found very useful is
that of Lister, a slight modification of which is shown in Figure
2, Plate XI.

In the apparatus as described by Lister a conical wine-glass
contains the culture-liquid, and this is covered by a circular glass
plate; the whole being protected from dust by a bell-jar which
rests upon a ground-glass plate.

When proper precautions are taken, a sterilized liquid may be
preserved in this apparatus for any length of time without under-
going perceptible change. I have used a bell-shaped glass cup
having a stem drawn out and sealed in the flame of a Bunsen
burner, in preference to a wine-glass, as it is more easily sterilized
l>y Iieat without danger of breakage. This is supported by a
I>ottle as shown in the figure, and I have commonly dispensed
xvith the use of a glass cover, the use of which is directed by
XLjister, as I have not found this to be essential to the success of
\y experiments.

Section II.

Description of Plates, Remarks upon Morphology, etc.

The most conspicuous vegetable organism found in the healthy

uman mouth, and the one which will usually first attract atten-

ion upon microscopical examination with low powers, is the well

nown Leptothrix buccal is, Robin. This I have never failed to

nd, in greater or less abundance, in material scraped from the

urfaee of the tongue, or in accumulations dislodged from between

he teeth. Often it is found in tufts and masses which indicate a

orous growth, and again it may only occur in the form of short

sparsely intermingled with the normal histological elements

f the saliva, as shown in Figure 2, Plate XIII. But in this case

t is probable that a careful search would reveal the presence in

he mouth of the microscopic plantations and garden-beds from

"^vhich these fragments were detached.

As might be expected, those who make frequent use of the

^.ooth-brush leave less soil upon the surface and in the interstices

^Df the teeth for the growth of this and other vegetable parasites.

^fo amount of care, however, will keep the mouth entirely free

from them, and the observations of Butlin (/. c.) show that the fur

Vipon the tongue, which is rarely entirely absent even in healthy

168 GEO. M. STERNBERG.

individuals, is in great part made up of this and other vegetable
parasites.1

In Figure 1, Plate XII, several filaments of Leptothrix are
shown in which evidence of breaking up into joints is seen; and
in Figure 2 of the same plate we have a mass of jointed filaments
that seem rather to come under the definition of Bacillus than of
Leptothrix, as given by Magnin in accordance with the classification
of Cohn. This author says: "The Leptothrix differ from the
Bacilli by their filaments being very long, adherent, very slender,
and indistinctly articulated/'

These characters seem to me to be very uncertain and unsatis-
factory, inasmuch as Bacillus subtilis and B. anthracis, in one
stage of their development, are very long and slender and indis-
tinctly articulated, and, on the other hand, we have here a Lep-
tothrix broken up into very distinct joints not distinguishable
from those of Bacillus.

The filaments represented in Figure 1 are from a specimen of
saliva obtained directly from my own mouth, while those in Figure
2 were developed in a culture-apparatus of special construction
(see below) in which a constantly renewed supply of pabulum —
chicken-6out/Zon — was passed through a small chamber, freely
supplied with air, containing saliva scraped from the surface of
my tongue. Biftlin did not succeed in his efforts to cultivate this
organism. He t\ays: "I made many attempts to separate them in
order to produce this fungus in a purer form by cultivation, but
did not succeed in doing so. Although this fungus did not de-
velop under artificial conditions in the presence of micrococcus
and other fungi, it is highly probable that its development takes
place freely upon the surface of the tongue."

It seems probable that my success in the experiment above
mentioned is to be attributed to the constantly renewed supply of
pabulum and the free access of oxygen, conditions which are cer-
tainly present in the mouth, where the surfaces upon which this
parasite grows are constantly bathed with saliva and supplied with
air. The author above quoted is of the opinion that the organism
in question is identical with Bacillus subtilis, and in certain cases
he observed "highly refractive spherical bodies which appeared to
be spores" in some of the filaments. I have also observed shor

i Butlin found "on 68 healthy tongues — fur on all except one. On 17
tongues of persons suffering from disease or accident — fur on all except two.1

BACTERIA IN HEALTHY INDIVIDUALS.

169

rods containing a single spore at one extremity in specimens of
my own saliva examined in New Orleans during the summer of
1880, But at this time similar rods with spores were abundant
in certain culture-fluids in my laboratory, and I supposed these to
be Bacilli accidentally present in my mouth and differing from
the common Leptothrix buocalis. This is a question, however,
which can only be determined by culture-experiments, and I
would suggest that the best way to settle it would be to cultivate
the Leptothrix in an artificial saliva constituted as nearly as pos-
sible like normal saliva — but, of course, without the histological
elements — and in a culture-apparatus such as was used in my
single experiment above referred to. This apparatus is made as
follows : A glass receiver A having two capillary tubes, one a to
&clcnit air, and one 6 to permit the gradual escape of the contained
culture-fluid, is supported by the bent tube C} which is maintained
in an upright position by being tied to the cork of a bottle -B,
hich answers as a support for the apparatus. Mercury may be
in this bottle to give it steadiness. The bent tube C
as a reservoir e, which is freely exposed to the air by means of
opening t. The organism to be cultivated is introduced into
*fcliis reservoir. The overflow from e is received in the beaker D.
o attempt is made to exclude atmospheric germs, as the object of
lie apparatus is to supply, as nearly as possible, the identical con-
itions found in the human mouth.

170 . GEO. It STERNBERG.

My observations have not been sufficiently extended to justify
roe in an attempt to describe all of the organisms which are occa
sionally found in the human mouth, and I shall only refer brief!;
to the fact that the recorded observations of microscopists indicat
that nearly every common bacterial organism known is sometime
found in this situation. This is no more than we should expect, a
the germs of these various organisms are widely distributed in th
atmosphere and must be deposited upon the moist mucous mem
brane during inspiration. Their development here will of conrs
depend upon whether the conditions are favorable or otherwise
As these conditions vary within certain limits, we naturally fin<
at different times and in different individuals a variety of organ
isms present iu the buccal secretions differing from those com mo:
forms which observations made at distant points * show to be con
stantly present under normal conditions.

Among the varying conditions found in the months of indi
viduals considered healthy may be mentioned, a greater or les
abundant flow of saliva, a difference in the reaction of this fluid
the presence of decayed teeth, various habits as to food, drink, us
of tobacco, etc The variety of odors to be detected in the breatl
is sufficient to show that conditions may vary, and it may be tha
a sufficiently thorough research would result in the establish men
of a euKal relationship between the presence of certain organism
and the peculiar and offensive odors referred to.

When enga^txl in the microscopical examination of foul guttei
water and in culture-<xx}x>riment$ with various puirifying organi
sahstanots. in Now Orleans La., daring the autumn of 1880,
not intTWjnentiv found neanv even* organism in mv own moot I
which was present in the puirlryi^ ;iqnid$ under examination
5iw\ad:T*g Bacicriin* trrm*\ i*.?c.,7w **&!...*. Sztirilitim vjk/u&i, an*
a rarksv of ir«;r.;:te >:>hvr!va, and i\xi-*ike forms difficult toclassif
c-xwTt uts-Sct the £*-r*or**! heaainc of mKcvvoora and bacteria.

* ^. •*.

A >7<-w/'*t.y ^;c r.Wi:*Tyr»t;shi:*;e frorr. i. Ol*rmacri of relaps
iiu: fevw lias bora rcp«:oj'.y cowrve*: iy z2icr«swp:sK* but I hav
nrc UTSf-.f nifi w:;>. it.

Tin. Jt/xu^W sii.-wr. *n Fica^c S, Pk:* XII, I have reason t<
luCicvf, fr*iit ;i« :reciK-iKy with *h>ri I hav* icarnd it, is almos

: &rih;T». &'•>{«} . Bi^r-.r . IU,rcv.t. tui u?t <*Uiar anrhT-s referred to «
1^!* J to.

BACTERIA IN HEALTHY INDIVIDUALS. 171

fcs commonly present in the healthy human mouth as is the larger
ind more widely known Lcptothrix, or Bacillus, already de-
ftcribed.

This minute organism, which would hardly be recognized with-
out staining and the use of high -power objectives, is also found in
normal feces, if we can trust to the morphological resemblance
.vhich will be seen by a reference to Figures 5 and 6, Plate XIII,
n which the amplification is the same (1,000 diameters).

Figure 3, Plate XII, is from a culture-experiment in which
icid malt-extract (sterilized and tested in culture-oven) was inocu-
ated with a little saliva from my own mouth.

In Figure 4, Plate XII, a fragment of an epithelial cell from
he mouth of Dr. K. is shown. The nucleus of the cell is seen at
he upper portion of the figure, near this some granules resembling
Micrococci, and on the margin of the cell a mass of rod-bacteria —
probably jB. latno. Referring again to Plate XIII, Figures 5
ind 6, we see that this form also is found in normal feces. To
iccottnt for the presence of these organisms in the alimentary canal
are have only to suppose that fully developed bacteria, or their
anrecognized germs, can withstand the action of the digestive
3uids in the stomach and the upper portion of the intestines, and
:hat those, found in the lower bowel, are the direct descendants of
ihose habitually present in the mouth, or of others taken into the
stomach with food and drink.

Another organism which I have found quite constantly in speci-
mens of saliva from healthy mouths, although never in any con-
siderable abundance, is shown in Figure 5, Plate XII. This
seems to be a Sarcina and is, perhaps, identical with S. ventriculi,
although it presents a somewhat different appearance as to form
and grouping from this organism, as shown in a specimen from the
8tomach in my possession. I have frequently observed little clus-
ters of this sarcina-like organism attached to the surface of epi-
thelial cells in my own saliva and that of others, but to obtain it
in abundance I Have been obliged to resort to culture-experiments.
The figure here given is from a specimen obtained by cultivation
in acid malt-extract. This organism, as well as the bacillus shown
in Figure 3 of the same plate, multiplies luxuriantly in this fluid
when kept at a temperature of 36° C. It may be remarked, en
passant, that acid malt-extract (a dilute solution) is not unlike the
acid fluid ejected from the stomach in cases of obstinate vomiting

172 GEO. M. STERNBERG.

attended with the abundant development of Sarcina venlriouli in
the stomach.

Figure 6, Plate XII, represents a micrococcus which possesses
an especial interest because of its abundant and constant presence
in the human mouth and because it has been shown to possess
pathogenic properties when injected beneath the skin of a rabbit.
This fact has been brought to light by recent experiments made
independently by Pasteur in France,1 and by myself in this country,2
and since confirmed by Vulpian.8

The plate accompanying the paper in which I give an account
of the experimental researches referred to is headed "Miorocooeus
septicus, Cohn." When this paper was written I thought it prob-
able that the organism represented in my photo-micrographs was
identical with the micrococcus described by Cohn and other ob-
servers under this name. I pointed out, however, that this micro-
coccus is larger than that described by Cohn as M. septicus, the
diameter of which is given as 0.5/*, while the organism in question
measures very nearly 1//. I have since met with a smaller septic
micrococcus which corresponds with Cohn's measurements, and am
now inclined to believe that the micrococcus found in the human
mouth is a distinct species, or at least a well established variety,
differing in size but having nearly the same physiological action
as the M. septicus of Cohn.4

i Oomptes rendus Ac. d. Sc, 1881, XOII, p. 159.

2 Bulletin National Board of Health, April 30th, 1881.

» Bull, de l'Aead. de Med., March 29th, 1881.

* The smaller septic micrococcus above referred to was found under the fol-
lowing circumstances:

Experiment No. 1, Baltimore, Md., July 9th, 1881. — Injected beneath the
skin of a small rabbit a little material scraped from the mucous membrane of
the intestine of a rabbit just dead. (This rabbit died from an experimental
injection, not yet reported, made for Professor Mallet of the University of Vir-
ginia. It presented upon post-mostem examination evidence of enteritis.)
Rfsult: Found dead at 8 A. M., July 10th. Diffuse cellulitis extending from
point of injection ; abundance of minute micrococci in serum from cellular
tissue and in blood from axillary vein ; liver, heart, and lungs, normal ; spleen
enlarged and softened, but contains no pigment.

Experiment No. 2, July 10th. — A hypodermic syringe point was dipped in
the blood — from femoral vein — of this rabbit and introduced under the skin of
rabbit No. 2. Result: This rabbit was found dead the following morning at
8. 30, and a post-mortem examination was made at once with the following
result: Diffuse cellulitis with hemorrhagic extravasations under the skin;
blood from superficial veins full of micrococci ; spleen enlarged, softened, dark

BA C TERIA IN HEAL THY INDIVID UALS. 1 73

In Figure 6, Plate XII, the micrococcus from the mouth is
seen as obtained by cultivation (in chicken-bouillon inoculated

ith saliva) in the form of apparatus described on page 169, in

Inch provision is made for a constantly renewed supply of the
culture-fluid.

A vigorous development is shown by the grouping in long
t,oru la-chains and in zoogloea masses. In Figure 5, Plate XI, the
same organism is shown as found in a culture-flask similar to those
shown in Figure 1, Plate XI. In this case the culture-fluid was
x noculated with a small quantity of blood taken directly from the
"'vessels of a rabbit just dead as the result of a sub-cutaneous injec-
tion of saliva.

The drop of blood used to inoculate the culture-fluid contained
"fc: he form shown in Figure 6, Plate XI, which differs from that
shown in Figure 5 and in Figure 6, of Plate XII, in having a
V>roader areole of transparent material. Identity is proved, how-
ever, by the fact that it is directly descended from the last form
C Figure 6, Plate XII) and that the first (Figure 5, Plate XI) of
"xvhich it is the progenitor is morphologically identical with that
^From which it originated. A reference to Figure 3, Plate VII, in
mnay translation of Magnin's work, "The Bacteria," will show this
Micrococcus upon an epithelial cell obtained directly from my own
:Knoutb. Here also I detect no morphological difference from the
brni obtained by cultivation in a bouillon made from the flesh of

chicken or of a rabbit.

The fact that this micrococcus is the most common organism
^Cbund in the human mouth and that it has been described by
Several observers at distant points may seem difficult to reconcile

^solored, has rounded edges ; liver light colored ; lungs congested and present
Numerous points of hemorrhagic infraction.

Experiment No. 8, July 11th. — A hypodermic syringe needle was dipped in
~fe>lood from left auricle of rabbit No. 2 and introduced under the skin of a small
-babbit (No. 8). Result : This rabbit died at 4.80 P. M , July 13th, but circum-
stances prevented me from making a careful post mortem examination, and I
^iave not since had an opportunity to make a more extended study of this form
«»f septicaemia, which, so far as I am able to judge from the experiments made,
differs somewhat from the form previously studied by me (l. c ). The spleen
"was not so much enlarged and was softer, with rounded edges, corresponding
'with the spleen of septicemia as described by Klebs and Totnmasi-Crudeli, in
"their memoir upon the nature of malarial fever (Studi sulla Matura della Ma-
laria, Roma, 1879). The inflammatory oedema or " diffuse cellulitis " was also
less marked.

174 GEO. M. STERNBERG.

with the fact, recently developed, that to its presence is due the
exceptional virulence of the saliva of certain individuals. It
accords, however, with the results of recent investigations, which,
as already stated in the introduction to this paper, indicate that
pathogenic organisms may differ greatly as to their virulent prop-
erties as the result of different conditions relating to their environ-
ment acting upon successive generations.

My observations lead me to believe that, having a suitable
medium, a proper temperature, and a sufficient supply of oxygen,
the development or intensification of pathogenic properties depends
to a great extent upon an abundant and constantly renewed supply
of pabulum. Now this is a condition which differs greatly in the
mouths of different individuals. In my own case there is, and has
been from my earliest recollection, a very copious secretion of
saliva. This, according to my view, accounts for the exceptional
virulence which my experiments show it to possess, and is in con*
fortuity with the principles of natural selection.

Rapid multiplication is, I infer, an evidence of vigor. Now it
is evident that in a natural culture-apparatus like the human
mouth the rapid flow of saliva by which contained organisms are
constantly washed away will have a tendency to sort out those
which develop slowly from those which develop rapidly, and that
the former will tend to disappear entirely, while the latter by
virtue of their rapid multiplication will survive and the tendency
will constantly be to a further development of this property of
rapid multiplication. My culture-experiments have shown me
that, in fact, this particular' micrococcus does multiply with great
rapidity, and that by virtue of this quality it has the precedence
over Bacterium iermo, the presence of which in any considerable
number seems to be fatal to it.

This rapidity of multiplication is shown by the fact that the
sub-cutaneous injection of a minute quantity of the material con-
taining it — in the rabbit — results within 24 to 48 hours in the
development of an infinite number of micrococci in the effused
serum in the cellular tissue, and in the blood of the animal, where
they far outnumber the normal corpuscular elements. In my
culture-flasks, also, a minute drop of this blood gives rise withiu
a few hours to the development of such a number of micrococci
that the fluid contents of the flask are invaded throughout and the
pabulum needed for a continued development is exhausted. I

BACTERIA IN HEALTHY INDIVIDUALS. 175

suspect, then, that this is the simple explanation of the phenome-
non in question— exceptional virulence — and I am inclined to
think that the modus operandi of the action of these pathogenic
organisms is also to be explained by the possession of this capacity
for rapid multiplication.

Nature has placed, or in other words evolution has developed,
in the living tissues of animals, a resisting power against the
encroachments of bacterial organisms invading and surrounding
them, which is sufficient for ordinary emergencies. But when the
vital resistance of the tissues is reduced, on the one hand, by
"Wasting sickness, profuse discharges, etc., or, on the other, the vital
.ctivity of the invading parasitic organism is increased, the balance
►f power rests with the infinitesimal but potent micrococcus. The
ipid multiplication of a micro-organism introduced beneath the
J» kin of an animal is also an advantage in its favor in the way of
^%restalling the restraining influence of the inflammatory process,
is a provision of nature for building up an impenetrable
rail around the invader and thus circumscribing its field of ope-

Experiment has demonstrated that, by some unknown mech-

inism, the ordinary bacteria of putrefaction and under certain

circumstances even pathogenic organisms — e. g. after protective

noculations with the micrococcus of chicken-cholera or the bacillus

f anthrax — may be introduced directly into the circulation with-

ut the production of evil consequences, and that after a short

nterval microscopical examination does not reveal their presence

n the blood. It is evident that here too a capacity for rapid mul-

iplication and the introduction in the first instance of a considerable

^dumber will be circumstances favorable to the parasite and may

oable it to get the start of nature's provision for getting rid of it.

Note. — It has occurred to me that possibly the white corpuscles may have
office of picking up and digesting bacterial organisms when by any means
%*hey find their way into the blood. The propensity exhibited by the leucocytes
^Fbr picking up inorganic granules is well known, and that they may be able
only to pick up but to assimilate, and so dispose of, the bacteria which come
n their way does not seem to me very improbable in view of the fact that
amoebae, which resemble them so closely, feed upon bacteria and similar or-

Reference has already been made to Figures 5 and 6, Plate
^XIII, representing the commou bacterial organisms found in

176 GEO. M. STERNBERG.

Dormal human faeces at the moment of their being discharged from
the rectum. The photo-micrographs tell the story of the abun-
dance and variety of these organisms, but the present state of
knowledge does not admit of an attempt to determine their phy-
siological r6le in the human economy. That their constant pres-
ence in the alimentary canal is a fact without import it is difficult
to believe in view of their demonstrated capacity for breaking up
complex organic substances external to the body in the process of
their growth and functional activity.

Figure 4, Plate XIII, shows an epithelial-cell and bacteria from
the orifice of the male urethra. By gently separating the lips of
the urethra and applying a thin glass cover to the moist mucous
membrane, good specimens are readily obtained of the organisms
commonly found in this locality.

The researches of Lister and my own experiments, shortly to be
detailed, indicate that the healthy human bladder is free from para-
sitic vegetable organisms, and it is probable that those organisms
found at the extremity of the urethral canal, being aerobic, do
not extend any considerable distance beyond the orifice.

Lister has shown that urine drawn from the healthy human
bladder with proper precautions may be kept indefinitely without
undergoing change, and Pasteur as long ago as 1862 (Ann. de
Chemie et de Physique, 1862, p. 52. Gomptes rendus Ac. de Sc.,
LVIII, 1864, p. 210) claimed that the alkaline fermentation oi
urine is due to the presence of a micro-organism — Microcoi
urae,-Cohu. This organism is described by Magnin as follows:
"Oval cells, isolated — diameter 1.5// (Pasteur), 1.2 to 2/* (Cohn)-
or united by 2, 4, to 8 (torula) in a line, straight, curved, zigzag,
or even in cross-form. In urine of which it transforms the
into carbonate of ammonia (Pasteur)."

My photo-micrographs, Figures 3 and 4, Plate XI, show whal
I believe to be the organism in question. The group in Figure
answers very nearly to the measurement given, while the arrange-
ment shown in Figure 4 corresponds with that in Cohn's drawin^^^ m&
(Beitrage zur Biologie der Pflanzen, Band I, Heft 2, Taf. TTT^T^ -),
although the micrococcus in this figure is smaller. It is poesibK ^K-*©
that we have here two different organisms, but I am inclined fc^- -^^°
believe that the difference in size is due simply to the fact th« * ""^

* The Bacteria. Little, Brown & Co., Boston, 1880.

BACTERIA IN HEALTHY INDIVIDUALS. 177

•

different stages of development are represented, Figure 4 showing
an active pullulating stage and Figure 3 a grouping of the micro-
coccus in masses after the completion of the transformation of the
wrea. A difference in the size of individual micrococci will be
noticed in Figure 4, and it must be admitted that in photographing
~£hese minute organisms with high powers a very slight difference
in focal adjustment makes a difference in the apparent size of the
organism. Too much stress should not, therefore, be placed upon
slight differences of measurement as reported by different observers
snd obtained by different methods.

I call attention to the fact that this micrococcus has a well de-
ifined outline and does not present the appearance of being sur-
~srounded by an aureole such as is seen in Figures 5 and 6 of the
same plate. This is an additional proof that this aureole is not
~fthe result of diffraction, but that it represents a transparent sub-
stance enveloping the micrococcus. (See remarks on page 18 of
Special Report on "A Fatal Form of Septicasruia," etc. Re-
jprinted from National Board of Health Bulletin, I c.)

The following experiments are reported here as relating to the
:r6le of this micrococcus, which, notwithstanding the researches of
ZPasteur, Lister, and others, is not perhaps generally admitted by
chemists and physiologists to be unfait ttabli.

Having repeatedly demonstrated the presence of micrococci at
^he mouth of the male urethra and knowing that Lister's experi-
ments indicate that urine as contained in the healthy bladder is
free from bacterial contamination, it occurred to me that in passing
urine from a full bladder the first portion of the stream might
wash away detached epithelial cells and bacterial organisms, and
that the last portion being received in a sterilized flask might give
evidence of freedom from these organisms by remaining unchanged.
Accordingly I made the following

Experiment, Baltimore, Md., June 25th, 1881. — Two bell-
shaped glass cups were sterilized in the flame of a Bunsen burner
and placed under clean bell-jars in the position shown in Figure
2, Plate XI (Lister's Apparatus). I then desired my assistant to
pass a small quantity of urine into No. 1 from the first portion of
the flow and into No. 2 from the last, removing and replacing the
bell-jars as expeditiously as possible. Result, June 30th : No. 1 is
turbid, has a considerable sedimentary deposit and is decidedly
alkaline. No. 2 remains perfectly transparent, has no sedimentary

BACTERIA IN HEALTHY INDIVIDUALS. 179

urine, the bell-jar being removed for an instant only for this pur-
pose. Five days later the contents of the four cups were carefully
examined. Nos. 1, 2 and 4 remained transparent, free from sedi-
mentary deposit, and acid. No. 3 was alkaline and contained an
abundance of Micrococcus urece.

A reference to Figure 4, Plate XIII, will show that the organ-
ism there seen in considerable abundance is not identical in ap-
pearance with Micrococcus urece as seen in Figures 3 and 4, Plate
IXI. Direct examination has not given me as satisfactory evidence
of the presence of this micrococcus at the extremity of the urethra
ms have the experiments above detailed. It may be, however, that
tinder different- circum^ances this organism assumes a different
appearance, and that the form shown in Figure 4, Plate XIII,
zfrom the surface of a mucous membrane exposed to the air, when
submerged in a liquid having the composition of urine undergoes
» transformation into the form seen in Figures 3 and 4, Plate XI.
"This is a question to be settled by carefully conducted culture-
experiments.

Figures 1 and 3, Plate XIII, are from the vagina of a healthy

zfemale at the termination of the menstrual flow. I shall not here

<lwell upon the possible import of the presence of micrococci in

such numbers in this situation, but from what has already been

said it seems evident that gynecologists may well be on their guard

"to prevent the invasion of wounds in this locality — accidental or

made by the surgeon — by these ever-present parasitic organisms,

and especially against the development of virulent varieties as the

Jesuit of profuse and long continued discharges — puerperal, etc.

Figure 7, Plate XI, is introduced for comparison with the other
micrococci upon the same plate. The amplification is the same in
each — 1,000 diameters. This figure is from a specimen obtained
by cultivation of micrococci found in gonorrheal pus, in chicken-
bouillon. The reader is cautioned against the inference that this
micrococcus is the cause of the virulence of the fluid in which it
was found. No great weight can be attached to the mere presence
of an organism under such circumstances in the absence of culture-
and inoculation-experiments to demonstrate its physiological prop-
erties. Such experiments I have had no opportunity of making
in this case, and the figure is introduced solely for the purpose of
showing that distinct morphological differences may be recognized
between these micrococci from three different sources, viz: from
the human mouth, from urine, and from gonorrhoeal pus.

180 GEO. M. STERNBERG.

DESCRIPTION OF PLATES.

PLATE XL

Figure 1. — Culture-flasks in position for introduction into culture -

oven.

Figure 2 — Lister's Apparatus (slightly modified).

FigueeS. — Micrococcus ureae X 1,000 diameters by Zeiss's t^ in.

horn. ol. im. objective ; aniline brown staining.

Figure 4. — Same as Fignre 3 (see remarks on p. 171).

Figure 5. — Micrococcus cultivated in bouillon (rabbit flesh) inocu-
lated with blood from septicemic rabbit, and descended
from common micrococcus found in the healthy human
mouth, X 1»000 diameters by Zeiss's Vf in* objective.

Figure 6. — The same micrococcus as in Figure 5, as it appears in the

blood of rabbit killed by the sub-cutaneous injection of
human saliva, X 1,000 by Zeiss's 11T in. objective.
Iodine staining.

Figure 7. — Micrococcus from culture-experiment with gonorrhoea! pus

X 1,000 diameters by Zeiss's ^ in. objective.

PLATE XII.

Figure 1. — Leptothrix buccalis, obtained directly from mouth, X

1,000 by Zeiss's -fr in. objective.

Figure 2. — Leptothrix buccalis from culture-experiment, X 1,00D

diameters by Zeiss's 1l1- in. objective.

Figure 3. — Bacillus (sp. ?) from culture experiment (saliva in malt-
extract), X 1,000 by Zeiss's T^ in. objective.

Figure 4. — Portion of epithelial-cell from mouth (Dr. K.) covered

with bacteria (B termof), X 1,000 diameters by Zeiss's
^j in. objective.

Figure 5. — Sarcina (ventriculi ?) from saliva-culture in acid malt-
extract, X 1,000 diameters by Zeiss's ^ in. objective.

Figure 6. — Micrococcus from saliva-culture in chicken-bouillon, X

1,000 by Zeiss's Jw in. objective.

. FATAL FORM OF SEPTICEMIA IN THE
RABBIT, PRODUCED BY THE SUB-CUTA-
NEOUS INJECTION OF HUMAN SALIVA. By

GEO. M. STERNBERG, Surgeon, U. 8. A. With Plate XIV.

In a report made to the National Board of Health in Fflhrnarv

The heliotype plate illustrating Dr. Sternberg's paper on
Septicaemia in the rabbit had not come to hand when the
rest of the present number of the "Studies" was ready. As
the issue had already been some time delayed, it was decided
to publish the number without Plate XIV, which will appear
with the next number, and subsequently on binding the
volume can be placed in its proper position.

m* -*^w iMiutiM iwuiio jvuvn tine iijjcv/l/iv/il V/l V/tliCl 11UUIB

containing organic matter in suspension or solution?

Answer. One c. c. of my own blood failed to kill a rabbit; 1 c. c.

putrid urine containing B. tenno in abundance failed to kill

small rabbit; 1 c. c. of liquid feces and distilled water (I

10) failed to kill two rabbits; 1.25 c. c. of bouillon undergoing

Putrefaction and loaded with B. termo, failed to kill a rabbit;

- c. c of sediment from Baltimore water, consisting of organic

1 1 have commonly injected an amount varying from 5 to 25 minims, accord-
^*g to the size of the animal, but in small rabbits have had a fatal result in
*^iree cases out of five follow the injection of 1 minim diluted with 5 minims of

FATAL FORM OF SEPTICEMIA IN THE
BABBIT, PRODUCED BY THE SUB-CUTA-
NEOUS INJECTION OF HUMAN SALIVA. By

GEO. M. STERNBERG, Surgeon, U. S. A. With Plate XIV.

In a report made to the National Board of Health in February
last, I have given a detailed account of certain experiments, made
in the first instance as a check upon experiments relating to the
eo-called Bacillus malaria of Klebs & Tomassi-Crudelli, which
show that my own saliva has remarkable virulent properties
"when injected into the sub-cutaneous connective tissue of a rabbit.
Further experiments, made in the biological laboratory of the
Johns Hopkins University, have fully confirmed the results here-
*tofore obtained, and the object of the present report is to place
xipon record these last experiments, which are of special interest
Just now because of the announcement by Pasteur, of "a new
disease,99 produced in rabbits by the sub-cutaneous injection of
"the saliva of an infant which died of hydrophobia in one of the
lospitals of Paris. (Comptes Rendus Ac. de Sc., 1881, XCII,
I>. 159.)

I have demonstrated by repeated experiments —

That my saliva in doses of 1.25 c. c. to 1.75 c. cl injected into the
jub-eutaneous connective tissue of a rabbit, infallibly produces death}
usually within forty-eight hours.

Query. Do similar results follow the injection of other fluids
containing organic matter in suspension or solution?

Answer. One c. c. of my own blood failed to kill a rabbit; 1 c. c.
of putrid urine containing B. termo in abundance failed to kill
a small rabbit; 1 c. c. of liquid faeces and distilled water (1
to 10) failed to kill two rabbits; 1.25 c. c. of bouillon undergoing
putrefaction and loaded with B. termo, failed to kill a rabbit;
1 c. c. of sediment from Baltimore water, consisting of organic

I I have commonly injected an amount varying from 6 to 25 minims, accord-
ing to the size of the animal, but in small rabbits have bad a fatal result in
three cases out of five follow the injection of 1 minim diluted with 5 minims of
water.

7 183

184 GEO. M. STERNBERG.

d6bris and organisms — chiefly Bacillus subtilis, Leptothrix pusitta,
Protococcus, and a few diatoms and flagellate monads — failed to
kill a rabbit.1

On the other hand, injections of a small quantity of surface mud
from the gutters of New Orleans during the month of September,
1880, invariably produced fatal results within forty-eight hours.
(See unpublished report above referred to.)

Query. Does the saliva of other individuals injected in the same
manner produce similar results?

Answer. The saliva of four students, residents of Baltimore (in
March), gave negative results; eleven rabbits injected with the
saliva of six individuals in Philadelphia (in January) gave eight
deaths and three negative results; but in the fatal cases, a less
degree of virulence was shown in six cases by a more prolonged
period between the date of injection and the date of death. This
was three days in one, four days in four, and seven days in one.

Query. Is there any recognizable peculiarity in the saliva which
exhibits the greatest degree of virulence ?

Answer. In the case of Dr. S., whose saliva shows an excep-
tional virulence, the teeth are sound, the secretions of the mouth
normal in physical properties and reaction, and the general health
good. There is, perhaps, an unusual flow of saliva, but no other
noticeable peculiarity.

Query. Is there any plausible hypothesis by which this difference
in virulence can be explained ?

Ansiver. This question will require for its solution more extended.
experiments. In the meantime it may be mentioned, as having
possible bearing upon the subject, that Dr. S. has been engaged
a considerable extent, during the past two years, in studies which—
have brought him in contact with septic material. Dr. F.f o
Philadelphia, whose saliva killed (after a longer interval) two^-
rabbits, is pathologist to a large hospital, and consequently is^»
constantly brought in contact with septic material. Mr. N. and—

1 Coze and Fcltz found, as the result of numerous experiments, that the blooX-
of healthy persons, and that of persons sick with non-infectious maladies, doe^»
not produce fatal results when injected into the sub-cutaneous tissue of rabbits- ~
(Clinical and Exp. Researches upon Infectious Maladies, 8°, Paris, 1872.^
Pasteur also has inoculated, without result, the saliva of asphyxiated rabbits
and of men dead with common diseases {I. c).

SEPTICEMIA. 185

It. B., whose saliva killed all the rabbits operated upon (four),
re residents of seaport towns in Cuba.1

Query. Is death produced in other animals by the sub-cutaneous
ojection of human saliva, which is virulent for rabbits?

Answer. Injection of 4 c. c. into each of two small dogs pro-
luoed local abscesses at point of injection, but no other noticeable
esult.2 Injection of 0.25 c. c. (each) into five chickens produced
10 result. Injection of 0.75 c. c. (each) into three guinea-pigs
>roved fatal to two— one in three and one in seven days. Injec-
ion of 0.5 c. c. into five rats resulted fatally to one only.3

Query. What is the nature of the fatal malady produced in rab-
)its by the sub-cutaneous injection of the saliva of certain indi-
viduals ?

Answer. The course of the disease and the post-mortem appear-
ances indicate that it is a form of septicemia. Immediately after
he injection there is a rise of temperature, which in a few hours
nay reach 2° to 3° centigrade (3.6° to 5.4° Fah.); the temperature
subsequently falls, and shortly before death is often several degrees
3elow the normal. There is loss of appetite and marked debility
ifter twenty-four hours, and the animal commonly dies during the
second night or early in the morning of the second day after the
injection. Death results still more quickly when the blood from
% rabbit recently dead is injected. Not infrequently convulsions
immediately precede death.

The date and mode of death corresponds with that reported by
Pasteur in the memoir referred to. Two rabbits injected with buccal
mucus from the mouth of a child recently dead with hydrophobia,
December 11th, were found dead December 13th. Other rabbits

* The possibility that this septic condition of the secretions of the mouth may
bear some relation to the protection which these Cubans and myself enjoy
against yellow fever, which is a disease presenting many points of resemblance
to septicaemia, has occurred to me, and without, at present, laying any great
stress upon this possibility, I think it worthy of further experimental con-
sideration.

* A dog succumbed, however, to an injection of 1 c. c. of serum from the
sub-cutaneous cellular tissue of a rabbit recently dead.

8 The results obtained by me in these experiments correspond with those
reported by Pasteur in the paper already referred to, viz: guinea-pig less sus-
ceptible than rabbit, complete immunity of the chicken, and susceptibility of
the dog to the "new disease" as the result of injections of blood from dead
rabbits.

184

I STERNBERG.

<M>r ^ ^-iTa of these died in still less time.

-fJ|"" * ^ dually produced death in less than

kill - %w

<
fn ^.lological appearance is a diffuse inflam-

1> _::>* extending in all directions from tin

(- ^ «.;vt-ially to the dependent portions of tin

■Kiv is a little pus near the puncture, bi

^ ^ before the cellulitis reaches the point cr— -^ (

v sub-cutaneous connective tissue contains a

xw% ktuiii, which possesses virulent properti*

.3..i> a multitude of micrococci. There is usual

.:;kjiuiaiory adhesion of the integument to the su

^ LLio liver is sometimes dark colored and gorg

wt more frequently is of a lighter color than norm =l~

fc ..x iuku fat. Tlie spleen is either normal in appearai "m

^ .iiid dark colored. Changes in this organ are ni«^»> Te

i chose cases which are of the longest duration. Z^Kn

jis** dark colored pigment has been found in the sple*E^- n,

V;n; that which has been supposed to In? characteristic of

^ ju tfver. The blood is dark colored, usually fluid, £lk.](1

> a tendency to agglutination of the red corpuscles.
'K- Wood commonly contains au immense number of niic«r- o-
.t itetiallv joined in pairs, and having a diameter of aL>cz»ut
. . These are found in blood drawn from sii|>ertieial vei »is,
fc ^ .urterio>. and from the cavities of the heart immediate] v af ^ter
.vU%;>% and in a few cases their presence has been verified durl ng
».\>; observations thus far made indicate, however, that it is aw^^lv
%it.:oig t'u* ki>t hour's of life ihat these parasites multiply in "die
%,:vui:iting fluid, and in a certain proportion of the cases a eare^i'ul
waivh has faiUd to reveal their pro>iiiee in i>:.st-r.\orUm exami**a-
;j.»n< made immediately upon uw d*a:h ol the animal. T"Im>*
%»s-g:inism. iu»we\ir. > i:i variably fou:ut in great abundance in zhe
m'I'uui ^Irl^'ii exuties in considerable qua:. :::iis from the (edematous
^v.unvlive :is*;:e w hi n an i:-.i i-h'i; i> :v.ade through the inteiTU-
men! over an\ poi:;: inwlvid in :\w r..:ia:ni::a:orv a-denia extent'-
irg from the or^i^a; puru:uiY.

A •.■;*:■.:>.:'. if -./.i" p:i]^.r e:" Va<\s\\t. s'-i :\,:y n:"irre«i :■«. has induced
r.n- :,- ;vo i^.\« :i". a::.:::.--. \\\ : W r* *•:*..: : >*-■:.■"•»••--•;.* t«» some
jv ::.:> ;^ w-.h.: '.*.• ■* a;;::-,e: :v.,:-- «:;.. :; I ; *.; :_; • ,.-. .-._■• in previous

SEPTICAEMIA. 187

examinations, viz : to the condition of the trachea, the lungs, and the
lymphatic glands in the groins and axillae.

Pasteur says, " The cellular tissue is almost always emphysematous."
(This has not been observed to be the case, except to a slight extent in
one instance in the rabbits operated upon by me.) " The lungs are
frequently filled with the noyaux of pulmonary apoplexy." (I have
found this to be the case in one out of three rabbits examined since
my attention has been directed to this point.) "A character more
constant than the last (not more constant, however, than that which
relates to the volume and color of the ganglions), is the state of the
trachea, which is almost invariably red, congested, with little hemor-
rhages from the smallest vessels." (I have found a marked congestion
of the vessels of the trachea in the three cases in which I have exam-
ined it, and in one case the lymphatic glands of the axillae were enlarged
and congested.)

Query. What constituent of the saliva injected produces the fatal
malady in question ?

Answer. The following facts demonstrate that the phenomena
detailed result from the presence of a living organism found in the
saliva — a micrococcus — which multiplies abundantly in the sub-
cutaneous connective tissue, and also in the blood shortly before or
after death.

(a) The poison is particulate. This is proved by numerous fil-
tration experiments. Example: March 15, 11 A. M. Injected
1 c. c. of filtered saliva (filtered through thin stratum of plaster
of Paris, by means of SprengePs pump) into left flank of rabbit
weighing 1 pound, and at the same time one-fourth the quantity
of unfiltered saliva into a rabbit of the same size. No harm
resulted to the first rabbit, while the second died the following
day, at 5.30 P. M.

(6) The virulence of the saliva is destroyed by boiling.

(c) The saliva loses Us virulence when kept for twenty-four hours
in a culture-chamber, at a temperature of 37° centigrade.

The presence of B. termo and an odor of putrefaction in saliva
kept for twenty-four hours in a culture-chamber shows that changes
are occurring which have heretofore been recognized as destructive
of the septic poison (organism), e. g., the virulence of the poison
which produces dangerous dissection wounds is lost when putrefactive
changes set in.

(d) The addition of one part of a 10 per cent, solution of carbolic
acid to two parts of saliva destroys its virulence.

188 GEO. M. STEBNBEBG.

(e) The effused serum from the sub-cutaneous connective tissue of
a rabbit recently dead, produces death attended with the same phe-
nomena as resulted from the injection of the saliva in the first instance.
But this does not contain epithelial cells or salivary corpuscles,
and we are, therefore, justified in excluding these as possible agents
in the production of the results indicated. Moreover, these are
present at all times in the saliva of all individuals, while viru-
lence, at least such an intense degree of virulence, is an exceptional
property of human saliva.

(/) This serum loses its virulence by filtration.

Un filtered serum from a recently dead rabbit has invariably proved
fatal in smaller quantity and in less time than is required by the saliva
in the first instance, showing an increase of virulence as the result of
successive cultivation of the organism in the body of a susceptible
animal. This corresponds with the results obtained by Davaine, Koch,
Pasteur and others. I have not attempted to ascertain the minimum
quantity which will produce death. Davaine says : M A rabbit may be
killed by the -1-bVtt Part °f a drop of septic blood." (Bull, de l'Acad.
de Med., 2 s., T. VIII, p. 121.) In my filtration experiments I in-
jected, however, quantities far in excess of the amount required to
produce speedy death if unfiltered serum had been employed.

Example: March 14. Injected 2 c. c. ef filtered serum (from sub-
cutaneous connective tissue of rabbit recently dead) diluted with dis-
tilled water (1 to 20) without result, while one-quarter the quantity
(0.5 c. c.) of the same dilution unfiltered, injected at the same time
into another rabb't, produced death in twenty-four hours.

{g) The micrococcus present in the serum from the connective tissue
of a rabbit which has succumbed to a sub-cutaneous injection of
saliva, may be cultivated in bouillon made from the flesh of a healthy
rabbit, or in blood serum obtained from a healthy dog, and these
fluids thereby acquire a virulence which they did not have before.

My first efforts to cultivate the micrococcus in urine, in gelatine solu-
tion, and in bouillon made from the flesh of a dog, all proved inef-
fectual, and these fluids after inoculation with blood or serum from the
connective tissue, showed a temporary virulence only, which was doubt-
less due to the presence of the micrococcci introduced, which preserved
their vitality for a certain time, although the conditions were not
favorable for their increase. After a few days the first culture lost its
virulence and successive inoculations gave negative results, both as to

SEPTICEMIA. 189

the presence of the micrococcus and as to noxious properties when
injected into rabbits.

(A) Successive cultures in which but a small drop is taken each
time to inoculate a fresh quantity of bouillon exclude the while and
red blood corpuscles (filtration-experiments have already shown the
poison to be particulate) as possible agents in the production of
this virulence, and prove conclusively thai the veritable cause is the
presence of a micrococcus, found first in the saliva, then in the
serum from the connective tissue, and (usually) in the blood of
the animal killed by the injection of saliva, and finally in each
successive culture-fluid inoculated (in the first instance) with a
small quantity of this serum or blood.

Within a few hours after inoculating sterilized bouillon made from
the flesh of a rabbit (first tested for several days in a culture-oven at a
temperature of 37° Cent.) with blood, or serum from sub-cutaneous
connective tissue of a rabbit recently dead, the fluid — previously trans-
parent— becomes opalescent, and upon microscopical examination is
found to contain innumerable micrococci, solitary, in pairs, and* in
torula chains. The same result follows upon inoculating a second
portion with a minute drop of the first, and so on. The continued
virulence of these successive cultures I have amply proved.

Example: April 13. Injected 1 c. c. of 6owtZ/on-culture, No. 6 (six
successive inoculations, the first with serum from sub-cutaneous con-
nective tisssue of rabbit), into left flank of a large rabbit. Result:
The animal was found dead on the morning of the 16th, and presented
the usual appearances upon postmortem examination. Its blood and
the effused serum in sub-cutaneous connective tissue contained, as
usual, an immense number of micrococci, like those already described.

Query. Does the micrococcus found under the circumstances
detailed differ from tbe Micrococcus septicus of Cohn, and is it
identical with the organism described by Pasteur, as present in
the blood of rabbits killed by the sub-cutaneous injection of the
saliva of an infant dead from hydrophobia (l. c.)?

Answer. Cohn describes the M. septicus, as follows :

"Little rounded cells of 0.5//, motionless and crowded in masses, or
united in chaplets in the secretion of wounds in cases of septicaemia
(Klebs), in zooglcea in callous ulcers, in isolated cells, united in pairs
or in chaplets in the serum of epidemic puerperal fever (Waldyer), in
all the tissues, vessels, etc., in cases of pyaemia and septicaemia." (The
Bacteria, Magnin: Little, Brown & Co., Boston, 1880, p. 76.)

190 GEO. M. STERNBERG.

Pasteur gives the following description of the micrococcus found
by him in the fatal disease described by him as new, and which
he evidently does not consider identical with septicaemia, a disease
which he had previously studied experimentally. It should be
noticed, however, that Pasteur recognizes several forms of septi-
caemia. Thus he says :

•

" And now we see why septicaemia has so often been confounded with
charbon; their causes are of the same order; it is a vibrio which causes
septicemia and a bacillus which produces charbon. * * * Septicaemia
and putrefaction in a living being are not the same thing. There are
as many different septicaemias as there are different vibrios. * * *
In septicaemia the vibrios do not appear in the blood until the last
thing, but in this liquid one of them takes a peculiar aspect, often
longer than the diameter of the field of the microscope, and so trans-
parent that it easily escapes observation ; when, however, it is once
perceived it is easily found again, flexible, climbing and removing the
blood globules as a serpent moves the grass in the bushes," etc.
(Charbon and Septicaemia, C. R. Ac. des Sc, LXXXV, 101-115,)

This septic vibrio of Pasteur I found in the blood of rabbits,
victims of my experiments, in New Orleans during the past
summer (Report to National Board of Health, not yet published),
but have not since met with it ; perhaps because it develops post
mortem and requires the hot weather of summer for its develop-
ment. Whether it is an independent organism or is developed
under special conditions from the Micrococcus septicus, being an
advanced phase in the development of this organism corresponding-
with the spore-producing filaments which have been shown tc*>
constitute one phase in the life-history of Bacillus anihracis9 Koch,
and of Bacterium termo, Ewart, is an interesting question foe-
further research. The vivid language of Pasteur describes it well,
and the wonderful vigor with which this extremely slender and
almost transparent organism thrusts aside the blood corpuscles in
its impetuous serpentine movements cannot fail to astonish the
observer. The micrococcus of Pasteur's " new disease " is, on the
contrary, quite motionless, and is described as follows:

"This organism is sometimes so small that it may escape a superficial
observation. Its form does not differ from that of many other micro*
scopic beings. It is an extremely short rod a little compressed towards
the middle, resembling a figure 8. and of which the diameter of each

SEPTICEMIA. 191

half often does not exceed a half a thousandth of a millimeter [== 0.5 fit
and corresponding with the diameter given by Cohn for the Micro*
coccus septic us, also with the micrococcus observed by myself in the
form of septicaemia described in this report]. Each of these little
particles is surrounded at a certain focus with a sort of aureole which
corresponds, perhaps, to a material substance." (Note. — The possi-
bility that this appearance is due to diffraction is considered, but Pas-
teur inclines to the opinion that in the case in question it is due to a
raucous substance which surrounds the organism.)

The foregoing descriptions answer as well for the micrococcus

observed by me as if they had been written especially for it, and

*fc is unnecessary for me to say more at present in relation to the

*X*orphology of this organism, which apparently is identical with

fcfciat of the Micrococcus septicus of Cohn, and with the organism

:'c>und by Pasteur in the " new disease" described by him. Does

*t then follow that the organisms are identical, and that the phe-

omena related by Pasteur, as resulting from the sub-cutaneous

ijection of saliva from an infant dead of hydrophobia, and by

yself, from saliva of a healthy adult, represent the same disease?

y no means.

The argument, that because a certain bacillus, or spirillum,
micrococcus, is morphologically identical with another, which
proved to be harmless as to its effects upon an animal organism,
uently it must be harmless, has no support from analogy
r experiment. The argument is: Bacteria are found everywhere,
eat them, we drink them, we draw their germs into our lungs
t each inspiration and without apparent injury. They are evi-
ently harmless. Your spirillum of relapsing fever does not differ
^the morphological resemblance is admitted) from a harmless
pirillum frequently found in the human mouth; your Bacillus
nthracis does not differ from Bacillus subtilis, etc. The answer
is plain. The fact that there are harmless bacteria does not dis-
prove the possibility of pathogenic bacteria; the fact that two
things look alike does not prove that they are alike; experi-
ment proves conclusively that the phenomena of anthrax are
*3ue to the presence and multiplication in the body of the affected
dnimal of the Bacillus anthracis, and that in the fatal form of
septicaemia described in this report, the efficient cause of the
morbid phenomena, and of death, is the minute micrococcus

192 GEO, M. STERNBERG.

Doubtless, harmless micrococci abound. Pasteur finds no differ-
ence, morphologically, between the organism which produces the
"new disease" described by him and that which produces the
cholera des poules. He says : " By the form which it has in the
blood the organism resembles the microbe of chicken cholera, but
it differs completely in its functions. We may inoculate fowls
with it without their experiencing the slightest ill effect." (The
same is true of the organism producing the form of septicaemia
described in this paper.)

"In the form of chaplets it resembles greatly many other organ-
isms which I have often observed," etc.

It will have been noticed from the account already given that
the fatal disease in rabbits observed by me and resulting from the
sub-cutaneous injection of my own saliva resembles in many par-
ticulars the disease described by Pasteur as new, resulting from
the sub-cutaneous injection of the saliva of a child dead with
hydrophobia. Another point of resemblance is the fact that the
saliva of one of my rabbits, recently dead, has the same virulence
as the blood and serum from connective tissue. A serous liquid,
which in some instances escapes from the bowels shortly before or
after death, also contains the micrococcus in abundance and pos-
sesses like virulence. All of these points of resemblance form a
strong probability in favor of the identity of the two diseases, but
I am not prepared to pronounce a positive opinion upon this
point, especially since Pasteur, who had previously given much
attention to the study of septicaemia, pronounces the disease ob-
served by him to be new, while I see no reason, at present, for
supposing that the disease observed by me differs essentially from
the experimental septicaemia produced by Davaine, Koch and
other investigators, who, however, obtained their first supply of
septic organisms from a different source.

In the light of what we already know, it seems very probable
that puerperal fever, hospital gangrene, and the various forms of*
septicaemia known to physicians and surgeons result from the
development of pathogenic varieties of harmless and widely-dis-
tributed species of micrococci, as the result of especially favorable
surroundings; such as are found in the lochial discharges of a puer-
peral woman or in the secretions from the surface of wounds in a
crowded and ill-ventilated hospital ward.

SEPTICEMIA. 193

Just as differences in resisting power to experimental septi-
caemia are exhibited by different species of animals, so doubtless
individual differences exist in man, especially as the result of
lowered vitality; and this want of resisting power, from whatever
cause resulting, must be counted as one of the conditions favorable
to the development and propagation of a pathogenic bacterium.
Thus we find that in experimental septicaemia the micrococcus
does not invade the blood until the vital powers are at a low ebb,
and death is near at hand.1

In the dog the vital resistance is competent to withstand the
assaults of a micrococcus — injected sub-cutaneously — having the
J>otency of those found in my saliva, and the result of such an
injection is simply a circumscribed abscess. But the increased
jDower (which is perhaps simply a more vigorous and rapid devel-
pment) gained by cultivation in the body of the rabbit, enables
lietse organisms to overcome the resistance of the dog, and a
iffuse cellulitis results of a fatal character.
The fact, observed by myself, that during the summer months
"t;he mud in the gutters of New Orleans possesses an extraordinary
ree of virulence2 shows that pathogenic varieties of bacteria
re not alone bred in the bodies of living animals. The more I
study this subject the more probable it seems to me that in this
direction lies the explanation of many problems which have
jDuzzled epidemiologists, and that the sanitarians are right in
fighting against filth as a prime factor in the production of epi-
demics— a factor of which the rdle is easily uuderstood, if this
'View is correct.

The presence of septic organisms, possessing different degrees of
'virulence, depending upon the abundance and kind of pabulum
furnished them and upon meteorological conditions more or less
favorable, constitutes, in my opinion, the epidemic constitution of the
rztmospherc, which wise men were wont to speak of not many years
o as a cloak for ignorance. It must be remembered that the

1 By virtue of some property or mechanism at present unknown, blood, which
external to the body is a favorable medium for the development of many spe-
cies of bacteria, resists their entrance or gets rid of them when they effect an
entrance, e. g., by injection, so long as it is circulating in the vessels of a
Ileal thv individual.

* There is no reason to suppose that this is peculiar to New Orleans, but I
have not yet had the opportunity to extend my experiments to other places.

194 GEO. M. STERNBERG.

gutter mud of to-day, with its deadly septic organisms, is the dust
of to-morrow, which in respiration is deposited upon the mucous
membrane of the respiratory passages of those who breathe the
air loaded with it. Whether the peculiar poison of each specific
disease is of the same nature or not — a question which can only be
settled by extended experimental investigations in the future — it
is altogether probable that this factor often gives a malignant
character to epidemics of diseases which, uncomplicated, are of a
comparatively trivial nature.

Part Second. — Morphology.

Since writing the report published in "The Bulletin/' of April
30th, my attention has been called to the fact that M. Vulpian has
arrived at similar results (Bull, de l'Acad. de Med., March 29,
1881) ; and I iufer that Pasteur has somewhat changed his opinion
as to the nature of the "new disease." described by him in his
communication to the French Academy (made January 26th),
from the following remark of Chauveau, which I find in his
recent address, as President of the French Association for Ad-
vancement of Science. (Revue Scientifique, April 16th, 1881.) He
says:

"For a moment we hoped that Pasteur had determined thus"
(by artificial cultivation) " the virus of hydrophobia, but he tells us
himself that he has only cultivated a new septic agent"

There seems, therefore, to be ho longer any reasonable doubt of
the identity of the "new disease," described by Pasteur, and the
fatal form of septicaemia in the rabbit produced by the sub-cuta-
neous injection of human saliva, which I first observed in New
Orleans, in September, 1880, and which I have since studied, ex-
perimentally, in Philadelphia (in the Medical Department of the
University of Pennsylvania, in January), and in Baltimore (in
the Biological Laboratory of Johns Hopkins University).

Having proved, experimentally, that the presence and multipli-
cation of a micrococcus is the essential feature in the etiology of
this disease, a further study of the morphology of this minute
organism becomes of interest.

In Plate XIV, Figure 1 represents the organism as found in
the blood of a rabbit recently dead ; Figures 2 and 3, the same
from a culture-solution {bouillon made from the flesh of a rabbit);

8EPTICJEMIA. 195

Figure 4, the same as found in human saliva, while Figure 5
represents a micrococcus from another source, introduced for com-
parison. The amplification in each case is 1,000 diameters, and
the photo-micrographs, which have been accurately reproduced by
the heliotype process, were all made with the same objective (Zeiss
TV horn, im.) and at the same distance, with the exception of
Figure 4, which was made at a different time with Zeiss TV in. and
a longer distance.

The first thing which strikes an observer upon an inspection of
this Plate will, doubtless, be the fact that the organism presents
very marked morphological differences in the figures given, and
the question will at once arise as to a possible mistake in identity.

So far as the form represented in Figure 4 is concerned, it must
be admitted that there is no positive evidence that this is really
the septic micrococcus as found in human saliva, whicfy is the
parent of the form developed in the blood of the rabbit, and rep-
resented in Figure 1. What is positive and invariable, so far as
my experiments go, is that the injection of my saliva into the sub-
cutaneous connective tissue of a rabbit is followed by the appear-
ance in the effused serum, and subsequently in the blood (usually)
of the micrococcus seen in Figure 1, and that this is a septic or-
ganism. As the saliva contains a variety of bacteria, including
rod and spiral forms, as well as micrococci, it may be supposed
that the form developed in the blood of the rabbit as the result of
the sub-cutaneous injection of this fluid is descended from any one
of these forms. But while there is no positive proof to this effect,
the abundant presence of themicrococcus represented in Figure 4,
and its morphological resemblance to the form shown in Figure 1,
makes it seem highly probable that this, rather than one of the
other forms referred to, is the parent form in this case.

It will be noticed that in Figure 4 the micrococcus is more
abundant over the epithelial cell than around it, and, indeed, the
cell seems to be invaded by the organism as if it were parasitic
upon it. This is a very common appearance, and in many cases
the epithelial cells are seen to be invaded to a greater extent even
than in the example which has served for this photo- micrograph.

It must be remembered that micrococci, morphologically resem-
bling these, are commonly found in saliva which does not possess
marked virulent properties as well as in that which does, and if
the organism is specifically the same iu both cases, we must admit

196 QEO. M. STERNBERG.

the existence of varieties possessing physiological peculiarities,
although morphologically identical.

It may be well to say here that the sharply defined photographic
image of these minute organisms which is seen in the figures, can
only be obtained by staining processes and by the use of first-class
objectives. A failure to demonstrate the presence of this micro-
coccus with a \ or i inch objective without the use of a suitable
staining fluid, cannot be accepted as proof of its absence.

In these photographs the staining was affected with iodine solu-
tion, as the yellow or brownish color which this gives is well
adapted for giving strong photographic contrast. When there is
a distinct cell wall, as in the larger bacteria, leptothrix, etc., a still
better effect can be obtained by first covering the organisms (dried
upon a thin cover) with strong sulphuric acid for a very short
time (one or two minutes), and after washing this off by a gentle
stream of water, immersing the cover in a weak solution of iodine
(iodine, grs. iii, potassic iodide, grs. v, distilled water, gro. 200) for
a few minutes.

For ordinary microscopical examination, I have found no stain-
ing fluid equal to a solution of aniline violet, first recommended
by Koch.

The most striking morphological difference between the micro-
coccus as shown in Figures 2, 3 and 4, and in Figure 1, is the
aureole which surrounds the well-defined dark central portion in
the latter figure.

Pasteur says of this appearance: "This organism is sometimes
so small that it may escape a superficial observation. ... It is an
extremely short rod, a little compressed towards the middle, re-
sembling a figure 8. . . . Bach of these little particles is surrounded
at a certain focus with a sort of aureole which corresponds perhaps to
a material substance"

Pasteur's inference that this aureole represents a material sub-
stance, and is not simply the result of diffraction, is fully sustained
by my observations and my photographs. The slighter aureole
seen in Figures 2 and 3 is probably a result of diffraction ; but
the use of aniline violet as a staining fluid promptly demonstrates
that in Figure 1 we have to do with a material substance. The
refractive index of this substance must be very nearly that of
blood serum, for it is with great difficulty that this aureole can be
distinguished without the aid of staining material. It may be

8EPTI0JSMIA. 197

seen by the practiced eye with a good immersion lens, but, as
already mentioned, even the darker central portion, which alone is
seen at first, may easily escape observation, and a false impression
is obtained as regards the real size of the organism. When, how-
ever, a small drop of blood, dried upon a thin glass cover, is
immersed for a minute or two in a solution of aniline violet, and
then washed and examined with, even, a good \ inch objective,
the observer will be astonished to find a multitude of organisms,
solitary, in pairs, and in chains, having a diameter of more than
1 fi, and mostly possessing an oval or elongated form, which might
lead to the inference that they should be referred to the genus
Bacterium, Duj., rather than to Micrococcus, Cohn.

The reason of this apparent change in dimensions a9 the result
of staining, is that the substance which constitutes the almost
invisible aureole is deeply stained by the aniline, and the central
portion, which was before seen because of its highly refractive
index, is now lost to view in the uniform and deep violet color
which the whole organism possesses.

A careful study of Figure 1 will show that the inference which
might be drawn from the examination of a specimen stained with
aniline violet as to the oval or rod form of the organism is not a
correct one. It will be seen that a certain number of spherical
(micrococcus) organisms are seen in the field, and that the oval
and elongated forms evidently represent successive stages in the
process of fission, which is seen on the point of completion in the
figure eight (8) form, in which two spheres are coupled together
and enveloped in a transparent matrix. It may be necessary to
explain that the large, dark colored, and ill-defined objects in
the field, are blood corpuscles changed in appearance by the
action of the iodine solution used for staining the micrococcus.

(Fig; 10

When a culture-tube containing bouillon made from the flesh of

the rabbit is inoculated with a minute quantity of blood taken

from a rabbit recently dead and containing the organism shown

in Figure 1, and placed in an oven at a temperature of 37° Cent.,

there is a rapid multiplication of the micrococcus, which, it is

proved experimentally, retains its virulent properties. While in

process of active multiplication the organism also retains, at least

to some extent, its characteristic form as shown in Figure 1, and

presents the appearance of being surrounded by an aureole as

198 GEO. M. STERNBERG.

already described. But in a limited amount of the culture-fluid
the process of multiplication by fission soon ceases.

Observations thus far made indicate that from six to twenty-four
hoars' time is sufficient to exhaust the capacity of the culture-fluid for
sustaining the development of the organism.

When the liquid is examined during the first few hours after
inoculation, it is seen to be slightly opalescent, and upon micro-
scopical examination is found to contain, distributed through it,
an abundance of micrococci, solitary, in pairs, and iu short chains.

At a later period (48 hours) the micrococcus will be found
chiefly at the bottom of the fluid in groups or zoogloea masses as
seen in Figures 2 and 3, and without the aureole of transparent
material which characterizes it, especially in the blood of the rab-
bit, during its active multiplication. There can be no question
that we have here the same organism for this culture-liquid in-
jected into the sub-cutaneous connective tissue of a rabbit produces
fatal septicaemia, and the blood of the victim swarms with the
form shown in Figure 1.

If a culture-liquid in which the micrococcus has been present
in abundance be examined at a later date (one or two weeks), the
organism will be no longer found, at least in a recognizable form ;
and, so far as my experiments go, the culture-liquid no longer
exhibits any virulence when injected beneath the skin of a rabbit.

My experiments thus far indicate that no germs are formed in
the blood or in culture-tubes, which may be preserved for an indefi-
nite time, and then employed for starting a new series of culture-ex-
periments, as in the case of Bacillus anthracis, etc. I design making
further experiments in this direction, however, and from what is known
of the life-histories of allied organisms, we have reason to expect that
permanent spores may be obtained capable of preserving their vitality
indefinitely, when the conditions of their development have been more
fully studied.

From what has been already said, and from a critical study of
my photo-micrographs, it will be seen that the measurement given
by Cohn for Micrococcus septicus, by Pasteur for the organism
described by him, and by myself in my first paper, viz : 0.5 fit, is
too small for the organism represented in my photo- micrographs.
The amplification in these is exactly 1,000 diameters, if the mi-

SEPTICEMIA. 199

crometer plate in my possession is accurate (Powell & Leland's).
But, according to my measurements, the micrococcus as shown in
Figures 2 and 3 is but little less than 1 /jl in diameter, while the
organism as shown in Figure 1 is of nearly twice this diameter,
when the aureole is included in the measurement. The most re-
liable measurement is, perhaps, to be obtained from the group
shown in Figure 3, in which the micrococci may be supposed to
touch each other. By measuring two or three lying in juxtaposi-
tion in a right line, we reduce the probable error which in a single
one results from the somewhat uncertain outline of the organism
as shown in a. photo-micrograph. Adopting this method, I obtain
an average diameter of ^tAtttt of an inch from Figure 3, and y*iinr
of an inch from Figure 1 (including the aureole). It must be
remembered that slight differences are likely to be deceptive, as it
is impossible to obtain exactly the same focus in every instance,
and the apparent size is influenced to some extent by the particular
focus at which the picture is taken, and possibly also by the stain-
ing material employed.

Figure 5 is introduced to show that there are micrococci and
Wiicrococci. The species (?) here represented was obtained in the
instance from gonorrhceal pus. A little of this pus, obtained
im a case of two weeks' duration, showed upon microscopical
examination, in a few of the pus corpuscles, an invasion by micro-
ti, while the majority of the corpuscles, as well as the liquid in
^vhich they were suspended, were free from organisms. A culture-
^ube containing sterilized bouillon (from rabbit) was inoculated
~^*ith a little of this pus, and an abundant development of micro-
^^occus resulted. A second tube was inoculated from the first, and
^^ third from the second. The organism was found in abundance
in all of these solutions (kept in a culture-oven at 37° Cent.),
Xnnchanged in appearance and unmixed with any other forms of
icteria. One cubic centimetre of the liquid from culture No. 3
injected under the skin of a small rabbit with an entirely
^negative result. It is evident, then, that physiologically this
^micrococcus differs from the deadly septic micrococcus which we
fcave been studying. It also presents slight morphological differ-
It is a little smaller and is more easily seen than the M. sep-
when examined without previous staining. This is because
it has a little color (?), or refracts light differently from the latter,
^ind not being surrounded by an aureole of transparent material, it
9

200 GEO. Jf. STERNBERG.

presents a more definite outline. A slight aureole due to diffrac-
tion will, however, be seen upon closely inspecting the photo-
graph.

The question will naturally be asked as to the possible relation
of this organism to the peculiar virulence of gonorrheal pus. I
have not yet found time to study this question experimentally,
but think it quite probable that this organism will be found to be
identical with the micrococcus found in pus from other sources,
e. g., open wounds, inflamed mucous membranes, etc. Whether
this common and widely distributed micrococcus is capable under
special conditions of cultivation of developing into various patho-
genic micrococci ; whether it is a distinct species from our septic
micrococcus, or whether the latter is a pathogenic variety devel-
oped from it, are questions which can only be settled by extended
and painstaking experimental investigations.

EXPERIMENTS WITH DISINFECTANTS. By GEO.
M. STERNBERG, Surgeon, U. S. A.

In experiments previously reported (National Board of Health
-I3ulletin, Vol. 1, Nob. 29, 30, 37 and 47), the comparative value
f certain well-known and commonly-used volatile and gaseous
isinfectants was tested. In these experiments vaccine virus was
lie substance exposed to the action of disinfectants, and the test of
isinfection was insertion of the disinfected virus into the arm of
b unvaccinated child, virus from the same source not disinfected
ing inserted at the same time at a different point. A positive
«8alt from the non-disinfected virus and a negative result from
liat exposed to the disinfecting agent was taken as proof of the
tency of this agent.

As an additional test some experiments were made upon the

cteria contained in putrid urine, the test of disinfection being

lie failure to multiply in sterilized urine after exposure to the

ction of a disinfectant. (See Bulletin Nos. 37 and 47, Vol. 1.)

The general results of these experiments may be stated 9s

bllow8 :

Chlorine. — In experiments upon vaccine virus, dried upon ivory

^joints, an exposure for six hours in an atmosphere containing 5

Volumes to 1,000 of air (J per cent.) was found to destroy the

potency of the virus. A still smaller quantity (J per cent.) was

^found to destroy the vitality of bacteria dried upon a piece of

filtering paper, and it is possible that further experiments would

lave demonstrated the efficiency of this agent in still smaller

quantities.

Nitrous acid gas (generated by pouring nitric acid on copper
filings and collected over mercury) destroyed the potency of vac-
cine virus in the proportion of 1 per cent. (1 volume to 100 of
air); time of exposure six hours. The experiments upon bacteria
showed this agent to be efficient in the proportion of £ per cent.,
but it broke down at J per cent. I should, therefore, place the
minimum amount which can be safely relied upon to destroy dried

201

202 GEO. M. STERNBERG.

films of virus and the bacteria of putrefaction (dried upon filtering
paper) at 1 per cent.

Sulphurous acid gas. — This agent was tested in various propor-
tions, and was found efficient in the proportion of 1 per cent, for
vaccine virus (no experiments made with a smaller amount), and
in the proportion of £ per cent, tor bacteria. Like nitrous acid,
it broke down at J per cent, in experiment No. 40, in which bac-
teria from putrid urine, dried upon filteriug paper, were exposed
to its action for six hours.

The conclusion reached is that these three agents, chlorine,
nitrous acid (nitrogen dioxide), and sulphurous acid (sulphur
dioxide) are reliable disinfectants in the proportion of 1 volume
to 100 of air. It is probable that a considerably smaller propor-
tion of the above disinfectants would be efficient in destroying the
potency of thin layers of virus in a moist state, or of virus ex-
posed to the action of the disinfectant in an atmosphere saturated
with moisture. It was my intention to determine the minimum
quantity of each of these agents which could be relied upon to
destroy the potency of vaccine virus, both in a dry and in a
moist atmosphere, but the difficulty of obtaining un vaccinated
persons upon whom to make the trial has prevented me from
making further experiments in this direction up to the present
time.

Carbolic acid. — The following remarks, quoted from Bulletin
No. 47, show the results reached in my experiments with this
agent :

The amount of pare acid required to destroy the vitality of bacteria
(10 grains, experiment No. 42) js equal to about 17 pounds in a room
12 feet square and 12 feet high (capacity 1,728 cubic feet), and to
fulfil the conditions of the experiment in disinfecting on a large scale,
it would be necessary to scatter this amount over the floor of a room
having these dimensions, and to suspend articles to be disinfected near
the floor for at least six hours, care being taken that all apertures were
closed so that the fumes of the acid might not escape. Experiment
No. 43 shows that four times this amount (68 pounds) of "crnde"
acid placed upon the floor of a room of the same dimensions would not
destroy the vitality of bacteria exposed in the room for six hours.
Experiment No. 24 (Bulletin No. 29) shows that an amount of the
impure acid equal to 46 fluid ounces volatilized in the same room will
not destroy the potency of vaccine virus in a moist state (rubbed up

EXPERIMENTS WITH DISINFECTANTS. 203

with glycerine) when the time of exposure is twelve hours. Finally,
these experiments show that the popular idea, shared, perhaps, by some
physicians, that an odor of carbolic acid in the sick-room, or in a foul
privy, is evidence that the place is disinfected, is entirely fallacious,
and, in fact, that the use of this agent as a volatile disinfectant is im-
practicable, because of the expense of the pure acid and the enormous
quantity required to produce the desired result.

Recent Experiments with Non-Gaseous Disinfectants.

Having ascertained that I have at hand a ready means of pro-
ducing a fatal form of septicemia in the rabbit (see special report
to National Board of Health, Bulletin No 44, Vol. II), and that
the blood and serum from the sub-cutaneous connective tissue of
a rabbit recently dead possesses still greater virulence than the
human saliva used in the first instance, the idea occurred to me
that this virus could be used to good advantage in further experi-
ments with disinfectants, the test being injection beneath the skin
of a healthy rabbit. As the virus so introduced produces death
in from twenty-four to forty-eight hours, it is evident that a nega-
tive result after treatment with a disinfectant is proof of its power
to destroy the virulence of the injected material or, in other words,
to disiufect it.

My results have been, in the main, very definite and satisfac-
tory, but my experiments have brought to light certain facts which
I did not fully appreciate at the outset, and which to some extent
detract from the value of the experiments herein reported.

These facts are :

(a) The action of certain substances may so modify the potency
of the virus that the fatal event is postponed from the fifth to
ninth day instead of occurring as usual during the first forty-eight
hours after injection ; consequently the assumption, upon which I
at first acted, that a rabbit which seemed in good health four or
five days after an injection, could be placed to the credit of the
disinfectant and used for another experiment, cannot be considered
a safe one, and it would have been better to allow a longer time
to elapse or to have used a fresh rabbit for each experiment. This
criticism only applies, however, to a small number of the experi-
ments made, as I have rarely given more than two injections to

204 GEO. M. STERNBERG.

the 8ame animal, and in cases where a negative result followed the
second as well as the first, the evidence is perfectly definite, the
doubt only occurring in those cases in which a fatal result followed
a second injection, which might possibly have been due to the pre-
vious injection, while credited to the last one made.

The following experiments will serve as examples of this post-
ponement of the fatal event as the result of the action of the dis-
infectant used :

June 13. — Injected 0.5 c. c. of virus, to which had been added
one-tenth of 1 per- cent, of iodine (in aqueous solution with potas-
sium iodide).

Remit — Died June 24. Post-mortem examination made imme-
diately after death (died in convulsions) showed hemorrhagic ex-
travasations under the skin in vicinity of point of injection, spleen
enlarged and dark colored, liver normal, blood from hemorrhagic
extravasations under skin and from mesenteric veins (no other
examinations made) contains an abundance of micrococci.

Same date (July 13). — Injected 0.5 c. c. of virus containing 10
per cent, of oil of eucalyptus globulus.

Result. — Died June 21 (was killed when evidently on the point
of death). Blood drawn into graduate measure coagulates very
firmly. Serous discharge from bowels (abundant) contains an
abundance of micrococci and other forms of bacteria ; no bacterial
organisms found in the blood; no cellulitis; liver and spleen
normal.

Remarks in this Case. — There is no evidence of septicaemia,
and it may be that the fatal result was due to the independent
action of the oil of eucalyptus, or to some other cause independent
of the injection made. Some of the serous discharge (0.25 c. c)
from the bowels of this rabbit was injected into a small rabbit
without result. An injection of 0.5 c. c. of blood serum (from
graduate measure after retraction of clot) into a small rabbit gave
also a negative result.

In the first of these cases the post-mortem examination gave evi-
dence of death from septicaemia. In the second the evidence was
to the contrary effect ; but it is very evident that either of these
rabbits, if made the subject of a second experiment, on the fourth
or fifth day after the first injection, although apparently in good
health at the time, would have given an uncertain or fallacious
result.

EXPERIMENTS WITH DISINFECTANTS. 205

The following experiment made at the same time and with the
same virus as the preceding is given to show that this virus was
reliable :

June 13. — Injected 0.5 c. c. of virus one part and camphor water
(aqua camphora of the Pharmacopoeia) one part into a small
rabbit

Result — Death occurred during night of June 15 with the
usual symptoms of septicaemia — diffuse cellulitis, enlarged spleen,
micrococci in blood, and effused serum in sub-cutaneous connective
tissue.

(6) Several small rabbits have died without any injection, and
from the appearance of the spleen and the presence of the micro-
coccus in the blood, I have concluded that these were cases of
septicaemia, not of traumatic origin, resulting from confinement in
cages in which other rabbits, tHe subjects of my experiments, have
died. These septicemic rabbits have very commonly a serous
diarrhoea shortly before death by which their cages and the food
remaining in them are soiled, and which contains an abundance of
septic micrococci. I have proved experimentally that not only
this serous discharge from the bowels but the saliva of an infected
animal possesses virulent properties and produces speady death
with the usual symptoms. (See speeial report to National Board
of Health, /. c.) 1 suppose, therefore, that these deaths resulted
from exposure in infected cages, a supposition which is supported
by the observations of Davaine, who affirms that septicaemia may
occur among rabbits as an epizootic independently of any wound
or contact with other rabbits suffering from septicaemia. (Re-
cherces sur quelques-unes des conditions qui favorisent ou qui
emp&hent le dSveloppement de la septicaemia. Bull, de l'Acad.
de M6d., 2 s., T. VIII, p. 121.)

That these rabbits died from an infectious septicaemia is further
proved by the fact that a small quantity of blood from one of
them (0.25 c. c.) injected beneath the skin of a large rabbit caused
death with the usual symptoms iu less than twenty-four hours.

As the companions of these rabbits of the same age (less than
two months and weighing about a pound) were subjected to ex-
periment and some died, doubt is thrown upon the result of these
experiments and I am obliged to exclude them from my record.

(c) The most important source of error, however, and one which
must be kept in view in future experiments, is the fact that a pro-

206 GEO. M. STERNBERG

tective influence has been shown to result from the injection of
virus, the virulence of which has been modified without being
entirely destroyed by the agent used as a disinfectant.

The following experiments will serve as examples of this:

May 2J/,. — Injected into a large rabbit (the subject of a previous
experiment, May 13, in which a negative result was noted and in
which 0.5 c. c. of virus treated with 1 per cent, of sodium hypo-
sulphite was injected) 1.25 c. c. of virus, not disinfected, from
rabbit recently dead.

Result negative.

Same date (May 24). — Injected into large rabbit (subject of pre-
vious experiment, May 13, in which 0.15 c. c. of a mixture of
virus three parts to alcohol, 95 per cent., one part was injected)
1.25 c. c. of virus not disinfected.

Result — This animal died June 2, nine days after the injection.
Post-mortem examination showed the spleen to be small and dark
colored ; liver contained numerous small abscesses ; no diffuse cel-
lulitis; no micrococci in blood. A small quantity of the blood
of this animal (0.25 c. c.) was injected into a small white rabbit
(weighing about one pound). This animal died June 6. Post-
mortem examination disclosed limited cellulitis without the presence
of micrococci; liver and spleen normal; no micrococci in blood,
which contains numerous granular white corpuscles.

Remarks. — These two animals probably died as the result of
the injections made, but they evidently did not die from the malig-
nant infectious septicaemia produced by introduction beneath the
skin of an unprotected animal of a small quantity of fluid con-
taining the micrococcus. In the latter case we not only have the
marked difference as to date of death, but the characteristic diffuse
cellulitis, the greatly enlarged spleen, and the presence of the
micrococcus, as distinguishing characteristics. It may be that
death in these cases resulted from the poisonous properties of the
sepcin, a chemical poison contained in the blood injected, but it is
evident that both of the large rabbits previously experimented
upon possessed an immunity from the action of the septic micro-
coccus, or rather that it could not multiply in the bodies of these
protected animals, and consequently that death did not result from
the infectious form of septicaemia, which has recently been the
subject of my studies (/. c). This immunity corresponds with
what has been proved to be the case iu charbon, chicken-cholera.

EXPERIMENTS WITH DISINFECTANTS. 207

and pleuro-pneumonia of cattle, in which diseases it has been
shown that protective inoculations may be practiced.

In the first case above reported the result was completely nega-
tive1 although the amount of virus injected was considerable
(1.25 c. c), and this virus was proved by comparative experiments
to be potent. Other evidence might be adduced in favor of the
view that protection results from the effects of inoculations made
with virus modified by the action of certain agents; but my object
here has simply been to show the importance of considering this
possible protective influence of previous injections in making dis-
infection experiments upon a virus of this character.

My method of collecting virus for disinfection experiments has
been to wipe up the bloody serum from the sub-cutaneous con-
nective tissue and from the thoracic and abdominal cavities, after
removal of the viscera and puncture of the large veins, with dry
cotton, which is then washed out in water. The potency of this
diluted virus has been amply proven and, indeed, in every series
of experiments made at the same time and with the same material,
1 have obtained evidence of virulence either from injection of non-
disinfected virus as a check experiment, or by the failure of one
or more of the substances undergoing trial as disinfectants. Thus
in the experiments just reported, the same virus killed a rabbit in
less than three days after having been treated with a 4 per cent,
solution of magnesia sulphas.

I have not attempted to determine the minimum quantity of
virus that would be effectual, but have kept on the safe side by
injecting quantities much in excess of the amount required to pro-
duce fatal septicaemia. In the experiments of Davaine (/. c), in
which the virus in the first instance was obtained from a different
source, fatal septicaemia was produced by injections of septicemic
blood in quantities as small as g^ part of a drop.

When we are dealing with a virus of which the virulence de-
pends upon the presence of a living organism capable of self-mul-
tiplication in the body of the animal into which it is introduced,
it is evident that the question of quantity is quite secondary to

1 In these experiments no temperature observations have been made, and by
a negative result failure to kill only is implied. No doubt slight indisposition
and a greater or less amount of fever might have been verified in many cases
by careful observations, but the object in view rendered such observations un-
necessary and want of time rendered them impracticable.

208 GEO. M. STERNBERG.

that of vital activity on the part of the pathogenic organism and
vital resistance upon the part of the living tissues of the animal
subjected to its action.

It seems probable, in the light of recent experiments, that patho-
genic properties in these lowly organisms depend upon rapidity of
development and adaptability to conditions such as are found in
the interior of the bodies of living animals, and that these quali-
ties may be developed in common and usually harmless bacterial
organisms as the result of specially favorable conditions, such as
high temperature, abundance of pabulum, &c.

That the virus which has been* used in these experiments is
capable of producing death in much smaller quantities than those
used, is shown by the following experiment :

June 2. — The needle of a hypodermic syringe was dipped into
the blood of a septicemic rabbit just dead, and proved by micro-
scopical examination to contain an abundance of the micrococcus.
It was then introduced under the skin of a small rabbit.

Result. — This animal died within 48 hours and presented all
the usual appearances of death from septicaemia.

An additional possible source of error will suggest itself as
arising from the extreme virulence and the small quantity of
material required to produce death. A vey little of this mate-
rial, not disinfected, adhering to the needle of the hypodermic
syringe from one experiment might be the cause of death in a
succeeding one and might improperly be ascribed to failure of the
disinfectant used in the last experiment. This possibility I have
had in view and have carefully guarded against by a thorough disin-
fecting and cleansing of my syringe after each injection. This has
been effected by means of a 10 per cent, solution of carbolic acid
or more frequently with diluted sulphuric acid, followed by re-
peated washings with pure water.

My practice has been to mix the different disinfectants to be
used at one time with separate portions of virus, obtained as
already described from the cellular tissue and blood-vessels of a
rabbit recently dead, in small beakers well cleaned, and to allow
a period of twenty minutes to half an hour for the action of the
disinfecting agent before making an injection.

Standard solutions of the different substances to be tested were
kept in glass-stoppered bottles, and at the outset of my experi-
ments these solutions were made of the strength of 5 per cent.

EXPERIMENTS WITH DISINFECTANTS. 209

Solutions of 4 per cent, were afterwards substituted for these
because of the greater convenience in reducing the quantity with-
out fractions. Thus one part of virus and one part of a standard
4 per cent, solution gave me the proportion of 2 per cent.; three
parts of virus and one of the disinfectant gave the proportion of
1 per cent., Ac.

Having fairly stated the possible sources of error in experiments
made by this method, I may be permitted to say that I believe my
results to be in the main reliable, and that the substances which
have best stood the test may be depended upon in practical disin-
fection in the proportions found to be efficient.

In but a single instance have I had a contradictory result in
which the greater quantity failed and the smaller did not. This
was in the use of zinc chloride, with which three experiments were
made. The rabbit injected with 1 per cent, died, while two others
injected with 2.5 per cent, and 0.5 per cent, gave a negative result.
To which of the possible causes of error, already pointed out, this
contradictory result is due, I am unable to say. The rabbit in-
jected with 1 per cent, may have died from some cause indepen-
dent of the injection, or from the remote effects of a previous
injection, or the rabbit injected with 0.5 per cent, may have been
protected by a previous injection. It is evident that in future
experiments by this method it will be desirable to use a previously
uninjected animal for each experiment.

After this somewhat lengthy preamble, which has seemed neces-
sary, I shall proceed to detail the results of these experiments,
placing first those substances which have proved most efficient.
For convenience each experiment will be recorded by placing after
the name of the substance used the figures representing the pro-
portion in which it was used. Death, or failure to disinfect, is
indicated by a full-faced figure representing proportion of disinfec-
tant used. The plain figure indicates a negative result or destruc-
tion of virulence by disinfectant (disinfection).

210 GEO. M. STERNBERG.

Group 1.

Disinfectants efficient in the proportion of 0.5 per cent, or less.

Iodine (in aqueous solution with potassium iodide), 1.25, 0.5,
0.25, 0.2, O.l.1

Chromic avid, 1, 0.5, 0.2, 0.1. (No failure.)
Ferric sulphate, 1.25, 0.5, 0.25, 0.12, 0.12.2
Cupric sulphate, 1, 0.5, 0.25, 0.1*
Thymol dissolved in alcohol, 1, 0.25, 0.1*
Caustic soda, 2.5, 1, 0.5, 0.25, 0.2*
Nitric acid, 1.25, 0.5, 0.25, 0.2.
Sulphuric acid, 1.25, 0.5, 0.25*
Ferric sesquicfdoride, 1, 0.5, 0.25.
Sodium hyposulphite, 1, 0.5, 0.25.
Hydrochloric acid, 0.5, 0.25*

Group 2.

Disinfectants which failed at 0.5 per cent., but proved efficient in

proportions below 2 per cent.

Carbolic acid, 2.5, 1.25, 0.5*

Salicylic acid (as salicylate of soda), 2.5, 1.25, 0.5.

Zinc chloride, 2.5, 1, O.5.3

Caustic potash, 2.5, 1, 0.5.

i In the experiment with 0.1 per cent, the animal did not die until eleven
days after the injection ; it is, therefore, hardly fair to consider this a failure of
the disinfectant, but in the absence of additional experiments 1 have thought it
best to mark this as a failure, and to assume that the limit of safety as to pro-
portion of the disinfectant required has been passed. It was my intention to
make a separate series of experiments with potassium iodide for the purpose of
ascertaining whether this agent should receive a portion of the credit for the
results obtained by the solution used. The scarcity of rabbits has prevented
me from making this experiment up to the present time.

2 Two experiments were made with 0.12 per cent, of ferric sulphate, in one
of which the result was negative (disinfection), and in the other the rabbit died
(failure to disinfect).

3 See remarks on page 209 for explanation of this contradictory result.

EXPERIMENTS WITH DISINFECTANTS. 211

Iron-tdum, 2, 1.

Zinc sulphate, 1.25, 0.5.

Potassium sulphide (sulphuret), 2, 0.5.

Tannic acid, 1, 0.5.

Boracic arid, 2, 1, 1,

Potassium permanganate, 2, 1, 1.

Sodium biborate, 2.5, 1.25.

Group 3.

Substances which failed to disinfect in the proportion of 2 per cent

Potassium nitrate, 4.
Potassium chlorate, 4.
Sodium chloride, 2.5.

:<4/um, 1.25, 4.

Ziarf acetate, 2.
Magnesia sulphate, 4.
Glycerine, 25, 12.5, 10.
Alcohol (95 per cent.), 25, 12.5, 10.

Camphor water. Equal parts of camphor water and virus were
injected with a fatal result.
PyrogaUic acid, L
Oil eucalyptus globulus, 10.1

Remarks. — It was my intention to make this experimental
inquiry as complete as possible before reporting, and to fix defi-
nitely the minimum quantity, which may be relied upon to destroy
the potency of septic virus (Micrococcus septicus), of those sub-
stances most commonly used as disinfectants, or prescribed inter-
nally, or as lotions, with a view to their antiseptic action ; also to
determine the time during which the septic virus will retain its

1 The rabbit injected with one part of oil eucalyptus to nine of virus did not
die until eight days after the injection, and the post-mortem examination showed
that it did not die of septicaemia. This cannot, therefore, be fairly considered a
failure to disinfect, and further experiments will be required to determine the
value of this agent, which is especially interesting just now from the fact that
Lister is using it in his antiseptic dressings to wounds.

212 GEO. M. STERNBERG.

potency in a dry state j the effect of gaseous and volatile disin-
fectants upon the dried virus, both in a dry and moist atmos-
phere; the comparative value of various proprietary disinfectants
now in the market ; the thermal death-point of Micrococcus aepti-

CU8, &c.

It will be seen that I have fallen far short of the accomplish-
ment of this purpose, but I have thought it best to report what
has already been accomplished, as practical sanitarians may obtain
some hints of value from the experiments recorded, and it is very
uncertain when I *will be able to resume my experiments, which I
have been obliged to discontinue on account of the pressure of
other duties and the difficulty of obtaining rabbits for experi-
mental purposes.

OBSERVATIONS ON THE DIRECT INFLUENCE
OF VARIATIONS OF ARTERIAL PRESSURE
UPON THE RATE OF BEAT OF THE MAM-
MALIAN HEART. By H. NEWELL MARTIN, M. A.,
M. D., D. Sc. With Plate XV.

The earliest observations on this subject, so far as I know, were
made by Marey (Recherches sur le pouls au moyen d'un nouvel
appareil enrSgistreur. IJemoires de la Soeiiie de Biologie, 1859) ;
but as the extrinsic cardiac nerves were not divided in his experi-
ments, and a rise of blood pressure is now known to stimulate the
medullary cardio-inhibitory and accelerator nerve centres, the
results obtained by him give really no information as to the direct
influence of increased aortic tension upon the rate of the heart's
beat. Since then others have experimented, previously dividing
the extrinsic cardiac nerves, Ludwig and Thiry in 1864 (Sitzb. d.
Akad. d. Wissensch. zu Wien) leading the way, but the general result
is that the matter has been left in a highly unsatisfactory state.
Some find that variations of arterial pressure have no effect on a
heart whose venous connections with other parts of the body have
been severed ; others that arterial pressure and pulse rate rise and
fall together; others that the pulse quickens when arterial tension
is lowered and vice versa. Finally, Tschirjew (Arch.f. Anat. u.
Physiologic, Jahrgang 1877, p. 116), the latest writer on the sub-
ject, finds all of the above effects in different cases: as the result
of an extensive series of experiments he conies to the conclusion
that after section of all the extriusic heart nerve paths, "any con-
siderable and rapid elevation of blood pressure may directly
stimulate either the inhibitory apparatus in the heart, or its motor
ganglia, and the pulse rate accordingly be increased or diminished,
or in more rare cases remain unaltered." Such contradictory
results obtained by a number of competent workers lead naturally
to the suspicion that some error is involved in the methods of
experiment, employed ; the nature of this error is not, I think,
far to seek. The methods used to vary arterial pressure have been
such as cause variations also in several other conditions which

213

214 K NEWELL MARTIN.

either are known to influence the heart, or may possibly do so ;
nevertheless all these secondary actions have been unheeded: their
relative prominence in any given experiment has not been noted,
and any change in the pulse rate has been ascribed solely to the
changed arterial pressure. Under such circumstances it need
cause no surprise that very inconsistent results should be obtained.

The higher aortic pressure is, the more force must be expended
by the left ventricle in forcing open the semilunar valves ; that is
to Ray, the higher will be intraventricular systolic pressure. It is
this influence only of increased aortic pressure which should be
meant when its direct action upon the cardiac rhythm is spoken
of; and to get pure rest Its all other consequences of increased
arterial tension which may influence the heart's rate of beat must
be eliminated. This, however, has not been the case in any series
of experiments with which I am acquainted.

Arterial pressure has commonly been increased by clamping the
descending aortic, either in the thorax or abdomen. When this is
done, however, we alter several other things in addition to arte-
rial pressure —

(1.) The amount of blood returned to the right auricle in a
given time is almost certainly altered, and therefore the rate of
filling of the heart during diastole.

(2.) The pressure under which venous blood enters the right
auricle is probably changed, and therefore intracardiac pressure at
the end of the diastole.

(3.) The temperature of the blood returned to the heart by the
systemic veins and, as a consequence, of the heart itself, is altered.
The blood returned to the right auricle by the inferior cava is
known to be warmer than that returned by the superior cava,
which has not flowed through the hot abdominal organs. When
the aorta is clamped the heart gets only the cooler superior cava
blood, as the capillary tracts tributary to the inferior cava are no
longer supplied with blood.

(4.) It is known that very slight chemical changes in the blood
profoundly influence the heart's beat. To quote no other instance,
Gaule has shown that the heart of a frog previously kept in the
cold and exhibiting deficient functional power, may be restored to
full vigor by circulating through it the extract of the heart of a
frog kept previously at a higher temperature. Blood in its flow
through the abdominal organs experiences important chemical

ARTERIAL PRESSURE ON PULSE RATE. 215

changes entirely differing from any undergone in other regions of
the body. If, therefore, we circulate blood through head, neck
and fore limbs only, and return it again and again to the heart
without exposing it to the action of kidneys, spleen and liver, we
very soon have a liquid to deal with which is essentially different
from that which flowed through the heart before the aorta was
ligated.

Of course when the arterial pressure is lowered by opening the
previously clamped aorta all of the above possible disturbing
actions occur in the opposite direction.

Another method which has been employed to raise arterial pres-
sure is to inject blood from another animal into the carotid of the
animal experimented upon. This also involves several possible
sources of error. (1) Venous inflow during cardiac diastole is
almost certainly changed. (2) Venous pressure and, therefore,
intracardiac diastolic pressure are probably altered. (3) The in-
jected blood may differ chemically from that already in the vessels,
and directly act upon the heart. (4) Unless extreme care be
taken the temperature of the injected blood will be less or greater
than that of the already circulating blood, and will alter the tem-
perature and, therefore, the rhythm of the heart. To the above
objections it may be added that only slight increase of arterial
pressure can be brought about in this way ; as is proved by Worm
Muller's experiments. (Arbeiten aus d. physiol. Amtali zu Leip-
zig, 1873).

When blood pressure is lowered by bleeding, diastolic inflow
and pressure are altered, as well as arterial pressure; and also
probably the chemical metabolisms experienced by the blood in
its flow through different organs.

As some one, at least, of the above secondary influences has
been present in all previous experiments as to the influence of
variations of arterial pressure upon the pulse rate, it is clear that
none of these experiments, interesting and important as their
results are in many cases, are really capable of affording an answer
to the question in hand, viz: what is the influence, if any, pure
and simple, of increased aortic pressure (*. e. of increased systolic
pressure within the left ventricle) on the pulse rate. It is, there-
fore, not necessary to consider in detail the experiments of pre-
vious writers. All are vitiated more or less by secondary changes
11

216 H. NEWELL MARTIN.

which have occurred along with the variations of arterial pressure;
and the number of these possible complications, and their varying
degree in different experiments, affords a sufficient explanation of
the contradictory results obtained.

As regards the frog's heart, there is more agreement between
observers, and the experimental conditions havq usually been more
satisfactory. Usually the auricle is supplied steadily with liquid
of constant composition and at constant pressure from a Marriott's
flask; but even here, so far as I know, the arterial cannula has
always been inserted into the ventricle and, therefore, beyond the
semilunar valves. As a necessary consequence of this, not only
systolic ventricular pressure (which normally is the thing changed
by varied arterial pressure), but also diastolic intraventricular
pressure has been varied. I accordingly suggested to two of
my pupils that they should undertake a fresh examination of this
question by better methods, on the hearts of frogs and chelonia.
Some results of their work will be found on subsequent pages of
the present number of this Journal.

The question involved is clearly one of great importance. In
almost every experiment relating to cardiac physiology arterial
pressure is altered : and it is essential to know exactly the direct
influence of this factor on the heart, before further conclusions can
be legitimately arrived at. I have, therefore, lately carried out a
large number of experiments as to the direct influence of variations
of arterial pressure upon the pulse, making use of the dog's heart
completely isolated physiologically from every other organ, but
the lungs: the method of isolation, which essentially consists in
closing the whole systemic circulation except that through the
coronary vessels of the heart itself, was described by me in the
last number of this Journal (Vol. II, No. 1, p. 119); as the ap-
paratus has since been modified only in some points of detail, I
liere reproduce, as Plate XV, the figure used in illustrating the
previous paper, in order to assist in the description of my more
recent experiments.

The right and left carotid arteries, o and r, have cannulas placed
in them, the right subclavian, w, is ligatured, and a cannula is
put in the left subclavian, m. Then the aorta is ligated imme-
diately beyond the origin of the left subclavian: the vena cava
inferior and the azygos vein are tied, and a cannula put in the
superior cava. Fresh defibrinated strained and warmed blood is

ARTERIAL PRESSURE ON PULSE RATE. 217

now ran in by the superior cava ; at the same time the cannula on
the right carotid is opened, and blood drawn from it until there is
reason to believe that all the blood originally in the heart and
lungs of the animal has been washed out; the carotid is then
again clamped, and the superior cava a few seconds later, when
the heart and lungs have been tolerably well filled with blood.
The animal is then transferred to the warm moist chamber, K,
the cannula of the superior cava is connected with one of the Mar*
riott's flasks, 27 or 28, from which a nutrient liquid is sent into
the heart under a uniform pressure, which in the experiments
described below was that exerted by a column of blood 10 centi-
metres in height. The left carotid, o, is connected with the out-
flow tube, 21, and the cannula in the subclavian with a mercurial
manometer, 26, the pen of which writes on the paper of a kymo-
graph ion in the usual manner. As soon as one Marriott's flask is
empty its connection with the heart is shut off, and that of the
other (which has been meanwhile closed) is freed by opening the
proper one of the clamps, 1 or 2, and closing the other. The
nutrient liquids employed in the experiments below described were
(1) fresh defibrinated strained dog's blood ; (2) the same diluted
with an equal bulk of 0.5 per cent, solution of sodium chloride
in distilled water. 1 may here state that in other cases I have
used with success (3) defibrinated dog's blood with one-third its
bulk of 0.7 per cent, sodium chloride solution; and (4) defibri-
nated calf's blood.

Under these conditions almost all of the ordinary collateral
results of increased or lowered arterial pressure can be elimi-
nated. By closing more or less completely the stop-cock, 22,
arterial pressure can be raised; by opening the stop-cock wider
it can be diminished. Meanwhile rate of supply to the right
auricle, the temperature of the liquid sent into it, and the compo-
sition of this liquid are unvaried ; all these disturbing elements
are thus got rid of. I have said above that "almost" all secon-
dary effects can be eliminated; the almost is due to the varied
coronary circulation ; when aortic pressure is high this must be
greater than when that pressure is low ; so far I see no method of
eliminating this possible source of error; but in recent years much
evidence has been accumulated to shew that if the flow of blood
through an organ is sufficient to nourish it (t. e., does not fall
below the starvation limit), and is under a lower pressure than
such as ruptures the vessels or otherwise mechanically impedes

218 H. NEWELL MARTIN.

the action of the organ, there is much reason to believe that varia-
tions in blood supply have no immediate influence on its functional
activity. The experiments detailed below give further support to
this view: as will be seen, variations of arterial pressure ranging
between 25 and 150 mm. of mercury have no influence whatever
upon the heart's rhythm, although considerably more blood must
flow through the coronary system under the higher than under the
lower pressure.

In the experiments described below the heart was always left in
the warm chamber at least half an hour before observations were
made, and longer if the thermometer did not shew that the tem-
perature was then uniform and had been for some five or ten
minutes. The animals during the isolation of the heart were
sometimes placed under the influence of morphia, sometimes of
curari, and sometimes of chloroform ; these various agents were
used to eliminate chances of error due the possible toxic action
of any one of them on a regulatory mechanism in the heart,
though when fresh unpoisoned defibrinated blood is run for hours
through the heart after its isolation, there can be little doubt
that any poison absorbed by the organ during the preliminary ob-
servation is thoroughly washed out. . The animals used were small
dogs, weighing from 6 to 7.5 kilos. Uniform artificial respiration
was kept up by means of a small water engine.

When temperature had become constant, the connection between
a full Marriott's flask (containing about 700 c. c. of liquid) and
the heart was opened. A minute or two was allowed to elapse, to
get a steady inflow current; then arterial pressure was raised by
partially closing the stop-cock, 22, or lowered by opening it wider.
Tracings were taken for from two to six minutes with arterial pres-
sures varied in this way ; then the observation ceased. Mean-
while the other Marriott's flask was filled ; and after some minutes
another observation was made while it was connected with the
heart ; and so on, so often as seemed desirable. In all cases the
experiment came to an eud long before the heart shewed signs of
abnormal or irregular action ; indeed in most instances it was sub-
sequently used for preliminary observations on the influence of
other conditions, as varied venous pressure or varied temperature
on the pulse rate.

The results arrived at may be summed up as follows :

1. When the pressure under which blood of uniform temperature
and composition is steadily supplied to the rigid auricle does not

ARTERIAL PRESSURE ON PULSE RATE. 219

exceed that due to a column of blood ten centimetres in height, no
variation of arterial pressure which can be brought about by opening or
closing more or less completely the outflow stop-cock, has any influence
whatever on the rhythm of a heart isolated from all other organs of
the body except the lungs, provided arterial pressure be not kept at a
very low level for a considerable time. In other words, within very
wide limits, changes in arterial pressure have no influence whatever
upon the pulse rate.

2. If the outflow stop-cock be widely opened and arterial pressure
lowered to less tlian twenty millimetres of mercury, Uiis has no direct
influence on the pulse rate; but it has probably an indirect influence.
For a minute or more the heart beats recur at the same intervals, but
after that time, if the low pressure be still maintained, the pulse some-
times becomes slower, probably from deficient nutrition of the heart
dependent on insufficient flow through the coronary vessels.

3. If the pressure at which venous blood enters the rigid auricle be
considerable [due to a column of blood forty centimetres in height),
and if simultaneously the arterial exit be greatly narrowed by closing
the outflow stop-cock, then arterial pressure at first rises greatly with-
out any alteration in the pulse rate; but ultimately attains a very
high level at which the cardiac rhythm becomes extremely irregular.
Beats occur which somewhat resemble those produced by feeble pneu-
mogastnc stimulation. If the arterial resistance be now diminished,
markedly dicrotic beats occur for some twenty or thirty seconds, until
arterial pressure again falls to a normal level, when the original
pulse rate is resumed. The conditions when the irregular beats are
observed are clearly pathological: a filling of the heart under a
pressure in the vence cava equal to forty centimetres of blood (twenty-
nine millimetres of mercury) probably never occurs normally com-
bined with great arterial resistance.

In the present article I shall confine myself to what may be
called normal variations of arterial pressure, that is to say, for
small dogs, variations between 25 and 160 millimetres of mercury.
The result under the above heading 2 is undoubtedly abnormal,
and due to commencing death of the heart; and the results indi-
cated under number 3 are probably due either to the reception by
the left ventricle in each diastole of more blood than, under the
resistance opposed to it, it can pump out in one systole, or to a
direct stimulation of inhibitory mechanisms in the heart by the
pathological pressure within the ventricle. This irregular beat

220 JST. NEWELL MARTIN.

with very great arterial resistance has been noted by Haidenhain,
and I may here state that Knoll's opinion that it really means
not a slowed heart beat, but a quick irregular beat which the
manometer does not properly record, is incorrect; direct observa-
tion of the exposed heart is conclusive as to the fact that the beats
are not quick and irregular, but really slow, and frequently dicrotic.

On the results numbered 2 and 3 above I desire to make further
observations before publishing detailed conclusions. Hitherto so
soon as I have observed indications of them I have at once raised
or lowered arterial pressure so as to prevent death or injury to the
heart. As regards point 1, the three tables below speak for them-
selves. They are selected from a dozen experiments which are
perfectly concordant, and they have been so selected that a dif-
ferent drug was given to the dog during the preliminary opera-
tion of isolating the heart in each case. The venous inflow was
a 1 way 8 so proportioned to the resistance to arterial outflow that pres-
sure in the subclavian during the intervals between any two obser-
vations was kept at a point from which arterial pressure could be
considerably raised without the variation passing beyond a physio-
logical limit; but at the same time, a pressure sufficient to keep
the heart in a functional condition for a long time.

Venous pressure in all the experiments recorded below was that
due to a column of nutrient liquid (defibrinated dog's blood, or
the same diluted with an equal volume of sodium chloride solu-
tion) ten centimetres in height, or very near that; it is not well
practicable to measure exactly in every experiment the difference in
level between the cannula in the superior cava and the lower end
of the tube for the entry of air into the Marriott's flask; but
errors of a few millimetres in this regard are of no importance: so
long as the pressure is constant during an observation a know-
ledge of its absolute amount within 5 or 6 millimetres of blood
is of no consequence.

The tables are constructed as follows: Temperature in the moist
warm chamber having become constant, the kymographiou was
started and tracings taken for from two to seven minutes. During
this time the stop-cock, 22, was opened wider, or more closed, or
opened and then closed, or vice verta, and consequently arterial
pressure was altered. A number of such observations having
been made the tables were constructed from the tracings obtained :
suppose the time to be 2 h., 20', 10", then arterial pressure is

AR TERIAL PRESS URE ON P ULSE RA TE. 221

measured at that time and at 2 h., 20', 20".. Half the sum of these
is taken as the mean pressure during the intervening ten seconds.
The pulse rate is counted for this ten seconds, multiplied by 6, and
the product given as the rate of heart beat per minute, with the
mean arterial pressure obtained as above. So far as absolute results
are concerned, it is seen that the mean arterial pressure arrived
at in this way is open to some error, and had changes in it been
accompanied by changes in the pulse rate, more accurate methods
of arriving at the true mean arterial pressure during each ten
seconds would have to be employed. But as very great variations
of mean arterial pressure were used and as the experiments shew
that none of them, within the limits described above as physio-
logical, cause any change in the rate of the heart's beat, it is clearly
unnecessary to resort to planimetry or other troublesome methods
so as to avoid possible errors of a few millimetres in the measure-
ments. When gross variations of arterial pressure from 30 to 150
mm. of mercury cause no change, it is not worth while to spend
time in endeavoring to exclude possible errors of ten or even
fifteen millimetres of mercury pressure; and the possible limits
of error in my measurements never reached the less of those
quantities. When the lungs are kept well extended and the arti-
ficial respiration apparatus works with tolerably slow powerful
blasts, marked respiratory waves are seen on the tracings of arte-
rial pressure, unless this fall to .50 millimetres of mercury or'
thereabouts, when they disappear. As these rhythmic rises and
falls of arterial pressure render it more difficult to correctly arrive
at the mean pressure, I have usually eliminated them by arranging
my water engine so as to work with rapid short strokes; then res-
piratory variations of arterial pressure entirely disappear from the
manometer tracings.

In the experiments recorded below the heart had been physio-
logically isolated from all other organs but the lungs for some
considerable time before the recorded observations were made;
the muscles of the body in general were often already in marked
rigor before the first observation was made and always long before
the last. When the words "no record" appear in the details of
an observation, some one or more of the pens was not writing, so
that either time, pressure, or pulse rate, could not be determined.
The temperature given is that of. the warm chest in which the
animal lay.

222

K NEWELL MARTIN.

Experiment A.

October 13, 1881. Small dog, narcotised with morphia during the
operation of isolating the heart. Nutrient liquid 1,400 cub. cent, of
defibrinated dog's blood drawn from two other animals. Arterial pres-
sure measured in left subclavian. Heart isolated and animal put in
warm chamber at 4 b. 10', P. M.

Observation.

Time.

Temperature in

Arterial Pressure

Pulse Kate

degrees C.

in mm. of mercury.

per minute.

4 h. 44' 00"

37°

137

147

" 10

134

147

" 20

131

146

" 30

132

147

" 40

116

147

" 50

147

4 h. 45' 00"

147

150

11 20

109

147

" 30

124

147

" 40

134

150

4 h. 46' 00"

149

150

" 10

142

149

" 20

120

147

" 30

147

11 40

150

11 50

147

II.

4 h. 58' 50"

37°

133

147

4 h. 59' 00"

134

149

139

147

" 20

143

150

" 30

144

150

11 40

142

149

" 50

138

149

5 h. 00' 00"

136

148

129

150

" 20

104

150

11 30

150

" 40

150

" 50

117

151

5 h. 01' 00"

123

148 .

129

151

" 20

133

150

" 30

130

150

" 40

110

150

" 50

151

ARTERIAL PRESSURE ON PULSE RATE. 228

Experiment A. — Continued.

Observation.

Time.

Temperature C.

Arterial Pressure.

Pulse Rate.

III.

5 h. 17' 00"

37°

112

150

" 10

119

150

44 20

102

150

44 30

150

" 40

No record.

11 50

100

150 (?)

5 h. 18' 00"

108

No record.

44 10

114

150

11 20

119

150

44 30

125

No record.

41 40

126

151

44 50

112

150

•

5 h. 19' 00"

150

IV.

5 h. 29' 40"

37°

150

11 50

' 81

153

•

5 h. 30' 00"

156

41 10

150

44 20

156

44 30

104

153

44 40

110

153

44 50

112

156

5 h. 31' 00"

111

153

44 10

112

150

44 20

150

44 30

156

44 40

156

44 50

153

5 h. 32' 00"

102

156

44 10

100

156

44 20

. 156

" 30

150

44 40

156

44 50

102

152

In observation I, arterial pressure varied between 74 and 149
millimetres of mercury (101 per cent.) and the pulse rate between
147 and 150 per minute (2 per cent.). In observation II, arte-
rial pressure varied between 82 aud 144 millimetres of mercury
(75.6 per cent.) and the pulse rate between 147 and 151 per
minute (2 per cent.). In observation III, arterial pressure varied
between 80 and 126 millimetres of mercury (57.5 per cent.) and
12

824 H. NEWELL MAS TIN.

the pulse rate between 150 and 151 per minute (0.66 per cent).
In observation IV, arterial pressure varied between 80 and 112
millimetres of mercury (40 per cent.) and, the pulse rate between
150 and 156 per minute (4 per cent.).

EXPEEIMENT B.

October 15, 1881. Small dog, curarised daring tbe preliminary
operation. Nutrient liquid 1,360 cnb. cent, of defibrinated dog's
blood taken from two other animals. Arterial pressure measured in
left subclavian. Operation completed and animal placed in warm
chest at 1 h. 50', P. M.

Observation.

Tim

Temperature in

Arterial Pressure

Pulie Rate

degrees C.

in mm. of mercury.

per minute.

2 h. 17' 50"

34.5°

53.5

120

2 h. 18' 0'0"

78.5

120

" 10

116.5

120

" 20

No record.

" 30

No record.

" 40

120

" 50

120

2 h. 19' 00"

120

" 10

120

» 20

80.5

122

" 80

102.5

122

» 40

114

121

" 50

121

120

."■

2 b. 41' 00"

35°

117

'■ 10

57.5

117

" 20

117

" 30

117

123

" 40

136

114

" 50

145

118.5

2 h. 45' 00"

104

114

" 10

118.5

" 20

114

" 30

117

" 40

117

" 50

35 '

117

2 h. 46' 00"

114

" 10

117

" 20
" 30

23
22

117
114

ARTERIAL PRESSURE ON PULSE RATE. 225

Experiment B. Observation II. — Continued.

Observation.

Time. Tempernturo C.

Arterial Pressure.

Pulse Bate.

II.

2 h. 46" 40" 35°

22.5

114

44 60

22.5

113

2 h. 47' 00"

111

41 10

111

" 20

114.5

" 30

110

III.

2 h. 64' 60"

35°

148

108

2 h. 55' 00"

116

108

" 10

112

14 20

108

11 30

109.5

" 40

108

11 50

108

2 h. 56' 00"

108

" 10

108

' " 20

112

" 30

131

111

14 40

143

110

IV.

3 h. 27' 40"

35°

72.5

44 50

87.5

102

3 h. 28' 00"

99.5

100

44 10

■

117.5

44 20

128

102

44 30

140

103

44 40

No record.

44 50

No record.

3 h. 29' 00"

No record.

•4 10

102

44 20

102

44 30

102

14 40

102

44 50

102

3 h. 31' 20"

35°

44 30

102

44 50

110

3 h 32' 00"

119

100

44 10

No record.

44 20

No record.

44 30

No record.

44 40

No record.

E NEWELL MARTIN.
Expebimbbt B. Observation V. — Continued.

Observation

Time.

Temperature C

Arterial Pressure.

Pulaa R«te.

3 h. 32' 50"

35°

127

101

3 h. 33' 00"

106

102

" 10

102

•' 20

101

" 30

" 40

100.5

*' 50

103

3 h. 34' 00"

108

" 10

104.5

102

" 20

ua.fi

101.5

" 30

122

103

" 40

130

102

" 60

131

103

3 h. 35' 00"

111

104

" 10

102

" 20

100

" 30

102

■ 40

102

" 50

102

VI.

8 h. 40' 55"

35°

50.5

102

3 h. 41' 05"

68.5

101

" 15

102

" 25

102

" 35

102

" 45

102

" 55

100

102

3 h. 42' 05"

114

102

" 15

101

102

" 25

102

" 35

102

" 45

75.5

103

" 65

98.5

102

3 h. 43' 05"

112

104

" 15

102

103

" 25

102

" 35

101

" 45

100

" 55

102

3 h. 44' 05"

102

" 15

102

" 25

104

" 35

18.5

102

" 45

17.5

103

ARTERIAL PRESSURE ON PULSE RATE. 22?

In observation I of the above experiment arterial pressure
varied between 53.5 and 116.5 millimetres of mercury (117 per
cent.) and the pulse between 120 and 122 per minute (1.6
per cent). In observation II, arterial pressure varies between
20 and 145 millimetres of mercury (625 per cent.) and the pulse
rate between 110 and 118.5 per minute (nearly 8 per cent.);
this it will be seen on closer examination is one of the cases
above referred to, which lead to the suspicion that a continued
arterial pressure (as measured in the subclavian) of less than
30 millimetres of mercury is insufficient to nourish the heart
and leads to a slowing of its beat. Arterial pressure was kept
below this limit for nearly one and a half minutes, and the pulse
rate fell from 117 to 110. In observation III, arterial pressure
varies between 38 and 148 millimetres of mercury (290 per cent.)
and the pulse rate between 108 and 112 per minute (3.6 per
cent.). In observation IV, arterial pressure varies between 43
aud 140 millimetres of mercury (225.5 per cent.) and the pulse
rate between 99 and 103 per minute (4 per cent.). In observation
V, arterial pressure varies between 40 and 111 millimetres of
mercury (177.5 per cent.) and the pulse rate between 100 and
104 per minute (4 per cent.). In observation VI, arterial pres-
sure varies between 17.5 and 114 millimetres of mercury (551.5
per cent.) and the pulse rate per minute between 100 and 104
(4 per cent.).

Experiment C.

October 26, 1881. Small dog, anaesthetised by chloroform daring
the operation of isolating the heart. Nutrient liquid 800 c. c. of de-
fibrinated dog's blood mixed with 800 c. c. of 0.5 per cent, solution of
pure sodium chloride in distilled water. Heart isolated and animal
placed in warm chest at 12 h. 50', P. M. When the series of obser-
vations detailed below was concluded the heart was still in good con-
dition and was used for two hours for other experiments.

Observation.

Time.

Temperature in
degrees C.

Arterial Pressure
in mm. of mercury.

Pulse Rate
per minute.

1 h. 23' JO"
" 20
" 30
41 40

37°

29
30
30
30

102
103
102
102

228

H. NEWELL MARTIN.

Experiment C. Observation L— Continued.

Observation.

Time.

Temperature C.

Arterial Pressure.

Pulse Rate.

1 h. 23' 50"

37°

103

1 h. 24' 00"

102

" 10

103

'• 20

102

11 30

102

" 40

103

" 50

101

1 h. 25" 00"

102

" 10

102

" 20

103.5

" 30

102

11 40

103

11 50

102

1 h. 26' 00"

105

" 10

104.5

11 20

No record.

" 30

105

11 40

105

" 50

105

II.

1 h. 33" 20"

37°

100

14 30

101

" 40

102

" 50

102

1 h. 34' 00"

102

" 10

102

" 20

102

11 30

101

11 40

101

" 50

101

1 h. 35" 00"

102

11 10

102

" 20

101

11 30

100.5

11 40

102

-< 50

102

1 h. 36" 00"

102

" 10

102

" 20

102

" 30

101.75

" 40

101

" 50

102

1 h. 37' 00"

102

" 10

103

" 20

103

ARTERIAL PRESSURE ON PULSE RATE. 229

Experiment C. Observation II. — Continued.

OUerration. j Time. Temperature C.j Arterial Pressure. I Pulae Rate.

II.

III.

lh.

37' 30"
40
50

38" 00"
10
20
30
40
50
00"
10
20
30
40
50

39'

1 h. 57" 30"
" 40
" 50

1 h. 58' 00"
" 10
" 20
41 30
" 40
11 50

1 h. 59" 00"
" 10
11 20
" 30
" 40
" 50

37'

105

102

101

102

103

102

101

100.5

102

100.5

24.5

29.5

14.5

24.5

37.5

96
94
97
95
96
96
99
95
96
96
96
96
98
99
96

IV.

2h.

02' 10"

" 20

" 30

" 40

" 50

' 2h.

03' 00"

" 10

" 20

" 30

" 40

1
i

" 50

2h.

04' 00"

" 10

51
54
64
76
87
94
89
56
30
37
54
70
81

100

100.5

102

103

105

102

105

108

104

a. NEWELL MARTIN.
Experiment C. ObsebTATION IV.— Continued.

•ervation.

Time.

Temperature C

Arterial Pressure.

PuUe IUte.

ir.

2 h. 04' 20"

31°

104

" 30

106

" 40

105

" 60

106

105

2 h, 05' 00"

105

" 10

105

" 20

105

'■ 30

104

" 40

105

" 50

108

2 h. 06' 00"

105

" 10

108

" 20

109

" 30

110

•' 40

108

" 50

108

2 h. 07' 00"

108

" 10

107

" 20

109

" 30

110

" 40

" 50

109

2 h. 08' 00"

No record.

So record.

" 10

109

" 20

109

" 30

109

2 h. If 20"

37°

105

" 30

108

" 40

105

" 50

105

2 h. 18' 00"

106

" 10

106

" 20

106

" 30

106.5

" 50

106.5

2h. 19' 00"

106.5

" 10

106

" 20

-!<";■,

" 30

108

" 40

106

" 50

107

2 h. 20' 00"

105

" 10

105

ARTERIAL PRESSURE ON PULSE RATE. 231

Experiment C. Observation V. — Continued.

Obsenration.

Time. Temperature C.

Arterial Pressure.

Pulse Rate.

2 h. 20' 20" 37°

105

" 30

105

" 40

105

" 50

105

2 h. 21' 00"

105

41 20

106.5

" 30

105

44 40

• 61

106.5

44 50

106.5

2 h. 22' 00"

106.5

No record.

" 20

No record.

" 30

106.5

41 40

106.5

44 50

106.5

2 h. 23' 00"

106.5

! " 20

106.5

14 30

•

106.5

In observation I of the above experiment, arterial pressure
varied between 26 and 80 millimetres of mercury (207 per cent.)
and the pulse between 101 and 105 per minute (4 per cent.). In
observation II arterial pressure varied between 24 and 103 milli-
metres of mercury (329 per cent.) and the pulse rate between 100
and 105 per minute (5 per cent.). In observation III, arterial
pressure varied from 12 to 38 millimetres of mercury (216.5 per
cent.) and the pulse rate from 94 to 99 per minute (5 per cent.).
In observation IV, arterial pressure varied between 18 and 106
millimetres of mercury (863 per cent.) and the pulse rate between
100 and 111 per minute (11 per cent.). In observation V, arterial
pressure varied between 20 and 78 millimetres of mercury (290
percent.) and the pulse rate between 105 and 108 per minute (less
than 3 per cent.).

A critical examination of the preceding tables will, I think,
shew conclusively that variations in arterial pressure within the
limits indicated in them have no influence on the pulse rate of the
13

232 H. NEWELL MARTIN.

isolated dog's heart. In the great majority of cases the variations
in the pulse rate fall clearly within the limits of error of the
experiment (2-3 per cent.), while arterial pressure is greatly varied.
Eliminating the obviously exceptional observations II, Expt. B,
and IV, Expt. C, the average variation of arterial pressure in an
observation was 204 per cent., and the average variation in the
pulse rate 3.3 per cent.

That the possible sources of error will readily account for the
pulse changes in most cases is clear — when it is remembered
(1) that a mistake of one-sixth of a beat in counting out the pulse
in any period of ten seconds appears in the tables as an error of
one beat per minute; (2) that the temperature of the air pumped
through the lungs and influencing the temperature of the blood
was often unavoidably altered during the course of an observation
as the doors of my present experiment room, which unfortunately
is somewhat of a thoroughfare, were opened by passers-by from
time to time. The latter influence is of great importance, as
experiments which I hope shortly to publish, have proved that
the dog's heart is, so far as its rhythm is concerned, extremely
sensitive to slight variations in temperature.

Whatever the cause of the slight pulse-rate changes observed
may be, it is at least clear that they are not dependent on varied
aortic pressure, for there is no possible relationship, direct or
inverse, to be detected between the two, when the whole series of
observations is examined. In most cases great variations of arte-
rial pressure are seen to occur without any change in the pulse
rate, and then, a little later in the same observation perhaps, the
pulse alters two or three beats a minute without any considerable
simultaneous change in arterial pressure.

If the relationship between pulse rate and arterial pressure were
invariable, even 3.3 per cent, of variation in the pulse per minute
might clearly be significant: but as there is no such constant rela-
tionship, and the known sources of error fully account for such
pulse-rate variations as were observed, they obviously mean nothing
in this connection : and we may safely conclude that within the
limits of aortic pressure indicated by pressures varying between 25
and 1J+0 millimetres of mercury in the subclavian, no change of
pressure has any direct action upon the rate of beat of the isolated
heart of the dog.

ARTERIAL PRESSURE ON PULSE RATE. 233

Before concluding it is my duty and pleasure to acknowledge
the willing and skilful assistance in the execution of my experi-
ments rendered to me by Mr. H. H. Donaldson and Mr. Mactier
Warfield, who not only undertook the tedious task of getting
ready the apparatus for each experiment, but gave me most im-
portant help in carrying it through.

THE INFLUENCE OP CHANGES OP ARTERIAL
PRESSURE UPON THE PULSE RATE, IN THE
PROG AND THE TERRAPIN. By WM. H. HOWELL,
A. B., and MACTIER WARPIELD, A. B. With Plate XVI.

At the request of Professor Martin we undertook some experi-
ments upon this subject, to see if the same results would be
obtained from these animals, as were obtained with the isolated
mammalian heart

We used substantially the same method as that employed by
Professor Martin in his experiments, described in the preceding
paper, keeping the venous pressure constant and varying only the
pressure in the outflow tube connected with the aorta, in a way to
be described presently.

As far as we have seen, no one has hitherto, in experiments
on these animals with regard to the effects of changes of blood
pressure, varied the arterial pressure alone.

Most of the work on the subject has been done with variations
of diastolic pressure. Luciani ! tried also the effects of variation
of systolic pressure. His method, however, did not furnish the
conditions which prevail in normal variations of arterial pressure.
He states that his apparatus was not suitable for studying the
effects of such changes, and does not give his results. With regard
to diastolic pressure, he says " that neither the frequency nor the
absolute height of the pulse was actually changed, when the pres-
sure (in the frog) was raised from 4 mm. to 13 mm. of mercury."

Tschiriew2 studied the effects of variations of both systolic and
diastolic pressure in the heart of the frog. He gets the same
result in both cases, viz: a quickening of the pulse rate with
increased pressure.

He does not describe his method of varying systolic pressure,
but it is evident that it was not the effects of varied arterial pres-

1 Luciani. Eine poriodische Function des isolirten Froschherzens. Lud-
wig's Arbeiten, 1872.

2 Tschiriew. Arch. f. (Anat. u.) Physiol., 1877.

235

236 WM. H. HOWELL AND MAC TIER WABFIELD.

sure alone that he got, since his arterial cannula was thrust beyond
the semilunar valves into the ventricular cavity and hence the
increased aortic pressure must have acted upon the ventricle during
its diastole as well as during its systole.

Ludwig and Luchsinger,1 in their experiments upon the entire
heart, appear to have varied venous pressure alone.

A cannula was put into the vena cava inferior, connected with
a pressure bottle, and the aortic arches cut through. Pressure was
varied by means of the pressure bottle. In this case pressure was
exerted upon the interior of the heart during both systole and
diastole, differing from true arterial pressure, which acts directly
upon the heart only during ventricular systole.

They found that increase of pressure caused an increase of pulse
rate.

It was the object of our experiments to leave the entire heart in
position in the body, cut off all external nervous influences, and
then, keeping up a constant venous pressure by means of a Mar-
riott's flask, to vary the arterial pressure alone.

Our method of operating with the terrapin, which we have used
in most cases, was to remove the plastron, slit open the peri-
cardium, bind the small ligament running from the ventricle to the
pericardium, the two superior cavae, the left hepatic vein, the pul-
monary artery, and put cannulas into the right and left aortas, (this
was done merely in case one should clot) ; one aorta, usually the
right, was connected with the manometer and outflow tube during
an observation, while the other was clamped. Finally a cannula was
put into the inferior cava and connected with the Marriott's flasks.
The animal's heart was washed free from all coagulable blood, the
vagi and sym pathetics cut, the latter below the middle cervical
ganglion, the head cut off, and the cervical spinal cord destroyed.
The heart was then allowed to run from half an hour to an hour
before any observations were made.

Essentially the same method was used with the frog; the arte-
rial cannula was put into one of the aortic arches before its external
division into three trunks.

1 Ludwig and Luchsinger. Zur Physiol ogie des Herzens. Pfluger's Archiv,
June, 1881.

ARTERIAL PRESSURE AND PULSE RATE. 237

We, at first, tried to feed the hearts with salt solution 0.6 per
cent, but found that the beat soon became weakened too much to
give a pulse in the manometer. Defibrinated calf's blood, filtered
through linen, and diluted with an equal bulk of 0.6 per cent, salt
solution was then tried ; it was found to work admirably. We
have kept the heart under experiment four or five hours, and it
was just as good at the end of that time as at the beginning ; it
was kept moist by lying in a small pool of the blood poured into
the visceral cavity of the animal.

Apprehending some trouble in the use of a mercury manometer
(which did not occur, however), we endeavored to make a water
manometer. We tried, at first, the one mentioned in the June
number of Pfltiger's Archiv, 1881, by Gruenhagen, but found that
it would not do for our purpose, since the paraffin stem floated so
little above the level of the water, that practically no variations
of pressure could be registered with it. We then, with the aid of
Dr. Sdwall, devised a water manometer which worked very satis-
factorily. The manometer we used (3/, PI. XVI) is made of
glass tubing having an internal diameter of 7 or 8 mm., the limb
in which the float works is about 40 cm. long, the other about
6 cm. The whole of the interior of the manometer is coated with
a thin layer of paraffin. For a float, S, we use a very light glass
stem, made by drawing out a thin test tube; this is also coated
with a layer of paraffin, and has a small bulb, 6, blown on the end
which is immersed in the water. A small cork float,/, well soaked
in paraffin, with a diameter a little less than the internal diameter
of the manometer, has a hole bored through its centre, and is then
slipped down the glass stem, so as just to touch the surface of the
water, when the stem is allowed to float freely in it. If the stem
sinks too low in the water, or is unsteady, one or more of these
little paraffined cork floats may be placed on that part immersed
in the water.

The top of the stem has a light glass pen fastened to it with
sealing *wax, and can be made to write upon a drum.

The manometer is provided with a glass cap, the opening
through which the stem works being well paraffined.

The stem in our manometer is about 38 cm. long, and sinks in
the water 17 cm., allowing us to register variations of pressure of
about 20 cm. of water. It is difficult to get a stem longer than
this that is not bent so much as to make it useless.

238 WM. K HOWELL AND MAG TIER WARFIELD.

The float follows very accurately every motion of the water,
and gives excellent tracings.

We used besides this a small mercury manometer having an
internal diameter of about 1.75 ram.

Plate XVI represents the apparatus used by us in our experi-
ments.

A and 2? are the Marriott's flasks, and are used alternately.
H is the heart, represented as separated from the body, though
such was not actually the case, a is a piece of stiff rubber tubing
leading from the aorta; at C there is a three-way tube, one branch
of which passes to the manometer, while the other (0) serves as
an outflow tube for the blood pumped out of the heart. By
raising or lowering this tube any desired arterial pressure can be
obtained. By means of a screw clamp on 0 we were also able to
change arterial pressure, to block the outflow entirely, or to alter
the height of the pulse wave. With very low arterial pressure,
for instance, it was very often found necessary to diminish con-
siderably the lumen of the outflow tube, in order to get a distinct
pulse wave in the manometer.

A pressure bottle, not represented in the drawing, was used to
fill the manometer and its connections.

Tracings were taken upon an ordinary revolving drum, upon
which wrote also a chronograph pen marking seconds.

In our later experiments before isolating the heart, we took
the blood pressure of the animal used, filling the cannula for this
purpose with 0.6 per cent, salt solution, or defibrinatcd calf's
blood. In the terrapin this pressure was taken in the left aorta,
in the frog in one of the aortic arches.

As the general result of our experiments, we can state that
variation of arterial pressure, up to the highest point of normal blood
pressure, has no direct effect whatever upon the pulse rate of the iso-
lated frog or terrapin heart.

In the terrapin we could carry the arterial pressure to more
than twice the normal blood pressure, without affecting the pulse
rate. Excessive pressure, however, caused in most cases a slight
slowing of the pulse, the slowing varying as a rule from 2.5 per
cent, to 9 per cent, of the normal pulse rate, in some cases more.

In the frog arterial pressure could not be carried much above
the normal without causing a slight slowing due to secondary

ARTERIAL PRESSURE AND PULSE RATE. 289

influences : very high aortic pressure may so distend the aorta as
to make the semilunar valves insufficient to close it : or may be
so great as to prevent the ventricle from carrying out a proper
contraction and maintaining the circulation. We are carrying out
further experiments with reference to these points; the latter of
which is probably the more important. With high pressure little,
and with the outflow tube completely blocked, no renewal of the
blood takes place in the heart, and Luciani found, that when the
serum in an excised frog's heart is renewed, the pulse becomes
more frequent.

The following tables give some of the results obtained. As a
general thing observations were made at intervals of five minutes,
of which two were taken up by the revolution of the drum ; the
pressure would then be raised ot* lowered, as the case might be, to
the next desired height, and the heart allowed to work at that
pressure for about three minutes, before another tracing was taken.
The pressure and rate of heart beat remained remarkably constant
for any one revolution of the drum. The tracings were divided
up into sections of twenty seconds each, and the average beat per
minute deduced from these.

Pressure was measured from a base line taken at the end of the
observation. In the tables " venous pressure " indicates the pres-
sure at which blood was supplied from the Marriott's flask to the
vena cava. The temperatures given are those of the room. The
blood supplied to the heart, and the animal experimented upon,
were always kept in the room a considerable time before com-
mencing an experiment. As will be seen, we could not always
keep the temperature constant during an experiment; and this
had sometimes a marked influence on the rate of beat of the heart.

240 WM. H. HO WELL AND MAC TIES WABFIELD.

P-l

T4f

-5,s

^November 30. Terrapin cnra-

4.20

215

35.6

rized. Head cut off at the

4.25

21.5

second or third cervical

4.30

21.5

vertebra. Vagi and sym-

4.35

21.5

pathetica cot, and cervical

spinal cord destroyed. Ve-

5.45

39.25

nous pressure = 2.1 cm.

5.50

31.5

Water manometer used.

5.55

11.5

38.5

6.00

31.5

88.5

6 35

6.40

21.5

6.45

6.50

6.55

7.00

7.25

7.30

7.85

36.33

7.40

7.45

7.50

8.05

86.4

8.30

8.40

8.50

ABTEBIAL PBESSUBE AND PULSE SATE. Ml

Table IL

g,E .

<SI

December 1. Terrapin cura-

4.05

36.1

rized. Head cat off at the

4.10

22.5

16.5

35.6

second or third cervical

4.15

22.5

35.4

vertebra. Vajri and sym-

4.30

35.5

pathetics cut, and cervical

spinal cord destroyed. Ve-

4.45

36.4

nous pressure = 4 cm. Mer-

4.50

37.5

cury manometer used.

4.55

37.5

5.00

37.7

6.20

40. 3

6.25

40.5

6.30

40.5

6.35

40.3

6.40

39.3

6.45

7.00

22.5

37.8

7.05

22.5

37.5

7.10

22.5

37.5

7.15

22.5

7.20

22.5

15.5

37.5

7.25

22.5

37.5

Table III.

|ll

December 8. Large Frog.

1.55

Brain and spinal cord de-

2.00

49.25

stroyed. Both water and

2.05

mercury manometer used.

2.10

33.5

Venous pressure, during the

first part of the experiment,

2.24

49.25

= 3 cm.

2.29

48.75

2.35

48.5

2.40

48.8

2.50

48.25

WM. K HOWELL AND MAC TIES WABFIELD.

Table III. — Continued.

Tenons pressure, mercury ma-
nometer used, =■ 1.5 cm.
In this case it was noticed
that with pressure above 29
mm. of mercury the ventricle
was never emptied, indicat-
ing a partial giving away of
the semilunar valves, or that
the tension in the aorta was
too great for the ventricle
to overcome.

S
fe

3.55
4.00
4.10
4.15

4 30
4.35
4.40
4.45
4.50

5.00
5.05
5.10
5.15
5.20

P °

21
21
21
21

20.5
20.5
20.5
20.5
20.5

00 3

« s

O C 3
> .5 O

2.5
31.5

25
28

3.5
10.5
19
26
33.5

5
13
23

28.5
33

«J8

42
38.5
39.5
39.15

40.25

40.5

39.25

40.5

41.25

40.5

37.75

Table IV.

December 10. Terrapin.
Blood pressure in left aorta
before commencing the ob-
servation = 20 mm. Hg.
Vagi and sympathetics cut,
head cut off and cervical
spinal cord destroyed. Ve-
nous pressure — 2.6 era.
Mercury manometer used.

feflS

> ft,

31.25

32 85

33.18

33.45

32.4

31.86

32.81

33.45

34.5

33.72

33.75

Heme iapabe rate that took
pbcc toward* tke«ad otitis
aeriea k - to

the me of taaao^raiare in
lie room. It u dearly in-
dependent of the pressor*

Id this case an aortic pressure
nearly hreetimesthat found
before the observations com-
menced (5" to 20) was ob-
tained by completely block-
ing the outflow tube — the
heart pumping into the ma-
nometer only. The heart wag
kept in this condition about
ten minutes— from 603 to
6.13. The very abnormal
pressure slowed tbc heart
and the slowing effect re- 1
mained some time after the
heart was relieved.

Water manometer put on.
Same venous pressure.

4.45

4.5

sies

4.60

21.5

13.7

455

21.5

355

3*5

5.00

31.5

33.1

34.5

5.05

34.5

5.10

34.5

5.20

235

34.5

5.25

225

ST.l

34.7

5.30

22 5

34.$

5.35

32.5

35.3

5.40

35.4

5.45

5 55

6.00

32.5

6.05

S3. ST

6.10

53.75

39 63

6.15

23.25

34.69

6.20

35.81

£.£*

6.50

22.5

6.53

22.5

42.35

7-00

22.5

S« WM. B. BOWELL ASD MACTIEB WASFIELD.

2£

Is.

6-3

December 13. Terrapin.

7.30

43.1

Blood pressure taken from

7.35

9.5

43.8

left aorta before isolating

7.40

24.5

205

43.5

heart— IS mm. "Vagi and

7.45

24.5

27.5

42.1

sympathetica then cnt, head

7.50

44.25

cnt off and cervical spinal

7.55

24.5

38.5

cord destroyed. Venous

pressure = 4.5 cm. Mercury

manometer used. The rise

of temperature at the end of

the series caused a slight

quickening of the pnlae.

M
ti

= 0

get.

I2'
a e

8>a

§"1

p c 0

b s.

E-i

&"*

December 14. Frog. Brain

2.00

4.75

55.5

and spinal cord destroyed.

2.05

55.5

Venous pressure = 4.5 cm.

2.10

55.5

Mercury manometer used.

2.15

'J 6. 5

55.5

2 20

2.25

52.6

3.10

23.5

67.7

3.15

23.5

57.7

3.20

28.5

25.25

3.25

23.5

56.4

3.30

23.5

35.5

55.6

3.35

23.5

65.5

3.40

23.5

56.4

ARTERIAL PRESSURE AND PULSE RATE. 245

Table VI.— Continued.

3.45
3.50
3.55

4.10
4.15
4.20
4.25
4.30
4.35
4.40
4.45
4.50
4.55

c ©

23.5
23.5
23.5

23.5
23.5
23.5
23.5
23.5
23.5
23.5
23.5
23.5
23.5

28.5
19
4.9

3.75
14
27
33.6
37

35.25
30.2
23
14

W)g

£ »-

► a.

57.6

56.7

56.4

55.6

Our thanks are due to Professor Martin, for advice and sug-
gestions during the course of the work, which we think shews
conclusively that within wide limits variations in aortic pressure
do not in the least influence the rate of beat in the heart in the
animals experimented upon.

SOME NOTES ON THE DEVELOPMENT OF
ARBACIA PUNCTULATA, Lam. By H. GARMAN
and B. P. COLTON. With Plates XVII and XVIII.

•

It was the privilege of the writers to spend some time last sea-
son at the marine laboratory of Johns Hopkins University at
Beaufort, N. G, and while there to make some observations on
the development of Arbacia punctulata which seem of sufficient
interest to warrant publication. The development, from the first
changes after fertilization of the ova to the formation of the young
sea-urchin and the resorption of the pluteus, was under constant
observation. Materials were thus accumulated for a complete
history of the development, as far as external changes are con-
cerned, but since the earlier stages do not differ essentially from
those of other Echini, and have been fairly well figured and
described by Dr. J. W. Fewkes,1 we shall not at present give
more than a few notes on some of the later stages of the pluteus
and on the young sea-urchins, thus supplementing, in some meas-
ure, the work already done. Our thanks are due to Dr. Brooks,
director of the laboratory, for facilities afforded us iu pursuing
the work, and for other assistance.

Arbacia punctulata appears to be the commonest sea-urchin at
Beaufort. Great numbers of them were brought up in the trawl
from the deeper water of Bogue Sound opposite Morehead City.
They were also taken in some numbers about the piers of wharves
at low tide. Strong ylocentrolus dr'dbachiensis was represented by
frequent examples of the form with white spines. Mellita tentu-
dituita was the only other echinoid at all common. The handsome

1 Mem. Peab. Acad, of Science, Vol. I, No. VI, 1881. In this memoir
Dp. Fewkes figured and described most of changes in the developing Arbacia
pluteus, but did not follow the development to the appearance of the young sea-
urchin, of which we were fortunate enough to rear a number of specimens and
would doubtless have obtained more could we have stayed longer at Beaufort.
The Figure '20 of Dr. Fewkes' plate is unlike any Arbacia pluteus we have seen.
"While our plutei varied within certain limits and were sometimes deformed, in
the many specimens examined we saw none that had more than two pairs of
arms on the oral lobe where Dr. Fewkes represents three.

15 247

248 H. O AMMAN AND B. P. COL TON.

bleached shell of this sand dollar was a common object on the
shoals. A single example of a fourth species was taken by Mr.
Rice.

The eggs of Arbacia were readily fertilized artificially. The
ovaries and testicles with ripe contents were taken from the living
animals, placed in a watch-glass containing sea- water and cut into
bits with a pair of scissors. The watch-glass was then emptied
into a beaker full of sea-water and the contents of the latter
gently stirred with a glass rod. Here they remained until the
pluteus emerged, an event which took place about six hours after
the fertilization of the ova. Portions of the water containing
plutei were then poured into a number of beakers of fresh
sea-water, leaving the undeveloped eggs and remnants of the ova-
ries and testes in the bottom of the first vessel. By this means
the plutei were given more room and materials likely to render
the water impure were got rid of. Afterwards, as the plutei grew,
individuals were dipped up from time to time with tubes and
transferred to separate glasses. It was found best to filter the
water used, thus removing creatures likely to prey upon the young
plutei. The vessels were usually kept covered to prevent dust
and insects from falling upon the water.

Ova and spermatozoa could be obtained from Arbacia at any
time while we were at Beaufort (from the middle of July till the
latter part of September), but after the first of September difficulty
was experienced in fertilizing the eggs. Many lots were tried, but
in most cases after one or two divisions of the egg-contents, the
development became abnormal and soon ceased altogether. The
spawning period seemed, to have passed, and the reproductive
organs, though still with apparently ripe contents, were much less
distended than earlier in the season.

The development of the egg takes place, under favorable cir-
cumstances, with great rapidity. In one instance the first segmen-
tation was noticed just twenty-five minutes after fertilization, and
at times divisions of the egg-contents took place within fifteen
minutes of each other. In another lot the first division was not
noted until an hour and a half after the eggs were fertilized, and
the periods between divisions varied to a similar extent. The eggs
from which the plutei were obtained upon which most of our work
was done, were fertilized on the 16th of August, at 9 o'clock,
A. M., and at 3.40, P. M., of the same day the first plutei had

ARBACIA PUNCTULATA. 249

emerged. The plutei obtained from these* eggs were under obser-
vation till the 22d of September, at which time a number of young
sea-urchins had emerged.

In these notes we follow the majority of authors in calling that
surface of the body on which the vent opens, ventral, and the
opposite one dorsal. The anterior part of the body is that in
which the mouth opens. The right and left sides will then be
those seen to right and left respectively when the pluteus lies with
the dorsal surface up and the mouth-lobe from the observer. This
explanation seems necessary to prevent misapprehension, as Dr.
Fewkes calls the surface in which the vent opens, dorsaJ.

The living plutei usually remained near the surface of the
water, the constantly moving cilia apparently serving chiefly as
means of maintaining the equilibrium. The anterior part of the
body is almost invariably carried uppermost, as one would expect
from the form of the body, the long, slender arms projecting
upwards, while the posterior portion being more compact, tends to
sink lowest. The movements of the perfect Arbacia pluteus are
not rapid. It is not built for speed, and when it moves forwards,
does so with the widest end foremost. But while the most
rapid movement cannot be effected in this way, the main object
of movement is attained most admirably. The large mouth and
movable lip are held in a position to collect the particles of food
brought into contact with the broad, concave front, and the animal
thus travels as a sort of self-acting surface net. The younger plutei
move more actively, though less steadily. The peculiar form causes
them to turn as they move, first on one then on the other side.

The matured pluteus is semi-transparent and is marked with red
pigment spots. This pigment is very abundant towards the tips
of the arms. It is also disposed in spots along the calcareous
rods of the arms, and occurs as patches and dots on other parts of
the body. The oral lobe bears two pairs of arms. The longer
pair (Fig. 1, 2) arises from the dorsal surface at the sides of the
lobe and extends forwards. The arms of the shorter pair (Fig.
1, 5) project obliquely downwards and forwards from the sides of
the anterior border of the lip. The lip (Fig. 1, b) is a large, freely
movable flap which partly covers the mouth in front, and at times
entirely closes it. The mouth (Fig 1, a) opens beneath the lip, is
very large, and gapes wide open. It opens into a large muscular
oesophagus (Fig. 1, c), by the peristaltic contractions of which food

250 H. O ARM AN AND B. P. COL TON.

is carried to the stomach (Fig. 1, d). Previous to the formation
of the young sea-urchin, the stomach occupies the greater part of
the body-cavity. The opening from the oesophagus into it usually
remains closed while that from the stomach into the intestine
stands open. The intestine (Fig. 1, e) arises from the posterior
under part of the stomach, extends forwards, and opens on the
ventral surface a short distance behind the anterior border of what
may be called the ventral lobe. Two pairs of very long arms
arise, one (Fig. 1, 1) from the anterior lateral angles of the ventral
lobe, the other (Fig. 1, 4) from the sides of the oral lobe about
opposite the first pair. Both pairs project forwards, the dorsal
obliquely upwards and the ventral obliquely downwards. The
remaining pair of arms (Fig. 1, 3) arises from what is termed the
anal lobe, and the arms project obliquely outwards and slightly
backwards. There is a thickening of the central portion of the
anterior border of the ventral lobe which continues backward on
each side and forms the margin of a pair of ciliated "epaulets"
(Fig. 1, i). A similar pair of epaulets (Fig. 29J) occurs on the
dorsal surface. They are supported anteriorly by prongs of
the skeleton. The size of the plutei and the relative length
of the arms is subject to considerable variation. The variation
in the arms does not appear to be due so much to resorption as
to a symmetrical development. The resorption of the pluteus
takes place chiefly, as will be seen later, within a short space of
time after the sea-urchin is protruded. This asymmetry may have
been due to the unnatural conditions in which the plntei were
living, and probably in nature little of such variation occurs.

Deformities were frequent; one pluteus observed was club-
c shaped, consisting of little else than a stomach and single arm.

The calcareous skeleton prevents any range of movement of the
arms, but in the later stages of development a frequent drawing
apart and closing of the long lateral arms may be observed, and
in some instances the posterior pair was seen to move back and
forth. The union of the rods of the arms of this pair prevents
independent movement, so that when one moves forwards the
other moves to the rear. The arms appear in the following order:
First, the arms of the ventral lobe (Fig. 1,1); second, the longer
pair of arms at the extremity of the oral lobe (Fig. 1, 2); third,
the arms of the anal lobe (Fig. 1, 3); fourth, the pair from the
sides of the oral lobe (Fig. 1,4); aud fifth, and last, the small

ABBA CIA PUNCTULATA. 251

pair on the anterior border of the lip (Fig. 1, 5). The arras, lip,
and membranous folds are supplied with cilia. The separate parts
of the skeleton appear as minute spicules. The first spicules to
appear are four- rayed. They develop near the ventral surface just
behind the anterior border. The lowest ray develops rapidly and
pushes down into the anal lobe, where it unites a little later with
a corresponding ray from the spicule of the opposite side. The
lateral ray pushes across the body near the ventral surface and
finally unites at the middle line with its fellow from the opposite
side. A third ray grows towards the dorsal surface, and curving
forwards, grows into the second pair of arms as they develop.
The fourth ray supports the long first pair of arms.

A crescent-shaped spicule next appears at the posterior end of
the stomach near the points of origin of the third pair of arms,
and sends a branch from near its extremities into each arm. The
third spicules are triradiate and appear with the fourth pair of
arms. One ray of each spicule supports an arm, and the remain-
ing rays project, one obliquely forwards and the other obliquely
backwards.

The only other spicule appears just above the oesophagus in the
middle line. The two lateral rays curve outwards and forwards
and support the fifth pair of arms. A small prong develops from
each of these lateral rays for the support of the anterior projection
of the dorsal fold (Fig. 2, k). The third ray does not develop.
The transverse bar formed by the union of rays from the first
spicules is, later in development, broken by the resorption of its
material at the point of union. The long rods of the first pair of
arms also unite at an early stage within the anal lobe, to be again
broken towards the close of pluteus life by resorption. Still
another change occurs. The rods which support the second pair
of arms originally form a part of the first spicules, but become
freed later by resorption. The meaning of these changes becomes
clear when we consider the sudden metamorphosis which closes the
pluteus stage. During the earlier periods of its existence there is
need of a strong support for the fragile body, but as the last change
approaches, greater freedom of movement is called for and can be
secured only by breaking the unions of the rods. The larger rods
of the skeleton are perforated by series of oval or round openings
and are usually more or less spinose. Figure 3 shows the skeleton
at a stage represented by Figure 1.

252 H. O AMMAN AND B. P. COL TON.

The rods are numbered to correspond with the arms they sup-
port. The prong for the support of the dorsal fold has not yet
appeared.

The young sea-urchins were first noted when the plutcus was
about two weeks old. At this stage the larvae rest on the bottom,
swimming but little. On the left side of the stomach the tube-feet
appear (Fig. 2, h). There are five of them in a circle extending
outwards from the abactinal disc which rests upon the side of the
stomach, their free ends approaching each other and forming a cone.
In a dorsal or ventral view usually only two or three can be seen,
but towards the left side the five are shown presenting a radiate
appearance. Over the apex of the cone, i. e. over the approxi-
mated free ends of the tube-feet is the opening to the exterior.
Already the growth of the young sea-urchins presses on the stom-
ach, flattening it, and later, pushes it towards the opposite side so
that it occupies the smaller part of the width of the body. The
tube-feet early show contractility. Around the circle of tube-feet
is a circle of flattish lobes, the beginning of the first developed
marginal spines. As the tube-feet develop, they are from time to
time protruded slightly through the opening, being thrust out far-
ther and remaining out longer as growth proceeds. By the
turning of the lateral arms outwards and backwards the opening
is enlarged and the feet pushed out. At first the feet only, later
the spines at their basis appear outside. Figures 4 to 7 show
the young sea-urchins in various degrees of exertion as seen from
different points. In the complete act of everting the lateral arms
turn back passing astride the apical pair swinging outwards like
the ribs of an umbrella turning inside out. At the same time the
oral lobe is drawn to one side. Then the arms return to their
former position, the feet are withdrawn, the opening is almost en-
tirely closed and the appearance is again as in Figure 1. Speci-
mens were observed to repeat this process about once an hour for
hours in succession, remaining everted a quarter or half an hour.
When extended the feet are moved about and by applying the discs
to the surface on which the pluteus rests, effect a slight degree of
locomotion ; but the movements are awkward, the long projecting
arms making the pluteus top-heavy. This process is kept up for
days and in some cases for weeks, the eversions becoming more and
more complete until finally the everted state becomes the perma-
nent one, the three principal pairs of arms extending from what

ABB A CIA PUNGTULATA. 253

was the apical end of the body, and the oral lobe is drawn out of
its original position as in Figure 7. Soon the rods pierce through
the ends of the arms and the softer tissues of the arras slide down
the rods and are withdrawn into the body. The bare rods are left
projecting. In endeavoring to crawl, the newly emerged sea-
urchin frequently topples over and some of the rods are broken off;
all of them soon disappear. The oral lobe also soon disappears.
The complete absorption of the arms and oral lobe occurs within a
very short time.

The young sea-urchin is semi-transparent and marked with pig-
ment spots similar to those of the pluteus. They are especially
abundant on the peristome. The shell is strong. About the
margin of the corona is a series of fifteen spatulate spines. From
above the five large tube-feet may be seen separating the spines
into sets of threes. On the actinal surface about the mouth many
small tube-feet are placed. The developing Aristotle's lantern
with its muscles and ligaments can be made out. The structure is
apparantly the same as in the adult. The plates of the periproct
are now relatively very large. A little later the five large tube-
feet are reduced in size, and above them appear five additional
spines. They eventually become longer than the original fifteen,
are narrower and more pointed. The radioles are articulated to the
corona as in the adult, and are so attached as to slope downwards.
At first they are serrate distally, but the serratures soon disappear.
The large tube-feet have a perforated calcareous plate in the disc,
and in the smaller feet the forming plate appears as minute
spicules. Large pedicellarise soon appear on the abactinal sur-
face, and the corona becomes studded with scattered tubercles.
Figures 8 and 9 represent the young sea-urchin soon after the
resorption of the pluteus.

The young sea-urchin aids its movements to a considerable
extent by pushing downward and laterally with its flattish spines,
thus early showing the use of the spines in locomotion which is
characteristic of the adult. The spines can be bent down at right
angles to the plane of the body, but can be bent upward but very
little, owing to the projecting rim at the base of the spine above
the obliquely inserted pedicel. At this stage the young sea-urchins
were frequently observed climbing up the sides of the glass vessels
in which they were kept.

254 H. O AMMAN AND B. P. COL TON.

EXPLANATION OF THE FIGURES.

PLATE XVII.

Figure 1. — Ventral surface of the Arbacia platens, a, mouth ; b, lip;

c, oesophagus; d, stomach; e, intestine; f, anns; g,
opening of stomach into the intestine; h, first appear-
ance of spines and tube-feet of the young sea-urchin ;
t, the ventral epaulets. The arms, 1-8, are numbered
in the order of their development.

Figure 2. — Dorsal view of pluteus at a little later stage of develop-
» ment than the preceding. Signification of letters and
figures the same, j, dorsal epaulets ; k, spine-like rod
supporting the anterior part of the epaulets.

Figure 3. — Dorsal view of the calcareous skeleton. 1, rods supporting

the first pair of arms; la, prong which at an earlier
stage was united with the corresponding one of the
opposite side ; 2, rods supporting second pair of arms
and at one time united with 1 ; 3, support of the third
pair of arms ; 4, supports of fourth pair ; 5, spicule
supporting the fifth pair of arms and sending up
branches for the dorsal epaulets.

Figure 4. — Appearance of the pluteus when almost ready to transform.

On the left side (right in the figure) is seen the invagi-
nation through which part of the tube-feet and spines
of the sea-urchin appear. The stomach is pushed to
the right.

PLATE XVIII.

Figure 5. — The pluteus with the tube-feet and part of the spines pro-
truded, as seen obliquely from the front. The lateral
arras are turned partly backwards and appear on each
side of the third pair. The oral lobe with the first and
fifth pairs of arms (I and 5) are drawn to one side.

Figure 6. — Side view, showing the feet and spines still more protruded,

with the lateral arms turned completely back, having
passed astride the third pair.

Figure 7. — Transformation nearly complete. The pluteus undergoing1

resorption. I, the remains of the oral lobe.

ABB AC I A PUNCTULATA. 255

'igure 8. — Ventral view of the young sea-urchin, as seen on the

twenty-fourth day after the fertilization of the ova.
a, the five large tube-feet ; b, the smaller, later-formed
tube-feet; c, marginal spines of the corona.

Fiqvrx 9. — Abactinal surface of an older individual, showing the five

additional radioles and the first pedicellariae.

Figures 2, 5, 6, 7 and 8 were drawn by Mr. Colton. Figures 1, 8,
4 and 9 by Mr. Garman. With tt)e exception of Figures 3 and 9 all
drawn from the living plutei.

ON THE STRUCTURE AND SIGNIFICANCE OP
SOME ABERRANT FORMS OF LAMELLI-
BRANCHIATE GILLS. By K. MITSUKURI, Ph. B,
of Tokio% Japan, Fellow of the Johns Hopkins University, Bal-
timore. With Plate XIX.

. The following contribution to the morphology of the Molluscan
branchiae is part of an investigation on which I have for some time
past been engaged, under the direction of Dr. W. K. Brooks, in
Professor Martin's laboratory at the Johns Hopkins University.
The gills, of which the description is here given, are those of
Nucula proxima and Yoldia limatula. They are extremely inter-
esting because of their simple structure, and this account of their
minute structure is published with the hope that it might throw
some additional light on the nature of Lamellibranchiate gills. I
wish to express here my sincere thanks to Dr. Brooks for his con-
stant advice and assistance. I am also deeply indebted for
specimens used in the investigation to Professors A. E. Verrill and
S. J. Smith, of Yale College, and to Mr. Richard Rathbone, of the
United States Fish Commission.

Nucula proxima, Say.

This Lamellibranch shows many departures from the structure
which is generally regarded as characteristic of the class. Figure
1 gives a fair idea of what is seen when the left valve of the shell
has been taken away, and the mantle of the same side removed
along the lower border of the visceral mass near the line x y. a. a.
is the anterior adductor muscle made up of several fasciculi; p. a.
is the posterior adductor. It will be noticed that Nucula possesses
one of the few shells in which the umbo is turned toward the pos-
terior end. In the specimen figured, the visceral mass (v. m.) shows
convolutions on the surface, which, under the microscope, proved
to be the male reproductive organ, probably enormously developed
for the breeding season, and this character enables one to distinguish

257

258 K MITSUKURl

the sex of a specimen without difficulty. All the males have these
convolutions, and, when preserved in alcohol, are of a greyish color.
The females show hardly any convolutions, and are much more
darkly colored. The foot (/) is folded longitudinally at its end,
and can accordingly be spread out into a flat circular disc. The
labial palpi (/) are unusually developed, and might at first sight be
taken for gills. The inside of the outer and the outside of the
inner palpus are raised into numerous parallel ridges, which,
as shown in the figure, can be seen from the outside, and do not
extend to the lower margin. At their posterior end there are two
remarkable structures. One of them is a hood-like structure (L 6.,
Figs. 1 and 2), which is the posterior prolongation of the united
upper edges of the inner and outer palpi. The other (Z. a., Figs. 1
and 2), lying immediately below the first, is a long* tentacular
appendage. It is a hollow tube, open, however, along a line on its
posterior aspect, and having its cavity continuous with the space
between the two palpi. As it has been seen protruded, with the
foot outside of the shell (Woodward's " Manual of Moll u sea," p.
426), and since, in alcoholic specimens, a great deal of dirt and sand
is found along its length and between the palpi from its base to the
mouth, it is no doubt a food-procuring organ, probably sending a
constant stream of nutritive matters to the mouth by means of its
cilia. It is interesting to notice in connection with this appendage
that in Nucula, the gills, unlike those of ordinary Lamellibranchs,
must be practically useless for obtaining food, as will be evident
from the following description of them.

The gill (g.y Figs. 1 and 3) is comparatively small. It is
situated quite posteriorly, and is suspended by a membrane (m.9
Figs. 1 and 3), which is attached to the body along the broken
line xy zw. It is united to the visceral mass (v. m.) from x to y,
and to the upper part of the foot (/, Fig. 3) from y to z (see
Figs. 1 and 3). At the last point (having come to the median
line of the body) it joins with its fellow of the opposite side, and
they continue in this way as far as w. Here they separate again,
each proceeding to the posterior tip (p) of the gill of its own side.
It should be remarked that, as the point x is further from the
median line of the body than the point y (Fig. 3), there is a con-
siderable free space beneath the suspending membrane of the gill.

When we turn to the gill itself, we find an altogether unusual
structure. Figure 4 shows it dissected out and seen from below

LAMELLIBRANCEIATE GILLS. 259

and slightly from one aide. In general appearance it resembles a
boat which is suspended by its keel, xcp, Figure 4 (seen in cross
section at ij, Fig. 5), is the line of attachment and corresponds to
the keel ; x dp, Figure 4 (seen in cross-section at d, Fig. 5), rep-
resents the bottom line of the hollow of the boat. The latter is
bounded by the two surfaces x a p d and x b p d (Fig. 4 ; seen in
cross-section at b d and a d} Fig. 5). The anterior end (x, Fig. 4)
is rather blunt, while the posterior end p} Figs. 1, 3, and 4)
is quite pointed. The resemblance of the gill to a boat is, however,
only very superficial, as the gill is not one solid mass, but is made
up of a series of paired plates of a peculiar shape, placed one after
another from the anterior to the posterior end. A little dissection
under a lens will show that the part above the line x dp (Fig. 4)
and below the line of suspension (x c p), is continuous along the
entire length of the gill, and that, with this part for the stem, the
plates are given off, one after another, in pairs to the two sides
(see Fig. 5). The plates constitute the proper respiratory parts of
the organ. They are largest in the middle, and diminish in size
toward the two extremities.

It is evident from this description that the gill in Nucula is of
quite an exceptional nature. It does not, as in most Lamelli-
branchs, extend along the whole length of the side of the body,
constituting the most conspicuous object of the mantle cavity, but
is comparatively insignificant, being pushed back and freely sus-
pended in the mantle cavity. It cannot, therefore, divide the
latter into the suprabranchial and infrabranchial chambers, and is,
of course, utterly devoid of any structure like the ciliated water-
passages in the ordinary gill, for driving water from the lower to
the upper. It cannot, also, as has been said, serve as an effective
food-procuring organ. The gill in Nucula must for these reasons
be of vastly less functional importance to the animal than it is in
common Lamellibranchs, and, so far as I am able to see, serves
only as the organ of respiration. It seems to me, however, that
the division of the mantle cavity into the upper and lower cham-
bers is begun in the posterior part. It has been seen that ventral
to the membrane suspending the gill (m, Figs. 1 and 3) there is a
large space continuous with the general branchial cavity, and there
certainly is a space dorsal to this membrane. These spaces seem
to be the rudiments of the supra- and infrabranchial chambers.
Moreover, the arrangement of the different parts at the posterior

260 K MITSVKUBL

end, as seen in Figure 3, recalls that of the corresponding parts in
many of those genera in which the mantle cavity is divided into
two parts. It is not difficult to conceive how the same division
might be brought about in the case of Nucula, by proper develop-
ment of the gill and the membrane.

Figure 5 shows a pair of opposed plates considerably enlarged.
The solid part (i dj) which I have called the stem, and which is
continuous throughout the whole length of the gill, together with
the suspending membrane (k ij l) is seen in cross-section in the
middle, and from this middle portion the paired plates (e. e.) are
seen to proceed. The colored part at the bottom represents the
complex chitinous framework. The membrane {k ij I) is made up
of fibrous tissue, the bundles of which this is composed crossing
each other in many directions. Its free surfaces are covered .with
columnar epithelium. The stem consists mostly of a solid mass
of large irregular cells with rather large nuclei. There are, I am
almost certain, two blood-channels excavated through it; a lower
larger (n), and an upper smaller (o). The latter seems to be in
connection with a free space (q.) found often in sections of the sus-
pending membrane. The large channel (n) sends a branch (r) into
each plate. The fibrous tissue found in the upper membrane dips
down into this part at regular intervals, viz: between every branch
(r) of the lower blood-channel (n). How these fibres end below,
when they reach the chitinous framework, I have not been able to
make out. A few fibres (it) are sent down into the plate a little
above the blood-channel (r), and gradually approach and finally
touch the latter near its lower end. A few more fibres {t) are seen
along the upper edge of the plate. Exactly what this fibrous tissue
is I am unable to make out, but it seems to be some sort of tough
connective tissue, with perhaps muscular fibres more or less inter-
mixed. That it is very tough and serves as a support to the whole
structure is seen by the fact that the fibres often stick out beyond
the broken edge of the soft tissue. The trough of the chitinous
framework is seen at s, in cross-section. It extends along the
whole length of the gill and sends out two branches into each
plate. I have obtained the appearances, in some sections, of a
bundle of fibrous tissue running in it and filling it. The frame-
work will be described more fully further on. The plates (e), the
proper respiratory organs, are comparatively speaking very broad
and quite thin, and hang down from the solid part of the gilk

LAMELLIBRANOHIATE GILLS. 261

The epithelium of the plates which is represented in the figure as
ending abruptly at the edges i d and j d, turns at a right angle at
these lines to cover the stem, and is soon reflected outwards again
to form the epithelium of the next plate in the series, This is
evident from an inspection of Figure 8. Each plate may be said
to be simply an enormously widened blood-channel (Fig. 6), and
as the blood is necessarily spread out in a thin layer over a large
area, the purposes of aeration must be admirably served. The
columnar epithelial cells seen at a d, Figure 5, are very character-
istic of the plates under a microscope, and are the cells (d a, Fig.
6) around the chitinous bars (A, Figs. 5 and 6) seen in optical sec-
tion. The surface of the irregularly rectangular cells placed just
inside these columnar cells in Figure 5, ought therefore to be conr
tinuous with the outer edge of the columnar cells, but in order to
avoid confusion is not so represented in the figure. This is also
the case with the cubical cells along the upper edge. The chiti-
nous support (A, Figs. 5 and 6) of the plate ruus near the lower
edge (Fig. 5) to its tip (a or 6, Fig. 5), and is made, up of two
entirely separate parts (seen in cross-seetion in Fig. 6) applied
closely together. Owing to the shape of these parts there is, how-
ever, a narrow oval space between them. This space, as will be
shown further on, is continuous with the space in the trough
(*, Fig. 5) of the stem. The cells along the lower edge of the
plate are columnar, and surround the chitinous support in a char-
acteristic manner shown in Figure 6. Their surface outlines are
irregularly rectangular, contrasting with the irregularly polygonal
cells covering the rest of the plate. The branch (r, Figs. 5 and 6)
of the lower blood-channel (n) in the stem, is seen to be circular in
cross-section and to bulge out the surface of the plate. These
points are not, however, constant, as the vessel is sometimes con-
stricted into more or less separate channels, while the amount of
bulging seems to depend on the quantity of blood present. The
remaining part of the plate (i, Figs. 5 and 6) is flat and quite
thin, enclosing a broad blood-channel between its two epithelial
surfaces. It is here no doubt that the aeration of blood is accom-
plished. The cells of this part are cubical, as seen in Figure 6.
Some of them send processes inward to join others from the oppo-
site side. This gives a labyrinthine appearance to this part of the
plate. The course of the blood is evidently from one blood-
channel in the stem to the other, through the space in the plate.

262 K. MITSVKURL

For instance, the blood may start from the upper channel (o) in
the stem, go to the broad flat part (e, Fig. 5) of the plate where it
gets aerated, then enter the branch (r), along its upper edge, and
run up this to reach the lower blood-channel (v e) in the stem.
This is, however, a purely hypothetical course. I have had no
means of determining whether the blood goes from the upper to
the lower channel or vice vtrsd.

The framework which supports the gill can be separated out by
heating it in dilute caustic potash, as it is insoluble in weak acids
and alkalies. It is stained by carmine and other coloring reagents.
Whether it is really formed of chitin I do not know, but as pre-
vious writers have described the substance as of that nature it will
be convenient to use the term "chitinous support1' for the present.
The framework consists of a trough (seen in cross- section at «,
Fig. 5; longitudinally from below in Fig. 8; diagramatically
represented in Fig. 7) which runs along the whole length of the
gill, and from which a pair of closely-applied parallel branches
(A, Figs. 5, 6, 7 and 8) is given off into each plate. The trough
is divided into two unequal parts : an upper larger and a lower
smaller, by a cross piece (c p, Figs. 5 and 7), which stretches from
one side of it to the other, a little below the middle. This cross
piece is not, however, continuous, but is pierced through by oral
openings (o vf Figs. 7 and 8) whenever branches are given off
laterally to the plates. The space enclosed between each pair of
closely-applied branches (see A, Figs. 6, 7 and 8) is connected with
the lower compartment of the trough by means of somewhat
circular openings (o p, and o' p', Figs. 7 and 8) found near the
bottom. In Figure 8 the letters a, a, a, are placed opposite each
pair of the branches that go into a plate. It will be seen how
one-half of the chitinous support of one plate, after forming an
arch at the trough, turns round to enter the next plate in succes-
sion, and to constitute there one-half of the support of that plate.
The framework treated with potash, and sometimes without any
treatment, shows marked longitudinal striation (Fig. 8), and some
of its fibres sticking out at the broken edge beyond the others
resemble in appearance the fibres found in the suspending mem-
brane, at t and u} Figure 5, and give reasons' for thinking that the
whole chitinous framework is nothing but the fibrous tissue found
in other parts cemented closely together and forming one cohering
mass.

LAMELLIBRANCEIATE GILLS. 263

Although, owing to the state of the specimens, I have obtained
only here and there evidences of cilia, it seems reasonable to sup-
pose that the whole gill is covered with cilia. On two rows of
cells (Lf., Fig. 6 ; d. a., Fig. 5) on the lower edge of the plate I
believe there are larger cilia than on the rest, as I have now and
then seen their remains, and as, without any question, cells in the
corresponding positions in Yoldia have long and conspicuous cilia.

Yoldia limatula, Say.

Yoldia resembles Nucula in several structural peculiarities — in
its well-developed labial palpi, with their peculiar food-procuring
appendage, in its feather-like gills, in the posterior position and
comparatively small size of the gills, and the consequent absence
of the division of the mantle cavity into the supra- and infra-
branchial chambers. It differs from Nucula in having a siphon,
and further^ shows its departure from the ordinary lamellibranchi-
ate structure in having a highly specialised tactile organ in the
siphon.1

The gill, although different in details from that of Nucula, is
essentially of the same structure as the latter. It is suspended by
a membrane, as in Nucula. Figure 9 shows it dissected out by
itself. The line of suspension is xcp; x dp is the ventral median
line, and corresponds to xdp in Figure 4. As in Nucula, the
gill is made up of a series of paired plate3, placed one after another,
and attached to the central solid stem continuous throughout the
whole length of the gill. The plates do not, however, project
downward, as we have seen in the case of Nucula, but here turn
upward (see Fig. 11). The plates are largest in the middle, and
gradually become smaller toward the extremities. At the front
end (x, Fig. 9) there is a rather interesting arrangement. Figure
10 shows diagramatically the relations of the various parts at the
anterior termination of the gill. It will be seen that the plates of
the gill gradually become smaller and finally die out toward the
front, and the gill is continued simply as a flat membraneous struc-
ture (x, Fig. 10), which goes into the visceral mass (». m., Figs. 9
and 10). A cross-section of this part shows that at its lower por-
tion at least, there is a blood-channel, probably continuous with one

1 W. K. Brooks. Proc. Amer. Ass. Adv. Sci., 1874 (end of note).
17

264 JT. MITSUKURL

of the channels in the stem of the gill. In some specimens this
membrane-like portion of the hranchia is longer than in others,
and goes some distance around the visceral mass.

Owing to the rather poor state of preservation of the alcoholic
specimens, I have not been able to make out the histology of the
Yoldia gill as fully as I should like, but the following description
I believe to be correct in essential points: — Figure 11 represents
an opposed pair of plates, and corresponds to Figure 5 of the
Nucula branchia. The suspending membrane (k ij I) consists of
fibres crossing each other in several directions, and is covered on
its two surfaces by columnar epithelium. The solid stem (i dj)
of the gill has two blood-channels, an upper (n) and a lower (o).
The latter seems to be in communication with a comparatively
free space (q) in the middle of the suspending membrane. Directly
below the upper blood- channel (o) there is a bundle of tissue,
which appears to be fibrous, running the length of the gill (seen
in cross-section at /, Fig. 1 1). It serves no doubt for support.
The floor of the lower blood-channel (r) is covered by a V-shaped
bundle of longitudinal fibres («). This would seem to be homolo-
gous with the trough-shaped chitinous structure in Nucula, but
seems to be formed of the same fibres already referred to several
times, which are found in the suspending membrane and other
parts of the Nucula and Yoldia gills, and I cannot establish any
connection between this bundle and the chitinous bars (A, Figure
11) in each plate. The latter, when they reach the longitudinal
bundle («), make a bend and turn out again to enter the next plate
in the series. In some sections I have obtained indications of a
very thin layer of chitin beneath the fibrous bundle («), which
may, therefore, correspond to the fibres found in the trough of
the framework in the Nucula gill (see above). If, however, this
V-shaped structure is really homologous with the trough of the
Nucula gill, it goes far in support of the view advanced above,
that the chitinous framework is really made up of the fibrous
tissue which is found fn other parts, here cemented into one com-
pact mass. In such a case fusion has gone further in Nucula than
in Yoldia, and we see in the first genus the trough well united
with the branches (A) in each plate. The plates (e> Fig. II) in
Yoldia spread themselves upward instead of downward, as in
Nucula. The chitinous bars (/i), of which there are two in each
plate, follow the curve of the plate and end rather bluntly about

LAMELLIBRANCHIATE GILLS, 265

half-way up, at the point a. That the part from d to a corres-
ponds to the lower inner edge of the Nucula plate (d a, Fig. 5) is
shown by the characteristic rows of columnar cells having longer
cilia than those found in other parts of the gill. There is another
system of chitinous structures (c/i, Fig. 11). Many fine chitinous
filaments come down together in a bundle on each side from the
suspending membrane, and as soon as each bundle reaches the
plate of its own side filaments spread themselves out like the
frame of a fan over the whole plate. Several fibres sometimes
proceed together, and then separating, give the appearance of
branching. They are found directly beneath the epithelial cells
that cover the plate. The effect of this framework must be to
keep the plate well spread out for the purpose of aeration. I
have not succeeded in obtaining any single section which shows
the structure of the plate well, but from the comparison of a good
many sections which I have made, I feel tolerably sure that the
whole space between the epithelial surfaces is pervaded by what
Peck1 calls "lacunar tissue" (Fig. 12). It is a loose trabecular
tissue with many nuclei and within whose network blood can flow.
The space between the chitinous bars (A, Fig. 11), which is quite
large in Yoldia, seems to be tolerably free from this lacunar tissue.
Figure 11, a, gives the outline of the plate seen from one side.

Theoretical Considerations.

The gills, here described, of Nucula and Yoldia are, I think, the
most rudimentary of any that have been studied so far. In fact,
at first sight, the resemblance to the ordinary Lamellibranch gill
is not apparent, and they suggest more the Cephalopod gill. But
I believe, the homology of their various parts with those of more
complex gills in Unioy Mytilus, Area, &c, is not difficult to make
out. After consulting the articles by Peck (foe. cit.), Posner,2
Lacaze-Duthiers,3 Bonnet,4 and others, and also after examining

1 R. Hoi man Peck. "The Minute Structure of the Gills of the Lamelli-
branch Mollusca." Quar. Journ. Micros. Sci., 1877.

2 Carl Posner. "Ueber den Bau der Najadenkieme." Archiv fur mikros.
An at, 1875.

8 Henri de Lucaze-Duthiers. u M6moire sur le Developpement des Branchies
des Mollusques Acephales Lamelli branches." Ann. d. Sci. Nat., Ser. IV,
Tome V, 1856.

* Robert Bonnet. " Der Bau u. die Circulations- Verhaltnisse der Acephalen-
kieme." Morphologischee Jahrbuch, III, 1879.

266 K. MITSUKUBL

the sections I myself have obtained of Unio, Modiola, Scapharcn,
<&c., I have no doubt whatever that the plates in Nucula and
Yoldia represent the descending or attached limb of the filaments
in the outer and inner gill-plates in forms like Mytilus, Modiola,
and Area, and accordingly are homologous with the folds on the
inner lamella of the outer gill-plate, and on the outer lamella of
the inner gill-plate in Uniof Anodon, and Dreissena. If a com-
parison is made of my Figure 6 with any of the cross-sections of
gill-filaments given by Peck, it will be seen* at once how similarly
the paired chitinous bars are placed, how almost identically the
epithelial cells are arranged around them, how two rows of those
cells (/./., Fig. 6) — called by Peck latero-frontal epithelial — have
longer cilia than the rest. In fact, Peck's Figure 12 (a transverse
section of a filament of the Anodon gill) agrees with my Figure 6
in all essential points. The left-hand figure in his Figure 5 (the
superficial view of the edge of a gill-filament of Mytilus showing
the latero-frontal and other epithelial cells) and the upper part of
his Figure 20 (the same view of a gill-filament of Anodon) would
pass very well for the corresponding part in Nucula. So far as I
can make out from rather poor specimens, the latero-frontal cells
in Nucula are strikingly like those represented in Peck's Figure
20. If, then, the plates in the gills of Nucula and Yoldia repre-
sent the gill-filaments in other genera, it follows from the embryo-
logical observations of Lacaze-Duthiers (foe. cit.), and from the
position of the chitinous bars in the plates, that they are homolo-
gous with the descending limb of the gill-filaments in ordinary
Lamellibranchs. Professor Huxley seems to have no doubt what-
ever of the homology stated here, as will appear from the quota-
tion given further on. Admitting, then, that this supposition is
correct, and that the gills in Nucula and Yoldia are in an unusually
rudimentary condition, what light, if any, do they throw on the
organogeny of the Lamellibranchiate gill? But, before proceed-
ing to the discussion of this point, let us review briefly what
theories have been advanced as to what is the most primitive type
of the branchiae of this group. Setting aside older authors like
Williams and Hancock, I consider the article*, already alluded to,
by Peck, Posner and Lacaze-Duthiers as having the most impor-
tant bearing on the subject. Posner, after a careful histological
examination of the gills of Anodon, Unio} Cardium, Mya, Mytilus,
Ostrea, Peden, Pholas, Pinna, Scrobicularia, Solen, Solecurtus, and

LAMELL1BRANCHIATE GILLS. 367

Venus, puts forward, although with hesitation, the theory that the
pouch-like gills of the Unionidsa are the most primitive type of
the Lamellibranchiate giil. StepanoflT,1 so far as I can gather,
inclines to this view. Peck, on the other hand, after an investi-
gation of Area, Mytilus, Anodon, and Dreissena, comes to the
conclusion that " the gill-plates of the Union idee are a highly
modified form derived from a simple condition in which the gills
consist not of plates but of a series of juxtaposed independent
filaments, such as we see in a less modified state in Area and
Mytilus." This view is the more generally accepted of the two.
The only complete history of the development of the Lamelli-
branchiate gill by Lacaze-Dutbiers (loc. eit.) and all the frag-
mentary embryological observations on the organs show that
the gills are at first of a tentacular or filamentary character.
Those who read carefully Mr. Peck's paper, will, I think, feel
convinced by the arguments he brings forward. So high an
authority as Professor Huxley is entirely of this view. He says:
" In its simplest form, the branchia of a Lamellibranch consists
of a stem fringed by a double series of filaments («. g. Nucula).
The next degree of complication arises from these filaments
becoming, as it were, doubled upon themselves at the free ends,
the reflected portions lying on the outer side of the outer, and on
the inner side of the inner, series of filaments . . . (Mytilus Pecten).
In roorit Lamellibranchs, the gills are four elongated plates, each
of which is in fact a long narrow pouch, with its open end turned
toward the heemal face of the body" (Invertebrates, p. 408-9, Am.
Ed.). My own observations lead me to the same conclusion. In
fact, it is difficult to see how the pouch-like gills of Unio can give
rise to such forms of branchiae as are found in Nueula and Yoldia.
By a very circuitous route they may have degenerated into their
present rudimentary state, it is true, but all recent observations
tend to show that while other organs in the La tnelli branch iata
have been steadily degenerating, the gills, on the contrary, have
become highly deyeloped and perform functions which the prob-
able change of the animal from the motile to the sedentary habits
of life has forced on these gills. If, then, there has been no con-
siderable degeneration, and if the homologies of different parts of

1 Paul Step an off. "Ueber die Geschlecbtsorgane und die Entwicklung von
Cyclas." Archiv f. Naturgcsch., 18G5.

268 K. MITSUKURL

these branchiae are, as I have stated above, the filamentary char-
acter of the primitive Lamellibranchiate gill is placed beyond
doubt.

I believe farther light is thrown on the subject by the gills of
Nucula and Yoldia. Peck shows that the gills primarily consisted
of a series of filaments, but does not attempt to account for the
fact that these filaments have come out in long rows on the side of
the body. I venture to suggest an explanation. If we reflect for
a moment, I think we shall see that the gills of Nucula and Yoldia
may be considered as a stem which, being folded on either side to
increase the surface of contact with the water, gives rise to the flat
plates which I have homologized with the descending limb of the
gill-filament of Mytifus and other like forms. The plates are,
strictly speaking, nothing but the epithelial covering of the stem
raised into folds and enclosing between the two sides of the folds
a blood-channel. In Jthe case of Yoldia mesoblastic lacunar tissue
is carried out into the folds. According to this theory, the gill of
the Lamellibranchiata was originally a longitudinal ridge on the
side of the body. Probably in this a blood-vessel ran, and must
have served as the organ of respiration. In course of time, how-
ever, this ridge became folded for the increase of the surface of
contact with the water and thus produced papilla on its two
sides — rudiments of the future gill-filaments. The gills of Nucula
and Yoldia have gone but little beyond this stage. I think there
is much to support this view. Stepanoff (loo. cit.) observed in
Oyclas that the gills arise first as a ridge on each side of the
body. Leydig1 makes the same statement. M. LoveVs2 observa-
tions have a still more important bearing on the point. He says :
"Nous avons, si je ne me tronipe, vu la premiere formation des
branchies ; nous en savons assez pour 6trc sur qu'elles se montrent
sous la forme d'un cordon fin, renflG & certains intervalles; que ces
renflements se contournent plus tard en anses, qui s'allongent de
plus en plus, et sur J esq ue lies se developpement les cils vibratiles

1 Franz Leydig. "Ueber Cyclas cornea." Muller's Archiv, 1855. He Fays:
«'Die letzto Hauptanderung im ausseren Habitus erfahrt der Embryo durch
die Bildung der Kiemen. Auch sie wachsen hIb Leisten von hi n ten nach vorne
und zwar gehen sio ureprunglich vom Mantel aus" (p. 62).

2 " Bidrag till Kamedornen om utvecklingen af mollusca acephala Lamelli-
branchiata." Memoirs of the Academy of Stockholm, 1848, lately reprinted
in an abridged form in German.

LAMELLIBRANGHIATE GILLS. 269

regulterement disposes et d'un forme particuKfcre." l " Un cordon
fin renflS h certains iutervalles" is, it seems to me, nothing but a
ridge with slight swellings or papillae. LovSn's figures are not
exactly clear to me, but what he designates as the gills are certainly
in favor of my view. In all the fragmentary embryological ob-
servations, the gills are generally seen as papillae, or nothing but
the folds of a blood-channel. I have already called attention to
the anterior part of the Yoldia gill where the plates die out and
the gill is continued simply as a ridge containing a blood-channel.
Whether this is a remnant of the primitive ridge or not it is diffi-
cult to determine, but the fact that there can be on the side of the
body a thin-walled ridge which, containing a blood-channel, must
serve more or leas for respiration, goes far in support of the view
here advanced.

To review the whole matter, the Lamellibranch gill was perhaps
originally a simple ridge on the side of the body, but to increase
the surface of contact with the water folds may have arisen on
two sides of this ridge. If such was the case, Nucula and Yoldia
are still in a stage only very little advanced from this primitive
condition. In course of time, however, as some of the Lamelli-
branchiata, either owing to degeneration or some other cause,
become incapable of extensive locomotion, these buds or folds were
perhaps prolonged to form tentacular filaments, which, going on
in their development, finally produced such complete gill structures
as we see in Mytilu8y Unto, (htrea, and other forms, taking on
at the same time functions totally foreign to their original one.
Between the simple gills of Nucula and most complex ones known, -
there are a great many intermediate stages, some going more in
one direction, others in another. For instance, Lucina and Corbis
are said to have only one gill-plate on each side (Owen* 8 Inverteb.).
According to Sars, Pecchiola is in the same condition (Remarkable
Foi-ms of Animal Life, G. O. Sars). Chamoatrea and Myochama
are described by Hancock (Ann. and Mag. of Nat. Hist., 1852-3)
as having the inner gill-plate complete, but the outer plate lacking
the outer lamella. In these tentacular filaments seem to be fused
with each other. On the other hand, although Area, Mytilus,
Modiola, have all the lamellae presqnt, the filaments composing

1 Translated by M. Young and quoted by Lacnze-Duthiers in the article
already referred to.

270 K. MITSUKURL

them have not fused with one another. It is interesting to notice
that Nucula and Yoldia, in which the gills have remained rudi-
mentary, have, as Dr. Brooks first pointed out to me, an unusual
power of locomotion, while forms wholly or almost wholly unable
to move, as Odrea, Photos, Ac., possess highly-developed gills.

For some reason the inner gill-plate seems to develop further
than the outer. For instance, in many genera, the inner is much
larger than the outer. In Chamostrea and Myochama, already
referred to, it is the inner gill-plate that is complete, and the outer
gill-plate that lacks a lamella. It will also be seen a little further
on that in Anodon the inner gill-plate has gone further than the
outer in its development. In the embryological study of the
branchiae of Mytilusy Lacaze-Duthiers observed that the filaments
of the inner gill budded out first.

It is very instructive to see the process of secondary folding
going on in higher varieties of the gill. The two lamellae of a
gill-plate are, in such a case, no longer parallel, but wavy, and the
surface of a lamella is thus considerably increased. In Anodon
this process is perhaps going on, for Peck shows that in that
genus the cross-section of the outer gill-plate has parallel and
straight edges, but that the outer lamella of the inner has a wavy
margin. Posner shows successive stages of secondary folding in
the gills of Pholas dactylusy Venus (sp.), My a arenaria, Ostrta
edulbij Solen vayina, Cardium edule, Pinna nobilis.

Diametrically opposite, as the views advocated by Posner and
Peck may seem, it is not difficult to reconcile the two.

If we look over the list of the genera examined by Posner, we
shall find all of them, except MytUus and perhaps Pecten, to pos-
sess more complex gills than Unio, and starting, as he did, from
the last genus, it is no wonder that he considered it to possess the-
primitive gill. On the other hand, Peck investigated forms
simpler than Unio, and arrived at the probably true conclusion.
Posner simply began where Peck ended. The two investigators,
therefore, supplement each other, and now, with the addition of
the extremely simple gills of Nucula and Yoldia, the series is fairly
complete, and it seems to me that the filamentary character of the
primitive Lauielli branch gill is made tolerably certain.

OBSERVATIONS ON THE EARLY DEVELOP-
MENTAL STAGES OF SOME POLYCEUETOUS
ANNELIDES. By EDMUND B. WILSON, Ph. D., Assistant
in Biology, Johns Hopkins University. With Plates XX, XXI,
XXII and XXIII.

In the course of two seasons' work at the Chesapeake Zoological
Laboratory, I made a few observations on the earlier stages of de-
velopment in a small number of marine Annelides. While these
observations are in many respects superficial and incomplete, they
nevertheless concern precisely those stages of development which
are least known and in regard to which current ideas appear to be
somewhat erroneous. I am therefore led to publish a brief account
of my observations, if only for the sake of affording a basis for
more thorough future study.

Although a number of writers have contributed to our knowl-
edge of the segmentation of the eggs of Polycbaeta, I have seen
no satisfactory account of the early stages of that process in those
forms characterized by an unequal segmentation. Most of the
statements in regard to it, as pointed out in the sequel, appear to
be somewhat erroneous and give no hint of the close similarity
which exists between the segmenting Polychsetous egg and those of
many other animals («.(/., some Oligochceta, Hirudinea^ Dendrocoela,
Pulmonate Gasteropods). Furthermore, the later stages of the
segmentation are of some interest, since the separation of the germ
layers is the result of a process which appears to be, in some
respects, intermediate between an epibolic invagination and a
kind of irregular or progressive delamination. In one or two
Polychaetous Annelides, as we know from Stossich's and Giard's
observations, the segmentation is nearly or quite equal, a large
segmentation cavity appears, and a gastrula is formed by embolic
invagination. In vastly the greater number of cases, however,
the segmentation is decidedly unequal, and a more or less modified
epibolic gastrula results. The passage from this mode of develop-
ment to a modified form of delamination is not hard to imagine,
and it appears to be actually exemplified, to some extent, in the
18 271

272 EDMUND B. WILSON.

development of Clymenella, Arenicola and Chcetopterus, as described
farther on.

The young stages of American Annelides are very imperfectly
known. Mr. Agassiz's valuable and well-known observations
stand almost alone, though there are a few scattered notes by other
writers. My observations relate to five genera, viz : Clymenelta,
Arenicola, Chatopterus, Spiochcetopterus, and Diopatra. Chcetop-
tenu is confined, so far as I know, to the region south of the
Chesapeake; the others have a much wider range. It will be
convenient to describe these forms in the above order.

Ctymenella torquata (Leidy), Verrill.

As in many other cases (c. g.y Terebella, ProtvUa, Dasychone,
Spio) the eggs are inclosed in a semi-fluid gelatinous substance
which forms in this species an ovoid mass about the size and
shape of1 a pigeon's egg. At one extremity the mass suddenly
narrows to form a peduncle which passes into the mouth of the
tube inhabited by the worm. These egg-masses were found
at Beaufort, N. C, from May until late in September, and in
such abundance as to form a very characteristic feature of the
beach and shoals. They extend from half-tide down to a depth
of two or three fathoms, and are consequently exposed to the air
for several hours each day; the embryos appear to sustain this
exposure without injury. The egg-mass contains several hundred
eggs, which are ovoid bodies measuring, on the average, about
.21 mm. in length and .16 mm. in diameter. The vi tell us,
which is surrounded by a very distinct chorion, is rendered very
opaque by the presence of a large quantity of granular food-mate-
rial or dentoplasm. For this reason I have not been able to study
the behavior of the segmentation nuclei and other internal phe-
nomena of the segmentation ; my observations relate, therefore,
almost solely to the external changes.

I was unable to determine when the eggs are fertilized, but
think it probable that this process takes place before the eggs are
laid, which is certainly the case with some other Annelides. No
direction cells are formed — at any rate, none which occupy any
definite position with respect to the vitellus.

Segmentation begins with the division of the vitellus into two
unequal parts (Fig. 2), after which the two spherules thus formed

P0LYCH2ET0US ANNELIDES. 2T3

become pressed together and a period of quiescence ensues ; this
continues twenty or thirty minutes. The spherules then assume
a more rounded form and are soon divided into four parts (Fig. 3)
by a furrow passing nearly at right angles to the first. This di-
vision takes place in such a plane as to divide the smaller of the
two primary spherules into two equal parts (6, c, Figs. 4, 5), while
the larger primary spherule is divided into unequal parts (a, d), of
which the larger (a) is'at the left side (the egg being viewed from
the upper or micromere pole). The four spherules soon become
flattened and closely pressed together, and a second resting-stage
ensues (Fig. 6). After a quiescence of about twenty minutes the
spherules again swell up (Fig. 4) and four smaller spherules are
separated from them by a horizontal furrow, passing in a plane at
right angles to the two preceding (Figs. 7, 8, 9). The four spher-
ules thus formed, being smaller, may be called micromeres, and
the four larger ones macromeres. The micromeres are not always
produced at the same moment (Fig. 7), but the difference in time
is very slight. The substance of the micromeres does not differ
in appearance from that of the macromeres, being still very opaque
from the presence of granular food-material. No difference in
in this respect between the macromeres and micromeres can be
seen until a much later period, although it is from analogy very
probable that the micromeres contain from the first a greater pro-
portion of protoplasm. As shown in Figure 8, each micromere
lies, at first, directly above the macromere from which it has sepa-
rated. In a short time, however, each micromere moves to one
side so as to come opposite the interval between two macromeres,
the spherules become closely wedged together and the egg passes
into a third resting-stage of about the same duration with the two
preceding. This shifting of the micromeres is a common occur-
rence in the similar eggs of other animals (e.g., Clepsine, BoneUia);
it appears to be a result of the mutual attraction of the spherules
since they are thereby enabled to fit more closely together. It is
worth noting that one of the micromeres is invariably a little
larger than the others ; this is the one derived from the spherule
marked d in Figure 4 — that is, the smaller of the two unequal
spherules resulting from the division of the primary larger spher-
ule (see Fig. 2).

After the formation of the first four micromeres we are enabled
to determine the relation of the parts of the embryo to those of

2U EDMUND B. WILSON.

the adult worm. The micromeres occupy the dorsal side, the
macromeres the ventral; and the largest macromere marks the
posterior end. The mouth is formed, long afterwards, at a point
nearly opposite to the micromeres.

The following figures are for the sake of convenient comparison
placed with the posterior end below. With the close of the third
resting-stage the segmentation loses its regular rhythmical char-
acter, and the spherules henceforth multiply independently of each
other, undergoing in their development, though less conspicuously,
alternations of activity and quiescence like those which have hitherto
been passed through by the entire egg. Thus it comes about
that while one part of the embryo is showing signs of rapid change,
another part may be almost stationary.

After the third resting-stage renewed activity is begun with the
division of the micromeres. Figures 12 to 17 represent the pro-
gressive changes of an egg. The lower or posterior micromere
was first to divide (Fig. 13) ; this was followed, two minutes later,
by the right-hand micromere (Fig. 14) and, after another minute,
by the left-hand micromere (Fig. 15). The upper micromere did
not divide, apparently, until considerably later. Five minutes
after the division of the last micromere the left-hand macromere
was divided (Fig. 16) into an upper smaller part, in all respects
like the micromeres, and a larger part lying below and at the side
of the egg (cf.f Fig. 18). Soon afterwards the large posterior
spherule divided, a smaller spherule separating from its left side.
The latter soon divided again into a smaller upper part (a, Fig.
17) and a larger lower part (6).

Examining the opposite side of an embryo at this stage, we find
(Fig. 18) five or six spherules considerably larger than those on
the other side. The spherule ab appears to have separated from
the large posterior macromere and to have produced the spherule a
of the preceding figure. (For a detailed study of the changes of
the lower pole the reader is referred to the account of the develop-
ment of Arenicola, p. 279).

Figure 19 represents the upper side of an embryo thirty minutes
later. The micromeres now form a layer of somewhat uniform
cells over the top of the embryo. If the egg is examined from
the lower pole the macromeres are found to have multiplied also,
though they are still considerably larger than the micromeres.
But at the sides of the embryo are spherules of intermediate size,

P0LTCHJST0U8 ANNELIDES. 275

and no dividing line between macromeres and micromeres can be
drawn. The only point where a definite limit to the layer of
micromeres can be assigned is at the posterior end where they
adjoin the large posterior macromere. The latter may sometimes
be observed in the process of division (Fig. 20) and it appears to
bud off smaller spherules which become incorporated into the
layer of peripheral cells.

Figure 21 is a somewhat oblique side-view of an embryo ninety
minutes later. To the left are cells resulting from the division of
the micromeres/ and these appear to graduate into the larger cells
to the right, which are derivatives of the macromeres.

Figures 22 and 23 represent ventral and dorsal views of an
embryo of a somewhat later stage. As before, the micromeres are
distinctly smaller than the macromeres, but the two kinds of cells
graduate into each other at the sides of the embryo. The pos-
terior macromere is still very large; dorsally it appears to be
overlaid, to some extent, by the micromeres.

At about this stage, or sometimes a little earlier (Fig. 24), the
large macromere divides into two (Figs. 25, 26). A side-view at
about this stage (Fig. 27) shows, as before, the micromeres passing
gradually into the macromeres, some of the latter near the pos-
terior end being in the course of active multiplication. From
this time the multiplication of the macromeres (if they can still
be so called) appears to be somewhat accelerated, so that the
peripheral (i. e., ectodermic) cells become more nearly of equal
size. The two large macromeres are gradually lost to view, being
in part, as I believe, overgrown by the ectoderm and in part used
up to supply smaller peripheral cells. Cells may often be observed
in active division at this part of the embryo while the other cells
of the ectoderm are quiescent.

At about this period (14 hours, Fig. 29) the anterior part of the
embryo becomes less opaque, and in some specimens the large
polygonal cells of the entoderm may be seen. The entoderm cells,
it is important to note, are distinctly larger than any of the periph-
eral cells, though the latter are as yet larger upon the ventral than
upon the dorsal side. Posteriorly, the limit between ectoderm and
entoderm is invisible. A section of the embryo at about this stage
(Fig. 28, a) shows the entoderm to consist of a solid mass of very
granular cells, the limits of which are ill defined in the section.
In the anterior and middle regions of the egg the entoderm is

276 EDMUND B. WILSON.

definitely separated from the ectoderm, the latter being clearer
bat with only obscure indications of the cell walls. Behind the
middle, however, this definite limit disappears, and we find large
cells with conspicuous nuclei which form no definite layer and are
continuous with the granular entoderm cells within. The ecto-
derm layer seems to abut against these large cells and not to overlie
them. It appears to me highly probable that the further back-
ward and downward extension of the ectoderm is, to some extent
at least, produced by the separation of the Outer ends of the large
cells as ectoderm cells, and furthermore that this process takes
place not only in the later but also in the earlier stages of develop-
ment. The spherule marked a in the figure is apparently under-
going such a division, but the section does not show this definitely
enough for certainty. It appears to be generally the case with
epibolic invagination that the micro meres receive, for a time at
least, constant additions from the macromeres, and the extension
of the ectoderm is due to this process as well as to the multiplica-
tion of the primary micromeres. So long as the two kinds of
spherules are of very unequal size and differ perceptibly in consti-
tution, the layer of micromeres can be seen to grow around and
envelop the macromeres. But if, as in the present case, the micro-
meres and macromeres differ little in size from the time of
their first appearance, the separation of a micromere from a
macromere must be the division of one of the larger spherules into
an outer ectodermic cell and an inner entodermic one, which is, so
far as it goes, a process of delamination. This process, however,
takes place progressively from above downwards and backwards,
so that the last parts of the ectoderm to be formed are those at
the posterior extremity of the embryo, where the anus, at a much
later period, is formed. While this appears to be the general
nature of the process in Clymenella, it is quite possible that the
four primary micromeres contain no entodermic part, and that a
part at least of the large posterior micromeres are at last actually
overgrown by the advance of the ectoderm without contributing
cells to that layer. Further discussion of this mode of develop-
ment is deferred until after a description of the segmentation of
Arenicola and Chcetopterus.

The embryo gradually elongates, and when about 24 to 30 hours
old (Fig. 31) acquires a broad band of short cilia surrounding the
anterior part; this is soon followed by a second much narrower

POL TCKE TO US. ANNELIDES. 2H

band (Fig. 32) near the posterior end. The ventral surface also
becomes uniformly ciliated except upon a narrow interval in front
of the posterior ring. The chorion has remained, during all these
stages, and now forms a very distinct cuticle which is perforated
by the cilia. This cuticle persists in later stages and from the
outermost layer of the body of the young worm.

The larvae now swim slowly through the mass of jelly, rotating
slowly about the longitudinal axis. In favorable specimens (Fig.
33) the large entoderm cells are visible and the layers are sharply
separated at the anterior extremity. Posteriorly, however, and in
the region of the ciliated bands, the ectoderm is very opaque and
cannot be clearly distinguished from the entoderm. Figure 1,
Plate XXIII, represents the larva of sixty hours, at which age it
sometimes leaves the jelly and swims for a time slowly about in
the water. More commonly, however, it remains in the egg-mass
during a much longer period.

The growth of the larva to the adult takes place in the usual
manner by elongation of the body and the continual formation of
somites in regular succession from the posterior region. The seg-
mentation is at first expressed externally only in the arrangement
of the setae; it is only in late stages that the external lines of divi-
sion between the somites become visible. About the third or
fourth day a pair of eye-specks appears just in front of the anterior
ciliated band. Figure 2, Plate XXIII, represents the larva five
days old, taken from the egg-mass. Four setigerous somites have
appeared, but the larval cilia remain as before, though they no
longer extend to the apex of the preeoral lobe and are disap-
pearing from the ventral side behind. The mouth is now visible
and lies behind the anterior ciliated belt. The setae number two
or three in each somite ; they are all setiform and belong to the
dorsal ramus, no uncini having as yet appeared.

The young ClymeneUas lived hiore than a month in the aqua-
rium, when they had acquired fifteen setigerous somites and some
of the characteristic external features of the adult. The anal
funnel is developed from a series of rounded papillae surrounding
the anal opening. The uncini (setae of the lower ramus) are not
developed until the setae of the upper ramus have appeared in a
number of segments. Like the latter, the uncini first appear in
the anterior segments, but the order of their development is less
regular than that of the upper setae.

278 EDMUND B. WILSON.

I have not observed how the month and anns are formed. At
the close of segmentation no blastopore is visible, and the mouth
appears much later on the ventral side in or behind the anterior
ciliated belt. The anus appears to arise still later at the posterior
extremity where the blastopore, if present, should by analogy be
found. The alimentary cavity is hollowed out in the middle of
the entodermic mass long before any communication with the
exterior can be found.

In the oldest larva observed the proboscis is well developed and
is protrusible; the last remnants of the anterior ciliated band still
persist in front of the mouth, and the larval eyes are still present;
there is still a single uncinus only in the anterior somite.

Arenicola cristata, Stimpson.

The segmentation of the eggs of this species is so similar to
that of ClymeneUa that it will be unnecessary to give so detailed a
description of it.

The eggs are embedded in huge gelatinous masses which assume
various forms as they are swayed to and fro by the tide. A com-
mon form is irregularly cylindrical, three or feet long and as many
inches in diameter. Sometimes they are rounded and shapeless,
lying flat on the sand ; in other cases they are as long as six feet
or more and from one to three inches in diameter. The size of
these masses is enormous, considering the dimensions of the adult
worm, and this is the more striking from the fact that the egg-
masses of A. marina (pi8catorumy Auct.) as described by Max
Schultze l art hundreds of times smaller, being scarcely a fourth as
large as those of ClymeneUa, and containing only three or four
hundred eggs. The number of eggs, in the case of our species,
must reach several hundred thousand. They are small (the
average diameter being about .13 mm.), nearly spherical or
slightly oval in form, very opaque, and are inclosed in a remark-
ably thick chorion which, seen by oblique light, appears to be
perforated by minute radiating pores. The vitellus is of a light
cinnamon color, so that the egg-mass appears of a decided reddish
brown tint.

1 Abhandlungcn der naturforschenden Gesellschaft zu Halle, 1865, printed
in 1856.

POLYCHMTOUS ANNELIDES. 27*

The early stages of development resemble those of Clymendla
so closely that one set of figures might almost answer for both.
No direction-cells were observed.

The vitellus first divides into two unequal parts which soon
flatten together somewhat, and a restiug-stage of about twenty
minutes ensues (Figs. 85, 36, 37). The second cleavage (Figs.
38, 89) takes place exactly as in Clymenella and is succeeded by a
second resting-stage (Fig. 47). The third cleavage takes place in
the horizontal plane, separating four micromeres from the upper
pole of the egg (Figs. 40, 49). As in ClymeneUa, the micromeres
become shifted so as to lie between the macromeres instead of over
them, and the egg passes into a third resting-stage (Figs. 41, 50).
As before, one of the micromeres is a little larger than the others.
It is a noteworthy fact that the micromeres, as compared with the
macromeres, are distinctly larger than in the Clymendla egg; that
is to say, the segmentation is less unequal.

After a quiescence of fifteen or twenty minutes activity is re-
sumed; we will first follow the changes at the lower or macromere
pole. The macromeres divided, in the specimen figured, almost
simultaneously (Fig. 42), each giving rise to a smaller anterior
and a larger posterior spherule (these are connected in the figure
by short lines to show their derivation). Very soon afterward
the micromeres also divide (Fig. 62, from another egg), and the
egg passes into another pretty marked resting-stage (Fig. 43) of
about twenty minutes duration. From the large posterior macro-
mere a smaller spherule then separates (Fig. 44) on the right side,
and the other macromeres divide in somewhat irregular succession
(Figs. 44 to 46). -

Figures 51 to 53 represent the changes (in anotHw egg) of the
upper pole. The micromeres form a cap of smaller cells which
are behind clearly separated from the macromeres, but elsewhere
graduate into the latter. Owing to the great opacity of the eggs,
I have not been able to follow the subsequent changes as fully as
in Clymenella, but so far as could be determined, they were essen-
tially similar to those of the latter. Two or three spherules at
the posterior end retain for a long time their predominance in size
(Fig. 54) and appear to be in part overgrown by the peripheral
cells. Elsewhere, the micromeres and macromeres graduate into
each other, and the macromeres appear to separate, in the course of
their development, into peripheral parts which become incorporated
19

280 EDMUND B. WILSON.

into the ectoderm and central parts which pass into the entoderm.
Figures 55 and 56 are opposite views of an embryo about nine
hours old. The blank space at the bottom of Figure 56 is the
posterior extremity; the cell outlines could not definitely be made
out in the specimen.

The subsequent external development differs from that of Cly-
menella only in matters of small detail. The embryo gradually
elongates, and when 18 to 24 hours old (Fig. 57), acquires a
broad anterior belt of cilia, in front of which appear two eye*
specks. Very soon a second belt, much narrower than the first,
appears near the posterior extremity, and the ventral surface
becomes covered with a broad band of short cilia, which, how-
ever, does not extend quite to the posterior ring (Fig. 58). The
first pair of setae appear during the third day (Fig. 59). The head
is now distinct and the mouth has appeared on the ventral side,
apparently in the middle of the ciliated belt; at a later stage it
lies behind this belt. The chorion of the earlier stages now forms
a very distinct transparent cuticle which is perforated by the cilia;
this cuticle persists in the latest stage observed.

At about this stage the larvae leave the egg-mass and swim
about actively at the surface of the water. They always swim
towards the lightest side of the vessel, where they crowd together
in such numbers as to form a cinnamon-colored scum on the water.
The free-swimming life is very brief, lasting commonly no more
than a day or two. Figure 3, Plate XXIII, represents a larva
of four days which was slowly swimming about near the bottom
of the vessel. The body now shows an obscure segmentation, an<
a new pair of setae has appeared behind the former pair. A nei
seta has also appeared on each side in the anterior setigerous
mite. In front of the two setigerous somites is a segment of th»^~
body without setae, and in front of this is the head. By the fiftT j
day another somite with a pair of rudimentary setae is developers
from the posterior region, and a second pair of setae appears if^
the next somite in front. These setae all belong to the uj
ramus ; the new ones appear below the older ones, so that the set
develop from above downwards. In a few specimens of this
an uncinate seta of the lower ramus has appeared in the ante-
rior setigerous somite. The cilia have begun to disappear, aa</
though many of the larvae are still swarming at the surface, tlmep
secrete a gelatinous substance which greatly impedes their move*

POLYCEMTOUS ANNELIDES. 281

raents. They soon, however, sink to the bottom, or attach them-
selves to the sides of the vessel. Here they secrete small masses
of a soft gelatinous substance in which they creep actively about.
Thus conditioned, they lived more than three weeks in the aqua-
rium. At the expiration of this time they were of a long vermi-
form shape, obscurely segmented, and possessed 11 or 12 setigerous
somites. Figures 60 and 61 represent the larvae of eight days.
The cilia have quite disappeared, and there are five setigerous
somites. Each of these has a single uncinus in the lower ramus
and in the upper rami 4, 4, 3, 2 and 1 respectively. The pro-
boscis is already developed as a thickened region at the beginning
of the alimentary canal and is actively protruded and with-
drawn.

Figure 4, Plate XXIII, represents the young worm of 15 days,
which possesses six setigerous somites. The larval eye-specks still
remain, and the head is distinct. The dorsal pseud haemal vessel
is well developed. There is an especially glandular region of the
stomach, extending from the third to the sixth setigerous somites.
Beyond this point I have not followed the development, since the
worms all died, probably from the lack of food.

Max Schultze has described (I. c.) the larvae of the European
A. marina (pincatorum) which are, in general, very similar to
those of our species. But there is a narrow ciliated ring poste-
rior and another anterior to the broad belt. The anterior ex-
tremity is much more acute, and the eyes lie in the broad belt of
cilia instead of anterior to it. The somites become very definitely
marked at an early stage, but the setae do not appear until a far
later period than in our species. Horst has also briefly described1
the larvae of a European species which agree closely with ours.
Unfortunately I have not his paper at hand, for more exact ref-
erence.

The larv83 of ClymeneUa and Arenieola are essentially alike, so
far as regards the distribution of the cilia; and they are of the
larval type originally called Telottocha by Johannes Muller, in
which the cilia are arranged in two belts, one being praeoral and
the other near the posterior extremity. The belts in this case are,
however, far less definite and concentrated than in the strictly
free-swimming Telotrochae, like the larvaa of Nerine, Capitella or

1 Tijdschrift der Nederl. Dierk. Vereeniguug, Deel I, bl. 61.

282 EDMUND B. WILSON.

Nephthys.1 The belts are broad and the cilia short and weak.
This modification of the Telotrochous type is evidently due to
the circumstance that the free-swimming life is very brief, the
larval development taking place chiefly in the gelatinous egg- mass.
A very similar larva is that of TerebeUa ncbxdosa? which likewise
is protected within a gelatinous egg-mass. Clapar&de and Metsch-
nikoff have also described a similar case — the larva of TerAdla
Meckelii — and have pointed out its significance. We find, in
accordance with this view, that in the Arenioola larvae which lead
a brief free-swimming life the belts, in the earlier stages at least,
are narrower and the cilia more powerful than in ClymeneUa, the
larvae of which never swim freely through the water. (This
difference is not well illustrated by the figures.)

•

Chcetopterus pergamentaceus, Cuvier.

This fine Annelide, for the identification of which I am indebted
to the kindness of Professor Verrill, of Yale College, is common
at Beaufort, and in the summer of 1881 I succeeded in fertiliz-
ing the eggs artificially and thus procuring the ciliated larvae.
The well-known researches of Johannes Muller, Busch and Max
Muller have made us familiar with the later larval stages, but
the segmentation of the egg and the early larval forms have not
hitherto been described.

Adult worms of full sexual maturity were found on the sand-
flats during the months of June and July. The ovaries are in
the form of convoluted masses of long narrow flattened bands of
a bright yellow color, which occupy a large part of the perivisceral
cavity in the posterior region of the body. The spermaries occupy
a similar position in the male (the sexes being distinct), but are of
a creamy white color. The spermatozoa are of the ordinary tailed
form.

The unfertilized ovum is a spherical body about .09 mm. in
diameter; the vitellus is granular and opaque, though less so than
in the preceding forms, and is surrounded by a very delicate mem-
brane, which only becomes distinctly visible after fertilization.

1 Claparede und Metschnikoff. Zeitsehrift fur wisa. Zoologie, Bd. XIX, 1869.

2 Milne- Ed wards. Recherches Zoologiques, etc., Ann. d. Sciences Nuturelles
Sex. Ill, T. Ill, 1845.

P OL YCHJE TO US ANNE L IDE S. 388

My observations upon the segmentation, especially it its later
stages, are somewhat fragmentary, but indicate a mode of devel-
opment very similar in the main to that of Clymenella. A few
minutes after fertilization the membrane begins to separate from
the vitellus (Fig. 63). After a period usually of about thirty
minutes the egg elongates slightly, becomes rather more trans-
parent towards one end, and soon produces in succession two small
clear direction cells (Fig. 64) at this end ; the egg then becomes
again spherical. The first cleavage, about twenty-five minutes
later, divides the egg, as usual, into two unequal parts, the plane
of division passing through the direction cells (Fig. 66). At the
same time a rounded opaque prominence appears on the surface of
the larger spherule, on the side opposite to the direction cells
(Figs. 65, 66, 6). This singular body, which is much larger and
more opaque than the direction cells, subsequently fuses com-
pletely with the vitellus, and the part played by it in the develop-
ment could not be ascertained. Its appearance is, possibly, due
to pathological changes, but it appears to be always present, and
I believe its formation to be a normal occurrence. Quatrefages
observed a somewhat similar thing in the segmentation of Sabel-
laria (Hermella),1 but normal and abnormal eggs are hopelessly
confused in his paper.

After the first cleavage the egg passes into a very marked rest-
ing-stage (Fig. 67) in which the spherules flatten together much
more completely than in the corresponding stage of Clymenella;
this continues fifteen or twenty minutes. The second division is
precisely like that of Clymenella or Arenicola (Figs. 68, 69); the
plane of cleavage again passes through the direction cells, which
therefore lie above the point where the four spherules meet. The
egg then passes into a second marked resting-stage (Fig. 70) dur-
ing which the peculiar prominence on the lower side of the egg
fuses permanently with one of the spherules.

The third period of activity results, as before, in the separation
of four micromeres (using this term for the sake of analogy) at the
upper pole, where the direction cells are situated (Figs. 71, 72).
These " micromeres " scarcely merit the name, for they are still
larger than in Arenicola, being, in fact, but slightly smaller than

1 M6moire sur 1'embryogSnie dcs Annelides. Ann. d. Sciences Naturellea,
S*r. Ill, T; X, 1848.

284 EDMUND B. WILSON.

the macromeres. They soon place themselves between the macro-
meres, and a third resting-stage results (Fig. 73).

In the next stage all the spherules divide at nearly the same
moment (Fig. 74) and the inequality in size between the spherules
is even less marked than before. The spherules again flatten
down, and a fourth resting-stage follows (Fig. 75). Further stages
in the segmentation are represented in Figures 76, 77, 78 ; in the
last figure the outline of the entoderm is faintly visible. The
inequality in size between the spherules becomes very slight after
the stage represented in Figure 76, and I have not seen the
large posterior macromeres observed in the other eggs.

The embryo gradually elongates and at some time between the
twelfth and eighteenth hours becomes everywhere covered with
cilia. These do not, however, perforate the egg membrane, as in
the cases already described. During the segmentation the mem-
brane separates more and more from the embryo and finally dis-
appears.

The cilia very soon assume the arrangement shown in Figure
79 (18 hours). The cilia are somewhat longer over an area behind
the middle of the body, thus forming an ill-defined belt in this
region. There is a very definite anterior apical tuft of long cilia ;
at the posterior extremity the cilia are longer than those of the
general surface, but they do not form a definite tuft. Six hours
later (Fig. 80) the body is still more elongated, the mouth has
appeared on the ventral surface in front of the belt of longer cilia,
and a pair of eye-specks are present still further forward ; they
are upon nearly opposite sides of the body, though somewhat
towards the dorsal side. The larva swims rapidly about, rotating,
at the same time, on its long axis. The ectoderm and entoderm
appear definitely separated, except at the extreme posterior end.

The larva of forty hours is represented in Figures 81 and 82
from the left and ventral sides respectively. The mouth is very
distinct and leads into a distinct oesophagus. The ciliated belt is
much more definite, and just behind it, on each side, is a long stout
flagellum (/) which is usually in a state of rapid vibration. The
anus is not yet formed. During the next twenty-four hours the
larva rapidly elongates, the prseoral lobe becomes distinct, and
the three regions of the alimentary canal are clearly defined.
The anus appears on the dorsal side just in front of a terminal
papilla (Fig. 5, PI. XXIII) which bears a tuft of long cilia* The

POLYCHJSTOUS ANNELIDES. $85

body becomes obscurely divided into three regions, viz : the very
large praeoral lobe, a middle region in which the stomach lies,
and a posterior region including the short intestine. The belt of
longer cilia encircles the middle region, but is now less definite
than in the last stage figured. A second belt, as yet ill-defined,
has, however, appeared on the posterior region. The cilia of both
belts graduate into those covering the general surface. I could
not see the flagella in the specimens figured, though they are cer-
tainly present in older specimens. The larva of five or six days
are shown in Figures 83 and 84. They are much like those of
three days, but the anterior belt of longer cilia has almost or quite
disappeared, while the posterior belt is well developed. The
flagella still remain, and in some specimens (Fig. 84) there are
two or three, instead of one, on each side. The alimentary canal
occupies almost the entire perivisceral space, which is reduced to a
narrow cleft. The three regions of the former consist of a very
wide rather thick-walled oesophagus, a large stomach with thick
and glandular walls, and a very short globose thin-walled intes-
tine. The mouth is enormously large and of a triangular form.
From its posterior angle a narrow groove, lined with rather long
cilia, runs backwards in the median line.

Figures 6 and 7, Plate XXIII, represent the larvae of twelve
days, the oldest raised from the egg. In general appearance they
.are considerably like the last stage, but the anterior belt of longer
cilia is entirely atrophied, and the cilia over this part of the body
do not differ from those covering the entire surface. The posterior
belt, however, is now perfectly definite and greatly developed. It
consists of a series of very long and powerful cilia, like those of
the characteristic belts of other Mesotrochal forms. Towards the
middle line on the ventral side these cilia gradually disappear, so
that the belt is open below ; it is very probable that new cilia are
formed at this point. The cilia of the belt, except in being
shorter, are much like the flagella of the last stage, and I was at
first inclined to believe that the belt is formed by an extension
of the* flagella around the body. This point was not definitely
settled ; I think, however, that the belt is not thus formed, but is
a further specialisation of the tract of longer cilia surrounding the
posterior region of the body in the preceding stages.

Soon after the last stage described the larvae all died. I have
only once succeeded in taking an older larva at the surface, and of

880 EDMUND B. WILSON.

this have, unluckily, no sketches. This larva agreed closely with
the Mesotrocha sexoculaia of Johannes Muller, which was shown
by Max Muller to be a young Chastopterus — probably C. Nor-
vegicus, Sars. In its general features it was similar to the larvae
just described, but there were two very distinct belts of powerful
cilia. Hence it would appear that our larva, in the course of its
further development, elongates and acquires a second belt of cilia
which probably arises behind the first.

The ventral longitudinal ciliated groove is a somewhat interest-
ing feature of the Chart opterus larva. It may perhaps be com-
pared with the ventral ciliated region of the larvae of Clymenella,
Arenicola, and a number of other Polychaeta, of Echiurus and of
OligochaBtous larvae generally. In the embryo of Euaxes Kowa-
levsky observed a ciliated groove very similar to that ofVhcetop-
terns. In the genus Protodrilus, recently described by Hatschek, a
similar groove is persistent throughout life; while in at least some
other Oligochseta, as we know from Hatschek's beautiful researches,
such a ventral ciliated groove becomes infolded to form a part of the
ventral nerve cord. It is, however, quite possible that the ventral
groove of Chcetopierus has no significance in this direction, but is
a special larval adaptation which simply plays a part in bringing
particles of food towards the mouth.

The appearance of a provisional belt of cilia, which afterwards
disappears and is replaced by another, may perhaps have some,
ancestral meaning.

Spiochcetopterus ocidatus, Webster. (?)

During the summer of 1879 numerous specimens of a ChaBtop-
terid larva were taken by the dipping-net at Fort Wool in the
southern Chesapeake. These larvae developed in the aquarium
until they could be recognized as SpiochostopteruSy or a closely
allied genus, and for the sake of comparison with other larvae of
this family a brief description of them may be given. The. larvae
were true Mesotrochae, closely similar in every respect to the free-
swimming young of Telep8avu8 as described by Claparfcde and
Metschnikoff in the paper already referred to; hence a very brief
description will suffice. My only reason for referring the larva to
Sars's genus is that Telepsavus is not known to exist in our waters,
while Spiochcetopteru8 is abundant at some places on the eastern

POLTOHJETOUS ANNELIDES. 287

shore of Virginia, as well as farther northwards. The " Tdep-
savus" larvae were not identified with certainty and might with
equal reason have been referred to Spiochcetopterus.

The larva (Fig. 8, PI. XXIII) is more or less elongated, though
of exceedingly changeable form. As in Chcetopterus there is a
distinct praeoral lobe, a fleshy bilobed lower lip and a large trian-
gular mouth. The eyes are two in number, placed on nearly
opposite sides of the body. Behind the eyes is a pair of very
contractile short tentacles. A little way behind the middle of the
body is a series of very powerful cilia encircling the body; they
are borne on a thickened ring of the body-wall. This ring divides
the body into two regions, the anterior of which contains nine
somites, as shown by the groups of set©; the posterior is imper-
fectly segmented and is terminated by a small appendage which
resembles the terminal papilla of Chcrtopterus, or is more usually
two-jointed. • Just below the ciliated ring, on the dorsal side, are
the rudiments of two pairs of branchiae. The alimentary canal is
very distinctly divided into the usual three regions— oesophagus, *
stomach, and intestine, the latter terminating in a dorsally placed
anus. On the ventral side of the body, about opposite the seventh
setigerous somite, is a glandular infolding of the body-wall.

In the oldest larvae observed (Fig. 85) the body is much more
elongate, and the anterior region and buccal segment have assumed
the appearance of the adult. Two pair of branchiae have appeared
on each of the two segments behind the thickened ciliated ring;
from the latter the cilia have nearly disappeared. The posterior
region is much elongated and is distinctly segmented. The young
worm has, in fact, attained practically the adult structure, though
the middle or branchiferous region contains as yet only two somites.
The oesophagus has elongated greatly, extending backwards nearly
to the middle region. Figure 87 represents the peculiar stout
seta of the fourth segment, and Figure 88 the ordinary form of
setae from the anterior region. The branchiae are bilobed, and
each lobe is furnished with a short series of powerful cilia.

The larvae of all of the Chcetoptaidce, so far as known, are
true Mesotrochae, and this type of larva is not known to occur
in other groups of Annelides. These larvae agree in so many
other respects besides the arrangement of cilia, that it is not easy
to avoid the conclusion that they must represent, to a certain
extent, the ancestral type from which the various forms in this
20

288 EDMUND B. WILSON.

very peculiar family have been derived. However this may be,
it is interesting to observe that the larvae of Spiochoetopterw (or of
Telep8avu8) with their single ciliated ring remain, throughout their
larval existence, in a condition which is only temporary in the
Chcetoptet*u8 larva. The larva of Phyllochcetoplerua, like that of
Chcetopteru8f has two ciliated rings (Claparfcde and Metschnikoff),
though the adult is far more like Spiochcetopterus. The case is of
some interest as showing how readily the ciliation of larvae may
undergo modification, even within the limits of a small and well-
circumscribed group, and (juite independently of such conditions
as parasitism or the special protection of the young in egg-masses
like those of Clymenella. It would be interesting to observe
whether the larva of PhyUochoetopierus likewise passes through a
temporary monotrochal stage.

Diopatra cuprea, Clapar&de.

A few observations upon the young of this species, made at
Beaufort, may perhaps be worth describing for the sake of com-
parison with other Eunicid larvae. The larvae are found embedded
in long strings of slimy jelly which may often be found attached
to the tubes of the worm. In spite of numerous efforts, I have
never procured the eggs in the early stages of development. The
youngest larvae observed (Fig. 89) were pear-shaped and without
indication of segmentation. There are two widely separated eye-
specks at the anterior extremity, and the body is peculiarly blotched
with irregular spots of dark pigment. A very broad band of short
cilia surrounds the middle region of the body, and just in front of
this is another very narrow and ill-defined band. At the anterior
end is a small apical tuft of cilia; the narrow posterior extremity
is surrounded by a narrow but pretty distinct ring. Thus the
larva appears to be a slightly modified Atrocha, like other Eunicid
larvae.

In the next stage observed (Fig. 90) the body has elongated
slightly, and the posterior region has become divided into five
pretty distinct somites, each of which, except the last, has a tuft
of cilia on each side. In other respects the ciliation is the same
as in the last stage. We observe that the broad ciliated belt lies
quite in front of the segmented part of the body and, therefore,
presumably in the head region. At the posterior extremity of

POLYCHJETOUS ANNELID ES. 289

the body are two slight protuberances which are rudiments of anal
cirrhi. Figure 9, Plate XXIII, represents a considerably older
larva from the same jelly-mass with the last. The segmented
region has greatly elongated, the somites are very distinct and of
about equal size, and the parapodia, with distinct dorsal cirrhi,
have appeared. The eyes have moved backwards nearly to the
middle of the head region. Nearly midway between them is the
rudimentary median antenna, and just in front of each is a lateral
antenna; the latter are much longer than the former. The cilia
are arranged as in the last stage, except that the apical tuft and
the anterior narrow ring have disappeared.

By the end of another week the young worms (Figs. 91, 92)
have lost their cilia, left the jelly-mass and crawl about on the
walls of the aquarium, especially on the side turned to the light.
The body is vermiform, three new somites have appeared behind
the five of the la3t stage, the parapodia are very prominent, the
anal cirrhi elongated, and the head and buccal segment well
defined. The cilia have disappeared, excepting a tuft in front of
each parapodium behind the first. The median antenna has elon-
gated considerably, the lateral ones still more so, and the latter
have acquired the short basal portion present in the adult. Below
the lateral antennae a second pair have appeared which are still
short and simple. On the upper side of the buccal segment are
rudiments of the "tentacles" (or "tentacular cirrhi"). A para-
podium in the anterior part of the body consists, at this stage,
of a somewhat bilobed protuberance with a cirrhus above and
below. The dorsal cirrhi are present in all the somites, diminish-
ing in size to the last; the ventral cirrhi, however, become reduced
behind the second somite to low rounded prominences. Rudiments
of the branchiae have appeared above the dorsal cirrhi of the sixth
and seventh parapodia. The branchia is at first a simple ciliated
cirrhiform appendage, which only in much later stages assumes
the spirally branched structure of the adult. The subsequent
development is very simple. The small frontal antennas appear,
two additional anal cirrhi grow out, the body elongates, and the
young worm of a month old, except for the simple branchiae,
resembles the adult; it has at this age twenty-three setigerous
somites. About the end of the second week the young worm
leaves the jelly-mass and secretes a membraneous tube in which it
thereafter dwells.

290 EDMUND B. WILSON.

The Diopatra larva agrees pretty closely with that of "Zumftrt-
oonereis sp." described by Clapar&de and Metschnikoff as one of
Muller's "Atrochee" also with the larva of Eunice sanguined (Koch),
and of the Eunicid larva described by Krohn and Schneider/
which appear to be likewise Atrochee. I did not determine the
precise relation of the mouth to the broad ciliated band, but the
latter is always confined to the head region and the mouth appears
at the posterior margin of the cephalic segment. Hence it seems
very probable that the band either passes in front of the mouth
or surrounds it, in which case the "Atrochal" larvae of at least
some Eunicidae may be extreme modifications of the Telotrochous
type, due, as in the case of Clymenella, Arenicola or Terebella, to
the absence of a free-swimming mode of life. It appears very
probable that Diopatra cuprea, like some other species of the
Eunicidtt, is viviparous, or, at least, that the segmentation of the
egg takes place within the perivisceral cavity of the parent. It is
hardly possible otherwise that I could have failed to discover the
segmenting eggs among the large number of egg-masses examined.
This, if true, would perhaps explain some peculiarities, of the
development, such as the modification of the ciliation and the
early and rapid development of the somites. In Eunice mn-
guinea? which is certainly viviparous, the development is much
abbreviated, and the somites and setae appear very early. It is
very probable that all of the "Atroch®" are either young larvae
which have not yet acquired their characteristic larval features
(e. g., Chcetopterus), or forms which, like Diopatra, have become
modified in accordance with special conditions, such as the absence
of free-swimming life.

The young Diopatra, at the stage represented in Figure 89, is
much like an unknown Eunicid larva observed by Krohn and
Schneider, agreeing in the number and arrangement of the an-
tennae, the number of setigerous somites and anal cirrhi. It resem-
bles also the young Autolytus, in the number, form and arrange-
ment of the antennae.3 In the latter genus, according to Agassiz,
the median antenna, though smallest, is first to appear. A further
likeness lies in the fact that the "tentacular cirrhi" (append-

1 Arch, fur Anat. und Physiol., 1867.

2 Koch. Neue Denkschrift dor Allg. Schweiz. Gesellsch. ges. Naturwiss.,
Bd. VIII, 1847.

3 Alex. Agassiz. Bost. Jour. Nat. Hist., Vol. VII, 1859-1863.

POLYOHJETOUS ANNELIDE8. 291

ages of the buccal segment) appear at a late period after both the
antennas and the dorsal cirrhi of the parapodia are well advanced
in development The same order in the development of the an-
tennsB is followed by Marphysa sanguined, according to Webster,
the median one appearing first, then the upper lateral ones, and
then the lower lateral. In view of the morphological importance
which has been attached to the head appendages of the Annelides,
it would be interesting to determine how far the order of their
appearance is constant in different groups.

Our knowledge of the segmentation of the egg in the Poly- •
chaeta is derived from the accounts of a considerable number of
observers, among whom may be especially mentioned Milne-
Edwards, Quatrefages, Sars, Clapar&de and Metschnikoff, Haeckel,
Von Willemoes-Suhra, Giard, Stossich. The observations of
Stossich have been already referred to. Those of Milne-Edwards
give little more than the result of segmentation, without consider-
ing its details. Quatrefages was entirely led astray by his failure
to distinguish between normal eggs and those which underwent
pathological changes, as often happens when eggs are artificially
fertilized. (I have unsuccessfully attempted this experiment with
eggs of the same genus [Sabellaria], studied by Quatrefages, and
can testify to the accuracy of his figures of the abnormal eggs.)
He was thus led to certain results which he very justly asserts to
be "tout nouveau dans l'histoire de I'embryogSnie."

The general statement in regard to the segmentation made by
Clapar&de and Metschnikoff in their admirable paper on Chaato-
pod development is as follows: " Bei alien Chsetopoden fuhrt der
Vorgang der Dotterkluftung zu der Bildung vop zweierlei Dotter-
elementen, die sich von einander nicht nur in Bezug der Grosse,
sondern auch (lurch das Ansehen, das Brechungsvermogen u. s. w.
sehr bedeutend unterscheiden. Die Bildung dieses Gegensatzes
der beiden Embryonal massen ruhrt von der allerersten Zweithei-
lung des Dotters her, indem die erste Kliiftungsfurche meist so
angelegt wird, dass der Dotter in zwei ungleiche Halften zerfallt.
Beide kluften sich zwar weiter fort, die kleinere jedoch viel
schneller als die grossere, so dass jene zur Bildung von sehr
kleinen Furchungskugeln oder Zellen fuhrt, welche die grosseren
aus der Kliiftung der anderen grosseren Halfte herruhrende

292 EDMUND B. WILSON.

Kugeln allmahlig umwachsen and einschliessen." l This account
is generally accepted, being given, for instance, almost verbatim in
Huxley's "Anatomy of Invertebrated Animals" and in the last
edition of Clans's " Grundzuge der Zoologie." Hffickel's figures
of the segmenting eggs of Fabricia agree with this account and
are wonderfully like Clapar&de and Metschnikoff's figures of the
eggs of Spioy a widely different Annelide. Without denying the
accuracy of these observations, it is nevertheless certain that in
the three genera described in this paper, one of which is very
different from the others, the ectodermic and entodermic parts of
the egg are not separated until long after the first cleavage. No
separation is effected until the third cleavage, at least, when the
spherules can be distinguished as macromeres and micromeres;
and the complete dissociation of the layers is not, as I believe)
effected until near the close of segmentation.

This error, if it be such, in regard to the early stages may
appear trivial, but, as already pointed out, it is one which ob-
scures the very close similarity between the Polychcetous egg and
many others, and as such perhaps needs further observations for
its rectification. Claparede and Metschnikoff drew attention to
the "sehr willkommene Uebereinstimmung" between the segmen-
tation of the Polychadta and of Hirudinea, but the correspondence
is much closer and more detailed than they supposed. Thus the
earlier stages in the segmentation of Clepsine are closely similar,
even in details, except for the greater inequality of the micromeres
and macromeres, to those of Clymenella. The egg of the Oligo-
chaetous genus Euaxes, so carefully studied by Kowalevsky, is
still more like the Polychcetous egg, since in this case the macro-
meres undergo division at a much earlier period than in Clepsine,
It may be worth while to compare the Euaxes egg with that of
Clymenella somewhat in detail. After the second cleavage, the
egg is exactly like that of Clymenella, there being a large posterior
spherule, two smaller spherules, and one of intermediate size.
This stage is, however, differently attained, if Kowalevsky's very
explicit account is correct, inasmuch as the large posterior spherule
is the undivided smaller spherule of the first stage, while the three
others are produced by two successive divisions of the primary
larger spherule. As in Clymenella four micromeres are produced at

i Zeitschr. fur wiss. Zool., Bd. XIX, p. 166, 1869.

P0LTCHJET0U8 ANNELIDES. 293

the upper pole of the egg, though not at the same time. The
posterior micromere is much larger and more opaque than the
others, and gives rise to two raesoblasts as well as to ectoderm
cells. Although it is not mentioned in the text, Kowalevsky's
figures show that the separation of micro meres from the macro-
meres continues for a considerable period after the formation of
the first four. The micromeres, with the exception of the pos-
terior one already referred to, are at all stages much smaller and
clearer than the macromeres, so that the limit of the ectoderm is
always plainly visible. At no time, however, does any segmen-
tation cavity appear, the invagination being typically epibolic.

As compared with the development of Euaxes, the peculiarities
of the Polychsetous segmentation depend upon the primary slight
difference in size and constitution between the macromeres and
micromeres and the speedy division of the former, so as to reduce
this inequality still further. In Glymenella the inequality between
the first four micromeres and four macromeres is much less than
in Euaxes, in Arenicola still less, and in Chcetopterus scarcely
exists. In Terebellides Stromii, according to Willeraoes-Suhm,1
the segmentation is equal, but I gather from his somewhat frag-
mentary account that no segmentation cavity and no invagi-
nation were observed. Putting these facts together it would seem
that the eggs of various Chaetopods, if carefully studied, might
show us within the limits of one group the actual steps in the
conversion of invagination into delamination (cf.} p. 276). Accord-
ing to this view, the egg of Serpula represents the primitive form
at the beginning of the series, having an equal segmentation, or
nearly so, a large segmentation cavity, and undergoing an embolic
invagination. As the entoderm ic portion of the egg became more
and more loaded with food-yolk, the segmentation became more
and more unequal, the segmentation cavity decreased in size and
at length disappeared. This condition is retained by the egg of
Euaxes, After this point, however, the segmentation becomes less
unequal, owing, perhaps, to changes in the distribution of the
food-yolk ; and the three genera described in this paper represent
three stages in the return towards an equal segmentation. This
return is not, however, accompanied by the reappearance of a
segmentation cavity, so that an invagination is not possible, and

* Zeitsch. fiir wiss. Zool., XXI, 1871.

294 EDMUND B. WILSON.

upon this fact has perhaps depended the acquisition of a mode
of development resembling delamination. While the develop-
ment of some Polychcetous eggs, if my account is correct, has
many of the features of an epibolic invagination, it. is, on the
other hand, nearly akin to a delamination like that of Tetrastemma1
or Clavularia? the main difference being that in the Annelide egg
there is only a partial delamination and that is effected by sue-
cesive steps. An almost precisely similar case is that of Tubularia
as described by Ciamician (ZeUschrift fur wissenschaftliche Zoologie,
XXXII, 1879), the eggs of which are more favorable for obser-
vation than those of Annelides, and which undergo a form of
development almost exactly midway between epibolic invagination
and delamination. His observations have not, however, been con-
firmed by other competent observers.

The only observer who has given a description of the later
stages in accordance with that given above is Willemoes-Suhm,
who says that in the eggs of Terebellides zostericola he could not
satisfy himself that the micromeres envelop the macromeres, and
adds: "In den ersten Furchungsstadien sah ich allerdings oft
ungleiche Furchungskugeln und mit grosster Deutlichkeit ....
aber niemals jene Stadien in denen das Vorhandensein der beiden
Dotterelemente scharf und klar hervorgetreteu ware." On the
other hand, in Spirorbis, "die kleineren Furchungskugeln um-
wachsen die grosseren."

Leaving this point, which must remain doubtful until a thorough
study by the aid of sections can be made, certain other points
are noteworthy. The greater size of the posterior spherule
in the first stages is a curious fact which calls to mind the seg-
mentation of many Molluscan eggs. The greater size of this
spherule may be in part due to the storing up of mesoderm ele-
ments in it. That this is not, however, the only cause is proved
by the case of Euaxes, where the preponderance in size of this
spherule is quite as marked after the separation of the raesoblasts
from it as before ; and where, moreover, a large part of the meso-
derm does not come from this spherule at all. The principal cause
seems to be a tendency towards the accumulation of food-yolk in
this spherule, which is thereby retarded in its multiplication.

1 Hoffmann. Niederlandisches Archiv, III, 187G-7.
Kowalevsky. Zool. Anzeiger, No. 38, 1879.

P0LYCHJET0U8 ANNELIDES. 295

This tendency, if pushed still further, might lead to the forma-
tion of a true food -yolk, as Rabl and Brooks have shown it to
have been formed in the Molluscan egg. It is, perhaps, worth
noting that the Annelide egg corresponds in this respect very
nearly to that stage in the evolution of a food-yolk which has not
3^et, according to Brooks, been discovered among the Mollusca,

The persistence in some cases of the chorion as the larval cuticle
is a remarkable occurrence, entirely confined, so far as known, to
"fclie Cbsetopoda and Gephyrea, and by no means universal among
tiliem. Some doubt has been cast upon the accuracy of observa-
tions relating to this point; but it has been seen in so many cases
and by so many different observers, that it is impossible not to
aocept it as a fact.

With regard to the nature of the various larval forms existing
among the Polych»ta, it is now generally admitted that, with
f>erhaps one or two exceptions, they have little morphological im-
Jx>rtance; and that it is impossible to form any classification of
t;liem, based on the distribution of the cilia, which corresponds
xvith the grouping of the adult forms. With the possible excep-
tion of the Mesotrochffi, which form a very distinct and well-
defined group, all of the larval forms appear to be readily derivable
from the Telotrocha; and in many cases the modifying conditions
hich have produced the change are obvious. The most impor-
nt of these is the absence of a free-swimming pelagic life, and
t.his, in turn, depends upon the provision made by the parent for
the care of the embryo or larva during its early life. In the
Oligochseta this provision is so perfect, both as regards food and
protection, that a larval stage is entirely dispensed with, the cilia
V>eing reduced to a mere remnant. This condition is, however,
but a step beyond such a larval form as that of Diopatra or Eunice,
and it seems evident that the embryological differences between
PolycbaBta and Oligochaeta are due to purely adaptive conditions.

Note. — While this paper was in press, I received an important
paper by Goette, entitled "Abhandlungen zur Entwickelungs-
geschichte der Tiere (sic). Erstes Heft, Untersuchungen zur
Entwickelungsgeschichte der Wurmer. Ill, Ueber die Ent-
wickelungder Chatopoden," [Leipzig, 1882, Leopold Voss.] The
observations descril>ed in the paper were made by the author at
Naples in 1880, and relate to the development of Nereis (Hetero-
21

JiD

dSp*

«fl D**" ~ - te *°

, WU-B0*" w**eot «<

p\»oe. * .^ tt> **|ai»t P»tl\ w be *°A*

*«ti*

teo

ci^T* *»*•

ittfo

ca^ *!? and *f J\***? S* ***

tf*"*^ *?£**•

ere* *~^ %>^*o«

\croflv

ctcB

i*^

tK«

~**rJZ* **

*■;*• 'e,j

*te° Y. a**"*- 1 «*** *«*«*• "Tot *ero' Z, a***"1

WOS - ^ * la* *"£, i^ e«*f V 0^6 rf *•

° «etioA «bc * ba» Cl"°7w true »>- , J)r
a** ** „ «ay,t0 *!» «p <* f •<,* ado?*** ;«a\ W*\ ate <*M

**» \C teSatA to *e J^t tbe * rf ^ fttu^ ^

be^^;c\eav^

€c\o^c0t 4

^%> oV>J^;^ to ^ "ft 9We- AeVCXoV- tV

o*et fott«^on . e

1 ^° -<-^«.^t:v-:tr.

it "^

r\to*ty

ti\\<

«ottt ?t^e>iV°°ai ^w W 8tftte
^elscb"^" .nation9

a p V a v>ec«

Mt««v,r^t»»wir,ift wt a.v«w»"

-rs: *2S5 ^

the

on
' a bV »^ ». 0V>servaW"''trad'^te'

dcv(

d."

coi

3»t\

\ft^«

POLYCHMT0V8 ANNELIDES. i\

EXPLANATION OF FIGURES.

PLATE XX.

iourx 1. — Unsegmented egg of Clymenella torquata, X 90.

IGURE8 2 to 4. — Formation of four primary blastomeres.

jgure 5. — Second resting stage from lower pole of egg.

iGURK 6. — Same, from upper pole.

igureb 7 tOa 9. — Separation of the micromeres.

igure 10. — Side view of an egg at the same stage.

jqure 11. — Third resting-stage, from upper pole.

nGURES 12 to IT. — Fourth period of activity, from upper pole.

J9GURE 18. — The same, from the lower pole.

GDRE8 19 and 20. — From the upper pole, thirty minutes later, to
show the separation of a small cell from the large pos-
terior spherule.

<atrRE 21. — Oblique side view, ninety minutes later.

ouau 22 and 23. — Lower and upper sides of another egg at about

the same stage.

Orb 24. — Lower side of another egg at about the same stage to
show division of posterior spherule.

* *aXJBES 25 and 26. — Lower and upper sides of a still later stage.

**<*Tjbe 2T. — Side view of last.

I**<^xjkb 28— The same, still later.
I*

be 28 a. — Longitudinal section of last stage, to show the large
posterior spherules and the absence of a segmentation
cavity.

be 29. — Embryo of fourteen hours ; the large polygonal entoderm
cells are visible in the anterior part.

x°xjbe 30. — Later stage with the layers well differentiated.

x**Tjbe 31. — Larva about twenty-eight hours old, with anterior belt

only.

*<*TJBE8 32 and 33. — A few hours later, viewed from side and from

dorsal surface respectively.

IqUke 34. — Head of young worm possessing eleven setigerous somites.

298 EDMUND B. WILSON.

PLATE XXI.

Figures 35 to 46. — Segmentation' of an egg of Arenicola crisiaia

viewed from lower pole; time, 1 boar, 43 minutes;
X 115.

Figures 47 to 53. — The same, seen from the upper pole ; time, 1 hoar,

30 minutes.

Figure 54. — Later stage, from lower pole.

Figures 55 and 56. — An older embryo from upper and lower sides

respectively.

Figures 57 and 58. — Larvae of 18 to 30 hours.

Figure 59. — Larva of three days.

Figures 60 and 61. — Larva of eight days, dorsal and lateral; in the

latter the proboscis is fully protruded.

Figure 62. — The same ; proboscis withdrawn.

Figure 89. — Young larva of Diopatra cuprea from jelly-mass ; X 60.

Figure 90. — Older larva from same jelly-mass with last, the body is

obscurely segmented ; X 40.

Figures 91 and 92. — The same; young worms a week later, dorsal

and ventral views ; X 30.

PLATE XXII.

Figure 63. — Fertilized egg of Chsetopterua pergamentaceus ; X 180.

Figure 64. — The same, after extension of the direction cells.

Figure 65 — The same, after the first cleavage, from the lower pole.

Figure 66. — The same, from the side.

Figures 67 to 70. — Further development, from lower pole.

Figure 71. — Separation of the micromeres, from upper pole.

Figure 71 (second). — The same, viewed from side.

Figure 72 — Third resting stage, from upper pole.

Figure 73. — Fourth cleavage, from side.

Figure 74. — The following quiescent stage.

Figure 75. — Fifth stage of activity, from side.

Figures 76 and 77. — Later stages of the same egg.

POLTCHJETOUS ANNELIDES. 299

Figure 79. — Larva of about 18 hoars, more highly magnified.

Figure 80. — Larva of 24 hoars, ventral view to show mouth.

Figure 81. — Larva of 40 hours, ventral.

Figure 82. — The same, from left side.

Figures 83 and 84. — Larvae of 5} days, lateral and dorsal views.

Figure 85. — Spiochaetopterus oculatus (?), lateral view of advanced

larva; X 40.

Figure 86. — The same ; dorsal view of buccal segment.

Figure 87. — Peculiar seta from fourth segment; X 360.

Figure 88. — Seta of ordinary form, anterior region.

PLATE XXIII.

Figure 1. — Larva of Clymenella torquata, 60 hours old; X 125.

Figures 2 and 3. — The same, five days old ; dorsal and lateral views ;

X 70.

Figure 4. — Larva of Arenicola cristata, four days old; X 215.

Figure 5. — The same ; fifteen days old.

Figure 6. — Larva of Chcetopterus pergamentaceus, 64 hours, from

right side, X 295.

Figures 7 and 8. — Larva of twelve days, lateral and ventral views.

Figure 9. — Spiochaeiopterus oculatus (?), free-swimming larva, from

left side ; X 50.

Figure 10. — Diopatra cuprea, somewhat advanced larva from jelly-
mass ; X 60.

[Note.— The cilia in Figures 2 and 3 are represented much too long ; in Figure 1 also they are
longer than in nature].

THE ORIGIN OP THE EGGS OP SALPA. By W. K.

BROOKS. With Pl^te XXIV.

In the summer of 1875 I enjoyed, through the kindness of Mr.
Alex. Agassiz, the privilege of spending several months at his
marine laboratory at Newport, R. I., and as specimens of Salpa
were very abundant, I devoted myself, at Mr. Agassiz's suggestion,
and with his assistance, to the study of their development. As
the result of my investigations I was led to the conclusion that
the eggs which undergo development inside the bodies of the
chain Salpse originate in an ovary contained in the body of the
solitary Salpa, and that the latter is therefore the female and the
chain Salpa a male; and that the life-history of Salpa is not a
case of alternation of generations. As this was my first effort in
the field of marine zoology, I should perhaps have hesitated to
publish a view so directly opposed to the conclusions of the many
famous naturalists who have contributed to our knowledge of
Salpa ; but as I was able to submit most of my specimens to Mr.
Agassiz's examination, I relied upon his judgment, and published
my results in a paper on "The Development of Salpa" in the
Bulletin of the Museum of Comparative Zoology, No. 14.

In this paper J showed that the eggs of the chain Salpse can be
traced back to a club-shaped ovary whidh lies inside the stolon of
the solitary Salpa, and which is present before the constrictions
appear on the walls of the stolon, to mark off the bodies of the
chain Salpa ; and on pages 386-337 I gave the following summary
of the subject: " Since the chain Salpa, at birth, always contains
a single un impregnated egg, organically connected with its body,
and since this egg, as well as the resulting embryo, is nourished
by the blood of the chain Salpa by means of a true placenta, and
since no reproductive organs have hitherto been described in the
solitary Salpa, it seems most natural to accept the view which
lias been generally held since the time of Chamisso's famous
paper ; that is, that Salpa presents an instance of ' alternation of
generations/ This view, in its most modern form, may be
stated as follows : 'It is now a settled fact that ike reproductive organs

301

302 W. K BROOKS.

are found only in the aggregated individuals of Salpa, while the
solitary individuals, which are produced from the fertilized eggs, have,
in place of sexual organs, a bud-stolon, and reproduce in the asexual
manner exclusively, by the formation of buds. Male and female
organs are, as far as we yet know, united in the Salpce in one indi-
vidual. The Salpce are hermaphrodite.' (LeuckarL Salpa u. Ver-
wandten, pp. 46, 47).

" When, however, we trace backward the history of one of the
individuals which compose a chain, and find that the egg is present
at all stages of growth, and has exactly the same size and appear-
ance as at the time when it is impregnated: when we find one
organ after another disappearing until at last we have nothing but
a faint constriction in the wall of the tube, indicating what is to
become the animal, the conclusion seems irresistible, that the ani-
mal, which has as yet no existence, cannot be the parent of the egg
which is already fully formed.

" The life-history of Salpa may then be stated briefly as follows :

"The solitary Salpa — female — produces a chain of males by
budding, and discharges an egg into each of these before birth.

" The eggs are impregnated while the zooids of the chain are
very small and sexually immature, and they develop into females
which give, rise to other males in the same way.

" Since both forms are the offspring of the female, the one by
budding and the other by true sexual reproduction, we have not
an instance of ' alternation of generations/ but a very remarkable
difference in the form and mode of origin of the sexes."

After I had finished my observations and while I was writing
my paper, Kowalevsky's paper on the development of Pyrosoma
(Zeit.f. wiss. ZooL, 1875) was published, and in this he also states
that the eggs of Salpa arise in an ovary which is contained in the
body of the solitary Salpa. " Bei den Salpen giebt es bekanutlich
zwei Generationen, in der einen entwickelt sich der aus vielen
Eikeimen bestehende Eierstdck, welcher in der Stolo hineingeht,
und sich hier zu je einem cinzigen Eie vertheilt, sodann die einzelnen
Knospen- resp. Ketten-Salpeu, in welchen weiter aus diesem Eie
ein Embryo entsteht, wieder mit einem aus mehreren Eikeimen
bestehenden Eierstock." Kowalevsky fails, however, to perceive
that the origin of the eggs in the body of the solitary Salpa ren-
ders this a female, as he goes on to say : " Bei Pyrosoma en thai t
jede Knospe auch wie die Kettensalpe das einzige groese Ei zur

ORIGIN OF THE EGGS OF SALPA. 808

unmittelbaren geschlechtlichen Verraehrung, und wie die Salpen-
Amme den Eierstock mit vielen Eikeimen zur Bildung der Ge-
$ehlechtsorgane der kiinftigen Knospen."

If Kowalevsky's statement and my own, that the eggs originate
in the body of the solitary Sal pa, are based upon sound observa-
tion, I do not see how my conclusion that the solitary Sal pa is a
female can be rejected, for that animal in which the eggs first
appear, as eggs, is certainly their mother; but soon after my paper
appeared Salensky published three very thorough and exhaustive
papers on the development of Salpa, "Ueber die Entwickelungs-
geschichte der Salpen," Zeit. f. wiss. Zool., XXVII, pp. 179-237,
Taf. XIV-XVI ; " Die Knospung der Salpen," Morph. Jahr-
buch, III, 4, and " Ueber die Entwickelung der Hoden und uber
den Generationswechsel der Salpen," Zeit. f. wiss. Zool., XXX,
276-293, Taf. XIII, and in these he says that Kowalevsky and I
are wrong in our statement that the eggs do originate, as eggs, in
the solitary Salpa. He acknowledges that the egg-cells can be
traced back to a mass of cells at the base of the stolon, but he
claims that they do not become eggs until they pass into the bodies
of the chain Salpa; that what I have called the ovary is not an
ovary at all, but simply a mass of undifferentiated embryonic cells,
which gives rise to the ovaries of the chain Salpse and also to their
digestive organs.

This discrepancy between his observations and my own has
rendered me very desirous of an opportunity to go over the ground
once more, to re-examine the subject for myself. For several
years I have been unable to do so, but last summer I requested
Professor Baird to try to obtain specimens of Salpa for me, and at
his request Professor Verrill collected a supply of specimens of a
very large new species, off Nantucket Island. These were care-
fully preserved for histological work, by Prof. Lee, and were sent to
me in the fall. I found that they were in excellent condition for
microscopic work, and I very soon obtained transverse sections
through the base of a very young stolon, showing fully developed
ovarian eggs in Salensky's "endoderm."

The great size and perfect preservation of the specimens enabled
me to obtain sections which have the greatest possible clearness,
and I soon found that while Salensky's figures give the general
anatomy of the stolon as shown in transverse sections, longitudi-
nal sections show that his account is very far from complete, and
22

804 W. K. BROOKS.

is, in some very important features, incorrect. I therefore at-
tempted to trace anew the whole history of the chain Salpa. This
involved the preparation of several thousand sections, and as it
was of the greatest importance that every section should be per-
fect, Dr. I. Bermann, of Baltimore, very kindly consented to stain
and imbed the specimens for me, by his process, and then, with the
greatest patience and interest in the work, to cut and mount the
necessary sections. I accordingly now have the material for a
very thorough description of the stolon and of the formation of
the chain Salpa, but as the preparation of a fully illustrated paper
will require considerable time, I have prepared this account of
those of my observations which bear upon the origin of the eggs,
and upon the question of alternation.

Salensky has given a very complete and clear statement of the
point at issue, and I will quote it, in full, as an introduction to the
description of the figures.

"Nachdem wir eben die Entwickelung der Salpenhoden kennen
gelernt haben, konnen wir nun auf Gfund der hier auseinander-
gesetzten Thatsachen, sowie deren, welche von mir an einem
anderen Orte fiber die Entwickelung des Eierstocks mitgetheilt
worden, zur Discussion der Frage fibergehen : gehort die Ent-
wickelung der Salpen zum Typus des Generationswechsels, oder
muss dieselbe an irgend eine andere Fortpflanzungsart angereiht
werden? Bevor wir aber zur Kritik der da ruber bestehenden
Meinungen schreiten, rafissen wir darauf Acht geben, dass bei der
Knospung der Salpen einige Eigenthumlichkeiten vorkommen,
welche der Salpenfortpflanzung einen ganz besonderen Character
geben. Das Wesentlichste von diesen Eigenthumlichkeiten bestebt
in der sehr frfihzeitigen Entwicklung der Eier in der Sal pen -
Knospe ; es ist bekannt, dass jede Kettensalpe noch lange bevor
die Kette vorn Mutterindividuum sich lostrennt, ein Ei bekommt,
welches be re its in einem ziemlich reifen Zustande vorhanden ist.
In keinem der bekannten Falle des Generationswechsels treffen
wir eine so fruhzeitige Entwicklung der Geschlechtsproducte, und
dieser Umstand hat, wie es scheint, als Beweggrund fur die An-
nahme gedient, dass die solitaren Salpen, welche man bisher als
ungeschlechtliche Formen betrachtet hat, weibliche Individuen
sind, dass sie aber ihre Eier in die von ihnen selbst prod ucir ten
Kettensalpen ablegen. Ist diese Annahme richtig, so muss die
Fortpflanzung der Salpen nicht als ein Fall des Generations-

w. 4

ORIGIN OF THE EGGS OF SALPA. 305

wechsels, sondern als eine ganz besondere Fortpflanzungserschei-
nung bctrachtet werdcn. Solche Meinung wurde von Brooks in
seinen von mir schon mehrmals citirten Auf'satzen iiber die Ent-
wicklung der Salpen ausgesprochen.

uNach der Meinung von Brooks hat die Fortpflanzung der
Salpen eine Analogie rnit der der Bieuen ; er findet diese Analogic
in der Art der Entwicklung der Geschlechter bei diesen beiden
Thiergruppcn.

"Wenn man selbst mit Brooks darin ubereinstimmt, dass die
solitaren Salpen weibliche, die Ket ten sal pen mannliche Individuen
darstellen, so kann man diese Analogic nur insofern bestehen
lassen, dass die Kettensalpen, wie die mannlichen Bienen, ohne
Befruchtung durch ungeschlechtliche Vermehrung entstehen, wah-
rend die solitaren Salpen, wie die weiblichen Bienen, aus dem
befruchteten Ei sich entwickeln. Weiter geht die Analogie nicht,
und der wesentlichste Punct der Sal pen vermehrung, namentlich
das hypothetische Ablegen der Eier von solitaren Salpen in die
mannlichen Kettensalpen, bleibt ohnedem ganz isolirt, denn ira
ganzen Thierreich treffen wir keine dem analoge Fortpflanzungs-
erscheinungen. Wo find en wir in der That eine Vermehrung, bei
welcher eine geschlechtliuhe Form ihre Eier in die Knospen,
welche sie selbst prcglucirt, ablege? Um eine derartige Fortpflan-
zungsweise fur die Salpen zuzu lassen, miisste man zuerst be we i sen,
dass die solitaren Salpen wirklich die Eierstocke oder deren Homo-
logon besitzen, und dass die Eier der Kettensalpen aus diesen
Eierstocken entstehen. Dies wurde durch keine Untersuchung
bewiesen. Brooks bestrebt sich zu beweisen, dass bei den Asci-
dien einige den bei Salpen vorkommenden analoge Fortpflan-
zungserscheinungen sich fin den, und dass die Eier dieser Thiere
genau in derselben Weise, wie er es fur die Salpen angiebt, von
einer Generation in die andere ubergehen. Er sagt dartiber Fol-
gendes: 'Die Zooiden der meisten Tunicaten sind hermaphroditisch
und entwickeln Eier aus ihrem eigenen Ovarium, aber, wenigstens
bei Pyrosoma, Perophora, Didemnium und Amauricium, ist das
Ei, welches die Befruchtung und Entwickelung in dem Korper
des Zooids erfahrt, nicht aus dem eigenen Ovarium, sondern von
dem der vorhergehenden Generation, und die Eier, welche im
Korper der zweiten Generation erzeugt werden, miissen in die
Korper der Zooiden der dritten Generation ubergehen, bevor sie
befruchtet werden konnen' (Arduf. Naturg., 1876, Heft 3, p. 353).

306 W. K. BROOKS.

"Ehe ich auf eine Behandlung der von Brooks angefuhrten
Ascidien weiter eingehe, will ich hier einige Beraerkungen iiber
die Analogie der Entwicklung der Sal pen und Ascidien im Allge-
meinen vorausschicken. Die Anald^ie, welche hauptsachlich die
Knospangserscheinungen dieser beiden Tunicatengruppen betrifft,
wurde von mir in meiner fruher citirten Schrift 'uber die Knos-
pung der Salpen' berucksichtigt. Sie besteht meiner Meinung
nach dar in, class an der Bildung des Keimstocks oder der Stolonen
der Salpen, so gut wie der Ascidien, die Derivate aller Keimblatter
theilnehmen. Diese Analogie wird aber bei der Bildung der
Athemhohle dieser beiden Tunicatenordnungen wesentlich gest5rt.
Bei den Ascidien bildet sich die Athemhohle als eine unmittelbare
Fortsetzung des gleichnamigen Gebildes des Mutterthieres, bei
den Salpen en stent dieselbe aus einer besonderen Anlage, welche
zugleich als Anlage des Eierstocks dient. Bei den Salpen giebt es
keine besondere Eierstocksanlage, und das ist ein sehr wesentlicher
Urastand, welcher den Grundsatzen der Brooks'schen Theorie
widerspricht. Wenn der Zellenklumpen, aus welchen die Eier-
stocke und die Athemhohlen der Kettensalpen entstehen, nur die
Anlage des Eierstocks darstellte, so konnte man denselben uuter
gewissen Umstanden als Eierstock der solitaren Salpen betrachten,
vorausgesetzt, dass er bei den solitaren Salpen im unentwickelten
Zustande existirt und erst in der Folge der Generation resp. bei
den Kettensalpen zur vol leu Entwicklung kommt; man konnte
aus diesem Grunde die solitare Salpe fur ein weibliches Indivi-
duum halten. Ist aber einmal bevviesen, dass im Keimstock der
Salpen keine besondere Eierstocksanlage existirt, so konnen wir
den Zellenklumpen, welcher nur theilweise in den Eierstock der
Kettensalpen iibergeht, nicht als Eierstock betrachten. Bei den
Ascidien ist aber, nach den Angaben von Kowalevsky u. A., eine
besondere Eierstocksanlage vorhanden, welche von der Anlage der
Athemhohle vollkommen different ist Das ist der wesentlichste
Unterschied in der Fortpflanzungsgeschichte beider Tunicaten-
gruppen, welcher geniigt, um zu beweisen, dass das Eierstocks-
rohr der Ascidien mit dem Entoderm der Salpen nicht homolog
ist.

"Aus altera oben Gesagten kann man den Schluss ziehen, dass
die solitaren Salpen keinen Eierstock besitzen ; da bei ihnen
gleichzeitig kein Hoden nachweisbar ist, so konnen dieselben als
Formen der uugeschlechtlichen Generation betrachtet werden.

ORIGIN OF THE E0G3 OF SALPA. 307

"Die Annahme der ungeschlechtlichen Natur der solitaren
Salpen kann schon allein fur die* Aufrechthaltung der fruherea
Theorie des Generationswechsels genugen, welche offenbar die
anderen Theorien, wie z. B. die von Brooks nnd Todaro, aus-
schliesst, und allein die Fortpflanzungsverhaltnisse der Salpen in
richtiger Weise darstellt." (Salensky. Entwicklung der Hoden
und vber den Generalionswechsel der Salpen, pp. 283-5).

Salensky's earlier paper (Die Knospung der Salpen) contains an
excellent account of the general anatomy of the stolon, so far as it
can be made out from transverse sections, but longitudinal sections
would have shown him that the digestive tracts of the chain Sal pee
appear very much earlier than he states, and that they are derived,
not from his "endoderra," but after the analogy of other Tuni-
cates from his " Athemrohr," with which, at first, they freely com-
municate.

If his specimens had been sufficiently well preserved to admit
of the examination of very thin sections with high powers he
would have found also that his "endoderm" was not simply an
"Eierstocksanlage" of embryonic cells, but a true ovary with fully
developed ova.

In support of these statements I shall now describe a few of my
own sections.

Figure 1, is a transverse section of the base of a very small
stolon, and represents the same stage as Salensky's Figure 3.
Figure 2 is a similar section of a somewhat older stolon., and is at
about the same stage as Salensky's Figure 10.

The stolon from which this section was cut had been a little
twisted, either by its own curvature or by the action of the pre-
serving fluid, so that the two sides are not symmetrical. Figure
3, is a transverse section of an older and larger stolon, upon which
the constrictions marking off the bodies of the chain Sal pas had
been formed, and it corresponds pretty closely to Salensky's
Figure 12. Like Figure 2 it is a little unsym metrical. Figure
4 is a vertical longitudinal section of the same part of another
stolon at the same stage, giving, as it passes through the bodies
of the chain Salpse, what is equivalent to a series of vertical
sections of Figure 3, along the numbered lines. That is the
line 1-2, in Figure 4, shows what would be seen in a section
of Figure 3, perpendicular to the plane of the paper on the
line 1-2. The line 3-4, in Figure 4, shows, in the same way,

308 W. K. BROOKS.

what we should have in a section of Figure 3, along the line
3-4, and so on. .

Figure 5, is a highly magnified view of a fragment of a section
through the ovary h9 of Figure 1, and Figure 6 is a longitudinal
section through the ovary of another stolon : the portion crossed
by the line 1, being of nearly the same age as the ovary A, of
Figure 1 ; the part crossed by the line 2, of about the same' age as
the ovary of Figure 2, and that crossed by the line 3, of about the
same age as that of Figure 3.

In all the figures, a, is the outer wall or ectoderm of the stolon
or of the chain Salp» ; 6, is the nerve tube, the Nervenrohr, Nf of
Salensky's figures and the tube yf of Figure 28 of my first paper.
1 there spoke of it as a second ovary, but my sections, as well as
those of Salensky show that this was an error: c and g are the
sinus tubes 1, 1, of Figure 28 of my first paper, Salensky's uBIut-
raume," Br. They seem to have been overlooked by Kowalevsky;
d, is the " central tube," Figure 28, 2, of my first paper, Salensky's
" Athemrohr/1 Arf and apparently Kowalevsky's "Darmrohr";
e and / are its thin upper and lower walls ; A, is the ovary, the
"ovary y" of Figure 28 of my first paper; Salensky's "en-
doderm En" and Kowalevsky's " Eieretocksrohr " ; i, t, the
"thickened edges 3, 3J of inner tube" of my original Figure 28,
the " Mesoderm if* " of Salensky, and the " KloakalrShren " of
Kowalevsky.

The greater part of the ovary h of Figure 1, is made up of a
granular ground-work in which are numerous transparent ovvidal
nucleated bodies, which at first sight appear to be cells. In Sa-
lensky's Figures 3, 4 and 7; they are represented as a compact
mass of cells, in contact with each other, and at first sight they do
appear to cover the whole surface of the section, but more careful
examination with a high power shows that only a few of them lie
in the plane of the section and that these are widely separated by
the granular substance, while between them, others at a lower level
are seen through this substance. As it is very difficult to represent,
at the same time, transparency and obscurity of outline, in black
pen drawing for reproduction by photo-lithography, I have only
drawn, in Figure 1, those which were in the plane of the section.
Between them there are very faint straight lines, mapping out the
granular substance into polygonal areas, with one of the transpa-
rent bodies near the centre of each. When a very thin section is

ORIGIN OF THE EGGS OF SALPA. 309

examined with a high power, Figure 5, each of the oval trans-
parent bodies is seen to be a germi native vesicle, with a nucleolus
suspended in its cavity by a protoplasmic reticulum of fine
branching threads; and surrounded by a granular layer of yolk
which is rendered angular and polygonal by the pressure of adja-
cent eggs. I have obtained a complete series of sections showing
the eggs at every stage, from the one just described, up to the time
when the single eggs are attached, by their gubernacula, to the
wall of the branchial sac of the chain Sal pa, and no one who ex-
amines the series, can doubt for an instant that the bodies in Figure
1, not only develop into eggs, but that they are actually eggs,
differing very slightly from the mature egg.

As we pass along the stolon, we find that the gerrainative vesicle
becomes a very little larger, the yolk grows more abundant and
the outline of each egg becomes more distinct and spherical, but
these slight changes are all, and before any traces of constrictions
appear on the surface of the stolon they have their mature form.
The ovary is surrounded by a layer of epithelial cells which are
thin and flattened at the sides, as shown at ra in Figure 5, while
at the point where the ovary touches the ectoderm they form a
thicker layer, Figure 6, m, which however, is only one cell deep.
This layer gradually increases in thickness, as shown in Figures 2
and 3, and when the constrictions appear and mark off the bodies of
the chain Salpae it becomes folded into a series of pouches, which
form the egg follicles, the so-called ovaries of the chain Salpae.
These pouches are what Salensky has wrongly interpreted as the
developing digestive tracts of the chain Salpae, but we shall see
farther on that the digestive tracts follow the analogy of the other
Tunicates and are developed from the walls of the large central
chamber of the stolon, Figure 1, d, Salensky's Athemrohr. Near
the internal surface of the ovary the epithelial layer changes its
character, as shown at m, in Figure 5 and 6, and on a smaller
scale at the top of h, in Figures 1 and 3. It becomes several cells
thick, and the cells become oval, transparent, with conspicuous
nuclei, and they resemble the germinative vesicles of the ova-
rian eggs in general appearance except that they "are smaller.
Figure 3 and n of Figure 6, show that this layer gradually dis-
appears as we pass towards the free end of the stolon, and when
the constrictions appear it is very thin or absent. Figure 1, «,
shows that the eggs nearest the internal edge of the ovary are

810 W. K. BROOKS.

smaller than those near its outer end, and this fact, together with
the fact that the layer n, is thickest at the base of the stolon, and
gradually disappears towards the free end, seem to show conclu-
sively that n is the germinal epithelium, the cells of which become
converted into eggs, which form a compact mass entirely filling
the lumen of the organ.

Full force cannot be given to the evidence without figuring the
eggs at all stages up to the time when the Salpa chain is discharged
from the body of the solitary Salpa, but I trust that the sections
which I have figured and described are enough to show con-
clusively that the body, h9 of Figure 1, is not an "Eierstock-
sanlage" but- a true ovary, arid that the cells, o, Figure 6, are not
undifferentiated embryonic endoderm cells, but ova. As no one
has ever claimed that the so-called ovary of the chain Salpa gives
rise to eggs, or ever contains more than a single egg, and as the
single egg which it does contain, is present, not as an embryonic
cell, but as an egg, in the ovary of the solitary Salpa, before the
chain Salpa comes into existence, I do not see how it is possible to
refuse to accept the conclusion that the solitary Salpa is the true
female, even if it were true that the ovary does also give rise to
the digestive organs of the chain Salpse, but this is not the case.

Figure 3 is a transverse section of a stolon on the sides of which
the constrictions are just beginning to appear. It is at almost
exactly the same stage as Salensky's Figure 12, although there are
slight differences, which are no doubt due to the fact that the two
sections are not from the same species. The most conspicuous
difference is due to the fact that the central tube, Figure 3, d, is
widely open, while in Sajensky's Figure 12, its upper and lower
surfaces are almost in contact and the cavity, Arf is nearly ob-
literated.

The sides of the stolon are formed by two thickened masses,
hj k} which, according to Salensky are masses of mesoderm cells,
Figure 12, Ms. In transverse sections they do have much the
appearance shown in his figure, but very careful examination of a
favorable section will show traces of a central cavity, shown on the
left in Figure 3, opening into the central cavity or " Athemrohr," d.
Longitudinal sections of the stolon show that, far from being an
unorganized mass of mesoderm cells, the body, A, actually has a
very complicated structure, and consists of a series of flat pouches,
the digestive tracts of the chain Salpee, which open into the central

ORIGIN OF THE EGGS OF SALPA. 811

tube, dy and which are separated from each other by infoldings of
the outer wall or ectoderm of the stolon.

These pouches are flattened so that it is almost impossible to
study them in transverse sections, but through the skill of Dr.
Bermann I have been able to get a complete series of sections
through the stage of Figure 3, in a vertical plane, perpendicular to
the paper. It is not necessary to figure all these sections for all
the points are shown in a single longitudinal section. A longi-
tudinal section passes, of course, through the bodies of a whole
series of chain Salpse, and as the stolon is always more or less
curved, such a section will not follow its central axis, but will cut
the bodies of the chain Salpae at different distances from the centre,
and it is plain that a section passing very obliquely through the
stolon from one side to the other, would give, on each side of the
central axis, what would be, in effect, a series of parallel and con-
secutive sections of the body of a single chain Salpa, although
actually, no two of these sections would pass through the body of
the same individual. Half of such a section is shown in Figure
4, and the vertical numbered lines indicate the axis of sections in
the planes of the numbered lines of Figure 3.

Along the line 1-2 we have first the ectoderm a of Figure 3;
then the upper blood-tube c; then the upper wall e of the central
tube or "Athemrohr;" then the cavity d of this tube; then its
lower wall /, and the lower blood-tube gf nearly filled by the
ovary A, which is made up, as in the transverse section Figure 3,
of an internal germinal epithelium, a mass of eggs, and a periph-
eral layer of epithelial cells. Along the line 3-4 we have the
same structures in the same order, but we also have at the top of
the figure a section of the nerve-tube b. As this is now broken
up, by the constrictions, into a series of chambers, it appears as a
tube, in longitudinal as well as in transverse sections. Near the
middle of the central tube d we also have a small slice from the
edge of the mass k of Figure 3, Salensky's mesoderm. Along
the line 5-6 we have the ectoderm a, the nerve-tube b and the
blood-tube c at the top of the figure, but below the latter we have,
in place of the central tube d, a section through the base of the
mass Jc, and this is now seen to consist of a central cavity p, which
opens into the tube d} and is bounded on each side by a single
layer of endoderm cells, which are continuous, around the edges
of the opening, with the walls of the central tube. -Along the
23

812 W. K. BROOKS.

line 7-8 we have this digestive pouch p as before! bat between it
and the pouch of the next chain Sal pa we have a fold of ecto-
derm, a, and between this and one side of the digestive poach, a
section of a structure, t, which is, in all probability, a portion of
the cloacal tube i of Figures 1 and 2. Along the line 9-10 and
the line 11-12, we have the same structures, but the bodies of
adjacent chain Salpse are more perfectly separated from each other
than they are nearer the axis of the stolon.

We have obtained hundreds of sections similar to the one shown
in Figure 4, and the presence of the digestive pouches at the stage
shown in Figure 3, and their communication with the central tube,
are points upon which there can be no doubt

Some of the sections show these points even more clearly than
Figure 4, and the only reason for selecting this section is that the
stolon from which it was cut was distorted almost exactly like
Figure 3, so that 3 and 4 not only resemble each other in general
structure, but in more minute features as well.

In the passage which has been qnoted, Salensky says, that if ^
the " Eierstocksanlage" did not also give rise to the digestives
tracts of the chain Salpse, and if it contained true eggs, instead o9
egg cells, the solitary Salpa might properly be regarded as me
female, and as I have shown that the digestive organs are realty^
formed from the central tube, while the ovary does contain tru
eggs, I think that the female nature of the solitary Salpa may
regarded as proven, and that we must conclude that we have i
Salpa not a case of the alternation of an asexual with an hernu
phrodite sexual generation, but simply a great and very anomalo
difference in the form and origin of the sexes.

While writing this paper I have received two papers on
development of Salpa (Neue Untersuchungen uber die embiy
Entwickelung der Salpen, Vorldufige Mittheilung, von Prof.
Salennky. Zool. Anzeiger, No. 97 u. 98, Nov. 28th, 1881, a
Mt moire mr les membranes embryonnaires des ScUpes, par le Dr.
Barrois. Journal de l'Anatomie et de la Physiologie, Dec. 28
1881), but as neither author treats of the origin of the chain Sa/I
or of the eggs, I have made no reference to them.

Baltimore, January 27th, 1882

ORIGIN OF THE EGOS OF SALPA. 313

EXPLANATION OF PLATE XXIV.

All the figures are from the stolon of an undescribed species of
ilpa, from the Atlantic, off Nantucket Island.
The reference letters 'have the following significance in all the
res:

a. Outer tube of stolon or ectoderm of chain Salpae.

6. Nerve-tube or ganglia of chain Salpae.

c. Upper blood- tube.

d. Central tube.

e. Upper wall of central tube.
/. Lower wall of central tube.
g. Lower blood-tube.
h. Ovary,
t. Cloacal tube.

k. Lateral thickenings of central tube of stolon to form the diges-
tive cavities of the chain Salpae.

m. Epithelium of ovary.

n. Germinal epithelium of ovary.

o. Eggs.

p. Digestive cavities of chain Salpae.

Iqure 1. — Transverse section through the base of a very young

stolon. Zeiss, D, 2.

**aURi 2. — Transverse section of an older stolon, a little further from

base. Zeiss, D, 2.

Iqure 3. — Transverse section still further from base.

*«3URE 4. — Half of an oblique vertical section through Figure 3.

Zeiss, D, 2.

*oure 5. — Fragment of a very thin section of the ovary of Figure 1.

Zeiss, F, 2.

^*oure 6.-— Longitudinal section of the ovary of a young stolon.

The line 1 passes through a portion which is in nearly
the same stage as h of Figure I ; the line 2 through
a portion like Figure 2, h, and the line 3 through a
portion like Figure 3, h.

'Nl8 REP 82^

> ■

SERVATIONS ON THE MEAN PRESSURE AND
THE CHARACTERS OF THE PULSE- WAVE IN
THE CORONARY ARTERIES OF THE HEART.

By H. NEWELL MARTIN, M. A., M. D., D. 8a, and W. T.
SEDGWICK, Ph.D. With Hates XXV, XXVI and XXVII.

tH&*

hile for a c6nsiderable number of years careful studies of the
Wood-flow in various arteries of the mammalian body have been
e under different conditions, the arteries of the heart itself
remained in an exceptional position. The average pressure
the pulse characters in them have been unknown, in spite of
recognized fact that great interest and importance belong to
r study.

lie following pages give an account of experiments under-

n with the object of gaining some knowledge of these points,

contain, we believe, a description of the first successful

xnpt to record graphically, as in other arteries, the blood-

isure and its variations in the arteries of the heart. They

e begun in the first place for the purpose of testing the theory

hebesius — a theory independently propounded and warmly

SriF>3ported in recent times by Briicke, and others, concerning the

P^^^eiology of the aortic semilunar valves. According to this

t«o<i>Ty, during ventricular systole the thin flaps of the valve are

eed upwards and cover the mouths of the coronary arteries,

jletely closing them, so that blood can enter those vessels

0X1 *,y during the time of ventricular diastole, and during that

8tri***H portion of the systolic period which is occupied by the

Va,*"V^ in travelling from its diastolic position across the mouth of

**° aaorta, to its systolic position against the aortic wall and over

r1^ mouths of the coronaries. Observations on the spirting of

°^>d from a cut coronary artery have shown this to be synchro-

^^^ with systole of the ventricle; but to the value of these

^^^rvations Briicke' has raised two objections. First, that

ir*^x*ely opening the pericardium is enough to destroy the normal

1 Vorlemngen, 1881, S. 185.

51 *

#•

tf£

WBLJt

JfA

fiftf

° 15, ">*4i W-*"^ Wool ** 0„«e •*»

V-

the

B"' "^^^e c**^ g^o»> ;:fl^ 1W* I^ cot»-

688^*

object

* tfr-XS* arzs* ?- oi

\&> *P ^ra

**-s. «s». -jeS- ?r£ » *^rc

T0«6t Tre o**eT*

;6tVOO»

cft»

detet

'^tSxe

the ^C
toog

AtU»R

tf**0"1, "^r ft ©

^t0tf e»t^ V—Aed for

ot»¥'

oTXtY^T^^txoe*-

»06»^

60T»e sy o effect* ^ „^u vu _aa» far

cp&

»t\ott

eft*

dW«

«#*

i<; "ftt^^O °* *t to *°* " y*****

tceo

,td8

o« tJ»e.

. e to ti^6 "Ttacfce ^

coto

jBO 6

lAoti^ -Jto »ttd ?Tftrt *»* »A for o** ''U H»«f

PULSE-WA YE IN THE CORONARY ARTERY. 317

Ration, however, have forced us to believe that the semilunar
"valves do not act as Briicke supposes, and that his theory is no
longer tenable. Apart, however, from this point, we venture to
"believe that the work possesses interest of its own ; and that the
<3iscovery that it is quite possible to get tracings of the blood*
^pressure in the arteries of the dog's heart, lays open a consider-
able field for investigations upon the mammalian heart in general

an organ which has hitherto been somewhat baffling to the

physiologist.

Our experiments have all been made on dogs placed under
^he influence of a full, or rather an extreme, dose of morphia —
:rfrom one to two grams of the acetate given subcutaneously in
eatery solution. While this drug greatly slows the respira-
tions, and somewhat later, to a certain extent, the rate of the
Xieart's beat, it seems in no way to impair the vitality of this
^Drgan ; if anything it appears rather to increase its capacity for
aring insults— a matter deserving of further investigation. The
nimal having been put very completely under the influence of
~^he drug, tracheotomy was performed, a cannula placed in one
carotid artery, and the pneumogastric nerve of the same side
exposed and divided so that its peripheral end was ready for
stimulation.

An incision was then made in the middle line along the
^nanubrium of the sternum; the muscles, &c, were dissected
^rom the first pair of costal cartilages, and (the apparatus for
•artificial respiration having been connected with the windpipe)
f;he cartilages of the first pair of ribs and the bit of sternum
'fcetween them were removed, thus laying bare the apex of the
^hest cavity, which was then opened. The artificial respiration
"was now stopped for a few seconds, so that the lungs might
collapse and thus expose on each side the internal mammary
wtery, running along the exterior of the mediastinum and the
remnant of the thymus, to the ventral aspect of the chest wall
opposite the second costal cartilage. These arteries having been
tied, the incision along the middle line was prolonged backwards
and the skin and muscles reflected on each side so as to expose
the rib cartilages. This operation is usually accompanied by
only an inconsiderable venous oozing, after the internal mam-
mary arteries have been secured in the manner just men*
tioned.

318 H. NEWELL MARTIN AND W T. SEDGWICK.

The sternum and costal cartilages were then removed, care of
course being taken not to injure the lungs. The next step is
to stitch the pericardium to the chest wall in order to support
the heart and prevent its receding too much when the lungs
empty during expiration.

Branches of the coronary artery can now be seen through the
pericardium, and a window is so cut in that membrane as to
expose a branch which seems suitable, while all the rest of the
heart remains protected and supported by its sac.

So far the operative procedures are tedious but present no
special difficulty; but to lay bare the coronary branch and to
fix the cannula in it while the heart continues to beat is much
more troublesome, since any carelessness in these operations is
apt so far to injure the heart as to destroy its normal beat and
throw the ventricles into incoordinate fibrillar contractions, from
which we ha««e never seen them recover. The success of the
attempt depends largely on the animal ; in the most favorable
cases the left coronary artery, after giving off its transverse
branch, which runs along the auriculo-ventricular groove, passes
along the septum ventriculorum on the ventral aspect of the
heart, and gives off near the base of the ventricle a considerable
branch to the right, which runs with a vein on each side of it,
and is covered only by the visceral layer of the pericardium and
some fat. Into this branch the cannula is inserted, and the
blood carried by the main trunk and its remaining branches
serves perfectly to keep the heart beating vigorously for several
hours, as we have repeatedly found. In other cases the artery
does not give off* this one main branch, but (especially in large
dogs) runs along the ventricles, giving off small twigs right and
left which are too minute for the convenient introduction of a
cannula, and are, moreover, often covered by a thin layer of the
musculature of the heart in addition to the pericardium. This
muscular layer adds greatly to the difficulty of successfully
isolating the artery, for any wound to the proper cardiac sub-
stance about the vessels seems more fatal to the organ than any-
thing else. Soon after such an injury it almost invariably
exhibits periodic beats for. a short time, and then the ventricle
passes into a state of fibrillar contraction. The well-known fact
that needles may be thrust into many parts of the heart without

PULSE- WA VE IN THE CORONARY ARTERY. 319

essentially influencing its beat for a long time, inclines us to the
belief that the result in the cases to which we refer is, perhaps,
due to the injury of nerve trunks which may run in the heart
near its arteries and which are torn with the muscle, rather than
to direct injury of the muscular substance; but we have not yet
had an opportunity to examine this point.

A suitable coronary branch having been found, the next step
is the most difficult in the operation, viz., to tear through the
visceral pericardium over the artery without opening that vessel
or its accompanying'veins; for the membrane is so smooth and
tightly stretched that it is not easy to catch hold of; and then so
tough that it is difficult to penetrate. Our method is as follows :
All being ready, the pneumogastric trunk is stimulated so as to
stop the heart's beat, and the artificial respiration simultaneously
suspended so as to avoid movements of the heart due to con-
tractions and expansions of the lungs. With a'sharp-pointed
pair of forceps the pericardium over the artery is seized and a
hole torn through it by means of a needle ; once this aperture is
made through the tough membrane without injuring any of
the vessels, the rest of the operation is comparatively easy. The
stimulation of the pneumogastric is stopped and the artificial
respiration resumed for a moment or two ; then the heart-beat
and breathing are again suspended, the edge of the hole is taken
in the forceps and the membrane over the artery slit up toward
the base of heart by a very fine-bladed knife. From time to
time, as the heart begins to beat in spite of stimulation of the
pneumogastric, the nerve is allowed to rest and the respiration is
resumed, and in this way the alternate stimulation and rest are
repeated as often as may be necessary in order to expose a suffi-
cient length of the artery, to place ligatures around it, and insert
a cannula in the manner adopted for any other artery. The
carotid was then connected with one mercury manometer, the
coronary branch with another, and, the pens being arranged so
as to write exactly over one another, tracings were taken on the
kymographion.

The mode of connection of the arteries with the manometers
demands a word. In the first place, the three inches of the arte-
rial end of the connecting tube between the coronary and its
manometer consist of highly flexible rubber tubing. This, no

320 H. NEWELL MARTIN AND W. T. SEDGWICK.

doubt, sligbtly modifies the pulse-waves on the tracing, but it
gives to the heart free play during each beat, since the flexible
tube offers no restraint, but yields readily. This soft tubing is
succeeded by a glass tube, which is firmly held by a solid support,
so that no locomotion of the tubing occurs beyond this point.

Movement of the bit of flexible tubing attached to the cannula
does slightly alter the level of mercury in the manometer, but, as
we have satisfied ourselves by careful examination, causes no
feature in the tracing which can be mistaken for a pulse-wave.
Beyond the piece of glass tubing mentioned above, the connect-
ing arrangement is similar for the two arteries.

To get a true base-line, or line of no pressure, for each mano-
meter gave us some little trouble. The base-line is often taken
as that drawn by the pen when the mercury stands at the same
height in both legs of the manometer, but this is seldom correct.
If the end of the connecting apparatus attached to the artery be
above the level of the mercury in the limb of the manometer
with which it is joined, the weight of the liquid in it will affect
that level, making it sink in the nearer and, of course, rise in the
farther limb which bears the pen. If, on the other hand, as is
more often the case, the arterial end of the connecting tube be
below the level of the mercury in tho gauge, the tube acts like a
siphon-tube ; the mercury rises somewhat in the proximal limb,
and sinks to the same extent in that which carries the pen, so
that in either case the baBe-line drawn with the two mercury
columns level will be incorrect.

As we wished especially to compare the amount of arterial
pressure in the coronary with that in the carotid, we had to elimi-
nate such errors, and the more so because the manometer attached
to the coronary artery was invariably above the one connected
with the carotid, and so the siphon action (for the ends of the
tubes farthest from the kymographion were always below the
levels of the mercury in the manometers) was considerably
greater. The method which we adopted gives, we think, abso-
lutely true results. Having finished an experiment, we stopped
the artificial respiration, and let the animal die of asphyxia, the
manometers being meanwhile shut off from connection with the
arterial system. When the animal was quite dead, and all traces
of arterial pressure had disappeared, the communication with the

PULSE-WA VE IN THE CORONARY ARTERY. 321

manometers was again opened, and the pens naturally fell with
the mercury to the level which corresponded to zero arterial
pressure: we, of course, satisfied ourselves that there were no
clots in the apparatus. The pens were then turned away from
the paper, which was next re-coiled on the drum until the begin-
ning of the record of the experiment was reached ; then, the pens
being turned back again, the kymographion was started once
more and each pen drew its own base line, being still connected
with its artery and the position of the animal being the same as
during the experiment. It has been suggested to us that the base
line so obtained may not be reliable, as some arterial pressure
might still remain in either the carotid or coronary vessel, or in
both, after general death ; but this objection we think will not
;ar examination. After death from asphyxia, as is well known,
e arterial system, at least in its larger trunks, is extremely
mpty ; a few minutes after its occurrence one may cut the aorta
without the slightest spirt of blood resulting, and, indeed, even
almost without bleeding at all ; and the carotids, snbclavians,
and other large arterial trunks are obviously collapsed and empty.
That under such circumstances there should be any arterial pres-
sure possibly remaining in arteries in free and direct connection
with the aorta is not conceivable.

A description of the tracings taken on the kymographion (Figs.
1—5, PL XXV, XX VI, XX VII) will serve best to show our
results. The tracings, in fact, speak for themselves, and have been
selected from a considerable number which all perfectly agree with
them as to the conclusions to which they lead ; we have never ob-
tained a single contradictory record. The pulse synchronism and
the similarity of the pulse-waves in the carotid and coronary under
different amounts of blood-pressure and with various rates of
heart-beat is remarkable throughout. In Fig. 1, PI. XXV, we
have a pulse-rate of 132 per minute, and complete synchronism in
the two arteries ; the mean pressure in the former being 62 mm.
of Hg. and in the latter 42. The verticals, vv, cut all the tracings
at points corresponding to the 6ame instant of time. In Fig. 2,
PI. XXVI, is a tracing taken with a quicker pulse, about 172 per
minute. At v', artificial respiration was stopped so as to get a
dyspnoeic rise of arterial pressure. As the verticals show, this
does not disturb in the least the synchronism or similarity of the

322 H. NEWELL MARTIN AND W. T. SEDGWICK

pulse-waves in the two arteries. Mean pressure in coronary, 46
mm. of Hg., and in carotid 56, at the beginning, rising to 100
mm. and 120 mm. respectively just before v'".

Fig. 3, PL XXVI, gives simultaneous tracings from the two
arteries during extreme dyspnoea, with greatly slowed pulse and
very high blood-pressure, rising in the part of the tracing given
to 120 mm. of Hg. in the coronary artery and to 132 in the
carotid. Ultimately the pressure rose still higher, and drove the
pen attached to the coronary vessel off the top of the paper, so
that a record could not be obtained. The accuracy with which
each tracing reproduced the other during all the variations of
pressure and pulse-rate which occurred during this observation
is very remarkable, and seems to make it certain that the pressure
in each artery is directly determined by the same cause, viz.,
aortic pressure. The contracting ventricle might conceivably
increase pressure in the coronary vessels by compressing them ;
but variations thus produced cannot possibly be imagined as
agreeing 60 perfectly with the variations in carotid pressure
(which, on such a theory, must be differently produced and sus-
tained) as do those given in this figure.

Unfortunately a seconds pen was not connected with the
kymographion on this occasion, so that the pulse-rate cannot be
stated accurately ; but by taking an average from the rate of
movement in other cases it may be set down, without any great
error, as about 60.

In Fig. 4, PL XXYII, is given a tracing taken soon after the
resumption of artificial respiration, which had been interrupted
long enough to produce (as seen to the right of the tracing) a
considerable dyspnceic rise of arterial pressure. Well marked
and similar Traube's curves are seen on each tracing, and also
the synchronous pulse in both arteries. This synchronism is
maintained throughout all changes of cardiac rhythm and blood-
pressure.

In Fig. 5, PL XXVII, is a tracing in which the coronary pres-
sure is higher than the carotid (76 mm. against 64 mm. Hg.) This
may perhaps be due to our having taken in this case a coronary
branch nearer the main stem than usual ; but it may be also,
and more likely is, due to the vasomotors. The heart arteries
have a very active system of these nerves, as any one who ex-

PULSE- WA VE IN THE CORONARY ARTERY. 323

periments with them will soon observe. Not unfrequently on
laying bare a coronary branch that seemed suitable for insert-
ing the cannula we have found it apparently so small that our
endeavor seemed hopeless; and then in a minute or two it
would dilate again to at least double its previous diameter. If
it be borne in mind that the coronary branch used was always
but a small twig of the whole coronary system, it seems pos-
sible that great constriction in the rest of the branches might so
oppose the blood-flow as to raise the pressure almost up to
that in the aortic arch, and so bring it above that in the carotid.1
In other respects the tracing illustrates the same points as those
reproduced in the preceding figures. The heart was beating 148
per minute.

We find then that whether the heart beats slow or fast, and
whether arterial pressure be high or low, every feature of the
carotid pulse is simultaneously given in the coronary. No doubt,
with a faster-travelling roll of paper the synchronism would not
be perfect, as the carotid vessel is farther from the heart, but the
pulse-wave travels so fast that this could not be expected to be
shown on the kymograph.

There is, however, no trace of any alternation in the pulse-
waves, such as would seem necessarily to follow from an occlu-
sion of the mouths of the coronary arteries during the ventricular
systole, and such as, if it existed, the kymograph would cer-
tainly show.

The argument which was used effectively against conclusions
drawn from observations upon spirting coronary arteries, may be
brought perhaps to bear upon our work, viz., that in the earliest
stage of contraction of the ventricle, the coronary shares with
the carotid the general rise of pressure in the arterial system,
because the valve has not yet closed over its mouth ; and that,
in consequence, it is to be expected that the two pens which have
travelled together during the diastole of the previous undulation

1 We have recently endeavored to discover the source of the vaso-constrictor
nerves of the heart, by connecting cannula with carotid and coronary arteries,
and then observing if a relative rise of coronary pressure could be brought about
by stimulating extrinsic cardiac nerves. So far our experiments have been con-
fined to the accelerators and have been entirely negative. We got the accelera-
tion of the pulse-rate, but no rise or fall in coronary pressure which was not
exactly duplicated on the tracing from the carotid manometer.

324 K NEWELL MARTIN AND W. T. SEDGWICK.

shall together begin thoir systolic journey on the new pulse-
wave. This is, no doubt, quite true, and we have no objection
to the argument as far as it goes. It leaves off, however, where
our work begins, and does not affect the real point of the ques-
tion, though it emphasizes the necessity for exact tracings which
can be studied leisurely.

Since the coronary artery is freely exposed to aortic pressure
during all of the diastole, and during the first fraction of the
systole of the ventricle, we are not surprised to find on the trac-
ings, at that time, complete agreement between carotid and coro-
nary pulses ; they are caused by the same thing and are there-
fore similar. If now we turn to the tracings described during
the major portion of the systolic period, and find them duplicates
one of the other, alike in form and synchronous in characters, it
is hard to believe that they also are not directly dependent on
the same immediate cause, i. e. aortic pressure. For if the
valve closes as Brucke believes, the forces acting upon the two
arterial contents are no longer identical; the carotid is still
marking an increasing pressure due to the outflow of blood from
the energetically contracting ventricle ; but the coronary, cut off
by the valve from influx of blood, is put under other conditions.
It is not snpposable that the ventricle acting upon the carotid
directly through the aorta should cause it to trace a pressure
curve precisely like one drawn at the same time by the coronary,
upon which it is acting only indirectly (i.e. by raising intraven-
tricular pressure, and so causing extra compression of the vessels
in the heart substance). Nor is it conceivable that the coronary
artery should have its mouth suddenly closed at one instant dur-
ing the period of rising pulse-wave, and still go on tracing un-
disturbed a uniform rise of pressure. Under such circumstances
some deformation of the coronary curve, some irregularity in the
tracing, must take place.

Again, after the systole is over and the valves rebound to
their position over the mouth of the aorta, a moment would
come (when the period of highest carotid pressure was just past)
when the coronary artery would suddenly be opened and blood
would be driven into it. An injection of blood into the pre-
viously closed coronary system at this moment ought surely
(even if it did not, as may be urged, raise arterial pressure in

PUL8E-WA VE IN THE CORONARY ARTERY. 325

•

the coronary artery, becauee the cardiac muscle was relaxing
and making the coronary circuit easier of passage) to show
itself in some break or rise, or other special feature in the pres-
sure-changes at that moment occurring in the vessel ; the tracing
from the coronary vessel (now for the first time receiving blood)
could not exactly agree in every respect with the tracing
from the carotid artery, which is simultaneously emptying itself
steadily and regularly under the force of arterial elasticity.
We find, however, nowhere any indication of such a difference
of events ; the coronary tracing is always a duplicate of the
carotid under all circumstances, and there is no sign of any
periods when great circulatory changes (such as are involved
in the supposition that the mouths of the coronary vessels are
alternately closed and opened) are taking place in the coronary
artery.

We are therefore forced to conclude that they are in the right
who have maintained that the flaps of the semilunar valve are
never pressed completely back against the aortic wall during
systole of the ventricle. Finally we may point out that the trac-
ings show the pressure-changesin the coronary system to be very
much like those in any other branch of the aortic system — the
carotid for example. It may be added in conclusion that though
forced to differ from Brucke, in regard to any interference of the
semilunar valve with the circulation in the coronary system, our
observations in no way contradict his teaching that during ven-
tricular diastole blood flowing into the coronary arteries aids in
distending the flaccid heart. This is probably true. The com-
plete "Selhststeuerung" is, however, no longer tenable; the
arteries of the heart are not emptied during the ventricular dias-
tole, so as to diminish the resistance to cardiac contraction, but
are at that time always tensely filled. Moreover, a6 our tracings
show, the little increment of pressure during the systole of
a single beat, when compared with the entire mean pressure con-
stantly at work in the coronary system, is so small that not much
would be gained by blocking the mouths of the arteries in order
to avoid it.

THE INFLUENCE OF DIGITALINE ON THE
WORK DONE BY THE HEART OF THE SLIDER
TERRAPIN, (Pseudemys rugosa, Sliawj By H. H. DON-
ALDSON, A. B., Fellow of the Johns Hopkins University, and
MACTIER WARFIELD, A. B.

The experiments described in this paper were undertaken as a
preliminary to an examination of the action of digitaline upon
the isolated heart of a mammal. On examining the literature of
the subject we found so much confusion and contradiction, and
so frequently methods of experimentation which seemed open to
objection, that we concluded it better to investigate afresh the
action of the drug on the isolated heart of a cold-blooded animal
before proceeding to study its influence upon the heart of the dog.

Though observations were made on the pulse-rate, arterial
pressure, and the changes in the form and size of the heart, we
wish now to develope only our results on the variations in the work
done under digitaline, and shall therefore only make use of the
above observations when they bear on that question.

It is quite agreed that digitaline has the same action on the
heart whether that organ be isolated from the central nervous
system or not. * 2 8

In mammals with the central nervous system intact, moderate
doses of digitaline are observed to cause a rise in mean blood-
pressure, which persists during the slow pulse.8 * 5 6 7 8*

Winogradoff 9 states on the other hand that the mean blood-
pressure found in dogs is not noticeably modified by moderate
doses of digitaline.

A contraction of the arterioles in the web and mesentery of
the frog and the mesentery of the rabbit under digitaline has
been observed by many s 7 10 u w ; others, however, have failed to
find it.

Brunton and Meyer 18 obtained curves from a dog under mor-
phia, which led them to maintain that the rise of pressure was
due solely to the narrowing of the arterioles.

327

328 H. H. DONALDSON AND MACTIER WARFIELD.

The suggestion having been made that this narrowing of the
arterioles was one cause, at least, of the rise in arterial pressure,
experiments were undertaken to test the point.

The results have been by no means concordant. Von Bezold"
cut the cord in an animal showing high blood-pressure under
digitaline. The pressure at once fell markedly, but it was yet a
question whether it was as low as it would have been without digi-
taline.

To answer this it was necessary first to sever the cord and then
inject the digitaline. Traube 4 crushed the cervical cord and
could get no rise of pressure by the subsequent injection of digi-
taline.

Bohm using the same method on rabbits obtained the same
results. But when, in an animal without its brain and spinal
cord, he first ligatured the thoracic aorta above its inferior
branches, and then injected digitaline, he obtained a decided rise
of pressure.15

Having thus cut off much of the arterial system he interpreted
the rise in pressure observed, not as a narrowing of the arterioles,
but as an increase in the work done by the heart.

Gorz 16 found after section of the cord a slight increase in
pressure under a subsequent dose of digitaline. This rise he also
attributed to an increase in work.

Ackermann u also states that he has often cut the cord and
then found a decided rise of pressure to follow digitaline.

Attention was then directed to the heart. In 1879 Bohm1
published an extensive article on the physiological action of digi-
taline. As has been mentioned, he could get in rabbits no rise
of pressure after section of the cord. He argues, however, in this
case, that the work done by the heart might have increased, and
yet the extreme relaxation of the vessels prevented its expression
as a rise of pressure. Moreover, he failed to observe a decisive
narrowing of the arterioles in the frog, and, finally, direct experi-
ments on the isolated heart of the frog led him to conclude that
the work under digitaline was increased by moderate doses
(.0005-.001 grm.), and decreased by large ones.

He used a " Lud wig-Coats " apparatus for feeding the heart
and the formulas of Blasius 1T for estimating the work. His
cannulas were tied in the vena cava and bulbus aorte. Phong

DIGIT A LINE ON THE HEART. 329

the arterial cannula in the bulbus almost certainly interfered
with the valves and thus introduced an important modification
into the circulation. The distance through which the blood was
raised varied in different experiments between 6 and 34 cm.
He does not state, however, his venous pressures, nor precisely
his doses of digitaline. Experimenting thus he obtained under
moderate doses of digitaline, an increase of work, lasting in one
case 23 minutes ; coincident with a slower pulse-rate. The work
then decreased and never again reached its original amount
Each experiment occupied about an hour.

When, however, so high an arterial pressure was U6ed that
the work began to decrease, digitaline was unable to prevent the
decrease. His final opinion as deduced from his investigations
is as follows :

"Jedenfalls abcr glaube ich durch die zuletzt mitgetheilten
Versuche bewiesen zu haben dass die bei der Digitalinwirknng
beobachtete Blutdrucksteigerung auf direckte Vemehrung der
vom Herzen geleisteten Arbeit zuruckzufuhren ist."

In 1880 Williams18 published some investigations on the
rise of pressure under digitaline. He finds, as did Bohm,2 that
the drug is incapable of increasing the maximal pressure against
which the heart can work. Moreover, he gets no evidence of
the narrowing of the arterioles under digitaline.

He observes that when the heart is working against a higher
pressure it undergoes a greater diastolic expansion, venous pres-
sure remaining the same, and does more work.

Digitaline he thinks affects the heart muscle like high pres-
sure, and then causes a rise in mean pressure through a variation
in the extensibility of the heart muscle. He made no direct
measurements of work.

This review of the literature indicates that further investiga-
tion of the question of work was not entirely superfluous.

The experiments which we have to record were made in the
following way.

Apparatus.

This was designed to keep the pressures (both arterial and
venous) and the temperature always constant, at the same
time to record the form and rate of the pulse, allow the estima-

330 H. H. DONALDSON AND MACTIER WARFIELD.

tion of the work done, and permit direct observation of the form
and movements of the heart.

The venous reservoirs consisted of three flasks, arranged as
Mariotte's bottles, each holding about 400 cc. The flow from
each flask was through a rubber tube. By the use of two Y
pieces, and two other bits of tubing, the three tubes are combined
so that the blood from all the flasks flows finally through one tube,
which is connected with the venous cannula. The flow from any
flask can be stopped by a clamp. In experimenting, the middle
flask, graduated for every hundred cc., always held the blood
which contained digitaline. The others held what we call "good "
blood, to distinguish it from the above ; only one flask was used
at any time. We had three pairs of cannulas varying in diame-
ter— the difference in the size of the terrapins making it impos-
sible to always use the same cannulas.

In order to reach the venous cannula the common inflow tube
passed through the end of a box. This was mainly of wood,
with a glass top. It always contained a thermometer. In this
box the animal rested firmly on its back. It was thus protected
from draughts, too rapid evaporation, and mechanical injury, and
yet always readily observable.

From the end of the box opposite the inflow passed the out-
flow tube. This was a tube of stiff rubber. First connected
with the arterial cannula, it then passed through the end of the
box, and at once branched.

One branch ended in a bit of glass tubing which was fastened
in a clamp that moved on an upright, and could thus be fixed at
any desired height ; this was the outflow. The blood as it was
pumped out was caught in small beakers for a known time and
then measured ; this gave us the means of estimating the work.
The other branch continued for about thirty cm. when it again
divided, one branch in this case being connected with a pressure
bottle filled with .75 per cent, salt solution, and the other joined
a manometer. In our earlier experiments a small mercury
manometer was used, but in all the later experiments a water
manometer, as described by Howell and Warfield,19 was preferred.
The manometer wrote on the smoked paper of a revolving drum.
Time was marked by an electric pen connected with a clock
beating seconds.

DIGITALINE ON THE HEART. 331

The apparatus being ready, a terrapin was weighed, a cord
tied tightly about the neck, the head cut off, the plastron re-
moved, and the pericardium opened. The animal was now
placed for a moment in the box, and the height of the heart
above the table measured ; then, while the operation was being
completed, one of us arranged the inflow and outflow of the ap-
paratus to give the desired pressures.

With the least possible handling all the vessels of the heart
except two were then tied. The two generally used were the
right aorta and the left vena cava superior. The left aorta and
the vena cava inferior were, however, sometimes taken. When
the cannulas were secured, one hundred cc. of pure defibrinated
blood were sent through the heart, in order to wash out the con-
tained blood which was liable to clot. Inflow and outflow were
then clamped, the animal pithed and put in the box. The inflow
tube through which blood from one flask was running was
slipped over the venous cannula, while the arterial cannula was
connected with the outflow. The circulation was thus estab-
lished.

For feeding the heart we used fresh defibrinated calf s or sheep's
blood mixed with its own volume of .75 per cent, salt solution.
This mixture we designate " good blood." When digitaline is
added to it, " poisoned blood."

The terrapins weighed between 437 grams and 1785 grams.
Initial temperature was between 13°-21°C, with a maximum
variation of 4°C. during an experiment.

The venous pressure varied between 2.7 cm. and 7 cm. The
arterial pressure was always 20 cm. of the blood circulated
through the heart. The feeding flasks sent through the medium-
sized venous cannula, under 3 cm. pressure, 1 cc. in 1 sec. when
the cannula was disconnected from the heart. The digitaline
used was prepared by Merck. It was amorphous, and gave a
slightly turbid solution with water or .75 per cent, salt solution.
The amounts administered were from .00035 gram to .005 gram
in 100 cc. of the diluted blood.

The digitaline was first dissolved — .001 gram in 1 cc. of water
or .75 per cent, salt solution, and then added with a pipette to
the flask half filled with diluted blood. The rest of the blood
was then poured in, and thus a fairly even mixture was secured.

332 H. H. DONALDSON AND MACTIER WARFIELD.

The time during which each observation lasted varied from 4
to 11 hours.

There is one unevenness in the apparatus to be mentioned :
When one of the flasks is to be refilled with blood its outflow
tube is clamped ; two tubes running through the cork are then
undamped, and through one the blood is poured, while the air
escapes through the other. The flask being filled, the two tubes
are again clamped.

During this operation the entire contents of the flask have been
under atmospheric pressure, and the liquid in the air tube has
risen to the level of the liquid outside of it. The flask, when
next connected with the heart, accordingly does not act at once
as a Mariotte's bottle ; it only becomes so when the air tube is
clear of liquid : hence the initial pressure when a new flask is
turned on is, for a few seconds, higher than it should be. This
higher venous pressure causes a slight increase in work for some
seconds. This error exists in our results, but it is practically too
slight to be important.

The heart having been placed in the apparatus, was allowed
to run until it did fairly even work per min. for half an hour
or more. It was soon found, however, that it was necessary not
only that the heart should do even work, but also that the work
should be near the normal amount, because if a heart which
under good blood was only pumping 2 cc, while later, under the
same conditions, it showed itself able to pump 10 cc. in the same
time, was in the first instance treated with digitaline, there was
an increase in work independent of the drug.

Having noted this fact in some earlier experiments, we always
waited until both the amount and regularity of the work done
showed that the heart was acting normally.

A tracing of the pulse was usually taken for one minute every
time the blood was measured.

As soon as the heart was working properly, good blood was
turned off and the poisoned blood allowed to run until the quan-
tity in the flask had decreased 100 cc. The time taken for this
was noted. Good blood was then turned on again, and the
attempt made to restore the heart to its previous condition.

DIOITALINE ON THE HEART. 333

If this was successful, the same operation was repeated until
either it was impossible to recover the heart or repetition was
deemed superfluous. This method, which allows of several
observations on the same heart, was suggested by Prof. Martin,
and is very satisfactory. In the earlier experiments we mea-
sured the blood and took a tracing once in five minutes. In the
later ones, however, this was done only once in fifteen minutes —
except when the poisoned blood was running through, when the
observations were more frequent.

The poisoned blood which had once circulated and that which
immediately followed it was always thrown away. The typical
effect of a moderate dose of digitaline given in this way was
primarily a slight acceleration of pulse joined with a sudden
decrease in work. Soon after the flow of poisoned blood had
ceased the pulse became normal ; and then the work increased
more slowly until the heart was doing, after an hour, for instance,
as much or more work than it had previously done. The second
dose appeared usually to take effect somewhat more quickly
than the first, but it was not until the third or fourth dose that
a slowing of the pulse usually became evident. As the number
of doses increased it became in most cases more difficult to re-
cover the heart.

Results.

We made fourteen series of experiments. Of these five must be
discarded ; two because of accidents during the observations ;
one because the three flasks did not give the same pressures ; one
because the blood was stale, and the fifth because the pressures
were varied during the experiments. We have then left nine
series, comprising thirty-four observations.

In order to express concisely what happens we have condensed
our observations in the following way : Taking the total number
of cubic centimeters pumped around in the 15 min. immedi-
ately preceding the giving of the poisoned blood, we found the
average number of cubic centimeters per minute during that
time. That number is our standard for the given experiment.

Now, the time being observed which it takes for 100 cc. of
the poisoned blood to pass through the heart, the number of
cubic centimeters pumped per minute for this period is calcu-

334 H. H. DONA LDSON A ND MA C TIER WA R FIELD.

lated; when the poisoned blood is turned off and the good
blood on, the number of cubic centimeters for the first 30 min-
utes is averaged, and the amount per minute found.

As the pressures in each experiment are constant, we can com-
pare the number of cubic centimeters per min. in the different
observations of the same series with one another just as well as
the absolute work. This we have done. Taking then one
amount pumped out, expressed in cc. per min., as a starting
point, we look to see how the amounts pumped in the same time
during the two subsequent periods compare with it. For brevity
we will call the three periods mentioned "before digitaline,"
" during digitaline," " after digitaline."

Out of the thirty-four cases there are twenty-four in which the
work " during " is less than that " before," and the work " after "
less than that " during " digitaline. That is, where the work has
remained decreased for at least half an hour after the digitaline;
but after that time, the heart being steadily fed with good blood,
has reached or nearly reached its original amount. Of the re-
maining ten cases there are six in which the work is less "dur-
ing " than " before," but rises in the period " after " above what
it was " during " digitaline. There are two in which the rise
" after " goes above what the work was " before " digitaline
(Series 3, No. 2 ; Series 7, No. 2).

In one of the remaining cases there is a slight and unaccount-
able rise " during " above the work "before " digitaline (Series
17, No. 1), while in one case the work increases from the first
to last period (Series 7, No. 1). The number of cases in which
less work is done " after " digitaline than " before " is then thirty-
one out of thirty-four. This leaves us three contradictory cases
to be explained.

The two exceptions in Series 7 (Nos. 1, 2) are cases in which the
experiments were made when the heart, though pumping evenly,
was doing an abnormally small amount before the administration
of the digitaline, and it was not till something near the normal
work was done that digitaline produced its usual effect. The
third case (Series 3, No. 2) was plainly a case where time
enough had not been allowed for recovery.

We conclude from these observations : 1. That where the
heart is doing normal work the influence of digitaline is to

DIOITALINE ON THE HEART. 335

decrease that work ; 2. That there is a rough relation be-
tween the size of the dose and the extent of the decrease ; 3. It
is further to be observed that with small doses of digitaline the
pulse-rate is at first increased.

This observation has a two-fold significance. It confirms
those of Jorg,20 Saunders,21 Hutchinson,23 and others, and at the
same time is a good indication that our doses were moderate.

An almost constant appearance under moderate doses was a
shrivelling of the auricles. This tendency, at first slight, became
at the end of a series of moderate doses very marked. With
a heavy dose the auricles became of course much distended.

During the period of accelerated or unaltered pulse rate the
volume of the ventricle appeared somewhat decreased, while
during the slow pulse it was plainly increased.

The question of dosage is one important in these experi-
ments. The dose is primarily the amount of the drug used.
But beyond that, the percentage in which it exists in the blood,
the length of time the heart is exposed to the poisoned blood,
and the surface of the heart acted on, are of the greatest import-
ance. For instance, our tables show (Series 8 and 11) that in the
course of an experiment much more digitaline can be given than
can be borne in a single dose. Indeed, in one series, not pub-
ished because vitiated by an accident, 10 doses of .0005 grm.
were given to a heart without any perceptible effect. This has
a bearing on the once held theory of cumulation of digitaline.
If it accumulated in the heart muscle one would expect, first, a
decided effect from numerous small doses, and second, a rather
tardy action of large ones. Neither occurs. The large doses act
with great rapidity, while the smali ones produce no effect pro-
portionate to their number.

Still, it makes a difference how long the poisoned blood
remains in the heart. If two hearts are taken, one pumping
100 cc. in 5 minutes, and the other the same in 10 minutes, and
the same weight of digitaline given in the same amount of blood,
the effects will be much more marked in the latter than in the
former case.

336 H. H. DONA LDSON A ND MA 0 TIER WA RF1ELD.

Finally, as the heart increases in size its capacity increases in
three dimensions, while the surface exposed increases only in
two; thus the larger the heart the less, proportionately, the sur-
face exposed to the poisoned blood. All these points are worth
consideration when the true dose is to be estimated.

It remains now to offer an explanation for those results which
are at variance with our own, namely, the direct results of
Bohm and the indirect ones of Williams. Roy23 has shown that
the curve of extensibility of the ventricular muscle is an hyper-
bola. In the case of the frog's ventricle it makes a sharp bend
at about 10 cm. of water pressure, and beyond that increase of
pressure produces little distension. In the ventricle of a frog
under a moderate dose of digitaline the elasticity is quite perfect,
but the distensibility is noticeably increased, or if it were repre-
sented graphically, the new curve would fall even more nearly par-
allel to the axis of ordinates and to a much greater distance before
bending than docs the old one. It is easy to see, then, that for
a time the curves would not differ much. That is, for moderate
pressures the much increased capability for distension caused by
digitaline would not be brought into play, but as soon as we
make the pressure more than moderate, as both Bohm and Wil-
liams did, this new factor is brought in. The distensions for
equal increments of pressure are now much beyond the normal,
the elasticity remains quite perfect, and the heart then does a
much increased amount of work.

The fact that when the heart is working against a maximum
pressure digitaline does not improve it, favors this view. If the
strength of the systole or perfection of elasticity were improved
by it, then we should get an increase in work ; but the heart
being already fully distended, and the tendency of digitaline being
to increase extensibility, it is here superfluous, and the work de-
creases in spite of the drug.

Thus it is plain that one important action of the drug is to in-
crease the distensibility of the heart muscle.

Following are the condensed records of our experiments given
in tabular form.

The table is constructed as follows :

DIGITALINE ON THE HEART. 337

At the beginning of each experiment are the most important
data : Time of observation — the weight of the terrapin — position
of the cannulas — the pressures used — and the temperature with
its variation.

Beyond these are eleven columns of figures. Column one gives
the number of the observation.

Column two: Gives in minutes and seconds the time which it
took the poisoned blood to pass through the heart.

Column three : The number of cubic centimeters of that blood.

Column four: The absolute weight of digitaline given.

Column five : The proportional weight in one hundred cubic
centimeters of blood.

Column six: The average number of cubic centimeters of
blood pumped by the heart each minute for fifteen minutes be-
fore the digitaline was given.

Column seven : The corresponding average pulse-rate.

Column eight : The average number of cubic centimeters of
blood pumped by the heart each minute while the poisoned blood
was running through.

Column nine : Corresponding average pulse-rate.

Column ten: The average number of cubic centimeters of
blood pumped by the heart each minute for thirty minutes after
the poisoned blood.

Column eleven : The corresponding average pulse-rate.

338 H, H

-ojw-o»|nd

DONALDSON AND MACTIER WARFIELD.

ssifio sislo sssSs! s«K8 ses= !■-" e'sss'

tM&B

iftljB era -av

33<*e

Jiw-

SS3S

^.U

■Mi-*

85°>s

^S5=c

o)w-9«inj

sasiss

|ss 1

&:^S'4

ssHsj

32^

as*

5S8'a

sis&sa

■Jfun np -oo Ay

3333

SiSS

KB.

aa=s

S£=£"

■BIHJ-OTIIId

Ssriii

-3

aaaa

.-<-«.«

S33S

fcSSSifi

S83S

tfeste

IJS*-

mms

•ajojaq 'oo ■■» v

sSSS

.553

=533

S=S§

tisas

siss

g„

SSss

ssiil

•oo ooi n| jonoinv

ill

§§sl

i!i-|

o*>a ojn[o«iv

lis

8S5S ls'§§ §§§

fef

§§§§

8§§§§

■00 JO 0\'

sa8S

2P3-

eggs

§888

888,8

888

ssss

88888

smij,

•"sa

a»g=

ESS"

saa"

*SS3

5-S

■sao|)TAje«io JO ""N

S* a,.

IS 13"

111! gs|

If'" II

Hi 111

iffev jl.

Jst.

Ill
is;

ClS Et-'E. E>

ts.
Ill

tig

■r|s

111

'ill

Hi |l>8 |lll |!>a j»& Js* |^a fjQ I

{►►3

*1bs

#«3

if-
la

4..

DIGITALINE ON THE HEART. 339

REFERENCES.

1. Dybkowsky and Pelikan. Zeitschrift f. Wissenschaft. Zo-
ologie. Vol. XI, p. 279.

2. R. Bbhm. Pfliiger's Archiv. Vol. V, p. 153.

3. Fothergill. Schmidts Jahrbiicher, Vol. 154, 1872. British
Med. Jour., July and August, 1871.

4. Trail be. Oesammelte Beitrdge zur Pathol, u. Physiologic

5. Legroux. Oaz. des HSpit. 37. 1867.

6. Brunton. On Digitalis, with some Observations on Urine.
London, 1868.

7. Gourvat. Gazette Medicals de Paris, 1871.

8. Eulenberg and Ehrenhaus. Allg. Med. Central Zeitung. No.
98. 1859.

9. Winogradoff. Archiv f. Pathol. Anatomw. XXII, p. 457.

10. Brie8emann. Schmidts Jahrbiicher. Vol. 153, p. 29.

11. Boldt Inaug. Dissert Schmidts Jahrbiicher. March, 1872.

12. Ackermann. Deutsch. Archiv f. klin. Med, Vol. XI, p. 125.

13. Brunton and Meyer. Jour, of Anat. and Physiology. Vol.
VII, p. 135.

14. Von Bezold. Untersuch. fiber die Inner vat io?i des Herzem.
II Abth., p. 205.

15. R. Bbhm. Dorpater Med. Zeitschrift. Vol. IV, p. 64. 1873.

16. Gbrz. Dissertation Dor pat. 1873. Archiv f. experiment.
Pathol, u. Pharm. Vol. II, s. 23.

17. Blasius. Yerliandl. der Phys. u. Med. Gesellsch. zu }\urz-
burg. N. F. II. Bd. 49.

18. Williams. Archiv f. ezperimen. Pathol, u. Pharm. Vol.
XIII, p. 1.

19. Howell and Warfield. Studies from Biolog. Lab. Johns
Hopkins University. Vol. II, No. 2.

20. Jbrg. Archiv de Med. Prem. sit. T. XXVII, p. 107.

21. Saunders. On Foxglove.

22. Hutchinson: Quoted by Homolle and Quevenne. Archiv
de Physiol, de ThSrap., &c. No. 1. 1854.

23. Roy. Jonrn. of Physiology. Vol. I, No. 6, p. 452.

ON A NEW FORM OP PILIDIUM. By E. B. WILSON,
Ph. B. With Plate XXVIII.

Among the many rare and interesting forms of pelagic animals
taken with the dipping net at Beaufort, N. C, during the sum-
mer of 1880, were two specimens of a Nemertine larva, which,
though belonging to the Pilidiurn. group, is very unlike any of
the species which have hitherto been described. It is a peculiar
and highly specialized representative of this larval type; and
though the scarcity of material prevented any careful histological
study of the creature, it is, perhaps, worth while to describe it in
order to point out its relations to 6ome other larvae of the same
group.

The full-grown larva (Fig. 1) is helmet-shaped, but the upper
or convex side is much more elevated than in most other species.
At the summit of the bell is a rather small flagellum. The an-
terior margin of the bell is produced into four short blunt arms
or lobes, of which two are seen in the figure. Behind these is a
deep sinus in each lateral margin followed by two lateral arms on
each side. The anterior of these, marked a in the figure, is con-
siderably the largest of all the arms; in the position most com-
monly assumed it is bent backwards, so as to assume roughly the
form of a sickle. All the lobes are very contractile, and the ap-
pearance of the margin of the bell varies greatly according to the
state of contraction. The walls of the bell are also contractile,
and the entire margin is sometimes drawn up so as nearly to
close the opening. The cavity of the bell, indicated in the fig-
ure by a faint curved line, is evenly rounded, and of great size as
compared with the corresponding cavity in other species. Be-
hind and below the bell terminates in a blunt point.

The bell is of glass-like transparency, and is covered with a
beautiful pavement of large epithelial cells. Scattered at inter-
vale among these cells are small highly refracting spherical bodies
which have the general appearance of oil-globules; they are
much less numerous than the cells, and are not therefore to be

341

342 E. B. WILSON.

confounded with the nuclei of the latter. Both the outer and
inner surfaces of the bell are covered with cilia, which are short
over the general surface, but become much longer and more
powerful along the margins of the lateral lobes. By the action
of these cilia the larva swims slowly and gracefully through the
water, at the same time revolving upon an axis passing through
the base of the flagellum and the centre of the lower surface. In
one specimen there was an accumulation of dark reddish-brown
pigment on each side of the bell near the base of the anterior
lateral lobe ; the other specimen was destitute of pigment.

The young Nemertines in both larvae were fully developed,
already exhibited some power of contractility, and within eigh-
teen hours after the stage figured, abandoned the Pilidium en-
velope. They were very opaque and granular, showing very
conspicuously and definitely through the transparent wralls of the
bell. Of their internal structure while within the larval envel-
ope little could be made out, save the ciliation of the alimentary
canal, which was rendered evident through the rapid rotary
movements of the contents of the stomach. The young worm lies
in the lower and posterior part of the larval envelope doubled up
in a peculiar way, so that the middle and anterior (?) part of the
body lies horizontally and transversely, while the remaining part
projects nearly vertically upwards. The recently escaped young
Nemertine (Fig. 2) is very contractile and changeable in shape,
and swims with some activity by means of the fine cilia covering
the surface of the body. Towards the posterior end a central
more opaque mass bordered by a clearer zone could be distin-
guished. The creature is somewhat remarkable for presenting
an appearance of distinct segmentation in the posterior part of
the body. The young worms, unluckily, soon died, and it was
therefore impossible to determine whether these " segments "
were permanent or simply the temporary result of contraction of
the body. It is highly improbable, in either case, that the ap-
parent segments are true somites.

For some time after the escape of the Nemertine the cast-off
larval envelope exhibited a striking, though deceptive, appearance
of continuing an independent existence. Portions of the bell
still performed well-marked contractile movements, and the
organism, through shrunken and distorted (Fig. 3), still swam

A NEW FORM OF PILIDIUM. 343

with considerable energy by the continued activity of its cilia.
These movements gradually ceased, however, and the remnant
of the Pilidium at length died.

This larva, which for the sake of convenience we may call
Pilidium brachiatum, is of interest, so far as its general features
are concerned, in two respects, viz. in the highly specialized
nature of the marginal lobes, and in the great relative size of the
larval envelope. Fig. 4 represents a species of Pilidium occa-
sionally found in the southern Chesapeake, which is closely
similar to the common European P. gyrans. In this species the
bell has a very different form from that of P. brachiatum, the
apical flagellum is much larger, and there is only a single mar-
ginal lobe on each side, which is, however, very large. The young
worm, which is shaded in the drawing, occupies a different
position in the larva, and the bell has scarcely any cavity. A
comparison of this form with Pilidium brachiatum shows that
the larval envelope in the latter species is proportionally three
times as large, at least, as in P. gyrans ; and the striking dis-
similarity of the marginal lobes in the two forms shows how
differently the tracts of locomotor cilia have been modified to
increase their extent and efficiency. If we extend our compari-
son to other species of Pilidium we find a rather interesting
series of modifications in the marginal lobes, to illustrate which
I have introduced figures of three European species, viz. P.
auriculatum (Fig. 6), after Leuckart and Pagenstecher, and two
species (Figs. 5 and 7) after Metschnikoff. It is clear that the
marginal lobes in these three species, although of different forms,
correspond with each other and with those of P. gyraiix (Fig. 4).
In each case, however, the lobe has acquired a character of its
own ; this is especially marked in Fig. 5, wThere the margin of
the lobe is crenate and the cilia are disposed in definite tufts
separated by bare spaces. In P. auriculatum the marginal
lobe assumes nearly the same form as the antero -lateral arm of
P. brachiatum (Fig. 1, a), and the two appear to be homologous.
A comparison with Fig. 7 strengthens this conclusion. The
lateral lobe has here assumed a slightly different form, being
intermediate between those of P. auriculatum and P. gyrans,
but the anterior margin is produced into two slight prominences
on each side which correspond in position with the anterior arms of

344 E. B. WILSON.

P. brachiatum. The outline of the bell, also, is to some extent
intermediate between the high-arched form of P. brachiatum
and the more flattened expanded form of P. yyrans. Thus the
form of this species (Fig. 7) shows clearly how the highly modi-
fied P. brachiatum may have been derived from the common
type represented by P. gyraixs, or how both may have arisen
from a common form. Attention may be drawn also to the
great variation in the size of the flagellum in the various species ;
in some cases it may even be replaced by a tuft of long cilia.

We are thus enabled in Pilidium to trace out in some detail
certain modifications which are due entirely to adaptation to
larval life, and which do not stand in any sort of relation to the
conditions of adult life. It is somewhat remarkable to find the
tract of locomotor cilia so variously modified ; for the conditions
of larval life, so far as locomotion is concerned, seem to be much
the same for all the species of Pilidium. The larvae would seem
to be capable of ready modification through the action of causes
apparently insignificant, or else the conditions affecting the life
of such a pelagic creature are more varied than appears at first
sight.

The rotation of the larva upon its vertical axis, which is char-
acteristic also of other species of Pilidium^ is worth noting on
account of the significance which Rabl has ascribed to this move-
ment in his well-known "Blastaea Theory" (Entwickelung der
Tellerschnecke, Morphologisches Jahrbuch, Vol. V, 1879). The
larvae of many Ccelenterates have been observed to perform
movements of rotation; and from this circumstance has resulted,
according to Rabl's theory, the acquisition of a radiate structure
not only in the larva but also in the adult. If, on the other
hand, the movement be unaccompanied by rotation, if it be
linear and not spiral, then, according to the theory, the tendency
will be towards the development of a bilateral instead of a radial
symmetry. Rabl's theory, it is true, considers especially the
movements and symmetry of the ancestral " Blastaea," from
which the Coelenterata and Bilateralia have been derived; but
his argument from the free-swimming larvae of Coelenterates is
based on the assumption that the particular form of symmetry
shown in these larvae stands in causal relation with their mode
of locomotion. In Pilidium is found, contrary to the demands

A NEW FORM OF PILIDIUM. 345

of the theory, a strict and pronounced bilateral symmetry co-
existing with a spiral movement, and the same is true of many
other larvae, as, for instance, among the Chsetopod annelides.
And further, some Coelenterate larvae — e. g., that of Renilla —
which perform marked spiral movements, are, to say the least,
as much bilateral as radiate. Hence, it seems probable either
that Rabl has attributed too much importance to the character
of the movements of the primitive Rlastsea, or that the argument
drawn from the locomotion of existing larvae cannot be sustained.

EXPLANATION OF PLATE.

Figure 1. — Pilidium brachialum, nov. sp., from Beaufort. N. (■. ;
from left side. X 60.

Figure 2. — The same ; young Nemertine soon after its escape from
the larval envelope, X 120.

Figure 3. — The same; larval envelope which has been cast off.

Figure 4. — Pilidium closely resembling Pilidium yyrans ; from
the southern Chesapeake, X 70.

Figure 5. — Pilidium with peculiarly modified marginal lobes;
after Metechnikoff.

Figure 6. — Pilidium auriculatum; after Leuckart and Pagon-
stecher.

Figure 7. — Pilidium, sp.; after Metschnikoff.

ON THE POLAR EFFECTS UPON NERVES OF
WEAK INDUCTION CURRENTS.— By HENRY
SEWALL, Ph. D.

More than a year ago I was engaged at Leipzig, in company
with Prof. v. Kries, in studying the action of two successive sub-
maximal stimuli upon each other in curarized muscle.

The results then obtained appeared to the authors, at least, to
recommend the simplicity of the methods employed ; and they
were accordingly used subsequently in a large number of expe-
riments performed with a view to discovering the physiological
interaction of rapidly succeeding stimuli applied indirectly to the
muscle through its nerve. It was soon evident, however, that
the method was quite inadequate to the task proposed, and that
portion of the work was for a time abandoned ; but not until it
was clear that the difficulty experienced was due to the specific
action of the electrical currents upon the nerve.

It is proposed in the present paper to consider the influence on
the nerve of very weak induction currents passing through one
pair of electrodes, as shown by their effect upon submaximal
muscular contractions excited through a separate pair of elect-
rodes.

The records were taken by means of an elaborate form of pen-
dulum myographion described'in another place.1 The recording
lever was very light, and magnified the contractions some eight
times. The weight hung upon the axis of the lever. The muscle
used was the gastrocnemius of the frog with its attached nerve.
The tissues were freed from blood before excision by a stream
of 0.6 per cent. NaCl through the aorta. The experiments were
made upon several different species of frog, and occupied the
month of February and part of May. Two du Bois induction
coils, without the iron cores, placed at right angles to each other
and several feet apart, supplied each pair of electrodes upon
which the nerve rested. The nerve was usually laid between two
strips of moist filter-paper upon platinum wires, which were

1 Journal of Physiology, Vol. II, p. 164.

347

348 HENRY 8EWALL.

stretched over an ebonite block, and this was then covered to
prevent evaporation. In this case no other moist chamber was
used, and the muscle was simply inclosed by the skin. Occa-
sionally the nerve was placed upon platinum wires without being
inclosed in moist paper. In such instances the electrode pairs
had to be moved much nearer together than previously, in order
to obtain the results to be described. A modified form of the
usual nonpolarizable electrode was also employed, which has
been found useful both in this and in general laboratory work.
Four glass U tubes were cemented, each by one limb, into holes
bored into an ordinary microscope slide. Clay plugs filled the
cemented limbs, and the amalgamated zinc wires dipped into the
free ends of the tubes. With a little care the zinc sulphate solu-
tion was prevented from rising to the top of the clay. After the
nerve was laid on the upper ends of the clay plugs, it was covered
by a glass slide borne by narrow glass slips cemented along three
sides, so as to cover in a little chamber which could be kept moist
by salt solution contained in a tube fastened into the lower slide.

In the experiments, unless otherwise indicated, the two keys
of the myographion were placed so as to be opened simultane-
ously by the swing of the pendulum. The shock from one induc-
tion apparatus was so far weakened that it just failed to call
forth a contraction from the muscle, and was, therefore, for itself
inefficient. The intensity of the stimulus from the second coil
was regulated to excite a contraction varying from about one-
tenth to three-fourths the height of a maximal contraction.

When the two pairs of electrodes are pretty far apart on the
nerve, one inch or more, the results from double stimulation are
not at all regular. There is good evidence of an interaction of
stimuli, however far separated on the nerve, but not in the sense
to be considered below. The results in such cases are too irreg-
ular, and their causes too obscure, to be treated at present.

When the electrode pairs are separated by a short distance on
the nerve, resting within three-quarters of an inch of each other,
the height of the contraction due to the single efficient stimulus
is profoundly and regularly altered under the influence of the
other stimulus (which is itself too weak to produce a contraction)
when both are let simultaneously into the nerve. This interac-
tion is more pronounced the nearer the two pairs of wires are

POL A R A CTION IN NER VES. 349

together. When the nerve is not covered by moistened paper,
the inner or adjoining wires of the electrode pairs must be
approached to within one-fourth to one-eighth of an inch of each
other. When the electrode pairs are moved farther and farther
apart the constancy of the results of double stimulation gradually
fails in an order which has not been studied; and, usually, when
the inner wires are separated by an inch or more, the effects to be
immediately described do not regularly appear.

The results obtained under the conditions described may be
conveniently arranged as below.

I. When the upper stimulus, that farthest from the muscle,
is able by itself to produce a contraction ; the lower stimulus,
that nearest the muscle, taken alone being inefficient :

a. When the upper is descending and the lower descend-

ing, double stimulation gives a strong diminution of
the single contraction obtained from the upper stim-
ulus.

b. When the upper is descending and the lower is ascend-

ing, double stimulation gives a strong increase in
contraction over the single.

c. When the upper is ascending and the lower ascending,

double stimulation gives slight increase over the
single.

d. When the upper is ascending and the lower descend-

ing, double stimulation gives diminution of the
single.

II. When the upper stimulus taken alone is inefficient to pro-
duce a contraction, the lower being by itself efficient :

a. When the upper is descending and the lower descend-
ing, double stimulation gives an increase over the
single contraction.

J. When the upper is descending and the lower ascend-
ing, double stimulation gives strong increase over the
single.

HENRY SEWALL.

c. When tbe tipper is ascending and the lower ascending,

doable stimulation gives strong diminution of the
single contraction.

d. Whou the upper is ascending and the lower descend-

ing, double stimulation gives a diminution of the

single.
Below is a table embodying the results of one experiment
made in May, with the use of platinum electrodes whose inner
wires were nearly one-third of an inch apart, the nerve resting
on moist paper. The numbers refer to the heights of the con-
traction curves measured in millimetres.

Lower Stimulus

ALON».

AUan-

HelBht

RIM.RKR.

■

•i

(ran

Double simulation

■=timiiLil

15 5

Strong increase.

1*J1

Olrnlnullon.
Strons Inerenso.
UimfnmlOTi.

It need hardly be pointed out that these results may be
described as due to the polar influences of the inefficient stim-
ulus.

The fact indicated in the preceding, that the excitation deve-
loped at the kathode of the efficient stimulus is depressed in
the neighborhood of the anode of an adjoining pair of electrodes,
and conversely, presents nothing essentially new; but it is inter-
esting to observe the results in cases 3 and 5 of the table, in which
it is shown that not only does an increase in the intensity of an
anodic area diminish an excitation wave passing through that
area from a kathode above, but even when the exciting kathode
is nearest the muscle, the contraction caused by it is lessened
when the intensity of the anodic phase higher up on the nerve
is increased. It ie seen at once that all these results follow the
general conditions of what is known as the " law of contraction."

POLAR ACTION IN NERVES. 351

In nearly all the experiments the two stimuli were let simul-
taneously into the nerve. When the myographion keys were so
arranged that the two shocks succeeded each other at different
time intervals, whatever the order of succession, the interaction
of the two stimuli gradually diminished with the increase of the
interval and failed altogether when this was still very small —
that is, about 0.001 second. Comparatively little attention was
paid to this aspect of the work ; but there was in no case evi-
dence of an oscillation of electrotonic condition at either pole of
the reacting current, such as occurs after the cessation of a gal-
vanic current in the nerve.

It is not very clear what relation the two phenomena, the
"actiou current" and the "electrotonic current" set up in a
nerve by an induction shock, bear to each other. The evidence1
goes to show that the two changes appear simultaneously on
stimulation and progress with equal velocity. Hermann2 is the
only investigator, as far as I know, who has made a definite
attempt to analyze the electrotonic phases of the induction cur-
rent in this connection ; and any one reading this paper must be
struck with the indissoluble character of the bond uniting the
purely electrotonic with the physiological excitatory changes set
up in a nerve by electrical stimulation. Griinhagen,8 starting
from some results of Harless, in which the latter fouud that by
the double stimulation of a nerve in two places he sometimes got
an increase and at others a diminution of the contractions from
the single stimuli, and working with constant currents, decides
that two effective stimuli applied simultaneously to the extremi-
ties of a nerve summate; but if one stimulus be by itself ineffec-
tive, then in no case does it influence the effective stimulus.
Griinhagen's work, however, has little in common with that
detailed above. Some of the experiments of Wundt4 touch upon
isolated points of the questions considered here. A short r§sum6
of the results of work upon the interaction of electrical stimuli
in nerve given by Hermann5 may be of use to one who is not
acquainted with the literature of the subject.

1 Helmholtz, Monatsbericht. d. Berlin. Akad. 1854, p. 329 PflUger, Electroto-
nns, p. 442. Tschirjew, du Bois1 Archiv, 1879, p. 525.
• Hermann. Pflttger's Archiv, Bd. XVIII, p. 574

3 Zeitechr. f. Nat. Med., 3te Seite, XXVI, 1866.

4 Wundt. Mechanik der Nerven.

8 Hermann. Hdb. der Physiologie, Bd. II, S. 109.

352 HENR Y SE WA LL.

It appears to the writer that a consideration of facts such as
those which have been detailed must affect to a great degree the
physiological significance of all results which follow the very
rapid succession of stimuli in nerve muscle preparations, and if
this be true the adaptability of the electrical method to such
experiments is extremely doubtful.

Some interesting conclusions of Dew-Smith1 from "double
nerve stimulation " have been kept in mind throughout this
work. That author found, essentially, that when a nerve was
simultaneously stimulated by submaximal induction shocks at
two different points, the muscular contraction ensuing did not
represent an addition of the contractions from the single stimuli,
as might have been expected, but about equalled the contraction
which was to be obtained from the lower single stimulus, that
nearest the muscle, acting alone. He suggests as an explanation
that the excitation-wave passing downward from the upper pair
of electrodes is " blocked " by the wave going upward from the
lower electrodes and is thus practically annihilated. The sug-
gestion was a valuable one as offering a possible clue to an expla-
nation of the difficult question of physiological inhibition, and it
seemed highly desirable to find the true meaning of the outcome
of the experiments.

These results, however, appear to be readily explained when
considered as a special case of the phenomena whose general
relations have been considered in this article. Let us consider
the effects brought about in the contractions from double stimu-
lation when the strength of one excitation is varied. When, for
example, the lower stimulus is ascending and efficient, the upper
being ascending and inefficient, double stimulation gives a con-
tration smaller than that obtained from the lower stimulus alone.
Let, now, the strength of the upper stimulus be gradually
increased ; there comes a point where the excitation from the
kathode of the upper electrodes balances the depressing effect
upon the lower "stimulus of the upper anode, and, as far as 1 have
observed, this point is reached when the single contractions are
not far from equal.

The resultant of the interaction of the two excitations depends,
of course, altogether upon their relative strength and direction
in the nerve.

1 Dew-Smitb. Journ. of Anat. and Phys., Vol. VIII. 1874, p. 74.

RESEARCHES ON THE GROWTH OF STARCH
GRAINS. By A. F. W. SCHIMPER, Ph. D.1 With Plate
XXIX.

The starch grains found in many growing chlorophyl-contain-
ing plant parts, show a constant structural peculiarity; these
grains, usually tablet-shaped in the observed cases, present
ragged edges, sometimes perforated. The broad surfaces are
very uneven, and present under the microscope a spotted appear-
ance, produced by superficial sculpturing, and, in many cases,
also by internal vacuoles. From the results of the following
researches, these appearances must be ascribed to partial solution,
due to the fact that some of the starch is used for the growth of
the organ. This conclusion rests, on the one hand, on the fact
that after the cessation or abatement of the growth of the organ
concerned, the starch granules deposited do not possess the above
characters ; on the other hand, on the fact that similar appear-
ances occur in germinating seeds (e. g. Zea mais.)

After the starch-bearing organs have ended or greatly slowed
their growth, the formation of normal starch begins ; usually
some new spherical starch granules appear, which show no trace
of the above described structure ; in addition, the already
present grannies increase in size. This increase does not occur,
as one would expect, in the interior of the grain, hut in the form
of an originally very thin and gradually thickening, shiny and
strongly refracting stratum, deposited around the original cor-
roded grain. This layer is not itself corroded, but shows, of
course, prominences and pits corresponding to those of the cor-
roded grain. The subsequently deposited strata agree in char-
acter with that first laid down, but the inequalities of the sur-
face become gradually obscured, so that it is often smooth in a
fully formed grain. In the centre of this complete grain, when

1 Translated from Botanische Zeitung, 1881, Nob. 12, 13, 14.

353

354 A. F. W. 8CHIMPER.

fresh, one can, however, with suitable illumination, still detect
the original corroded grannie.

The appearances just described may be seen in many different
species of plants.

Among others, I have seen them very beautifully in the seeds of
of some Leguminosce. The starch grains of the cotyledons Doli-
chos lablab (Figs. 1-3), which is one of the plants most suitable
for the purpose, first appear when the seeds have attained one-third
of their full size. They are then flattened corpuscles, with very
lumpy surface, surrounded by chlorophyl. The starch grains
retain the same form and structural peculiarities, though increas-
ing considerably in size, so long as the cotyledons are growing and
possess a vivid green color. With the cessation of growth and the
diminution of the chlorophyl the formation of the final " reserve "
starch commences. First appear, in most cases, glistening bluish-
shimmering spots on single prominences or on one side of the
starch grain ; soon the whole grain is surrounded by a thin layer
of dense non-corroded substance. The starch formation thence-
forth proceeds uniformly. In the completed grain one clearly
recognizes the corroded uneven kernel.

Starch formation in the seeds of Vidafdba agrees essentially
with that observed in Dolichos. In Phaseolus the grains are
originally spindle-shaped, and with a less uneven surface than
that of the plants above named. Nevertheless, the same mode
of development may be recognized in them.

In the medullary parenchyma of Cereus speciosissimus (Figs.
4-7), the starch formation is like that in Dolichos. The tops of the
stems examined contained, close beneath the "punctum vegetatio-
nis," many large starch grains with smooth surface. The develop-
mental processes which are briefly described below refer to actu-
ally growing stems, in which starch formation is easily observ-
able. One or more starch grains arise in the chlorophyl granules
accumulated around the cell nucleus. Here, also, they appear
as minute angular tablets, but with a not very greatly corroded
surface. With the diminution of the chlorophyl grains, which
towards the last form only thin membranes around the starch
grains, the definitive starch development commences, and pro-
ceeds as in Dolichos. Here also the primary corroded tablet is
clearly recognizable in the completed grain.

GROWTH OF STARCH GRAINS. 355

There can, therefore, be no doubt that the inner parts are not,
as Nageli maintains, the youngest, and the outer the oldest ; the
exact opposite is the case. The growth of a starch grain occurs
by deposition on its exterior.

More careful examination of the development of starch grains
results in many other facts which are incompatible with Nageli's
theory. The starch grains of Dieffertbachia seguvna are, for
example, very instructive ; l in contact with a second chlorophyl
granule they obtain a new system of layers, deposited on the
primary. In the following sections we shall meet with still other
phenomena conclusively showing the untenability of Nageli's
doctrine ; those described in the present section are, however,
sufficient. My immediate object is to examine more closely and
explain those properties of starch grains which have been
regarded as proving a growth by intussusception.

These properties are generally known, and will have, more-
over, to be closely discussed in the course of this article. I con-
tent myself, therefore, for the present, with briefly stating them
in the order in which they are discussed in the following pages :
1. The differentiation into regions containing different amounts
of water. 2. The differences in regard to percentage of water,
and sometimes of shape, between the small granules and the inner
strata of the larger. 3. Unlike rate of growth in different direc-
tions. 4. The mode of growth of compound and partially com-
pound grains.

One might be inclined to assume, as Dippel2 has for the cell
membrane, that there occurs an intussusception growth of layers
first deposited by apposition, but in such cases the original ker-
nel would certainly be lost, which is by no means the case. On
the other hand, we shall see that all the properties of the starch
granule may be explained without the assumption of any intus-
susception.

So far as concerns the objection which one might raise before-
hand in opposition to the whole drift of these researches, viz.,
that cell membranes undoubtedly grow by intussusception, and
that consequently the so similar starch grains must do likewise,

1 See Schimper. Untersuchungen tlber die Entstehung der StftrkekOrner. Bot.
Ztg., 1880. Taf. 13, Fig. 13.
8 Die neuere Theorie tlber die feinere Structur der Zellhtllle, etc., 1878.

356 A. F. W. SCHIMPER.

it is no longer tenable, after the well-known researches of Sachs,
Traube and De Vries on the influence of tnrgidity on the growth
of the cell membrane. These researches have completely eluci-
dated the surface-growth of the membrane by intussusception,
since they have shown that it only occurs under the action of
cell turgidity, and consists in a constantly repeated exceeding of
the elasticity limit, with an immediately following deposition of
solid particles in the interstices. Increase in area and increase in
thickness of the cell membrane are, therefore, to be attributed
to quite different causes ; from the fact that the former takes
place by intussusception, the conclusion is not justifiable that
the latter occurs in the same method. Still less can it be applied
to the starch grains where there is no question of turgidity.

II.

In Nageli's theory the part concerned with the developmental
history of the kernel and of the layers in simple starch grains
is undoubtedly the best thought-out portion. The facts that the
kernel consists of a soft material, while starch grains of like
size in the same plant are dense; and that the outer layer is
always poor in water, even duriug the deposition of layers con-
taining different proportions of water (which necessarily would
lead to an equally frequent appearance of a peripheral layer,
rich in water), appear entirely inconsistent with growth by appo-
sition, while they find a satisfactory explanation through the
intussusception theory.

It seems desirable, before stating the results of my own re-
searches, to present extracts from Nageli's great work, giving his
view of the history of the differentiation of starch grains into
kernel and layers.

According to Nageli the developmental history of a simple
starch grain is as follows : *

" All starch grains are spherical in the earliest stage and consist of
a dense material. Then in all cases a spherical kernel of softer mate-
rial separates, and after it has increased in size divides again con-
centrically into a new small spherical kernel, and an inner dense
and outer softer stratum, the latter strata forming spherical shells

1 Die St&rkekOrner, 8. 280.

GROWTH OF STARCH GRAINS. 357

around the kernel. The process may be repeated once or several
times. Less frequently a small spherical, denser kernel is deposited
in the large, more watery original one. The outer stratum as well
as those which have been formed by division of the kernel divide
from time to time concentrically after they have attained a certain
thickness. Usually one dense stratum splits into two of similar
character, with an intermediate soft one : more seldom a soft stratum
is divided by a denser. In addition a thickening occurs ; it may be
observed in the soft strata and in the kernel even when the hard
strata have attained considerable density. If, however, the strata
differ from one another so little in consistency that the whole mass
appears homogeneous, it is the dense parts which first appropriate
more material."

Nageli's theoretical explanation of these processes is as follows : l

" If we conceive the spherical beginning of a starch grain as con-
sisting of similar concentric molecular layers, then any nutritive
liquid entering will first lay down new particles in the surfaces of
these layers. This results from the fact that the resistances are there
less than those which would be met with on deposition between the

layers Let us assume that the molecular layers in the whole

grain simultaneously and uniformly increase : then any two neigh-
boring layers will exhibit a tendency to separate from one another,
since the radius of the outer would with unimpeded growth increase
more than that of the inner. Since the adhesion does not allow a
separation this tendency results in a tension, positive in the outer,
negative in the inner layer. Since all molecular layers in the entire
grain behave similarly, the positive tension in the particles of each
one must decrease from the surface to the centre and the negative
tension increase. The tension in a given layer must act on the next
outer layer as a contracting, on the next inner as an expanding force.
In fact, however, the nutritive fluid does not nourish all molecular
layers simultaneously and uniformly. Its concentration diminishes
as it approaches the centre. The condition that the outer molecular
layers are earlier and more richly nourished than the inner must in-
crease the tension between them. That the outer lavers have a
greater tendency to expand than the inner is proved by various facts.

" So soon as the tension under which, in consequence of growth,
the molecular layers find themselves has reached a certain degree
they separate from one another, and new layers are deposited between
them. This will occur most frequently where the tension most

1 Loc. cU. S. 289.

358 A. F. W. SCHIMPER.

easily overcomes the adhesive force. The adhesion is directly pro-
portional to the superficial area of the molecular layers. The tension is
primarily present as a surface force, and it is merely a question how it
is changed into a radial or separating force. Calculation shows (1)
that the radial force which holds in equilibrium a tangential or sur-
face force, in a system of spherical shells or cylindrical envelopes of
like thickness and similar property but of different size, stands in
inverse proportion to the leugth of the radii ; and (2) from this first
fact it results that when two spherical or cylindrical shells in contact
with one another and of like thickness and elasticity grow by like
quotients superficially, the force which tends to separate them is
inversely proportional to the square of the radius. The molecular
layers are so much the easier separated from one another as they are

nearer the centre of stratification

" The larger the young dense grain becomes the greater becomes
the unlikeness in density and cohesion between surface and centre,
and so much the greater becomes the negative tension in the inner-
most part of the mass, and the tendency to deposit material there.
When these ratios have attained a certain value, a space filled with Soft
material is rapidly formed in the central point of the grain. A
similar process occurs subsequently in the dense cortex, and later
repeatedly in the dense strata. These grow thicker : so soon as they
have attained a certain thickness the dissimilarity of tension in outer
and inner molecular layers produced by surface growth, and the
effort to separate from one another, become so considerable that it
cannot longer be met by deposition of new material of similar density.
These results, therefore, are actual separation ; a space filled with soft
material appears."

I believe that I have above given the most important points on
the theory of growth of simple starch grains. Subsequently
Nageli endeavors to explain the occurrence of dense strata in
soft, and in the kernel ; also the condensation of soft strata
throughout their whole thickness. I have failed to completely
understand these parts of his work, and since they appear to me,
for reasons to be immediately stated, much less essential than those
dealing with the formation of the kernel and of the soft strata, I
must, so far as they are concerned, refer the reader to the original.

Some of the phenomena regarded by Nageli as undoubted
are only assumptions facilitated or made probable by the theory,
namely (1) The occurrence of new strata in the kernel. (2) The
occurrence of new strata in the soft ones. (3) The condensation

GROWTH OF STARCH GRAINS. 359

of the soft strata throughout their whole thickness. (4) The
absence of any increase in thickness in the outermost strata.
With reference to these points observations are entirely wanting.
They could only be established if the development of a starch
grain could be directly watched, or if it was so far the same for
all the grains of an organ that the comparison of specimens of
different ages could give an accurate notion of the developmental
history of a single grain. As is well known, neither of these
alternatives is the case.

Moreover, Nageli himself concedes, with reference to the
formation of dense strata within soft, that he has made no sure
observation on the point. " Like the dense strata, the soft, with-
out doubt, also split, forming two superficial soft strata and a
median denser. However, this process is only seldom and to a
partial extent to be clearly seen ; much less frequently than the
division of the dense strata, which in innumerable cases presents
itself with all certainty."1 He seems also to have seen no very
clear picture of the occurrence of dense strata in the nucleus.
He says rather, at the end of his description, referring to this
point, " No grains were drawn which give an accurate picture of
it. One can, however, form a tolerably accurate idea by the aid
of Figs. 20 and 21, plate XVII."2

The most important of the phenomena upon which Nageli's
theory is based are, however, undoubted facts. The develop-
mental history of a starch grain, as deduced with certainty from
a comparison of specimens of different ages, is as follows : (1)
The appearance of starch grains in the form of strongly refract-
ing corpuscles, poor in water; (they are by no means always
spherical, as Nageli assumes). (2) Differentiation of the originally
homogeneous grain into a central kernel, rich in water, and a
peripheral dense stratum. (3) In later conditions the kernel is
surrounded by three strata, of which the middle one is always
rich in water ; such a layer never appears as peripheral, and it
must, therefore, be formed through a cleavage of the first dense
stratum. (4) The number of strata increases ; but the outer one
is always dense. (5) As the starch grain increases in volume,
the proportion of water in its inner parts increases.

The explanation of these appearances I find in certain long-
known physical properties of starch grains, to the consideration
of which I now proceed.

1 S. 234. * S. 233.

360 A. F. W. SCHIMPER.

The compression of a starch grain leads to the production of
numerous clefts, which in a simple grain usually run in directions
perpendicular to the surfaces of the strata ; never parallel to
them. Careful crushing of the grains under water does not as a
rule split them into fragments. They appear even after the
action of very strong pressure coherent, greatly flattened struc-
tures, traversed by numerous radial fissuree.

The cohesion of a starch grain, therefore, varies very remark-
ably with the direction ; it is small tangentially, very great radi-
ally. In the latter direction its substance is very extensible,
while extensibility in the tangential direction seems almost en-
tirely absent.

The formation of clefts and the flattening cure not the only
results when starch grains are crushed; on the co?itrary, the
grains experience a change consisting in a more or less marked
jelly-like sioeUing.

That mechanical means bring about the tumefaction of starch
grains, has been observed by Nageli and Schwendener.1 Accord-
ing to them, the phenomenon occurs very clearly when starch
grains are cut ; the parts adjacent to the cut surface assume a
swollen character. According to W. Nageli2 this swelling, which
occurs whenever a starch grain is subjected to mechanical injury, is
to be regarded as dependent on a slight degree of the same process
which takes place when starch is boiled in water.

The phenomenon, both as regards amount of swelling and the
place where it occurs, differs with the strength of the pressure ex-
erted. Weak pressure leads only to swelling of the innermost parts
of the grain. In this case the kernel appears as if considerably
increased in size, since, in consequence of the pressure, the
layers immediately surrounding it have become entirely like it in
light-refracting power. The outer layers only swell on exposure
to stronger pressure.

The swollen part contracts on drying; its light-refracting power
becomes again like that of uninjured grains, except the most
strongly swollen parts, which remain less refracting. A second
moistening brings about, renewed swelling.

1 Das Mikroskop. 2 Aufl. S. 433.

7 Beitr&ge zur n&heren Kenntniss der St&rkegruppe, S. 25. After very strong
swelling organic coloring matters are imbibed in small quantity.

GROWTH OF STARCH GRAINS. 361

Mechanical actions are, therefore, capable of imparting to the
water-poor parts of the starch grain the characteristic properties
of the water-rich parts, namely, greater wateriness and less light-
refracting power.

Could the proof be furnished that in the growth of starch
grains by simple surface deposition, forces were set in play
which must result in the swelling of different parts in such a way
that the known differentiation of the grains would be produced,
then the question as to its origin might, without doubt, be
regarded as solved.

According to Nageli, we must regard tensions as the chief
active forces in the differentiation of the kernel and of the layers.
These tensions, as shown by the not unfrequent presence of
fissures in starch grains, may attain considerable intensity. We
have to more closely examine the cause of these tensions and
their possible r6le in the development of the starch grain.

That starch grains swell in water has been generally known
for some time. Nageli, however, first showed that the deposition
of water did not occur in all directions, but is much greater
parallel to the stratification than perpendicular to it.

Among other things this conclusion is based on the direction
of the fissures that take place on drying; this is always perpen-
dicular to the stratification. If the water were uniformly dif-
fused in the starch grain, then clefts must occur in other direc-
tions also. The great extensibility of the swollen grains in a
radial direction diminishes very considerably with loss of water,
and would, therefore, oppose no hindrance to the formation of
fissures.

The unequal deposition of water shows itself most clearly
when one suffers the starch grain to swell strongly under the
influence of acids, or potash, or heating. It then comes out in
the clearest manner that the maximum water deposition is par-
allel to the stratification ; the least, perpendicular to it. Nageli
has instituted a series of measurements on the starch grains of
Canna and of Curcuma zedoaria, which indeed (since the strata
are not even, but curved in an hour-glass form) only express
the relations approximately, but, nevertheless, give some idea of
the greatness of the difference ; they may, therefore, be here
repeated.

362 A. F. W. SCHIMPER.

Canna.1

The starch grains of Canna have, as is known, a flattened form

and very excentric kernel ; most of the strata are incomplete.

On. swelling a deep pit is formed on the side where the kernel lies,

in consequence of the preponderating extension in the cross

direction. The length of grain I was measured to the bottom of

the pit ; that of grain II to the points of folds on each side of

the hollow. In the latter, therefore, the difference given is too

sm all.

Length of grain.

Before swelling, 61

After swelling, 100

Katio, 1 : 1.6

Increase per cent., 64

Curcuma zedoaria?

The starch grains of Curcuma zedoaria have, as known,
essentially the same structure as those of Canna, and as regards
swelling behave similarly.

Length of grain. Breadth.

eadth.

Length. Breadth

74 55.5

150

240.5 203.5

:11

1 : 3.2 1 : 3.7

971

225 267

- - -»

Unchanged,

6«

After swelling,

105

Ratio,

1:1.4

1:1.3

1:1.4

1:3.1

1:3.5

1:3

Increase per cent.

• TT

211

250

200

The appearance is, however, so conspicuous that direct meas-
urements are not necessary in order to convince oneself of the
want of uniformity in the swelling. Figs. 21 and 22 show a starch
grain of Canna before and after swelling.

Another noteworthy appearance, brought about by the pre-
ponderating swelling in the transverse direction, is the concave
folding of the cut surface of starch grains which have been
bisected through the kernel. One readily obtains such grains on
cutting a Canna rhizome with a sharp razor.

The predominance of the tangential directions when com/pared
with radial as regards water deposition, brings about tensions.

1 Loe. at. S. 76. » S. 77.

GROWTH OF STARCH GRAINS. 363

// the starch grain consisted of loose molecular layers these would
separate from one another when the grain swelled up; but since
the layers actually cohere firmly, each layer is strained positively
with reference to the one on its inner side* and negatively with
reference to that on its outer side. These readily comprehensible
consequences of uniform surface increase of the molecular
layers, without corresponding radial increase, have been arrived
at by Nageli by means of calculation.

If the tensions have reached such intensity that the limit of
elasticity is exceeded, and the layers can in consequerice follow
their tendency to separate, this cannot occur through the formar
Hon of fissures running parallel to the stratification, as Nageli as-
sumes. The earlier described appearances of compressed grains
show, on the contrary, that a traction acting vertical to the layers
can produce an extension, but not a tearing in that direction.
The starch grains can be extended by pressure to the extent of
several diameters without the formation of tangential fissures. The
stretching, however, causes, as shown by the same experiments, a
swelling-up of the substance, which assumes the characteristic
properties of the more watery parts of the normal starch grain.

If we seek to take into account the effect of these tensions on
the developing starch grain, we find that the formation of the
kernel and of the soft strata actually occurs where these tensions
must exhibit themselves.

The developmental history of a starch grain is, without doubt,
as follows. It consists originally of homogeneous, dense material.
When, in consequence of non-uniform water deposition, the in-
creasing tensions have attained such a degree that the elasticity
of the grain can no longer resist them, the material in the centre
of the grain must be extended and brought into a condition of
greater swelling and less light-refracting power. Observation, in
fact, shows that when a starch grain has exceeded a certain size,
a less refracting strongly swollen spot, the kernel, appears in its
centre.1 The central formation of the kernel depends, as Nageli
has proved by calculation, on the fact that action of the tensions
is there most powerfully exhibited. As regards this point it

1 Compare the representation of the formation of the nucleus, as given by
N&geli, I. c. S. 309.

364 A. F. W. SGHIMPER.

naturally amounts to the same thing whether the tensions,
as Nageli assumes, depend on an uneven deposition of starch
molecules, or, as I (basing my belief on observation) contend,
upon a non-uniform deposition of water molecules.

The formation of the kernel causes, of course, a diminution of
the tensions. Through the deposition of new material they
soon, however, increase again in the dense stratum surrounding
the kernel, and finally become sufficient to overcome the elas-
ticity. For reasons already stated there then occurs, not a tearing
of the layer into an inner and an outer part, but a straining, in
consequence of which the starch substance in the middle of the
layer becomes swollen and less light-refracting. The simple
dense stratum becomes, in other words, differentiated into three ;
a median soft, and an inner and an outer dense.1

The peripheral dense stratum now behaves exactly like that
which first arose through differentiation of the homogeneous grain.
When the tensions have attained a certain intensity it experi-
ences a strain in its middle, through which a soft stratum is pro-
duced— and so on.

Through the deposition of new material the inner parts of
the starch grain, as a whole, become constantly more expanded
by the outer. On the one hand there results from this a drag
on the inner soft strata, in consequence of which they increase
in bulk and in tendency to swell up. On the other hand it is
also probable that dense strata are likewise affected, and the
water in them increased.

The radial fissures, often present in fresh starch grains, as well
as the partially compound grains, to be later discussed, are to be
ascribed to the strain exerted by the outer parts upon the inner.
That these clefts only depend upon non-uniform distribution of
water is taught by the appearances which such grains present on
slow drying. Those of beans, for example, which commonly ex-
hibit gaping clefts, completely lose them on drying, the loss of
water bringing about a diminution of the tensions. Since, how-
ever, the inner parts are richer in water and poorer in solids than
the outer, they contract more than the latter on complete or
nearly complete drying; they pass again, therefore, into their

1 Compare Nftgeli, /. c. S. 310.

GROWTH OF STARCH GRAINS. 365

earlier state of negative tension, and this is associated with the
reappearance of the fissures.

More powerful swelling reagents, of course, bring about an
increase of the tensions in each layer.. A priori it is highly
probable that the effort of the molecular layers to separate from
one another would thereby be increased sufficiently to overcome
the elasticity in fresh places. In other words, the original dense
strata would experience in their middle a strain, and in conse-
quence a stronger swelling of their substance; that is, would
differentiate into three strata. This view again stands in full
agreement with the fact that stronger swelling is associated with
the occurrence of numerous new soft strata where none were
previously visible ; that is to say, where the tensions previously
had not been strong enough to overcome the elasticity.

Strong swelling, however, is also associated with a consider-
able increase of the strain exerted by the outer strata upon the
central parts of the grain; these, therefore, experience a stronger
drag. We see in fact that the inner parts at first become greatly
extended in a radial direction by means of the outer, and that at
last they are forcibly torn from one another, so as only to present
swollen fragments in a large central cavity.

In accordance with the foregoing, the differentiation of starch
grain 8 into regions of unlike wateriness presents itself as the
necessary result of certain of their physical properties, and
requires for its explanation no assumption of a growth by intus-
susception.

It need not be pointed out that cohesive or elastic properties,
the action of mechanical influences upon the swelling power of
starch grains, and finally the unequal extension in tangential or
radial directions, are properties which the grain may acquire
through growth by apposition as well as by intussusception ;
they alone are the grounds upon which my explanation rests.

That the capacity of the starch grain to lay down water in
no way proves that it is permeable also for the dissolved sub-
stances out of which the grain is built up, needs no special
discussion. We find, indeed, that the starch grain is not per-
meated by many solutions (for example of organic coloring mat-
ters), which are absorbed readily by cell membranes and protein
crystalloids. Even assuming such permeability, we would be

366 A. F. W. SCHIMPER.

far from justified in believing, on that ground alone, that there
also occurred a change of the amylaceous substances into starch,
and a deposition of the molecules so formed between those already
present. We are just as little justified in assuming, before it has
been definitely proved, that along with apposition-growth in
starch grains some little growth by intussusception occurs, as we
would be in inaTring the same assumption with reference to a
crystal of quartz or calc-spar.

In connection with the difference as to water contents between
small starch grains and the inner set of layers of larger grains,
the differences of form sometimes observed may be mentioned.
Nageli does not appear to have laid much weight upon these
appearances, and only speaks of them very briefly.1 According
to him, in Pimm and other Papilionacese, the small grains are
broader than the kernels of the full-grown grain. This is qui to true
if one compares the small and large grains of ripe seeds. The
younger developmental stages of the larger grains have, however,
no resemblance to the spherical or sub-spherical small grains
which are present in ripe or nearly ripe seeds ; they are thin,
spindle-shaped, corroded, and resemble in form the nucleus of
the large grains. In the root-stock of Carina are sometimes
imbedded in the large grains "sets of layers of lancet-like or
linear spindle form, such as no grains resemble in shape or struc-
ture." A figure is not given, but reference made to a similar
structure depicted from Cereus variabilis. From this figure and
from the description I believe myself able to conclude that we
have here to do with an inner set of layers such as shown in one
of my figures (Fig. 20); but we find in this case independent
grains of the same form present, as Fig. 19 shows. I have not
been able to examine the root-stock of Dent aria.

III.

According to Nageli the unequal growth in different diame-
ters of many starch grains is not compatible with external depo-
sition. " It would be incomprehensible that free floating starch
grains should increase seventy times more on ono side than on
the other."

lLoe. at. S. 219.

GROWTH OF STARCH GRAINS. 367

The explanation which he gives of this phenomenon is some-
what indefinite. Its cause is to be sought in the arrangement
of the smallept particles, and in the fact that on account of dif-
ferences of cohesion in different places more material is deposited
in some than in others.

The intussusception theory can give no very satisfactory expla-
nation concerning the causes of this unlike arrangement and
cohesion, which is a regular phenomenon in certain plants, and as
regularly is absent in others ; and for each species is so constant
that in it only forms of one and the same type appear. With
reference to it Nageli's words are, "Since the nutrition depends
not on external relations but on internal causes, the deviations
which the starch grains show later in structure and form must be
already present in their earliest beginning in the spherical
smallest grains ; this is conceivable, as the original spheres are
formed under different specific relations. They show accord-
ingly in the arrangement of their smallest particles, and in the
nature of these, specific modifications from which of necessity
the entire peculiar growth results."

The mode of growth of starch grains is, according to Niigeli,
dependent only on internal causes; external influences could
not bring about an uneven growth, but only exert a determining
influence upon the direction of most or least growth. Excentric
starch grains would grow most where they obtained the most
dilute solution. This is, according to Nageli, especially clearly
the case in compound and partially compound starch grains, in
which the directions of greatest growth are directed towards the
centre of the grain. So also should be explained CriigerV state-
ment that excentric starch grains are attached by the hinder end to
the primordial utricle or the protoplasm. " The close agreement
of secondary grains and simple grains as regards increase of
volume and density of the material along the long and short
radii, supports throughout the view that the plasma in contact
with the hinder end acts like starch substance, and, therefoie,
either entirely prevents the access of nutritive liquid or only
allows a more dilute solution to pass." 2

That the form of starch grains is primarily determined by the
mode of nutrition, I have pointed out in a former work.8 1 have

1 Bot. Ztg. 1854. * N&geli, I. c. S. 327. « Bot. Ztg. 1880.

368 A. F. W. SCHIMPER.

shown that centric starch grains arise when they are surrounded
ring-like by starch-producing plasma (chlorophyl grain or
" starch- former ") ; and that excentric grains arise at the peri-
phery of the formative centre, and grow fastest at the points
in contact with it.

The flat grains with central kernel originate in lens-shaped
chlorophyl grains, and their broad -sides are, as Nageli has already
pointed out, parallel to those of chlorophyl grains. The elon-
gated starch grains of beans and some other Papilionaceae are
formed in spindle-shaped chlorophyl grains, with their long axis
parallel to that of the latter. Flat excentric starch grains (e. g.
Carina, Phajus grandifolius) are nourished by a formative mass
(" starch-former " or chlorophyl grain) which courses along their
hinder end. These phenomena can only be explained through
unequal nutrition.

The relations between the growth of the starch grain and the
supply of nourishing liquid are, finally, exactly what they should
be in a body growing by apposition.

Excentric starch grains only touching the formative organ
with one part of their surface, increase not only at this point ;
all or nearly all of the grain is recognizably in growth. This
growth is fastest at the point of contact, and diminishes rapidly
as the distance from this increases, so as to become extremely
small at the anterior end of the grain, at least when the latter
has attained a tolerable size. This point calls for more, minute
discussion.

If we seek to form a conception as to how a starch grain is
nourished by its mother material, we can hardly conceive of the
latter except in the form of a solution which impregnates the
formative organ. We may leave it, however, undecided whether
it is uniformly distributed through this organ, or (what is, per-
haps, more probable in the case of peripherally originating
starch grains) is limited to certain parts of this. In either case
capillarity will lead to the accumulation of a layer of mother
liquid between the starch granule and its supporter. The further
necessary condition, that the nutrient matter shall not rftmain
confined to this spot, is also fulfilled. A starch grain and its sup-
porting formative organ, as we know, do not lie in the cell sap,
but imbedded in protoplasm, and, as Hanstein first recognized,

GRO WTff'VF ST A RCH OR A INS. 369

the protoplasm is especially dense where in contact with the
starch grain. If we imagine for a moment the starch grain and
its nonrisher surrounded, not by protoplasm, but by a jelly-like
substance, then, through capillary action all around the starch
grain, water would be drawn from the jelly and collect between
the two in a thin stratum. This layer of water would neces-
sarily be continuous with that collection of nutritive liquid sepa-
rating the starch grain from its formative body, and would con-
sequently obtain the properties of a nutrient liquid, and afford
the starch grain with material for growth, diminishing in quantity
with distance from the nutritive organ.

If, however, we assume that the water or watery solution im-
pregnating the jelly is so combined with it as not to be capable
of extraction by capillarity, then the layer of nutrient liquid
between the starch grain and its formative body will, under the
influence of the same force, spread all over the grain. In this
case also the rate of growth would diminish with increase of
distance from the formative focus.

Protoplasm, however, cannot, without much qualification, be
compared to an ordinary jelly-like substance, and I, therefore, do
not maintain that either of the above given explanations of the
mode of nourishment of the starch grain is the correct one ;
though I think it highly probable. The illustrations mainly
serve to show that so far as analogies are concerned we arc led
necessarily to a phenomenon of the same kind as that which we
do actually find, and that for its explanation the assumption of
an intussusception is by no means essential.

IV.

It is well known both partially compound and perfectly com-
pound starch grains have yielded to Nageli different points of
support for the theory of intussusception. The following are
those phenomena which, according to him, are not in harmony
with the theory of growth by apposition.1

1. The difference of form between the secondary grains of a
partially compound and perfectly compound starch grain on the
one hand, and simple grains of- the same size on the other.

1 p. 228.

370 A. F. W. SCHIMPER.

The former have hemispherical, angular, discoid, or elongated
shapes, while the latter— the simple grains — are spherical. A
development of these forms through fusion of simple isolated
grains cannot be admitted, because the grains float free in liquid.

2. Whenever the secondary grains possess excentric kernels,
these lie upon the outer side, away from the surfaces of contact
of the secondary grains.

This position, aside from some peculiar exceptions which
stand in precise relation with the irregular Gratification of simple
grains, is everywhere constant. The regularity would be in-
explicable, however, upon the theory of apposition ; at least the
reason why the grains are always united by their posterior ends
would not be apparent.

3. The occurrence of clefts between the secondary grains. The
latter could not have been from the first enveloped in this man-
ner by the external substance ; the splits must have arisen sub-
sequently— a point which, according to Nageli, can be explained
only through internal growth.

4. Specially important, according to Nageli, are the differences
in substance between the secondary grains of partially com-
pound grains and simple ones of similar size. The latter are
composed of comparatively anhydrous, the secondary grains, of
watery substance.

The explanation which the theory of intussusception gives of
these phenomena seems to me far from clear; at any rate, like
Sachs,1 1 have been unable to grasp it.

u The conditions which disturb the concentric and radial arrange-
ment of the component parts may reach such a degree in certain parts
of the grain that the molecular forces of the surrounding stratified
substance may be no longer able to control the new depositions. The
latter then proceed in the same way as if they took place free in the
cell fluid, wrhere starch-forming goes on undisturbed by external in-
fluences. In this way is formed a complex of component parts which
begins to stratify concentrically, and results in a secondary grain
similar in its development to a perfect starch grain. These dis-
turbing conditions find freest play where the molecular layers ex-
hibit the greatest tendency to separate one from another, namely, close
to the periphery in the neighborhood of sharp corners, edges and

1 Exp. Physiologie, S. 421.

GROWTH OF STARCH GRAINS. 371

elevations, as well as in the centre of stratification itself, where, in-
stead of one, two or more new kernels may arise."1

The formation of clefts between the secondary grains is to be
regarded as a consequence of the weak cohesion at this spot, in
which the arrangement of the molecules in the course of these
secondary formations has suffered the greatest disturbance. The
outer portions of the secondary grain are fed by a more concen-
trated solution than the inner portions; for this reason, the
latter possess less cohesion and consequently exhibit more rapid
growth.

The proofs that compound and partially compound grains
originate by division and not by fusion of simple grains are in
part no longer cogent. First of all, as regards the angular form
of the secondary grains, it is indeed clear that it cannot be
accounted for by compression of the grains. Nevertheless, the
same phenomenon occurs in numerous organized bodies whose
origin by apposition is undoubted ; for example, to mention a
case with which botanists are familiar, in the sphero-crystals of
inuline, for which a correct explanation has been given by Sachs.2
The flattening is due simply to this, that growth naturally stops
at the surfaces of contact of two or more bodies which touch one
another.

As to the greater softness of the contents of the secondary
grains of partially compound forms in comparison with the con-
tents of simple grains of equal size, this is the necessary conse-
quence (precisely as for the contents of large simple grains) of
the tension exerted upon the inner strata by tho outer, and
requires, after what has been said in the second section, no fur-
ther remark or explanation. The splits between the secondary
grains are doubtless to be ascribed to the same thing. That
originally separate secondary grains which by subsequent
growth come to touch one another and to be surrounded by com-
mon stratifications, adhere but feebly to their fastenings, and hence
can easily be separated by mechanical conditions, is, A priori^
probable, and is proven by this, viz. that compound grains
which have undoubtedly arisen by the fusion of free simple
grains break up easily under pressure into their secondary
grains.

1 p. 294, cf. also p. 323 et %eq. » Bot. Zeitung, 1864.

372 A. F. W. SCHIMPER.

On the other hand it is far more difficult to reconcile with the
theory of growth by apposition the statement that in the excen-
tric secondary grains the kernels always lie on the periphery.
If it were shown to be indeed trne that forms such as those
shown in Fig. 9 e have developed from forms like 9 a, the
developmental history wonld furnish the most enigmatic con-
tradictions. The rhizome of Canna,1 where partially compound
grains are very common, offers a superior field for investiga-
tions. In this plant the attempt was made to obtain a pic-
ture of the developmental history of the partially compound
grams (Figs. 10-16). Close to the punctum vegetationis one
finds in the first place only simple grains which, approxi-
mated in pairs or threes (at this level seldom more), are seated
upon the "starch-formers." At somewhat lower levels, com-
pound forms, having two or three members and usually a clear
kernel, are abundant. That these have originated by the fusion
of simple grains is put beyond doubt by their position upon the
compound grains, by the corresponding disappearance of groups
of simple grains made up of two or three members, and, finally, by
the total absence of forms which might be looked upon as devel-
opmental stages between a simple and a compound grain. Strati
fieation is present very early, but this is difficult to detect in the
secondary grains on account of the marginal shadow. A little
farther from the apex of the rhizome, however, will be found
some few stratifications (to some extent shared in common)
upon the secondary grains. The most vigorous growth will be
found to have taken place, however, contrary to the statement of
Nageli, perpendicular to the axis joining the kernels and cor-
responding to the position of the "starch-former" The com-
parison of the stages of development in sections taken further
and further from the growing point demonstrates a steady sub-
sequent development; the axis of strongest growth and the
position of the kernels remain unchanged, at least in so far as
the average distance of these latter can be made out in the
partially compound grains of sections of varying age. In the full
grown parts of the rhizome, finally, grains occur like that
depicted in Fig. 16.

1 In Strasburg I used C. giganiea, in Baltimore a species unknown to me,
but agreeing throughout with C. giganUa in respect to the starch grains.

GROWTH OF STARCH GRAINS. 373

TJie large grains having multiple kernels found in Canna and
the potato, are, according to Nageli, those which have just
undergone division of their kernels ; but one does not discover
anywhere in the works of Nageli upon what this statement is
founded, and we are, therefore, justified in assuming that it is
not a result of the comparison of starch grains found in sections
of tissues of various ages, but that it rests upon purely theoretical
assumptions, which, when once the theory of intussusception
peemed to be proven by other phenomena, were justified, for then
any other explanation was quite impossible.

Partially compound forms are indeed found in the rhizome of
Canna having kernels far apart (Fig. 17 £), or again some in
which the axes of strongest growth of the secondary grains are
turned toward each other (Fig. 18.) ; these, however, are scarce
in comparison with those having approximate kernels, and are
easily accounted for by the fusion of two grains which either lay
upon a single "starch-former," but at some distance from each
other, or were produced from different " starch-formers." The
starch grain depicted in Fig. 17 b is, for example, to be regarded
as a more advanced developmental stage of a twin grain like
that in Fig. 17 a.

The partially compound grains of the rhizome of Canna arise,
therefore, from the fusion of originally free simple grains. The
same explanation suffices for the grains of the pith-parenchyma
of Cereus speciosixsimus, which afford the most beautiful illus-
tration of the same mode of development, since they actually
exhibit the two angular corroded original masses imbedded in
denser and not-corroded substance (Fig. 6 b).

Taking into account the phenomena above described, it seems
almost certain that the partially compound grains, which are
much more abundant in the potato than in Can?ia, and which
have kernels removed far apart, have originated by the fusion of
simple grains. Unfortunately, it is not possible in the potato, as
it is in the Canna rhizome, to get a complete developmental
history of these grains by the comparison of sections made
through regions of different ages, so that we must be contented
with the endeavor to answer the question how a fusion of two or
more grains by their posterior ends could conceivably occur. In
most organs of plants which possess exientric starch grains, the

374 A. F. W. SCHIMPER.

chlorophyl grains or the " starch-formers " frequently exhibit, as
I have pointed out in my earlier work, starch grains lying at
two or more points of their periphery. Wherever two starch
grains lie opposite to one another, their posterior ends will natur-
ally be turned toward each other. The formative area gradually
diminishes when the starch grains have surpassed a certain size ;
after a certain time only a thin stratum exists between them, and
this finally wholly disappears. Both grains have now fused into
one compound grain whose kernels are remote from each other.
The separate stages of this process can be followed without diffi-
culty in the rhizome of Irte fiorcntina. Developmental stages
like those represented in Fig. 8 prove that the compound grains
in the potato which have kernels remote from each other, have
arisen in this way ; the figure is taken from the rind of a young
greened potato. At a the greened "starch-former" is seen
reduced to a thin disk between the two grains ; at the periphery it
extends beyond the grains as a thick swollen ring. Between the
secondary grains of the grain depicted in 8 b there is found no
longer any trace of the starch-forming organ ; except a swollen
remnant of it which remains like a girdle around the basal
parts of the grains. This outer part of the starch-forming
organ will continue to form starch ; since both grains touch each
other the newly-formed strata will be common to both, in other
words the compound forms will have been converted into
partially compound forms (Fig. 8 c). So far as concerns grains
like those depicted in Figs. 9 a — rf, they can have originated
only by the early fusion of two simple grains, which lay upon
the starch-forming organs in an approximate rather than remote
position, as, for example, we have proven to be the case in the
grains of Canna.

So far as my observation goes, in Phajus grandifolius occur only
partially compound grains which have the direction of strongest
growth perpendicular to the line of union of the kernels. This
depends on the fact that in this plant the rod-shaped starch
forming organs bear starch grains only on one side ; rarely, also,
upon their ends. A starch-forming organ may develop as many
as six starch grains, and these always lie in a line parallel to its
longer axis, never in the opposite direction.1 The reason is the

1 Cf. my paper in Bot. Zeitung, 1880, Figs. 36, 37, 39.

GROWTH OF STARQH GRAINS. 375

same for the rare occurrence in Canna of partially compound
grains having remote kernels. Here also there is a localization of
the starch formation upon one side of the starch-forming organ,
so that I have observed only very rarely young secondary grains
in an accidentally opposite position.

We have thus subjected to a closer examination all the phe-
nomena advanced by Nageli as points of support for his theory,
and have seen that without the assumption of growth by intussus-
ception they may all be explained in a simpler manner ; while,
on the other hand, there is a series of facts quite inconsistent
with the theory of intussusception. We are, therefore, no longer
able to ascribe to starch grains a molecular structure similar to
that of protoplasm. Consequently our problem is next to deter-
mine to what category of bodies do starch grains belong.

Starch grains possess no single peculiarity which justifies us
in assuming for them a physical constitution very different from
that of other rigid bodies ; there are both among amorphous and
crystalline bodies numerous examples of that characteristic
peculiarity of starch grains, the power of swelling in water. The
investigations of Schmiedeberg x and of Drechsel2 as well as my
own investigations3 have shown that the protein crystalloids,
which have so much resemblance to starch grains, can be pro-
duced artificially and represent the crystals of albuminoid sub-
stances. We have therefore merely to endeavor to decide
whether starch grains are amorphous or crystalline bodies.

Those peculiarities which allow us best to distinguish crystal-
line from amorphous bodies, when definite crystalline form is
absent, are cohesion and the optical properties. Hence in
starch grains the solution of this question may be expected
through the investigation of these peculiarities.

The peculiarities of cohesion (with which we may begin) have
been already described in the second section ; it has been shown
there that starch grains are very brittle parallel to the strati-

1 Zeitschrift fur phys. Cliem. Bd. I.

* Journal fur praktische Chemie, Bd. 19.

'Untereuchungen Uber die Protein Krystalloide der Pflanzen. Ioaug. Diss.

376 A. F. W. SGHIMPER.

tication, and vertical to it are very ductile. The difference is
so great that while radial fissures easily arise under the influence
of pressure, tangential splitting even by a destructive pressure
never occurs. A difference of cohesion in different direc-
tions has never been observed in amorphous bodies and is quite
inconceivable in them, since their chief characteristic is the
irregular arrangement of their parts. The splittings which arise
by crushing amorphous spherical bodies (for example dried gum
or caramel drops) take place very irregularly. The crushing
or bruising of fibrous crystalline bodies occasions, on the other
hand, the formation in the first place of fissures parallel to the
fibres, which means that the forces binding them together are
more easily overcome than the cohesion within the individual
crystals; the easy separability of the latter from each other
produces the striated structure which the surfaces of fibrous
crystalline bodies present, and which are also exhibited in a
striking manner by fragments of starch grains. Hence, starch
grains behave in respect to cohesion precisely like radially
fibrous crystalline aggregates (sphero-crystals), and differ entirely
from amorphous bodies.

The optical peculiarities are in full agreement with those of
cohesion ; they are to be referred to the crystalline composition
of the starch grains, and not, as has been frequently assumed, to
tensions. These peculiarities have been the subject of several
erroneous statements, and on account of their importance for our
purpose must be more fully described here in respect to certain
details.

Nageli has already sought to show that the cause of the double
refraction of starch grains is not the tensions ; he believed that
he was justified in drawing the conclusion that double refraction
is not brought about by the tensions of stratification, because sec-
tions of the grain polarized light in the same way as when they
were a part of the intact grain.

This conclusion is, however, not justified, since doubly refrac-
tive bodies, which owe their polarizing peculiarities without ques-
tion to tensions of the same kind as we have in starch grains,
preserve these properties even when they have been broken into
little pieces (for example alum and analcim).1

1 Marbach, Pogg. Annalen, Bd. 94.

GROWTH OF STARCH GRAINS. 377

That alum owes its doubly refractive peculiarities to tensions
has been shown by Reusch,1 who found that he could increase
the double refraction as he liked, could diminish or could make
it entirely disappear by an increase or diminution in pressure or
traction.

Hence in alum the phenomenon depends on this, that the strata
during solidification undergo a contraction in consequence of
which the optical elasticity parallel to the surfaces of the
crystal becomes less than it is perpendicular to them. A sus-
pension of the tension, brought about by pressure, is accompanied
by the disappearance of the double refraction, while traction in
the direction of the surfaces brings about, on the other hand, an
increase of tensions and hence also of the double refraction.

I have carried oat similar investigations on starch grains ; trac-
tion of the outer strata in the direction of the surface (since
these, unlike alum, are in positive tension) must bring about a
diminution of the tension, and in proportion as this takes place
will the double refraction get weaker or wholly disappear, if it
is dependent upon the tension. Starch grains which have been
treated with very dilute potash undergo in the first place only a
swelling of their inner softer substance, while the outer layers,
remaining unattacked by the reagent, are nevertheless stretched
by the swelling inner portion ; the outer layers of grains treated
in this manner did not thereby alter their optical properties,
although the formation of numerous radial fissures must neces-
sarily have brought about a marked decrease of positive tension.

Tensions therefore cannot be the cause of the doubly refrac-
tive properties of starch grains. Closer investigation, however,
teaches, on the other hand, that the interference-figure in parallel
polarized light in each individual case is exactly that which
starch grains must exhibit if they were composed of fibrous crys-
talline (uniaxial or rhombic) elements whose course was similar
to that of the splits, that is to say, perpendicular to the strata.
Essentially, this conclusion has been put forward already by
Baily.2 On the other hand, the statement of Mohl,s that the

1 Monataberichte der Berliner Akad. 1867 ; und Pogg. Annalen, Bd. 132.
Groth, Physicalische Krystallographie, S. 117.

•Philosophical Magazine, 1876. Compare also V. v. Lang, Pogg. Annalen,
Bd. 128 (and Carl's Repertorium, Bd. III.)

a Bot Ztg. 1858.

378 A. F. W. SCHIMPER.

arras of the cross of interference always run perpendicular to
stratification, applies only to regularly symmetrical spherical
grains; in excentric grains these often cut the strata at a very
acute angle. In order in each individual case to determine
before hand the interference-figure, one needs only to draw from
the kernel to the periphery, lines perpendicular to the stratifica-
tion. The dark bars will contain the parts of these lines which
are parallel (or perpendicular as the case may be) to the direction
of vibration of the Nicol.1

In regularly centred spherical grains, just as in the axis of
excentric ones, the doubly refracting elements are straight and
extinguish the light simultaneously along their whole length ;
on the other hand the case is different in the lateral parts of the
excentric grains, where the fibres, as is shown by the splitting,
take a bent course, and hence for every position, throughout a
greater or less part of their length, dependent on the curvature,
fulfil the conditions for the extinction of the polarized ray.

These peculiarities can be explained like those of cohesion,
only upon the assumption that the starch grains are composed
of crystalline fibres running perpendicular to the strata.

Starch grains differ from common sphero-crystals in respect to
their power of swelling, hence we must call the fibrous crystals
composing them crystalloids, as it is desirable to unite under
this head all crystalline bodies which have the power of swelling.

As a result of these investigations, it turns out that starch
grains are composed of radially arranged crystalloids, and ex-
hibit the crystallization of starch substance, C6H10O5, of which
there are probably several isomers.

That the starch crystalloids always occur in the form of fibrous
aggregates, and never single, can be referred to various cir-
cumstances. Previous investigations upon sphero-crystals have
shown that the conditions for their appearance instead of
separate crystals are, difficult solubility, feeble power of crystal-
lization, and viscosity of the solution ; to which, however, it
should be added that a single one of these conditions is, in many

1 Still more simply by constructing a striation parallel to thoj-e parts of the
layers whose course agrees with one of the directions of vibration of the Nicol's
prism ; this gives for each case a precise picture of the interference-figure, cf.
Bailey, I. c.

GROWTH OF STAROH GRAINS. 379

cases, sufficient.1 We must leave it to be determined to which of
these circumstances the regular occurrence of starch in sphero-
crystals is to be ascribed ; we can, however, with some probabil-
ity, assume that all three conditions are fulfilled.

That the strata in the tangential directions deposit more water
can, I believe, in lack of a better explanation, be explained as
conformable to the familiar hypothesis of Nageli2 concerning the
form of the molecule, which he supposes to be longer in the
radial direction than perpendicular to it. That the strata are
always formed perpendicular to the long axis of the fibres, as the
course of the fissures shows, can also be simply explained by
taking account of the fact that stretching is always easiest par-
allel to these fibres ; that in crystals the hardness varies with the
direction, and has its maximum and minimum parallel to the
crystallographic constants, may be assumed as already known.

Baltimore, January, 1881.

EXPLANATION OF THE FIGURES.

All the figures drawn with a magnifying power of 850 diameters.

Figure 1-3. — Starch grains from the cotyledons of the seeds of

Dolichos lablab.

Figure 1. — Corroded starch grains from young seeds.

Figure 2. — Beginning of final starch formation around the cor-
roded grains.

Figure 3. — Almost fully grown starch grains.

Figure 4-7. — Starch grains from the pith parenchyma of Cereus

speciosissimus.

Figure 4. — Corroded grains from young cells.

Figure 5. — Commencement of final starch formation around the cor-
roded grains.

Figure 6. — Grains surrounded by a continuous dense layer.

Figure 7. — Fully formed grains.

Figure 8-9. — Starch grains from potato.

Figure 8. — Chlorophyl grains with starch grains from the rind of a

greened potato.

1 O. Lehman n, Ueber das Wachsthum der Krystalle (Zeitschrift Air Krystal-
lograpnie, Bd. I.)
iLoe. tit. p. 355.

380 A. F. W. SCHIMPER.

Figube 9. — Partially compound grains from the interior of the same.
Figure 10-20. — Starch grains from the rhizome of Canna gigantea.
Figure 10. — Young starch grains on " starch-formers."
Figure 11-17. — Developmental stages of partially compound grains.
Figure 18. — Partially compound grain, with separated kernels.
Figure 19. — Narrow starch grain.

Figure 20. — A similar grain surrounded by strata of different direc-
tion.
Figure 21-22. — Starch grains from rhizome of Canna sp.
Figure 21. — Fresh.
Figure 22. — After swelling in dilute potash.

SOME OBSERVATIONS UPON THE FORM OP
THE PULSE WAVE, AND THE MEAN ARTERIAL
PRESSURE, IN A DOG WITH PATENT DUCTUS
ARTERIOSUS. By WILLIAM H. HOWELL, A. B., and
F. DONALDSON, Jr., A* B.

In the course of some experiments which we were making
upon the isolated mammalian heart, a dog evidently suffering
from some form of heart disease came under our notice. We
supposed that either the mitral or aortic valves were diseased,
and Prof. Martin suggested that it would be of some interest to
take tracings of the arterial pressure and the form of the pulse
wave. It was especially desirable to know the arterial pressure,
since such an observation, of course, could not be made upon
the human subject except in a very indirect way.

A post-mortem examination which Dr. McLane Tiffany was
kind enough to make for us, revealed the fact that there was a
patent ductus arteriosus, establishing a very wide communication
between the aorta and the pulmonary artery. There was also ap-
parently some slight insufficiency of the mitral valves and of the
pulmonary semi-lunar valves.

The aorta from its origin to the end of its arch was consider-
ably dilated, though there was no evidence of any atheromatous
changes in the walls of the artery.

The heart weighed 97 grams, and, upon comparison with the
hearts of other dogs of about the same weight (from 15 to 18
pounds), showed general enlargement, together with some hyper-
trophy of the walls of the left ventricle. The heart of a dog of
about the same weight from which tracings were taken for com-
parison, weighed 66.5 grams.

At the opening of the ductus arteriosus into the aorta there
was a small valvular fold, not nearly large enough to cover the
opening, but so placed as to direct the stream of arterialized
blood sent out from the left ventricle at each systole along the aorta,
and impede its passage into the pulmonary circulation ; in form and

381

382 HOWELL AND DONALDSON.

mode of action this valve somewhat resembled the eustachian valve
of the foetal heart. After the completion of the systole, however,
when the elastic recoil of the aorta had set in, this valve could
have offered no obstacle to the passage of blood from the aorta
into the pulmonary artery ; indeed, would rather have guided any
backward current in that direction.

The only recorded case of this form of heart disease that we
have beei* able to find, is the one reported by Dr. Hilton Fagge
in the Guy's Hospital Reports, 1873.

As we were not competent to make a satisfactory auscultation
of the case, we requested Dr. Frank Donaldson to examine the
dog for us. This he very kindly consented to do, and gave us the
following written report of the symptoms observed :

" I carefully auscultated the dog and found the heart beating
at about 140 per minute; the impulse as compared with that of
a healthy dog was much increased ; the apex of the heart
extended much further to the left of the sternum, showing
marked hypertrophy. Over the whole cardiac region there was
a loud, rasping, systolic murmur, with the maximum of intensity
over the base ; there was also a slight murmur with the second
sound."

In our observations we endeavored in the first place to obtain
tracings of the form of the pulse wave. The dog was tied down
firmly upon a dog-board, and sphygmographic tracings were taken
from the femoral artery by means of a Marey's sphygmograph.

The most favorable tracings obtained, when the 'animal lay
perfectly quiet, and any irregularities resulting from psychic in-
fluences were excluded, were found, upon comparison with
sphygmograms taken from the same artery in a healthy dog, to
be entirely normal.

The femoral arteries were then laid bare and a cannula intro-
duced into each of them ; one of the cannulas was connected in
the usual way with a mercury manometer, which served to regis-
ter arterial pressure; the other was connected with a Fick's
federhymographion. The object in using this latter instrument
was to obtain some idea of the form of the pulse wave in the
opened artery.

The pens of these manometers wrote upon the roll of paper of a
Lud wig's kymograph and on the same vertical line; a chronograph

PA TENT D UCTUS ARTERIOSUS. 383

pen marking seconds and a Marey's tambour for registering res-
piration were also made to write upon the same roll of paper.

The animal was not at first under the influence of any anaes-
thetic, the operation of laying bare the femorals being too slight
to cause any serious pain ; afterwards chloroform was given. It
was noticed that when the animal was deeply under chloroform
the heart beats lost entirely an arhythmic character which had
been very marked when the dog first came under observation,
indicating that this irregularity had been caused before by
psychic influences.

The arterial pressure as given by the mercury manometer was
good, ranging from 140 mm. to 150 mm., which is within the
limits of what can be called the normal blood pressure of a dog.

The pulse wave given by the Fick manometer showed a sud-
den rise of pressure at the beginning of the wave, corresponding
to the sudden ejection of the contents of the left ventricle into
the aorta at each systole, and then a much more gradual fall of
pressure as the excess of blood in the arterial system was gradu-
ally forced through the capillaries into the veins, corresponding
to the description given by Fick of the pulse wave as obtained
by his manometer from normal animals. The descending limb
of the wave was marked by a strong indentation. This indenta-
tion or dicrotism is, according to Fick, who has made a careful
study of the tracings obtained from dogs by means of his mano-
meter, a characteristic of every true tracing, sphygmographic or
manometric, of the pulse wave. Roy, on the other hand, from
some experiments made upon rabbits with his sphygmo-tonometer,
says that the pulse wave in the opened artery is not, in a healthy
animal, dicrotic.

From a comparison of the tracings obtained from this dog
with others obtained from normal dogs, it was seen that the in-
dentation was more strongly marked in this case. In all other
respects the tracing was normal.

The pulse rate varied from 156 to 180 per minute.

The results of our observations, though mainly of a negative
character, are not on that account devoid of interest. The fact
that the animal kept up such an excellent arterial pressure is
especially worthy of notice. The normal pressure in the pul-
monary arteries of a dog, as observed by Beutner, Chaveau and

384 HOWELL AND DONALDSON.

others, is not more than one-third as great as the pressure in the
carotids. Knowing this, and remembering that the lung vessels
possess great distensibility — can accommodate, in other words, a
much larger quantity of blood than they usually contain without
any rise of pressure in the pulmonary arteries resulting — and,
further, that they are probably subject to vaso-motor influences
to a much smaller extent than are blood-vessels in other parts of
the body, one would conclude, from & priori reasoning, that when
this very extensive and distensible vascular region was thrown
into free communication with the systemic circulation, there
would be a marked and permanent lowering of general blood
pressure. That this did not occur must be explained by a com-
pensatory increase in the force of the heart beat, or by an increase
in the amount of peripheral resistance ; or possibly by an increase
in the total bulk of blood in the body.

The pressure in the pulmonary arteries must have been from
two to three times greater than the normal pressure, requiring
an increase in the force of the systole of the right ventricle to
overcome this extra resistance, and causing a greater amount of
blood to flow through the lungs in a given time into the left
side of the heart. From the nature of the conditions governing
the flow of the blood current in the aorta and in the pulmonary
artery, it is not probable that there was any serious escape of
venous blood into the systemic circulation ; the abnormal flow
must have been in the other direction — from the aorta into the
pulmonary tract.

ON VARIATIONS OP REFLEX-EXCITABILITY
IN THE FROG, INDUCED BY CHANGES OF
TEMPERATURE. By W. T. SEDGWICK, Ph. D.

Physiologists aro by no means agreed as to the effects upon
reflex actions of changes in temperature. It is generally
admitted that a cool temperature is favorable either for preserving
or working upon reflex preparations, and that a warm tempera-
ture is equally unfavorable ; but beyond and between these
general and indefinite ideas there is a wide difference of opinion
both as to facts and causes. This is the more surprising because
looked at & priori nothing should be simpler. The organs com-
bined to make up a reflex apparatus though now, in the adult,
physiologically and structurally unlike, have all directly descended
from similar protoplasmic masses in the embryo. Their tissues
are composed, even in their highly differentiated conditions, of
protoplasm more or less modified, and they should, therefore,
obey less or more closely those laws which govern protoplasmic
activity.

Every one knows that protoplasm wherever found behaves very
definitely in respect to temperature. From almost complete
inactivity at a low temperature it passes, with a gradual rise of
temperature, little by little into a phase of greatest activity, beyond
which under excessive heat its functions fall rather quickly back
to zero, or if the temperature be raised still higher, pass beyond
and disappear with the occurrence of coagulation and death.

It is agreed that most of the tissues and organs of the frog,
taken separately, do obey the laws which govern their protoplas-
mic basis. Muscles, afferent and efferent nerves, and glands
exhibit nearly the same series of events which may be observed
in an amoeba or in a white blood-corpuscle. Even the heart — by
no means a simple protoplasmic organ — is subject to the same
laws when free from nervous disturbances. One fact of extreme

385

386 W. T. SEDGWICK.

importance must not be overlooked. Various protoplasmic
combinations exhibit their periods of greatest activity at very
different degrees of temperature. In some cases it might be
supposed, therefore, that one portion of an apparatus would, per-
haps, pass beyond its own period of activity before some other
part would have reached the temperature best suited to it, thus
causing the apparatus as a whole to behave in a contradictory or
exceptional manner. It must be granted, however, upon the
theory of the correlation of parts that it would be ordinarily
more advantageous to the organism to have come to possess
organs made up of harmonious than of discordant tissues ; so
that, unless evidence to the contrary is brought forward, we may
reasonably expect to find in the parts of any apparatus no such
dissimilarity in respect to their behavior toward changes of
temperature.

It is within the experience of every physiologist that the frog,
which, even in the normal state, is now admitted to be to a great
extent a reflex mechanism, exhibits a noteworthy increase of
functional activity as the temperature of winter gives way
before that of summer. That the energetic movements wit-
nessed in the summer, in the animal keenly alive to external
stimuli, pass over in the autumn into the drowsy repose of the
winter " sleep," is also known to every one. It is, therefore,
somewhat surprising as well as confusing to read that in the brain-
less frog (a much more perfect reflex apparatus than the normal
one) the motor and sensory nerves, according to most authors,
obey the laws of protoplasm, while others state that the spinal
cord exactly reverses them ; to find that gentle heating of a
reflex frog, in the opinion of one writer heightens the reflex
excitability, and lowers it according to another; that packing
of the body in ice increases enormously the reflex-excitability,
and the same thing done with hot sand gives the same result ;
that the spasms of strychnia poisoning, commonly supposed
to indicate a high-grade excitability, and which have disap-
peared in a room at the ordinary temperature, may be developed
again in full force by laying the frog upon ice ; while we are told
that in spite of the fact that thermal stimuli are powerful agents
for exciting reflex movements, a brainless frog will sit motionless
until boiled, in water whose temperature is gradually raised. A

TEMPERATURE AND REFLEX ACTIONS. 387

brief review of some of the literature of the subject will show
that these apparent contradictions actually exist.

I. Historical.

Brown-Sequard l seems to have been one of the first to consider
the effects of temperature upon reflex frogs. Having once suc-
ceeded in June in keeping such an animal alive much longer
than usual, he was led to observe again in September and later,
and found at length that he could keep frogs, etc., in good condi-
tion during these months for days and even weeks after the
destruction of the medulla, while previously an hour or two was
the longest time observed. He also noted the effects of destruc-
tion of portions of the cord ; and when the objection was raised
that very likely the prolonged vitality detected by him in the
autumn was due only to the same actual amount of energy fading
out more slowly (owing to the retardation of functional activity by
the lower temperature), he replied by advancing experimental
evidence that there is really more energy exhibited in the fall
than in the summer — a more prolonged and vigorous vitality
rather than a longer exhibition of an enfeebled vitality.

Kunde? writing a revised account of his previous work, states
that if a frog be cooled, an electric current which, when the frog
was warmer, produced tetanic movements, now either produces
them later or not at all. He investigates the effects of temper-
ature upon the spinal cord by giving frogs strychnia and then
placing the animals in water at different temperatures. From his
researches lie concludes that frogs under small doses of strych-
nia, lose their spasms in the cold and regain them when
brought back into a warm room. A dose just large enough
to produce spasms in a warm room having been given, the
animal was put upon ice and the spasms disappeared. If
the animal were held in the hand of the observer or carried
back into the warm room they returned. Large doses have
precisely the opposite effect. The spasms in this case having dis-
appeared under heat, will reappear in the cold. His work, then,
indicates that cold depresses reflex excitability, except in severe
strychnia poisoning.

1 For titles see list of references at the end of this paper.

388 W. T. SEDGWICK.

Cayradef writing in 1864, states that heat shortens the dura-
tion of reflex movements, but increases their energy. When the
increase of temperature is gradual, "as in nature," the reflex
functions also increase gradually their functional activity ; " move-
ments are more speedy, more energetic, and contractions last
longer." When the temperature is " very high, 29°-30°, for
example," section of the medulla produces tetanus and convul-
sions; from which it appears that his conclusions given above
are drawn, in part at least, from intact frogs.

He believes that a sudden rise of temperature is depressing in
its effects upon the reflexes, an opinion derived from his considera-
tion of Kunde's earlier work (1857), in which a frog poisoned to
tetanus at the ordinary temperature, lost the spasms and re-
covered at 34° : also from this observation ; if two cats of equal
weight be poisoned with the same-sized doses of strychnia, and
if, when tetanus has appeared, one be left in a room at the or-
dinary temperature (16°-19° 0.) while the other i6 put in a room
at 40° C, the former speedily dies, while the latter gradually re-
covers. He closes the subject with the remark that in order to
work upon frogs in the summer, one must keep them covered
with wet linen, which keeps them both cool and moist.

Weir-Mitchell* and Richardson* published in 1867 communi-
cations on the effects of extreme cold (freezing by ether and
rhigolene spray) upon frogs and some other animals. Incident-
ally they remark that the freezing, if not too sudden, was the
cause of a preliminary stage of increased excitability, though
this speedily passed into total loss of function, if the whole ani-
mal were frozen, or if all of the cerebro spinal axis were affected.
They observed that frogs and rabbits having frozen brains behaved
in respect to their reflex actions precisely as if they had been
decapitated, i. e, the reflex-excitability rose enormously.

For the purpose of demonstrating a striking difference between
the normal and the brainless frog in respect to conscious sensa-
tion, Goltz6 in 1869 recalled an experiment described by him
long before that time. Though employed by Goltz for a quite
different purpose, it is nevertheless of great interest to us, since
it has given rise to no small difference of opinion concerning
the effects of heat upon reflex excitability. Goltz's experiment
is as follows: A normal frog if immersed in water which is

TEMPERATURE AND REFLEX ACTIOSS. 389

gradually heated, speedily becomes violent in his attempts to
escape. In striking contrast to this phenomenon is the behavior
of the brainless frog, which, on the contrary, save for a few un-
important twitches, sits motionless until it is dead from the ex-
cessive heat. Though Goltz makes no definite statements as
to the cause of this singular quiet of a highly excitable reflex frog
(a matter which has been studied by Foster et #/.), it seem** fair
to conclude from the context that he refers it to a dullness of
perception which is not present in the frog possessing a cerebrum.

Tarckanow 7 studied in the first place the effects of heating
and cooling sectional areas of the central nervous system. For
this purpose he used either high or low temperatures (heated oil
or ice) and thus applied powerful stimuli. His results indicate a
marked coincidence between chemical or electrical and thermal
stimuli. Besides, he devised the following important experi-
ment :

" If the spinal cord of a decapitated frog be laid bare along
its length and covered with ice or snow, a definite depression of
the tactile reflexes will be noticed. If, on the other hand, the
cooling take place upon the intact trunk of a frog similarly de-
capitated and without any opening of the neural canal, we ob-
tain results diametrically opposed to the foregoing, L e* a quite
clearly pronounced increase of reflex excitability."

In order to effect this, he recommends that the trunk of the
frog be packed in ice, by means of a bag or sack .having holes
below for the hind legs.

Tarchanow has also studied upon the normal frog the effects
of heating and cooling, and employed for the purpose, apparently
not knowing of Goltz's work, the same method which was de-
vised much earlier by that observer. He notes the period of
unrest through which the animal passes as the temperature rises,
and also the period of prostration which finally ensues. He calls
attention to the fact that since the cause of this prostration can-
not lie in the nerves or muscles (these being found intact), it
must be sought in the brain or spinal cord. By certain experi-
ments not wholly free from objection, he concludes that the cause
lies in the brain and not in the spinal cord. He points out
again that direct cooling, by ice or snow, of the exposed cord, as
described above, gives a depression of excitability. Indirect

390 W. T. SEDGWICK.

cooling by ice-packing gives an enormous rise of that excitability,
but he omits to explain this difference, as early in the paper he
promises to do, and leaves it without further remark.

Heinzmann* working under the guidance of Preyer, published
in the next year (1872) a paper of very great interest to the student
of this subject. Starting from the fact that a motor nerve may
be subjected to stimulation (chemical, electrical, pressure, and heat
and cold stimuli are mentioned) too feeble to excite movement of
the connected muscle, and that this stimulation may be gradually
increased in intensity so far as to produce finally destruction of
the nerve and yet without causing the least movement in the
muscle, Heinzmann raises the question as to whether or not the
same thing is true of sensory (afferent) nerves.

Thermal stimuli seemed to offer the best opportunity for the
examination of this question, and by means of a carefully
arranged apparatus the work was begun.

Normal frogs and frogs destitute of cerebral hemispheres were
heated very gradually both " locally " and " totally." The local
heating was by dipping one leg of a frog hung by the jaw from
a hook, in water whose temperature could be gradually raised
or lowered. In the " total stimulation " the whole body was
heated by allowing the frog to sit upon cork floating in a cylinder
of water which could be heated gradually. A uniform result was
obtained.

The frog destitute of cerebral hemispheres could be heated
easily, the normal frog for obvious reasons with some difficulty,
until death ensued ; often passing from, perhaps, 22° C. to 40° or
45° ; or could be cooled as many degrees with a similar absence of
movement. This result seemed to Heinzmann satisfactory. It
put the sensory alongside the motor nerve in this respect, and
seemed only to add another support to a well-established law.
Heinzmann's conclusions in regard to the " total" heating of the
normal frog must be compared with those of Goltz and Tarchanow,
who both found, unlike Heinzmann, that gradual, heating of the
normal frog produced most violent movements. Heinzmann
does not refer to the work of either of these observers, and appar-
ently does not know that in recording the quiet of the headless
frog under a gradual rise of temperature he is but repeating a
much earlier experiment of Goltz. It must not be overlooked

TEMPERATURE AND REFLEX ACTIONS. 391

that his explanation of the phenomenon differs widely from that
which might be inferred from Goltz's paper. The latter's work
seems to imply that the quiet of the brainless frog is due to
dullness of perception, so to speak, while Heinzmann sees in the
phenomenon a failure to secure movement due merely to a lack
of stimuli succeeding each other with sufficient rapidity.

Heinzmann has also undertaken to fix the nearest tempera-
tures at which reflexes appear in frogs of known warmth under
heating or cooling of fixed rapidity (Reflex8chwelle\ and the
rapidity of stimulation needful to provoke movement at various
temperatures ( Unterschiedsschwelle).

In 1872 appeared in the Russian language a paper by
Tarchanow* on the physiology of thermal reflexes. I have not
seen the original, but have been obliged to depend for an abstract
of it upon the Jahreshericht of Hofman and Schwalbe for 1872.

The author compared with each other the sensibility of the
skin and afferent nerve, and concluded that special end-organs
for, the detection of thermal stimuli must be located in the
skin. Setschenow had already advanced the idea of special
end-organs for the detection of chemical stimuli, and others have
located there tactile end-organs, so that Tarchanow remarks that
it only remains for* the microscope to detect the structural
peculiarities of these three kinds of nervous end-organs. He
has noticed the unrest of the frog destitute merely of the cere-
bral hemispheres, already observed by Goltz long before. Finally,
having observed that warm dilute acid (in Tiirck's method)
called forth reflexes sooner than the same acid when cool, he
proceeds to draw from the fact two interesting conclusions:
1. "This result can be explained by the hypothesis that by the
higher temperature the irritability of the nerve-endings in the
skin is increased." 2. "In this way, probably, is to be explained
the well-known fact that on passing from a warm into a cooler me-
dium the animal reacts more quickly than when passing from a
cool into a warmer medium ; in the former case the end-organs
are in a more irritable condition."

Dr. M. Foster,10 in 1873, raised the question why, in the ex-
periment of Goltz described above, the brainless frog (a far bet-
ter reflex machine than the normal one) remains undisturbed in
water the temperature of which is gradually raised. Goltz

392 W. T. SEDGWICK.

argues for a greater dullness of perception in the brainless frog,
because it sits quiet under conditions which throw the normal
frog into violent movements, viz. under a rising temperature ;
but he does not mention that we have a strange anomaly in the
fact that the normal frog, whose reflex functions are feebler than
those of the decapitated animal, reacts much sooner upon the
application of heat-stimuli. I shall shortly return to this paper,
so that at present it need only be said that Dr. Foster, who
apparently had not seen the paper of Heinzmann, published a
year before, came to a result wholly different from that author.
Heinzmann believes the quiet of the reflex frog to be due to a lack
of stimuli-changes succeeding each other with sufficient sudden-
ness ; Foster, on the contrary, believes the spinal cord to be
directly depressed in function by the hot circulating blood.

In the same year (1873), appeared in the Russian language an
article by Archangelshy^11 on the influence of warmth upon the
nervous and blood-vascular systems of the frog. Of this paper I
have seen only an abstract given in the Jahresbericht of Hofman
and Schwalbe for 1S73. Archangelsky used in his work, as a
convenient means of regulating the temperature, a wooden box
having two windows, and provided inside with tubes arranged
zigzag upon its walls, so that they presented a large surface
to the air of the chamber and could be filled with hot or cold
water at will. He seems to have worked first upon normal frogs ;
and he found that when these were warmed to 29°-34° C, cramps
were readily observed, succeeded by weakness, inaction and heat-
rigor. He remarks that it is a matter of indifference whether the
air be moist or dry, the end-result being the same. When it is
moist, however, a high temperature is much sooner reached.

Upon decapitated frogs he has investigated according to Turck's
method the variations of reflex-excitability. He does not say
whether the air was dry or moist in this case. Under rapid
heating he finds the excitability, at first, heightened ; under slow
heating he was able to discover no change in the irritability at
the outset. He finds " in many cases " that when the temperature
has reached 25°-30° C. the reflexes evoked by acid become gradually
weaker and finally cease, though the tactile reflexes remain some-
what longer (33°-34°). The acid and the water for removing
it from the foot were kept inside of the warm chamber.

TEMPERATURE AND REFLEX ACTIONS. 393

Archangelsky has also studied, in respect to its behavior toward
warmth, the reflex apparatus analyzed into its separate parts ;
having sought in this way to locate the cause of the failure of
reflex power under heat.

(a) The end-organs. " It proved to be the fact that the higher
the temperature of the acidified water, the sooner were the feet
withdrawn. Hence the excitability of the end-organs is height-
ened by heat."

(5) The afferent nerve. Like other observers, the author finds
the afferent nerve to be more irritable when warmed ; he says,
however, that slow warming has no perceptible effect.

(c) The spinal cord. Two needles having been thrust into
the cord of a decapitated frog, were connected with an induction
apparatus. As a measure of the excitability, that distance of the
secondary from the primary coil which was just sufficient to pro-
duce a minimal contraction of the muscles, was employed. The
result proved to the author that the irritability gradually falls
and becomes zero at 34° — the very point at which the reflexes,
under similar conditions, also disappear. No preliminary phase
of increased activity is mentioned ; nor is it stated whether or
not the air was saturated with water. As a check upon this
experiment, more evidence was sought in this way : a brainless
frog was hung up in a glass tube, which covered only the upper
part of the trunk and left the pelvis and legs not covered in
any way. Including the tube just mentioned and connected
tightly with it was a larger glass tube of the same form. Thus a
hollow jacket was formed around the frog and yet not touching
him, and through this jacket could be passed water of different
temperatures. It turned out that rapid heating produced at first a
rise of excitability (measured by Turck's method) which speedily
passed over into a fall even to zero ; while gradual heating pro-
duced a steady fall, with no previous phase of heightened excita-
bility.

(d) The efferent nerve. This was investigated with the results
already reached by numerous observers. Like the afferent nerve
it preserves its irritability at a temperature above that at which
the reflexes fade away.

(e) The connected muscles. These were investigated with the
well-known result. The author found, however, that in dry air

394 W. T. SEDGWICK.

a muscle did not pass into rigor before 45°-50° had been reached ;
whilst in moist air it perished at 33°-35°. Its irritability in a
moist room quickly increases from 20°-30° and then gradually
decreases to 34°.

Archangelsky's conclusion is easily foreseen. The loss of reflex
excitability under heating is due, according to him, to a weak-
ening of the spinal cord alone.

Like Heinzmann, Fratscherf* working in 1875 in the Jena
laboratory under Preyer, does not at the outset undertake to
contribute to the discussion of the present problem. Heinz-
mann having reached the extremely interesting results described
above, it was an important question to ask if acids and alkalies
might also be so stealthily administered to a part of a living
animal (either brainless or normal) as to cause destruction of tis-
sue without having ever produced movement. This question
Fratscher took up under the direction of Preyer, and he had
already succeeded, as he believed, in demonstrating the truth of
the hypothesis, when, by Dr. Foster's paper, his attention was
called to the explanation of the effects of thermal stimuli gradu-
ally applied, and to the need for a repetition of Heinzmann 's
work. This he undertook, and he reiterates all of Heinzmann's
results, contradicting some of Foster's statements in a way
which will shortly be described. He finds that heat stimuli, as
well as acid and alkali stimuli, if only applied slowly enough,
may be concentrated so far as to produce tissue-death without
giving even a solitary movement.

Rosenthal™ in a brief summary of his " Studies on .Reflexes,"
published in 1875, states as one result of his work, that cooling
depresses reflex-excitability. This result, it will be observed, is
practically opposed to the conclusions of Tarchanow, Foster, etc.

In the same year, Freusberg 14 makes use of the experiment of
Tarchanow7 quoted above, and verifies it. He endeavors to
explain it upon his theory of " latent stimulation," and, what is
of great interest, shows that not only will ice-packing enormously
raise the reflex-excitability, but that packing in hot sand will do
the same thing, (cf. Archangelsky.11)

He distinctly affirms that an explanation is not to be sought
for in a general reduction of body temperature, " for this, on
the contrary, effects a general inactivity of the organism ; and

TEMPERATURE AND REFLEX ACTIONS. 395

besides, the phenomenon is so quickly produced by the ice-pack-
ing that it cannot be ascribed to that cause. "

Freusburg was at once attacked by Tarchanow,15 M who refused
to accept his explanation of the increased excitability seen in a
frog packed in ice. Tarchanow states that he prefers an explan-
ation offered (he omits to say where) M already some years ago,"
#nd states, as showing the falsity of Freusberg's theory, that an
exsanguinated frog does not show the same phenomenon which he
observed in 1871. From this observation he concludes that the
blood is an essential element in the experiment, and seeks to
account for the facts by supposing that the heightened excitabil-
ity is due to an excess of oxygen, the result of cessation of rapid
oxidations, or by considering a deficiency of CO* as the active
cause, etc., etc.

Freusberg11 returns the attack by showing defects in Tarcha-
now's method and obscurity in his results. He seems to me to
have decidedly the best side of the question.

The second part of Wundtfs18 Untersuchungen appeared in
1876, and contains one section devoted to the influence of tem-
perature and the time of year upon the reflex-excitability of the
frog. He has worked, however, only upon the effects of lower-
ing, and not upon the effects of raising the temperature. By em-
ploying methods similar to those of Tarchanow7 and Freusberg 14
(mentioned above) he has substantiated and somewhat extended
their results. By ice-packing of the trunk he obtains, like them,
an increased reflex-excitability, which, however, passes over
speedily into gradual depression, and finally into a condition of
complete inactivity under stimulation.

His explanation of the phenomenon is somewhat unlike
Freusberg's, which apparently he had not seen, and cannot be
given in full at this point. He considers it, however, as due
partly to heightened activity of the motor nerves and partly to
central nervous changes. He further points out a singular effect
of cold upon the spasms caused by strychnia. He affirms that,
as is well known, a small dose will produce violent spasms at the
ordinary temperature, while he adds that even large doses
produce no effect upon a frog in the cold. It is interesting to
compare these results with those of Kunde 3 given above.

396 W. T. SEDGWICK.

In closing, Wundt suggests that his experiments seem to offer
a sufficient explanation of the various changes which frogs
undergo in respect to their reflex-excitability during the various
seasons.

The latest contribution to this subject, so far as I know, is
embodied in a suggestion offered by Langendorff}* This
writer found that stimuli appear normally to travel along
the optic nerve, and to inhibit reflex actions to some extent,
perhaps by exciting the so-called " inhibitory centres " of Sets-
chenow. He recalls an observation of Fubini, that after
blinding the reflex-excitability of a frog is increased, sees in
it a confirmation of his own idea, and adds that he is inclined to
believe the rise of irritability after ice-packing, observed by
Tarchanow7 and Freusberg,14 to be due to an anaesthesia of
the skin, which, if I understand him, no longer sending in
exciting stimuli to the inhibitory centres, allows them to relapse
into quiet, and thus brings about heightened excitability.

The writer of the present paper was led to take up this sub-
ject by a perusal of Dr. Foster's article referred to above, and
more especially by certain evidence and conclusions recorded by
Dr. Foster which seemed to be scarcely harmonious with well
established physiological laws. The results of his investigations
(which have now extended over a considerable period) have jus-
tified him, he believes, in making still further studies. The pres-
ent communication will be devoted chiefly to a review of certain
parts of Dr. Foster's paper, and to the description of 6ome new
observations bearing upon the problems at stake.

II. The Experiment of Goltz.

As has been stated above, it was pointed out long ago by
Goltz that the brainless frog, if allowed to rest in water the
temperature of which is gradually raised, behaves wholly unlike
the normal frog under the same circumstances. The normal frog
leaps away, or, if confined, becomes violent in his attempts to
escape as soon as the temperature of the water reaches 30° or
thereabouts, while in the same vessel the brainless frog sits
motionless until death supervenes.

TEMPERATURE AND REFLEX ACTIONS. 397

This observation was repeated and verified by Dr. Foster,10
who saw, however, in the behavior of the brainless frog a new
problem which had not been touched by Goltz. Goltz's experi-
ment no doubt demonstrates as clearly as he meant to have it,
a difference between the normal and the brainless frog ; bnt, as Dr.
Foster observes, it presents a new difficulty, viz., " why the brain-
less frog is not excited to reflex action by the stimulus of the hot
water ? "

It might have been expected that a frog in full possession of
his faculties would be more acute than a brainless frog in per-
ceiving a temperature which was gradually rising to a painful
pitch, and more prompt and skilful in his endeavors to escape
than his neighbor destitute of a brain and scarcely recovered
from a recent profound operation ; but it would not have been
predicted that a decapitated frog, whose reflex functions are well
known to be keenly alive and even more delicately adjusted and
more easily aroused than those of the normal frog, would sit
unmoved in the presence of abundant stimuli until it perished
from excessive heat.

It is a surprising fact that although provided with a delicate
reflex apparatus, ordinarily responding to small heat stimuli
quite as well as to acids or mechanical injury, the brainless frog
remains perfectly calm in the presence of multitudes of power-
ful stimuli which are attacking large areas of his sensitive skin,
and makes not a single reflex movement worthy the name. Still
more astonishing is it when we learn that during this period of
calm, very complex and orderly reflex movements can be evoked
by a gentle touch or a drop of dilute acid, proving that the
reflex apparatus is not paralyzed, but, for some reason, though
wide awake to other and apparently feebler calls, is deaf to thoso
of the heat stimuli.

This problem which Dr. Foster has pointed out he has also
endeavored to solve. He has extended and modified Goltz's
experiment, using for the purpose brainless frogs suspended by
the jaw, and immersing the hinder parts to various depths in
water whose temperature could be gradually raised. In this way
various definite areas of the body-surface could be exposed to the
action of gradually-heated water, and his results are described by
him as follows :

398 W. T. SEDGWICK.

" Observation 1. If a frog, from which the brain has been removed,
be suspended by the jaw, with the legs hanging freely down and the
toes dipping into a vessel of water, on gradually heating the water
the toes are withdrawn by reflex action as soon as the temperature of
the water reaches a little over 30°. The result does not essentially
depend on the rapidity of the rise. However slowly the water be
heated, the feet are always withdrawn at a temperature of 35° or
earlier. Rapid heating may possibly lower the degree at which the
feet are withdrawn ; but to this I have not paid particular attention.
"Whether heated slowly or rapidly, the feet are withdrawn at about
35° C. or at a lower temperature.

Observation 2. If the whole body, thuB suspended, be similarly im-
mersed and heated, no movements (or only the very slightest spasms
of the muscles of the legs) take place ; and on still further raising
the temperature, the body becomes rigid (rigor cahris).

Observation 3. If both legs be immersed up to the anus and simi-
larly treated, they also become rigid without movement either of
ihe legs or of any part of the body, gave only a few spasms.

Observation 4. If one leg only be immersed and similarly treated,
it also becomes rigid without movements, or with only slight move-
ments.

Observation 5. If both legs (or one leg) be immersed up to the knee,
they are sometimes withdrawn, but sometimes no movements take
place, and the portion immersed becomes rigid. The results in this
case are not so constant as when either more or less of the body is
immersed.

Observation 6. If the feet only are immersed, they are invariably
withdrawn at 35° or under.

Observation 7. If a frog be suspended over a vessel divided by a
partition, with water at unequal levels on the two sides, so that one
leg is wholly immersed and the foot only of the other leg, and the
vessel be surrounded with water the temperature of which is gradu-
ally raised, neither the leg nor the foot will be withdrawn, if care be
taken that the water on both sides of the partition be equally and
uniformly raised in temperature. If, in this last observation, the
water on both sides be reduced to the same level, both feet are with-
drawn. This result shows that warm air and vapor have not the
same effect as warm water, and that the absence of movements is not
due to the unavoidable contact of the thighs of the animal with the
top of the partition giving some support to the legs, and .thus
diminishing the tendency to the withdrawal of the feet."

TEMPERATURE AND REFLEX ACTIONS. 399

It is not difficult to repeat these experiments and to arrive at
about the same results. It iB, indeed, my own experience that if
no special attention be paid to the rate of heating, and that if it
be not too rapid, one will obtain results agreeing essentially with
Dr. Foster's. If, for example, a brainless frog be immersed as
above described in water at 20p or 18° C, and the temperature
be raised to 40° (by a lamp below the vessel) in ten, fifteen or
twenty minutes, events will juetify the above statements.

If, however j a powerful burner be used and the water be heated
in muck less time than ten minutes, not even the frog immersed
to his fore limbs will remain quiet, but, like the frog with only
his feet immersed, will exhibit violent movements. It is easy to
prove, and is practically admitted by all observers, that under heat-
ing which is at all entitled to be called " gradual," the immersion
of an actively reflex frog suspended as described above and im-
mersed to the fore limbs or to the anus, will bring about such a
state of things that the animal will pass into heat rigor without
making a single movement of consequence.

A year before Foster's work was published, Heimmann* had
found that by gradual heating of an entire frog, or even of only
one hind leg, the temperature of the animal or of the part might
get to be so high as to produce rigor and yet without the least
disturbance of its general repose. He, however, puts special
stress upon the effects of very gradual heating, and makes the
important discovery that even a normal frog may be made to
perish in the same way without a struggle, provided only that
the increase of heat be gradual enough. This statement in-
volves a direct contradiction of the statements of Goltz,6 Tarcha-
now 7 and Foster,10 who have all agreed that under gradual heat-
ing the normal frog becomes violent in his attempts to escape.
The contradiction is only partial, however, for any one in half
an hour can prove to his satisfaction that the three observers
are correct ; while FraUcher u has fully justified Heinzmann.
The truth appears to be that if the heating be sufficiently grad-
ual, no reflex movements will be produced even in the normal
frog ; if it be more rapid, yet take place at such a rate as
to be fairly called " gradual," it will not secure the repose of the
normal frog under any circumstances, though it will do so for
the reflex frog if only enough of his skin be immersed, while

400 W. T. SEDGWICK.

it will fail if only a small portion be dipped ; again if the
temperature rises so rapidly as scarcely to be called "grad-
ual " iD its upward progress, not even the reflex frog will
remain quiet, though wholly immersed, but, like the normal
frog, will exhibit violent movements.

Heinzmann did not experiment with immersion of the feet
only, so that an interesting question was left after the paper of
Dr. Foster appeared, as to whether or not Heinzmann would
have succeeded in keeping the frog quiet by his extremely
gradual heating had he immersed only so small a portion of the
animal as the feet. This question has been answered in the affir-
mative by FratscherJ* who found that he could warm even the
normal frog to the point of rigor by immersing merely the feet.
My own work points in the same direction ; and we may take it
as settled that Foster was mistaken when he came to the conclu-
sions laid down in Obs. 1. I believe that I can explain, how-
ever, the result which Dr. Foster obtained. In my own case, at
least, I found that it was due to reflex movements, caused by dry-
ing. When the feet only are immersed a very large part of the
body is exposed to the dry air of the room, and the naturally
moist skin of the frog dries, producing reflex movements. In a
moist chamber it is not very difficult to raise the temperature of
the water in which the feet are dipping, higher than 35° without
causing movement.

It is plain from what has been said that the smaller the por-
tion of the animal immersed the more difficult it is to heat with-
out producing movements, and the more gradual must be the
rise of temperature. Moreover, since, as the part immersed gets
smaller, the surface exposed to outside stimuli gets larger ; while,
at the same time, the heating must be more gradual (thus pro-
longing the period of exposure) and the tendency to movements
gets greater, the slight stimulation due to drying, and perhaps
to the coincident cooling of the not immersed parts, becomes an
important factor in the experiment ; a factor which is less im-
portant and can be neglected when much of the body is
immersed, but which may lead to error when the feet only are
dipped. At least one safe conclusion maybe drawn at this point.
It is plain that if Goltz had slightly varied the conditions of his
experiment; if his brainless frog had not been in contact with

TEMPERA TURE AND REFLEX A CTIONS. 401

the heated water by a tolerably large surface, he would have
failed to demonstrate by this experiment that difference between
the normal and the headless frog for which he was seeking.

"We have next to consider why it is that the reflex (and the
normal) frog, in fall possession of healthy end-organs to detect
and sensory nerves to transmit painful impressions, may never-
theless exhibit total indifference to temperatures which are
gradually raised so high as to kill the tissues immersed. Differ-
ent explanations have been offered by Goltz,6 Heinzmann8 and
Foster10 respectively, and that of Goltz may be conveniently
referred to first. I have not seen his original communication
upon the subject, but if one may judge from the context in the
description of the experiment given in 1869, it appears that
Goltz 6 considers the lack of movement to be due to lack of " per-
ception." He regards the failure to move under abundant
stimuli as showing this lack of perception, not wanting in the
normal frog, which therefore displays movements. If this be
the theory of Goltz to account for the quiet of the reflex frog it is
plainly defective, since the reflex functions of the brainless frog
surpass in delicacy those of the normal one.

The theory of Heinzmann, 8 who approached the subject from
an entirely different standpoint and while endeavoring to solve
a different problem, may conveniently be deferred until the theory
of Foster, 10 who was, I believe, the first to raise the point at
issue and who has given the subject its most exhaustive treatment,
shall have been reviewed.

After describing the results of his investigation in the passage
quoted above, Dr. Foster writes as follows (p. 46):

" The above observations show that when the toes (alone immersed
in water) begin to be affected by the high temperature, say 30° C,
the stimulus of the hot water causes a reflex action which results in
the withdrawal of the foot. When the whole leg or body is immersed,
the same stimulus is still at work, but no reflex action occurs. What
is the reason that reflex action is absent ?

The following explanation is, perhaps, the first to offer itself. The
warmth applied to the leg diminishes the irritability of the nerves or
of the muscles, or of both ; and thus the impulses generated by the
warm water in the sensory terminations of the nerves of the foot are
not carried up to the cord, owing to the diminished irritability of the

402 W. T. SED0WI0K.

sciatic trunk, or, being so carried, the reflex process taking place in
the cord cannot manifest itself on account of the diminished irrita-
bility of the muscles or motor nerves.

But this view is clearly untenable. It requires that the nerves and
muscles, covered and protected by the skin, should be affected before
the sensory terminations in the skin itself. Moreover, no appreciable
difference in the irritability of the nerves, trunks or muscles of a leg
thus exposed to 35° C. could be detected. And it is directly contra-
dicted by Obs. 7, where the immersion of one leg prevents movements
in the other.

Two other views then suggest themselves.—-^!.) The blood return-
ing from the legs being wariner than the normal, raises the tempera-
ture of the spinal cord above thfe normal ; this reduces the irritability
of the cord, and hence reflex actions set going by a feeble stimulus,
tfhich in a normial cord would manifest themselves, are here absent,
(2) From the stimulation of the Whole leg as compared with that of
the foot, a multitude of impulses, arising from ail parts of the skin
exposed to the warm water, reach the spinal cord. These produce
such an effect upon the oord that the simpler reflex action resulting
from the stimulation of the toes alone is prevented."

It will bo observed that the question raised in the first part of
the passage here quoted, in view of what has been said above,
would now have to be stated somewhat differently ; nevertheless,
the question is at bottom much the same, viz. why the frog is
not excited to reflex action by the stimulus of the hot water. It
may be well also to recollect, at this point, that the rapidity with
which the temperature may be raised without causing reflex
movements seems to depend largely upon the amount of surface
immersed.

It will be instructive to follow the evidence which leads Dr.
Foster to accept as the principal cause of the phenomenon in
question, the former of the two views which he has suggested*

III. Is it true that the brainless / rog tits motionless in water
which is gradually heated, because the irritability of his spinal
cord is depressed by heat brought by the Mood from a remote part
of the body t

This is the theory finally adopted by Dr. Foster; hence it
must bo specially examined. It involves oH6 very conspicuous1

TEMPERA TUBE AND REFLEX A CTIONS. 463

objection, however, which Dr. Foster haa not overlooked, but
which he dwells upon in these words (p. 50) :

" In all observations on the effect of a rise of temperature on living
animal tissues, the state of exhaustion or depression which ultimately
ensues is preceded by a stage of exaltation in which the functions of
the tissue are raised above the normal. This is well shown in the
case of muscles, nerves and the heart In none of the observations
recorded above was there any indication of such an initiative stage of
ihcf eased action. Had there been it would naturally have led to the
withdrawal of the feet in all cases. And the absence of this presented*
a great difficulty to considering the results obtained as being merely
due to a depression of the powers of the spinal cord by reason of the
increased temperature.

" Some observations, however, made in the laboratory here by Mr.
T* O.. Harding, afforded a clue, by pointing out a distinction between
simply and directly raising the temperature of an organ or a tissue^
and indirectly heating it by supplying it with blood heated beyond
the normal in some distant part of the economy. Thus the heart of
a frog, either empty or filled with serum, when heated beats with a
more frequent rhythm and, at first, with greater force. But the same
heart when indirectly heated by the immersion of the legs of the frog
in hot water (the heart remaining in the body and the brain and
spinal cord being destroyed) is lowered at once both in the force and
frequency of its beat, by reason of the heated blood with which it is
supplied. This result leads us to expect that in the same way the
spinal cord, if heated by being Supplied with blood heated beyond
the normal, would be depressed without any preceding stage of exal-
tation, and thus reflex actions which otherwise would have occurred
be prevented. The observation, Obs. 7, where the heating one leg
prevents reflex action in the other, seems to point distinctly to sncfr
an explanation/'

These " observations " of Mr. Harding were what drew my
attention to this subject in the first place; If true they are of
extreme importance. If an organ, either empty or full of bloody
is to behave in one way when directly heated, and in another
way, exactly the reverse of the former, when blood heated ia a
remote part is passed through it, it ie certainly a very surprising
fact, well worthy of thorough investigation, I hare not fotmd,
however, any other reference to Mr. Harding's work, and am

404 W. T. SEDGWICK. •

forced to believe that he pursued it no further. I am also in the
dark as to his exact method of experimentation, but I assume
that his frogs were hung up by the jaw and the legs only were
immersed in water, as seems to be implied in the passage just
quoted ; if they were not, it is possible that some of the remarks
I am about to make may be irrelevant.

The results of Mr. Harding's work which have given Dr. Foster
a " clue " seemed so novel that I set to work to see if 6ome expla-
nation of the facts could be obtained which would not compel
us to believe that heating of the frog's heart from the inside by
blood warmed in a remote part has an effect upon it diametri-
cally opposed to the effects of heat directly applied from the out-
side. I began by making preliminary experiments, and employ-
ing the method that I suppose Harding to have used, viz. hang-
ing the frog by the jaw after destroying the brain and spinal cord.
The heart was exposed by a small hole cut in the chest wall, and
I, like Harding, saw that the heart beat slower as the tempera*
ture of the water about the legs in the vessel below, rose.

Bearing in mind the work of Cyon,30 who has shown that
passing hot blood through the mammalian brain slows the heart
beat by stimulation of the vagus, I was led to inquire if it
might not be that in the experiments of Harding, the hot circu-
lating blood acting as a common heat stimulus, irritated directly
the trunk of the vagus somewhere along its course, and so over-
came (by ordinary vagus inhibition) that increase of function
which the heated blood might be supposed to induce in the heart
itself. It was but a forlorn hope ; for aside from the fact that the
hot blood pouring through the cavities of the organ would be
presumably the more powerful stimulus, it might also be expected
that the vagus would soon get wearied ; though between the two
antagonizing forces we should look for intermittent or irregular
pulsations — which we never get. Still it was possible, and so I
tested the idea by making another experiment, after previous
administration of a small dose of atropine sulphate. This doubt-
less paralyzed the vagus, but the result of the experiment was
exactly the same as before : the heart beat steadily slower as the
water about the legs grew warmer.

In repeating Harding's experiments with the frog suspended
by the jaw and his legs in heated water, I was struck, however,

TEMPERATURE AND REFLEX ACTIONS. 405

with the emptiness of the heart. Its paleness and feeble beat were
conspicuous; the aortic arches were white and empty, while
the vessels of the thighs seemed gorged with blood. To the
eye there appeared to be little or no circulation, and I was thus
led to ask : Does the blood in these cases really circulate so as to
heat the heart ?

If we reflect upon the conditions we must admit that they are
highly unfavorable for a good circulation. The brain and spinal
cord having been destroyed, all vasomotor centres are out of the
question, and their influence in maintaining blood pressure is
lost ; hence " resistance" is removed, arterial pressure falls, and
the blood flows freely from the heart and arteries into the veins ;
here it settles slowly into the legs and viscera, and remains there
(respirations and movement — the conditions requisite for adequate
venous pressure — having long since ceased) under the simple
influence of gravity. If the web be examined with a microscope,
no circulation will be detected in a frog destitute of spinal cord
and hanging by the jaw. I can scarcely suppose that Dr. Foster
and Mr. Harding have overlooked so elementary a fact, if indeed
their experiments were conducted in this way ; but the heart
certainly does beat slower in these cases, while the legs are
gradually heated, though, contrary to Mr. Harding's belief, no hot
blood passes through it. A thermometer placed upon the heart
or among the viscera, or in the stomach near by, if the heart goes
slower, never shows any rise of temperature, though the tempera-
ture of the water about the legs may be raised from 20° to 40°
while the observation is being made; conversely, if in any such
case the heart does beat faster it will always be found to be
warmer than before.

Moreover, quite aside from temperature changes, I have
repeatedly seen the heart-beat, in a frog destitute of brain and
spinal cord, fall as much as from forty-four to twenty-eight beats
per minute on simply changing the position of the animal from
the horizontal to the vertical. In short, from numerous experi-
ments I am forced to conclude that in cases similar to those
described by Dr. Foster as observed by Mr. Harding, the heart
beats slower, not because of heat nor from heated blood, but
owing perhaps to starvation ; possibly to a zero, or even nega-
tive venous pressure; or to some cause equally remote. In every

408 W. T. SEDGWICK.

ease where the heart was actually heated by warm blood its
beats were increased in frequency, often to a surprising extent.

It must not be forgotten that we have, in the vessels of the frog
destitute of brain and spinal cord, a system of flaccid tubes, only par-
tially filled with a fluid which, with little or no hindrance, obeys the
laws of gravity. It will be found that with no vital spinal cord, the
frog's* heart behaves very differently according as the animal's body is

(a) horizontal ;

(S) vertical, with head highest; or,

The heart of a frog lying horizontal in a pan of water, and having
the spinal cord intact, beats regularly and powerfully, driving into
the arteries with considerable energy the blood which goes to keep
up the head of arterial pressure. In the frog destitute of a spinal
cord, however, it will often be found — particularly if care has been
taken to destroy all of the cord — that the arches springing from the
base of the heart are white and empty ; while if the beats be counted,
it will frequently, indeed usually, appear that the rate per minute is
growing less and less. Now, the truth is that in this case there is
little or no circulation!. The heart is really the highest organ in
the prostrate flaccid body ; and on the familiar principle that liquids
will not freely run up hillT the venous blood subjected to no vis a
tergo'm. the muscles, and free from the pull of thoracic aspiration,
lags behind and gravitates* into the lowest veins. If, perchance, any
venous blood gets crowded up through the auricles and into the ven-
tricle, it is speedily pumped down the hill again through the arteries,,
and by their elasticity is driven on into the veins.

I have repeatedly seen cases like the following, which may serve
as a type ; it is an actual observation.

A small bull frog had his medulla divided and his brain destroyed
at 10 A. M. He lost very little blood and rested quietly in a pan of
water until just before the beginning of the observations, when the
heart was exposed (but left in the pericardium) by a small hole made
in the ventral chest-wall. Observations were made once in three
minutes.

TEMPERATURE AND REFLEX ACTIONS.

407

No. of
Observation.

Time.

Rate of Heart Beat
per i minute.

Remarks.

11.28

Arches very red and full.

11.31

11.34

Bulbus arteriosus beating

11.37

very powerfully.

11.40

11.43
11.46

Spinal cord destroyed at

11.44. Arches paler.

11.49

11.52

Arches white and appar-

11.55

ently empty.

11.58

12.01
12.04

Hung up by the feet.

12.07

Arches very full.

12.13

12.16;

12.19

(J.) With the body vertical and the head highest, the condition of
things just described is aggravated. The fore part of the body
becomes exsanguinated, and even the ventricle, which, in the case last
mentioned, usually contains some small amount of blood, may get to
be perfectly pale and white. The blood settles away into the hind
legs and visceral veins, the great capacity of which is well known.
It may be of interest to recall in this connection the fact that after
the administration of certain drugs (e. g. quinine) which depress the
reflex-excitability of the cord and hence impair the circulation, the
blood will in the same way be found after a time chiefly in the hind
legs and viscera.

Whether the downward pull of the fluid due to gravity may or
may not cause a negative pressure in the heart I have not ascertained.
When the legs are gently heated in water of a rising temperature the
vessels probably relax somewhat, thus further robbing the heart of
blood ; at any rate we may take it as certain that in these cases the
heart is not filled with warmed blood coming from the legs, and its
retarded beating cannot be considered as due to that cause.

(c.) With the body vertical and the feet uppermost^ the blood previ-
ously contained largely in t}ie hind legs and viscera flows freely down
into the heart This organ fills, gets very red, and beats much more

408 W. T. SEDGWICK.

powerfully. The beats seem also to become more frequent, though
the original rate is seldom or never attained.

No better demonstration of the meaning of "resistance" as an
element in blood-pressure, or of the fact that this resistance is due to
nervous influences residing chiefly in the cord, could be desired for
laboratory use. The frog which has been made to pass through
stages (a) and (J), can be turned with now the head and now the
feet highest; and the demonstration is complete of a system of
partially-filled tubes through which blood flows most freely, and
which in spite of an active pump and abundant arterial elasticity is
nevertheless, without "resistance," no circulation at all.

From the foregoing considerations it seems clear that if my
observations are correct, the u clue " of Dr. Foster leads to noth-
ing. As au u analogy " supporting the theory under considera-
tion it is worse than useless, for it leads to results which tend to
weaken that theory. It is plain that the heart of thd frog has
never yet, when freed from all extrinsic nervous influences, been
made by heat or by heated blood to beat at first more slowly ; on
the contrary it always beats faster when fed with heated blood.

The theory that any organ or tissue having a protoplasmic
basis may so far depart from obeying the laws of protoplasm as
to reverse them completely, and under gentlq heating may suffer
loss of functional power with no preliminary phase of increased
activity, if true in the case of the spinal cord of the frog (which
is plainly protoplasmic), stands now wholly unique, and must be
proven beyond all question if it is to stand at all.

The examination of the evidence for and against this theory
will be reserved for the second paper.

AUTHORS AND PAPERS REFERRED TO IN THE TEXT.

1. Brown- Sc'quard. — (a.) 1847. Note stir la durce de la vie des
grenouilles en automne et en hiver aprcs l'extirpation de la moelle
allong£e et de quelques autres portions du centre nerveux c^rtfbrora-
chidien. Comptes rendus, tome 24, p. 363.

(b.) 1851. De la survie des batraciems et tortues apr£s l'ablation
de leur moelle allong^e. Gazette mtfdicale de Paris, 1851, p. 476.

TEMPERATURE AND REFLEX ACTIONS. 409

2. Kunde, Dr. F. 18G0. Der Einfluss der Warme und Electricitat
auf das Riickenmark. Virchow's Archiv, 18, 357.

3. Cayrade, Jules. 1864. Recherches critiques et expirimentales
eur les mou Yemen ts reflexes. Thtse pour le doctorat en m^decine,
Paris, 1864.

4. Weir- Mitchell, S. Jan., 1867. On retrogressive motions in birds
produced by the application of cold to the cervical spine, with
remarks on the use of that agent as an aid in physiological investiga-
tion. Amer. Journal of the Medical Sciences, No. 105, p. 102.

5. Richardson, B. W. May, 1867. On the influence of extreme
cold on nervous functions, etc. Medical Times and Gazette, 1867,
p. 489.

6. Goltz, F. 1869. Beitr'age zur Lehre von den Functionen der
Nervencentren des Frosches. Berlin, 1869, p. 127, etc.

7. Tarchanow, J. 1871. Ueber die Wirkung der Erwlirtnung resp.
Erkiiltung auf die sensiblen Nerven, das Hirn und Ruckenmark des
Frosches. Bulletin de Tacademie imp£r. des sciences de St. Peters-
burg. Tome XVI (1871), p. 226.

8. Heinzmann, A. 1872. Ueber die Wirkung sehr allmiiliger Mn-
derungen thermischer Reize auf die Empfindungsnerven. Archiv
fur die gesammte Physiologie, Bd. VI (1872), p. 222.

9. Tarchanow, J. 1872. Zur Physiologie der thermischen Reflexe.
Original paper in the Russian. Abstract in Hofman & Schwalbe's
Jahresbericht, Bd. I (1872), S. 520.

10. Foster, Dr. M. 1873. On the effects of a gradual rise of tem-
perature on reflex actions in the frog. Journal of Anat. and Physi-
ology, viii, 45 ; also, Studies from Physiological Laboratory, Univ.
Cambridge, 1873, p. 36.

11. Archangelsky, P. 1872. Ueber den Einfluss der Warme auf
das Nerven- und Blutgefuss-System des Frosches. Original paper in
the Russian. Abstract in Hofman & Schwalbe's Jahresbericht, Bd.
II (1873), S. 555-559.

12. Fratscher, C. 1875. Ueber continuirliche und langsame Ner-
venreizung. Jenaische Zeitschrift, N. F. II (1875), S. 130.

13. Rosenthal, J. 1875. Studien uber Reflexe. Monatsberichte der
Berliner Akad., 1875, S. 419.

14. Freusberg, A. 1875. Ueber die Erregung und Hemmung der
Th'atigkeit der nervosen Central organe. Pfliiger's Archiv, Bd. X
(1875), S. 181.

15. Tarchanow, J. 1875. Augmentation des actes rt flexes sous rinflu-
ence du froid. Gazette m£d. de Paris, 1875, p. 287.

410 W. T. SEDGWICK.

16. Tarckanow, J. 1875, De l'influence de l'augmentation de Poxy-
gene ou de l'acide carbonique dans le sang sur lee actes reflexes de la
grenouille. Gazette m£u\ de Paris, 1875, p. 426.

17. Freutberg, A. 1876. Kalte als Keflexreiz. Archiv fiir exper. Path,
nnd Pharm. Bd. VI, S. 49.

.18. Wundt, Wilhelnu 1876. Untersuchungen aur Mechanik der
JServem und Nervencentren. Zweite Abtheilung, S. 59. Stuttgart,
18.76.

19. Zangetulorff, 0. 1877. Die Beziehungen des Sehorganes zu den
reflexhemmenden Mechanismen des Froschgehims. Archiv fiir
(Anat. und) Physiologie, 1877, S. 435.

20. Cyon, E. Ueber den Einfluss der Temperaturveriinderungen
auf die centralen Enden der Herznerven. Pfluger's Archiv. Bd.
VIII, S. 340.

NOTES ON THE DEVELOPMENT OF PANOP.2EUS

S ATI (Smith). By B. A. BIRGE, Ph. D., Professor of Zoology
in the University of Wisconsin. With plates XXX, XXXI,
XXXII and XXXIIL

The observations on which the following paper is based were
made at the Johns Hopkins Summer Laboratory in the summer
of 1878. The paper was written in the following college year
1878-9, and was lost in transmission to Baltimore. Absence
from the country and press of other work have deferred the
reproduction of the paper until this late date, when it seems best
to print only such parts as are directly concerned with observa-
tions, leaving out all general considerations.

To Dr. W. K. Brooks, the director of the laboratory, I have to
offer my thanks for the generous opportunities for study fur-
nished by the laboratory under his charge.

PanopcBus sayi (Smith), and P. depressus (Smith), are both
very common in the neighborhood of Crisfield, Md. They
swarm along the muddy shores, under stones and oyster-shells,
and, especially P. depressus, in the interior of sponges in deeper
water.

In spite of this abundance of material I found it impossible to
raise any one crab from the egg to the adult stage. Specimens
were raised from the egg to the second zoea stage, and the moults
observed from stage to stage throughout. As, however, many
moults do not alter the form of the zoea, it has been found impos-
sible to determine the number of these operations during the
larval life.

Egg-development — As my plan of study did not include
observations on the intraovular development of the Crustacea, few
notes were made on the eggs during the first part of my stay at
Crisfield. When I discovered to which crab my zoeas belonged,
it was too late to trace in order the development of a single set of

412 E. A. BIRGE.

eggs, and those crabs which I left behind in different stages of
development unfortunately died. I can therefore present only
detached notes on this part of the life-history.

The diameter of the egg is about 0.2 inch. The yolk is com-
posed of olive-green globules of various sizes. The nauplius
stage was the first observed (Plate XXX, Figs. 1 and 2). The head-
and tail-folds first appear, and, within a few hours, the first three
pairs of appendages in rapid succession from before backward.
At the same time the telson becomes divided and the labrum is
marked off from the head. The branch of the antenna soon
.-appears in the form of a lobe on its posterior side. Both antennae
are from the first directed toward the dorsal side of the embryo,
and m their growth soon cover the mandible (Plate XXX,
Figs. 1 and 2).

In the next stage all the appendages of the young zoea are
present. They rapidly appear — the entire change from Fig. 1 to
Fig. 3 (Plate XXX) taking place in less than 18 hours. The
antennule has become wrinkled, showing its rapid growth in
length ; the antenna has gained four small points on the main
stem, the rudiments of the four great lobes of the larval skin. The
mandible is larger, otherwise unchanged. The first maxilla shows
traces of its future lobes, while the second maxilla is bilobed
.from the start This last appendage is also crowded out of place
and partly concealed below the abdomen. The two maxillipedes
are as yet simple outgrowths of the blastoderm, showing no trace
of division ; their long axis lies parallel to that of the abdomen.
The telson shows six lobes on each side, the rudiments of the
future cuticular appendages. The clefts separating the head- and
tail-folds from the underlying blastoderm are much deeper, and
the cephalic lobes are more definite in shape (Fig. 6, intermediate
between Figs. 3 and 4).

The next stage is represented by Figs. 4 and 5, Plate XXX.
Fig. 5 is from a slightly older embryo than that shown in Fig. 4.
Here the larval skin is quite firm and is easily demonstrated*
The eye is'clearly marked out, though still without pigment. The
antennule has its cuticular appendage, and the rudiments of
spines and hairs appear on all appendages which are to bear
them. The third maxillipede has appeared and is apparently
larger proportionally than after hatching, All the maxillipedee

DEVELOPMENT OF PAN0P2EUS SATI. 413

are divided into endopodite and exopodite, and the segmenta-
tion of the abdomen is plainly marked. The carapace is present.
The yolk still fills the whole dorsal part of the egg.

Development now goes on more slowly. The appendages
take on their proper form within the larval skin. The abdomen
grows forward between the eyes and reaches nearly to the heart.
The larval skin grows out into the long cuticular appendages of
the antennule, antenna and telson. The hairs of the maxillipedes
develop and are invaginated into the terminal joints of those
appendages. Figment is deposited in the eye and the macula
nigra appears. Both come at the same time, or the eye a little
sooner. This order of appearance reverses that of Palemoneies
vulgaris as observed by Faxon {Bull. Mus. Comp. Zool. Vol. 15,
p. 308).

The yolk is absorbed and only a few globules are left at the
time of hatching. The rudiments of the spines of the carapace
appear.

First Zoea stage. Still m larval skin. — Plate XXX, Fig. 9,
and Plate XXXI, Fig. 1.

When hatched the young Panopseus is still enveloped in the
larval skin, which it retains for several hours. The time varies
with the activity of the specimen from two or three hours to
as many as twenty-four. The shorter times are the more com*
mon.

The skin is unsegmented and takes no part in the fold of the
carapace, nor is it prolonged for any of the hairs of the appen-
dages or the spines of the telson. It bears, however, numerous
hairs itself, and has peculiar prolongations, which will be spoken
of in detail under the description of the appendages.

The dorsal, lateral and frontal spines can be detected under
the skin, and these, as well as the long hairs of the maxillipede,
are ready to push their way out as soon as the larval skin is cast
off— or rather in the act of moulting. The same is true of the
invaginated antenna and the spines of the telson. The abdomi-
nal spines are merely indicated. Numerous spots of black
pigment are present in carapace, abdomen, mandible and
maxillipedes. The labrum is enormous, projecting downward
between the mandibles. No trace of the third maxillipede was

414 E. A. BIRGE.

seen, although it is probably present. The thoracic legs have not
appeared.

The mnscles of the animal are still weak, as is also the skele-
ton. The animal is sluggish in its movements, and usually
carries the abdomen bent, as in Fig. 1, Plate XXXI.

Second Zoea *%tf.— Plate XXXI, Fig. 2.

With the casting of the larval skin the regular zoea form is
assumed. It is characterized chiefly by the great length of the
dorsal, and, especially, the frontal spines, and by the correspond-
ing length of the antennae, a feature in which according to Faxon
{Bull. Mu8. Comp. Zool. Vol. VI, No. 10) this species stands
alone. The structure is shared, however, by the sister species
P. depressus, although neither spine nor antenna is so long {cf.
Plate XXXII, Fig. 12). The maxillipedes bear four long, jointed,
plumose hairs on the exopodite. The abdomen has four joints
besides the telson, of which the first has a short spine on each
side, which bears against the sides of the carapace when the
abdomen is flexed. The telson is developed into a long fork,
bearing on the inner side six spines, and one on the outside of
each arm. The animal is very active, swimming and kicking
vigorously.

The beautiful figure of Faxon {Bull. Mus. Comp. Zool. Vol.
VI. No. 10, Plate II, Fig. 4) represents this stage.

Third Zola stage.— Plate XXXI, Fig. 3.

The zoea moults a large number of times, some moultings mak-
ing little or no change of form. The stage represented in Plate
XXXII is reached after as many as three moults. I have kept a
zoea which moulted twice from the first stage without causing
any change of form.

The third stage is characterized by a greater size, greater pro-
portional length of the frontal spine, which may be slightly
longer than the antennae. The last abdominal segment before
the telson has developed two long spines, and the spines on the
second segment are larger. The maxillipedes bear six swimming
hairs, instead of four, and there are thick-set hairs on the edge of
the carapace. The abdominal legs can now be distinguished as
masses of cells lying under the skin. They cause no elevations of

DEVELOPMENT OF PANOPJEXJS SA TL 415

the skin as yet. The eye is larger and more movable. The
thoracic feet are unchanged.

Later jZoea stages.

Numerous changes of minor importance occur in the moultings
between the third stage and the last. The abdominal legs appear,
first as simple elevations, then becoming divided by a joint and
gaining a rudimentary endopodite. The sense-hairs of the
antennule increase in number to six or seven, and the swimming
hairs of the maxillipedes to eight and nine respectively. The
rudiment of the permanent antenna appears as a small lobe on
the inner side of the larval antenna. The number of abdominal
spines increases to three pairs.

Last Zoea stage. — Plate XXXI, Figs. 4 and 5.

The chief characteristics of this stage, apart from greater size,
&c, are the division of the telson, the appearance of the mandi-
bular palpus, and the segmentation of the antennule.

The abdominal feet have grown and the endopodite is plainly
marked (Plate XXXIII, Fig. 81). The swimming hairs have
increased to twelve or fourteen on each maxillipede. The animal
is, however, very sluggish, lying for hours quiet at the bottom of
the jar. This habit makes the zoeas of this stage rather rare in
the open water, so that it is easier to raise them than to find
them. They moult less frequently and accumulate all sorts of
d&yris and parasites upon their shells, making their study more
difficult.

From the antenna has grown the projection in which the per-
manent antenna is developed. The outgrowth holds the same
relation to the permanent antenna that the larval skin holds, i. e.
it is a mere sheath — unsegmented. Inside of this sheath the
segmentation of the true antenna goes on. The antennule, on the
contrary, becomes segmented into three or four joints, and de-
velops a small outgrowth on the basal joint.

A new abdominal segment is formed by the separation of the
anterior part of the telson, whose forks are now at their maxi-
mum size. A small unjointed palpus appears on the mandible,
the thoracic legs are developing inside their skins, and the gills

1 From an earlier stage, bat essentially like this.

416 E. A. BIRGE.

and epipodites are present. It is worthy of note that all the
appendages of the crab appear as nnjointed projections of the
skin, inside of which the segmented appendage develops. In the
case of the thoracic legs this is especially marked. Several succes-
sive sacs are formed for the developing leg, all nnjointed even
where joints are distinguishable inside the sac. Joints, however,
appear before the zoea stage is left.

Further peculiarities of the appendages will be considered
under the appropriate head.

First Megalops stage. — Plate XXXI, Figs. 7 and 8.

With the moult from the last zoea stage to the first megalops an
enormous change takes place both in the form of the body and
of appendages. All of the long spines are entirely lost and
leave no trace behind. Panopseus thus differs from Cancer as
figured by Smith ( U. S. Fish Com. Rep., 71-2, Plate VIII),
where the frontal and dorsal spines persist in the megalops. The
form of the carapace is changed from one horizontally compressed
to one vertically flattened. The abdomen suffers the same
change in proportion, and the telson loses its fork and becomes a
simple plate. No less marked is the change in the appendages.
These will be spoken of in detail later. All are profoundly/nodi-
fied. The maxillipedes and thoracic legs undergo the greatest
change, the former losing greatly in size and the latter gaining.
The abdominal legs get their hairs. The abdomen is usually
carried stretched straight out or slightly bent down, and is used
in locomotion. The ear-sac can be seen in the base of the anten-
nule, and the permanent antenna replaces that of the zoea. Both
are partly concealed by a broad flat plate, projecting forward on
the carapace. In the middle of this a small notched projection
is the only suggestion of the frontal spine. The animal is covered
by scattered coarse hairs.

Subsequent changes in the megalops affect the proportions of
the carapace, which becomes broader proportionally, and that of
the abdomen which becomes smaller, and is permanently flexed
under the sternum. The appendages undergo many changes,
gradually approximating them to the adult form.

The last megalops stage is reached after several — at least
four — moultings.

DEVELOPMENT OF PANOP^US 8A TI. 417

aving this stage the megalops assumes the form shown in
) XXXI, Figs. 9, 10, the first crab stage. Here the carapace
oet most of the broad notched projecting plate in front, and
Ige has assumed a curve not greatly different from the adult
• Each side of the carapace bears three teeth which persist
le adult. The abdomen is also nearer the adult form,
9 the appendages have not greatly altered from the last
ilops stage. No specimen was reared beyond this stage.
le youngest crab found is figured in Plate XXXI, Fig. 6.
is already the adult proportions. The erenulation of the
rior border of the carapace is more distinct than later. The
ne of the carapace of a large male is figured in Plate
£1, Fig. 11.

/d.— First zoea stage, Plate XXX, Fig. 9 ; XXXI, Fig. 1.
nd zoea stage, Plate XXXII, Fig. 1. First megalops stage,
b XXXII, Fig. 2. Adult, Plate XXXII, Fig. 3.
te eye undergoes few changes during the zoea stage. It
Dies larger and more movable as development progresses,
in form and proportion alters little. It is, however, longer
he older zoeas. The eye has the same general form in
negalops stage and through the first crab stage. When it
tnes divided into two joints, and when the sinus in the
ea is developed, I cannot say. The adult form is present in
stab of Plate XXXI, Fig. 6.

ntmnule. — First zoea, Plate XXXII, Fig. 4. Second zoea,
B XXXII, Fig. 5. Third zoea, Plate XXXII, Fig. 6.
r zoea, Plate XXXII, Fig. 7. Last zoea, Plate XXXII,
8. First megalops, Plate XXXII, Fig. 9. Adult, Plate
HI, Fig. 10.

lie antennule in the first zoea stage is enveloped in the larval
h which extends out in a very long plumose expansion, and

I on one side a short and slender branch. This carries a tuft
airs at its end. Into this branch the sense-hairs of the per-
out zoea antennule extend. The long projection of the

II skin is plainly homologous to the seta of the antennule of
Arval Callianassa, as figured by Claus,1 although that zoea is
later stage of development.

JKlenochungen zur Genealogischen Grundlaga dee Crustaceen System, T.
1g. 2.

418 E. A. BIROE.

In the second zoea the antennnle has the ordinary elongated
conical form, and bears one short and two long sense-hairs. In
later stages the number of hairs increases to three and finally to
six or seven. A lobe also appears on the inside of the antennnle.

In the last zoea stage the antennnle has divided into three
( ? four) joints, and the lobe is attached to the second from the
base. One hair among the sense-hairs seems much stouter than
the rest, but its subsequent fate was not traced. The number of
sense-hairs is greatly increased.

In the first megalops stage the lobe has formed a distinct joint,
bearing two hairs on its end. The terminal joint shows traces of
segmentation which afterwards disappear. The basal joint is
enlarged for the ear.

In the adult the joint formed from the lobe is divided into six
parts, and the expansion of the former terminal joint is smaller,
although its hairs have greatly increased. The basal joint is
also larger.

Antenna.— First zoea, Plate XXXII, Fig. 11 ; Plate XXXI,
Fig. 1. Second zoea, Plate XXXII, Fig. 12. Later zoea,
Plate XXXII, Fig. 14. Last zoea, Plate XXXII, Fig. 15.
First megalops, Plate XXXII, Fig. 16. Adult, Plate XXXII,
Fig. 17.

The larval skin enveloping the antenna is much shorter than
the permanent organ of the zoea, and bears on one side a very
large four-lobed appendage. Into the base of this projects the
minute " squamiform appendage " of the zoea antenna. The
lobes of the cuticnlar expansion are covered with short fine
hairs. The spine which forms the main part of the future antenna
is greatly wrinkled and invaginated, so as to be only about one-
third as long as in the next stage.

The second zoea stage shows the proper zoea antenna — an
enormously long spine, smooth and gently curving, extending to
the tip of the frontal spine. This is the spine of the ordinary
zoea antenna — the " stachelfortsatz " of Claus, "exopodite" of
Balfour. It is probably the epipodite, while the " squamiform
appendage " — " ramus exterior " of Claus — is the exopodite, and
the permanent adult antenna is clearly the endopodite.

The squamiform appendage — apparently overlooked by Faxon —
is a minute joint, situated near the base on the inner side of the
antenna, and bearing a single terminal hair.

DEVELOPMENT OF PANOPJETJS SAYI. 419

There is no trace of the adult antenna. This structure appears
in the older zoeas after numerous moults, as a small lobe on the
inner side of the spine. This extends and increases in size during
the later changes of skin, and finally the joints of the megalops
antenna can be plainly seen within it.

In the moult to the megalops stage the spine and ramus exterior
are lost, and the permanent antenna, consisting of about eleven
joints, takes its place. The third or fourth joint from the end,
as in Carcinus maenas^ is enlarged and bears large sense-hairs.

In the adult the antenna has 18 to 20 joints, and the sense-
hairs are about equal in size. When the opening of the green
gland is formed was not determined.

Mandible. — Second zoea, Plate XXXII, Fig. 18. Second
zoea, Plate XXXII, Fig. 19. Last zoea, Plate XXXII, Fig. 20.
First zoea, Plate XXXII, Fig. 21. Young crab, Plate XXXII,
Fig. 22. Adult, Plate XXXII, Fig. 23.

The larval skin of the mandible presents no features of especial
interest.

In the second zoea stage the mandible bears at each end two
projections. Of these the anterior one at the proximal end serves
as the articular point, while to the other is attached the main
muscle. Of the two distal projections, the outer — lower — is thinner
than the other, which is toothed, and serves as the main instru-
ment in chewing. The axis of the jaw passes through this sur-
face and the articular projection ; and the appendage is rotated
on this axis by the muscles. As the zoea becomes older the
attachment of the muscle (in Fig. 18) extends further toward the
distal end of the appendage.

No marked change in the form of the mandible occurs before
the last zoea stage, when the palpus shows itself as a small eva-
gination on the anterior edge. This feature is diagnostic of the
last zoea. The notch in the posterior side, in which the labium
lies, becomes deeper.

In the first megalops stage the palpus is three-jointed, and the
appendage differs only slightly from the adult form. The two
proximal projections are larger proportionally, the cutting sur-
face is less sharp and its tooth is not so clearly marked, the
whole structure is broader proportionally. As the carapace
grows in breadth the mandible lengthens and acquires the adult
form.

420 E. A. BIRGE.

The cutting surface of the adult mandible is the lower pro-
jection of the zoea mandible, and the flat surface back of the
edge corresponds to the grinding surface of the zoea.

The upper lip is enormously large in the first zoea stage, and
becomes smaller, covered with hairs and enclosed within the
mandibles. The shape is little altered during development.

First Maxilla.— First zoSa, Plate XXXII, Fig. 24. Third
zo§a, Plate XXXII, Fig. 25. First megalops, Plate XXXII,
Fig. 26. Young crab, Plate XXXII, Fig, 27. Adult, Plate
XXXII, Fig. 28.

The larval skin of the first maxilla shows three elevations
corresponding to the parts of the appendage. It bears no hairs
or setae.

With the second zoea stage the regular zoea maxilla appears.
It consists of three parts, of which the outer one is two-jointed.
This bears on its basal joint one spine, and five or six on the
terminal one. These spines appear to be smooth. Those on the
other lobes are bearded with short stiff hairs. There are about
six of these stout spines on the middle lobe, and four on the
inner.

With the change to the megalops the outer branch is bent
proximad and outward and loses most of its hairs. The middle
and inner lobes are greatly elongated, and the number of their
spines is much increased. Those of the middle lobe are the
larger. During the transition from the megalops to the adult the
inner lobe becomes curved toward the middle one, and the joint
in the outer lobe becomes more distinct than in the early mega-
lops stages.

The sudden outward bend of the outer branch of this appen-
dage at the change to the megalops, recalls the inward bend of
the exopodite of the maxillipedes, and suggests a possible
homology for the part. The fact that the appendage is bilobed
at a very early stage also looks in the same direction.

Second MaxiUa.— First zoea, Plate XXXII, Fig. 29. Second
zoea, Plate XXXII, Fig. 30. Last zoea, Plate XXXII, Fig. 31.
First megalops, Plate XXXII, Fig. 32. Young crab, Plate
XXXII, Fig. 33. Adult, Plate XXXII, Fig 34.

The alterations of the second maxilla during the zoea state
Are much more considerable than are those of the first maxilla.

DEVELOPMENT OF PANOPJEUS 8ATI. 421

In the larval skin at hatching there are four lobes over this
appendage, of which the three median correspond to the lobes of
the first maxilla.

In the second zoea stage the appendage has four main divisions.
Each of the three median parts is bilobed at the end, and bears from
six to eight spines, of which but few are obviously plumose. There
is a trace of a joint at the base of the outer of these three lobes.
The outer part — the scaphognathite — is the most interesting.
This plate is much extended in two directions from the point of
attachment. The shorter extension extends distally and outward,
and bears four or five long slender projections, hardly to be called
hairs. The other and longer projection passes downward, curv-
ing toward the median line, bearing very fine hairs on its edges.
It is impossible to avoid noting the resemblance of this plate to
the epipodites of the adult maxillipedes, especially the first. It
forms the entire scaphognathite, and neither at this nor any other
time shows a trace of segmentation. Its subsequent changes are
merely to fit it in shape to the broadening cavity in which it is
to work, and to increase its efficiency by means of hairs on its
edge. It is difficult to believe that this plate is composed of
epipodite and exopodite united, as asserted by some authors.

The zoea life causes changes mainly in the scaphognathite,
which becomes more oval in shape by shortening its projections,
loses its fine hairs, and gains new, long setae, which become more
hair like and more thickly set.

In the first megalops stage the outer of the three median lobes —
the pro bable exopodite — is a good deal changed. It loses its terminal
hairs and becomes fringed with fine hairs on its edges. It no
longer shows the terminal lobes, which at one time even hinted
at two joints, but is a single slender plate. The other parts
are little changed. The scaphognathite is becoming rhomboidal
and its hairs are more numerous.

In the first crab stage these hairs have greatly increased in
number and are plumose, forming a real extension of the plate
so far as work is concerned. The exopodite is also wider at the
base.

These features are accentuated in the adult. The exopodite
is much broader at the base, the two median lobes are deeply
cleft, the scaphognathite is nearly rhomboidal and densely
fringed with plumose hairs.

422 E. A. BIRQE.

First Maxillipede.— Fmt zoea, Plate XXXIII, Fig. 1. Second
zoea, Plate XXXIII, Fig. 2. Last zoea, Plate XXXI, Figs.
1 to 3. First megalops, Plate XXXIII, Fig. 3. Young crab,
Plate XXXIH, Fig. 4. Adult, Plate XXXIII, Fig. 5.

In the first zoea this appendage is closely invested by the
larval skin, and the hairs are all more or less invaginated in the
joints to which they belong. The hairs are extended during the
molt to the second zoea form ; and the exopodite is then fur-
nished with four long, tri- articulate, densely plumose swimming
hairs. The endopodite has the normal five joints, each having
one hair, except the last, which has several. The long and stout
protopodite is covered for its basal half by the carapace. During
the zoea life, few changes take place in this functionally impor •
tant appendage, or its fellow, the second maxillipede. The
swimming hairs increase in number to six, then eight, and finally
twelve. The exopodite in the older zoeas shows marks of a divi-
sion into two joints.

With the change to the megalops, the appendage greatly alters
in form. The epipodite, not seen before, makes its appearance.
The exopodite bends abruptly at its middle joint, and the long
swimming hairs are much reduced in size. The exact fate of
endopodite and protopodite is not clear. They are much reduced
and consolidated, and opportunity was lacking to trace the his-
tory of each part. Probably the part a, Plate XXXIII, Fig. 3, is
formed from the two terminal joints of the endopodite, and two
or three of the median lobes from the rest of the endopodite,
while the protopodite is greatly reduced in size and importance.

The only noteworthy changes in this appendage from the first
megalops to the adult form are in the terminal joint of the exopo-
dite, which segments into numerous joints and gains a correspond-
ingly great number of hairs; and in the epipodite, which develops
an anterior — lower — lobe homologous to that of the scaphogna-
thite.

Second Maxillipede. — Second zoea, Plate XXXIII, Fig. 6.
Last zoea, Plate XXXI, Fig. 6. First megalops, Plate XXXIII,
Fig. 7. Young erab, Plate XXXIII, Fig. 8. Adult, Plate
XXXIII, Fig. 9.

The history of this appendage in the zoea closely resembles
that of the preceding. The main difference is in the endopodite,

DEVELOPMENT OF PANOPJEUS SAYL 423

which is smaller than that of the first maxillipede, consisting of
three joints, of which the terminal one shows in the last stages a
trace of division into two parts.

In passing to the first megalops stage the changes of the exo-
podite are much the same as those of the corresponding part in
the next anterior appendage. The endopodite now has five
joints, the protopodite has greatly diminished in size, and the
epipodite appears.

Third Maxillipede.— -Third zoea, Plate XXXIII, Fig. 14.
Late zoea, Plate XXXIII, Fig. 15. Last zoea, Plate XXXIII,
Fig. 16. First megalops, Plate XXXIII, Fig. 10. Yonng crab,
Plate XXXIII, Fig. 11. Adult, Plate XXXIII, Fig. 13.
Comb -hair, Plate XXXIII, Fig. 12.

The third maxillipede appears before hatching as a simple pro-
jection, which condition it retains until the later zoea stages, when
the exopodite, epipodite and gill appear as unsegmented pro-
jections.

In the first megalops the appendage has a five-jointed endopo-
dite, directed forwards, and the exopodite resembles that of the
corresponding stage in the other maxillipedes. The protopodite
is not anchylosed to the endopodite. The subsequent changes in
the endopodite consist in the enlargement of the two proximal
joints, while the terminal then become relatively smaller and
bend inward. Finally they become a sort of palpus for the
broad plate formed by the basal joints.

In the later megalops stages comb-hairs appear on the terminal
joints and are used in cleaning the other mouth appendages.

Walking Legs.— Third zoea, Plate XXXIII, Fig. 14. Late
zoea, Plate XXXIII, Fig. 15. Last zoea, Plate XXXIII, Fig. 16.
Megalops, Plate XXXI, Fig. 8.

These limbs during the life of the zoea closely follow the for-
tunes of the third maxillipede. like it they first appear as
rounded lobes on the sides of the body. At first two appear in the
second zoea (see Plate XXXI, Fig. 3), and no more are present in
the third zoea. In the later stages all are present. The posterior
two legs grow forward beneath those already present, and the
joints are clearly marked. Gills and epipodites are present.

The legs of the megalops are more slender and joints more
cylindrical than are those of the adult. They are sparsely and

434 E. A. BIRGE.

evenly covered with coarse hairs, and there is no obvious differ-
ence between the right and the left chela.

The segment of the fifth pair of legs is anchylosed to the pre-
ceding one at the change from the megalops to the young crab
of Plate XXXI, Kg. 8.

Abdominal Appendages. — Fourth zoea, Plate XXXIII, Fig. 17.
Fifth zoea, Plate XXXIII, Fig. 18. First megalops— Third
appendage, Plate XXXIII, Fig. 19. First megalops — Last
appendage , Plate XXXIII, Fig. 20. Adult ? , Third appendage,
Plate XXXIII, Fig. 21.

The abdominal legs appear quite early in the larval life. In
the third zoea they may be distinguished as cell-masses below the
skin, and in the fourth (with eight swimming hairs) they appear
as elevations. They then acquire a small endopodite, and are
two-jointed. This condition they retain till the last zoea, when
the hairs are visible, invaginated in the joint. The legs appear
first on the fifth abdominal segment, then on the anterior segments,
last on the sixth.

In the megalops the exopodite becomes a broad flat plate,
which bears from eighteen hairs in the second to six in the last.
These are long, tri-articulate and plumose. The endopodite is
still a small elevation, and is unjointed.

No series of forms connecting this stage with that of the adult
was found. In the adult female theprotopodite is much reduced
in size, the exopodite much elongated, and the endopodite has
six joints.

Characteristics of stages.

First Zoea. — In larval skin.

Second Zoea. — Moulted from larval skin, four swimming hairs.

Third Zoea. — Six swimming hairs. First appearance of ab-
dominal legs under skin. Long spine on fifth abdominal segment.

Fourth Zoea. — Eight or more swimming hairs. External ab-
dominal legs. Spines on anterior abdominal segments.

Last Zoea. — Twelve or more swimming hairs. Divided telson.
Mandibular palpus.

First Megalops. — Immediately after moult from last zoea.

First Crah. — Three spines on each side of carapace. Anchy-
losed segment for fifth walking leg.

DEVELOPMENT OF PANOP^UB SA YZ 426

Measurements of Panopcsus sayi {from single specimens of the
stage indicated) given infractions of an inch :

Past Measbbed.

"Ed

§1

if
Bo,"

Ill-.MAflKS.

Total length

" height

Carapace length

■ breadth

" height

Breadth between eyes. .

Abdomen length

Frontiil spine

Dorsal "

Tel son length.

An ten oa

.043
.01
.01
.01
.01
.019

.018
Al

.048
.068
.018
.018

.01 r

.014

.64

.017
.017
.04

.079
.108
.088
.088
M
.08
.067
.067
.082
.088

.064

.(*>

.018
.081
.081

.071

.081
M

.088
.04

M
JOB

.008

.047

.018

.01

.028
jDII
.01

.081

Across lateral spines.

CUTICULAB APPENDAGES

.081!

Plate XXXI, Fig. 12, shows the second zoea of Pompous
depre&ws (Smith), and Plate XXXII, Fig. 13, its antenna.

The aoea is readily distinguishable from that of the allied species
by the following characteristics :

The spines of the carapace are much shorter proportionally, espe-
cially the frontal spine; the antenna are shorter, more strongly
curved, and armed at the tip with short spines ; and the telson is
much shorter.

Otherwise the zoeaB closely resemble each other, and their develop-
ment is nearly parallel.

The first megalops of P. depres&us was not found. Nor indeed
was there any megalops which could be certainly referred to P. de-
pressus. The megalops of P. sayi was raised from the zoea.

October, 1882.

426 E. A. BIRQE.

Plate XXX, Pigs. 1-2, Nauplius stage. 3, Stage 2. 4-5, Stage
3. 6, Stage 2, from side. 7-8, Just before hatching.

The yolk is shown only in Figs. 5, 7 and 8.

T=telson, Z=labrum, a* = antennule, a" = antenna, md= man-
dible, maf= first maxilla, ww?' = second maxilla, mpr = first max-
illipede, mp" = second maxillipede, a=eye.

Fig. 9. First zoea stage from above.

Plate XXXI, Fig. 1, First zo^a stage. 2, Second zoea stage.
3, Third zoea stage. 4, Last zoea stage. 5, Monlt to megalops stage. 6,
Young crab (carapace). 7, First megalops stage. 8, First megalops
stage. 9, First crab stage (carapace). 10, First crab stage (carapace).

11, Adult crab (carapace). 12, Second zoea, P. depressus.

Plate XXXII, Fig. 1, Eye, second zo8a; 2, Eye, last zoea; 3, Eye,
adult 4, Antennule, first zoea; 5, Antennule, second zoea; 6, An-
tennule, third zoea ; 7, Antennule, late zoea ; 8, Antennule, last zoea ;
9, Antennule, first megalops ; 10, Antennule, adult. 11, Antenna,
first zoea; 12, Antenna, second zoea; 13, Antenna, second zoea, P.
depressus ; 14, Antenna, late zoea ; 15, Antenna, last zoea ; 16, Antenna,
first megalops; 17, Antenna, adult 18, Mandible, second zoea;
19, Mandible, third zoea; 20, Mandible, last zoea; 21, Mandible, first
megalops; 22, Mandible, late megalops; 23, Mandible, adult. 24,
First Maxilla, first zoea ; 25, First Maxilla, second zoea ; 26, First
Maxilla, first megalops ; 27, First Maxilla, first crab ; 28, First Max-
illa, adult 29, Second Maxilla, first zoea; 30, Second Maxilla, second
zoea ; 31, Second Maxilla, last zoea ; 32, Second Maxilla, first mega-
lops ; 33, Second Maxilla, young crab ; 34, Second Maxilla, adult.

Plate XXXIII, Fig. 1, First Maxillipede, first zoea; 2, First Max-
illipede, second zoea; 3, First Maxillipede, first megalops; 4, First
Maxillipede, first crab ; 5, First Maxillipede, adult 6, Second Max-
illipede, second zoea; 7, Second Maxillipede, first megalops; 8,
Second Maxillipede, first crab ; 9, Second Maxillipede, adult 10,
Third Maxillipede, first megalops; 11, Third Maxillipede, first crab ;

12, Third Maxillipede, comb-hair; 13, Third Maxillipede, adult
14, Thoracic Legs, third zoea ; 15, Thoracic Legs, late zoea; 16, Thora-
cic Legs, last zoea. 17, Abdominal Leg, fourth zoSa; 18, Abdominal
Leg, late zoea. 19, Third Abdominal Leg, first megalops. 20, Last
Abdominal Leg, first megalops. 21, Third Abdominal Leg, adult 9 .

STRUCTURE AND GROWTH OP THE SHELL OP
THE OYSTER. By HENKY L. OSBOEN, Late Fellow
in Biology of the Johns Hopkins University. With Plate
XXXIV.

All modern accounts of the formation of the Lamellibranch
shell accord well with the statement of Huxley that "the shell
itself consists of superimposed lamellae of organic matter hard-
ened by a deposit of calcareous salts. It is a cuticular excretion
from the surface of the mantle and never presents any cellular
structure."1

Dr. Wm. B. Carpenter in 1844 published in the "British Asso-
ciation Reports " a full account of the structure of adult shells in
many mollusca. He did not study the development of the shell,
but gave it as his opinion, based upon inference from adult struc-
ture, that the lime prisms are internal casts of prismatic cells,
these cells being layers of cuticle, stripped from time to time
from the surface of the mantle. This view of Dr. Carpenter's is
taught by Siebold in his Anatomy of Invert ebrata,' and Bronn
leaves the matter an open question, but so far as I can learn the
current view is the one quoted above from Huxley.

Since the history of the shell's growth in Lamellibranchs
does not seem to have been directly studied by any one, Dr.
Brooks suggested last summer at the Beaufort laboratory that I
should work upon it ; proposing a modification of the method
long ago in vogue among the Chinese for growing images of thoir
gods inside the shell of the pearl oyster. He supposed that the
study would orily confirm general opinion upon the subject, but
that observations would be valuable.

The method used was this : the edge of the shell was snipped
away with a pair of bone forceps until a gap was produced wide
enough to permit the insertion of a thin circular glass cover be-

1 Anat. Invert p. 406.

* P. 191, edition 1854. Bronn : Classen and Ordnang, UL I Abtheil. p. 846.

428 HENRY L. OSBORN

tween the outside of the mantle and the inside of the shell. This
cover was carefully pushed well back from the gap — it could be
done with no appreciable injury to the mantle surface. From their
abundance, oysters were at first used. Several of them were
taken from the flats, where they grow in enormous numbers, and
were provided with glass slips ; they were then placed inside a
strong, fine wire-net cage and replaced upon the flats. By this
means the natural conditions were very nearly obtained, and the
protection of the oyster from the army of predaceous Crustacea
was secured. Under these conditions the oysters apparently went
on thriving, and I could from time to time open individuals
and learn what had taken place. Studies upon other forms
beside the oyster were attempted, but these were not successful.

Pinna is abundant in the waters where the oyster grows, and I
attempted to study it in the same manner as the oyster, but
without success, since the presence of the cover seemed to irritate
the animal. It is quite free from the shell except at the attach-
ment of the adductor muscle, and always succeeds in scrubbing
away the glass cover. Other forms were also tried, Siliqua and
Venus j but the attempts were not successful, apparently from the
impossibility of closely imitating their conditions of life.

Examination of the glass slips left twenty-four hours inside
the oyster, showed a thin gummy deposit. It formed a faintly
yellowish brown film, which had hardly consistence enough to
hold together. After treatment with staining reagents, haema-
toxylin, picrocarmine and eosine, the film would show a faint
color, but this was diffused evenly in every part and absolutely
no structural characteristics could be observed. In some in-
stances lime crystals were already formed, though sparingly.
From the character of this young film it is perfectly apparent
that it is a viscid excretion poured out from cells upon the sur-
face of the mantle. If one make vertical sections of the mantle
properly hardened, it will be seen that the surface is formed of
columnar cells. These stand closely packed and are stained
intensely. They are glandular and very full of granules ; it is they
that pour out this very viscid and very abundant secretion. Sur-
face views, also, of mantle stained with silver nitrate show a close
pavement all over the mantle formed by the outer ends of these
secreting cells.

SHELL OF THE OYSTER. 429

The hardly consistent film of twenty-four hours has by forty-
eight hours become a tough, leathery membrane. Its color is
brown. It already forms a definite envelope about the animal,
and has shut in the glass cover between itself and the previously
formed shell. It resists all attempt to demonstrate any structure
in itself by means of the ordinary histological reagents, and is a
structureless cuticular or horny envelope, the organic basis of
the shell ; it is this which is evident as the epidermis in many
shells, and which, as may be shown by treatment with dilute acid,
forms the skeleton of all shells. In later growths of the shell
this membrane or film waxes, fresh supplies of the gummy excre-
tion being spread over its inner surface continually, so that this
surface is never so hard and brittle as the outside may become.

In solution in the gummy excretion there is held calcium car-
bonate, and this, as the film hardens, crystallizes, and gives rise to
the various stony structures to which many shells owe much of
their beauty. These crystals take on various forms. In one
preparation (Fig. 2) they are flat scales with not very sharply cut
edges. They are obscurely hexagonal, have an average diameter
of t^Vc inch, and fill the membrane as thickly as indicated by
the figure. If a film of forty-eight hours be placed in dilute acid
(acetic was used in this instance), the lime is completely dissolved
away and the spaces occupied by the crystals are plainly seen.
Such a film is represented in Fig. 3, it is a beautifully tessellated
pavement after treatment with the acid, and shows the more or
less hexagonal spaces occupied by the crystals. It seems scarcely
doubtful that these spaces were formed by lime crystals. Their
resemblance to the cells in decalcified Pinna shell is so extremely
close that two drawings would look identically the same except
in respect to the size of the spaces.

Besides the scaly crystals these regularly formed films of forty-
eight hours show many crystals which assume forms represeu ted in
Figs. 4 and 5. Some are acicular, tapering away from an oval centre,
and these are often united into a large nodule, many having
formed about some common nucleus. In the figures, which are
accurate camera lucida drawings, these crystals are seen to have
not as yet formed a continuous layer, and the membrane, being
perfectly structureless and almost transparent, cannot be shown.
These acicular crystals are generally about TBVrr of an inch in

430 HENRY L. OSBORN.

length. They are, however, much less numerous than a second
form (Fig. 5), which is perhaps built upon them. These are ob-
long crystals somewhat swollen at either end and slashed into
many fine points, suggesting striated epithelium cells in the ani-
mal body. These are often compounded into twin and higher sys-
tems, and are frequently seen forming large spiny-looking masses.
They occur in other parts of the same film in which the acicular
crystals may be found, and seem to be the most common condi-
tion of the film after forty-eight hours' growth.

Another film, six days old, has almost completely lost its
leathery character and become stony, from the great amount of
lime present in it. The most of this layer is a thick pavement
of flat cells so closely packed that they are perfectly continuous
over an area of a square inch or more, with here and there small
breaks where the shelly formation has not gone on as regularly.
In these places one sees such crystals as are shown in Fig. 5, but
they are not numerous, also crystals of the sort figured in Fig. 6.
These seem to have a core, which is striated lengthwise, or, as
they finally broaden out at the tip, radially, surrounded by an
outer shell in which the same radially striated appearance is
very strongly marked. These are not common and I hardly
think they can be normal. The size of these nodules is as fol-
lows, viz. in one marked a, greatest diameter of central core
tfo inch, diameter of the peripheral part a little less than 24}00
inch. These nodules are thus very much larger than the average
scales of the 48 hours' film whose diameter may be placed at
y^^ inch. They are, however, only about twice the size of the
average scales which make up the bulk of the film at this time.
Dr. Brooks informs me that he has found nodules almost exactly
like these in the shells of Mya. It may be noticed that the
peripheral columnar layer bears a very close resemblance to the
prismatic layer as figured by Pagenstecher1 in his study of the for-
mation of pearls.

I have no studies of the oyster shell later than films of one week
old until we reach films of three or four weeks. By this time
the glass cover is completely shut into the stony shell, and can
no longer be seen, and its place is only to be traced by its form,

1 Zeitechr. f. wifls. ZoOl. IX, p. 496, plate XX, 1858.

SHELL OF THE OYSTER. 431

preserved perfectly upon the inner surface of the shell. By
breaking out this cover very carefully it is seen to be coated
with a thick plate of white shell, which is beautifully smooth
upon the side nearest the cover slip. Examination shows this
plate to be made up of many lime scales not arranged in any
definite system, but with the many layers laid on quite at
random. It is of such crystals as these that the bulk of the
oyster shell is formed. The inner layer of the 6hell, or as it is
called in Bronn's account, the mother of-pearl layer, forms most
of the stony shell, the prismatic layer is almost entirely absent.
It is to be regretted, so far as concerns the present purpose, that
this is the case, for the oyster is such a quiet animal that the
prismatic layer could be readily studied in it were this layer
developed in any such beautiful manner as it is in Pinna, while
Pinna, so far as I was able to experiment, did not make a
favorable subject.

Upon edges of the oyster shell elongated cells may be seen
placed very obliquely; these may represent the prisms of the
shells where a prismatic layer is strongly developed. These
cells, however, shade off directly into cells which form a close
pavement like those of Fig. 3, and seem to be undoubtedly
formed in a manner similar to the ordinary polygonal cells of my
forty-eight hours' films.

There can be no doubt, I think, on these observations that the
shell is formed by the crystallization of the lime in the chitinous
sheet as has been generally supposed, and that the older view,
that the forms assumed by the lime show that it has been laid
down as internal casts, is not at all sustained by the facts in the
history of the shell's growth.

It is worth while to mention here a few observations upon
young growing oysters as illustrating the wonderful rapidity with
which the shell increases. Since the wire cage, in which the
oysters were confined and protected, was placed among the
growing oysters upon their native flats, it will be seen that not
only a favorable place was afforded for the embryonic oyster to
attach himself and grow unmolested, but enormous numbers of
spawn would be likely to be at hand, and the inside of the cage
to be well supplied with them. And such was the case. In a
month the box, the stones put into it for ballast, and the oysters

432 HENRY L. OSBORM

themselves, were literally paved with young oysters about the
size of an old-time three-cent piece. In two months these had
grown so that only about one fourth of the original number now
survived, the others having been literally " shoved out," and the
survivors now had shells averaging from three-fourths of an inch
to an inch in length, strong and solid, and weighing often as
much as three or four grammes.

EXPLANATION OP PLATE XXXIV.

Fig. 1. Glandular epithelium from the outer surface of the
mantle, xi 3D.

Fig. 2. Lime scales in film of 48 hours, xi 3E.

Fig. 3. Decalcified film of 48 hours, xi 4D.

Fig. 4. Acicular prisms from film of 48 hours, x} 3D.

Fig. 5. Prisms in film of 48 hours, xi 3E.

Fig. 6. Peculiar crystals in film of 6 days, xi 3E.

The figures were all drawn of the size they appeared with the
Zeiss oculars and objectives indicated and reduced one-half in the
process of their reproduction.

THE NERVOUS SYSTEM OP PORPITA. By H. W.

CONN, and H. G. BEYEB, M. D., U. S. N. With Plate XXXV.

The discovery of a nervous system among the Ocslenterata has
been one of the important results of modern histology. Start-
ing with Kleinenberg's neuro-muscle cells,1 which later obser-
vations have shown to have been wrongly interpreted, many
observations upon the subject by excellent histologists have been*
made, and to-day it is known that a very primitive, and there-
fore very interesting nervous system exists in many of the Coelen-
terates. The brothers Hertwig found and described such a sys-
tem in Medusae.2 From their observations they drew some
interesting theoretical conclusions as to the origin of the nervous
and muscular systems. Later8 the Actinia were studied by the
same histologists with similar results. The Ctenophorae have
been found by Chun4 and again by the Hertwlgs* to possess the
same nervous system, with a central nerve ring and peripheral
scattered ganglion cells. More recently the Hydroids have been
the object of special investigation in this regard. Jickeli6 found
in Endendrium and Hydra certain cells, which he considers as
nerve cells, scattered quite widely over the animal. Lendenfeld7
independently discovered the same cells, and extended his obser-
vations to include Campanularia. He also discovered in Cam-
panularia what he considers as a central nervous system, in the
form of an endodermal nerve ring around the proboscis inside the
oral opening.

Our knowledge of the nervous system of Siphonophores is
nearly all contained in a short article by Chun8 upon Yellela.

1 Kleinenberg. Hydra. Leipzig.

9 O. and R. Hertwig. Medusen. Leipzig.

8 Hertwig. Actinia. Jenaisches Zeit vol. 18.

4 Chun, Monograph on Ctenophorae of the Gulf of Naples.

5 Hertwig. Ctenophorae. Jenaisches Zeit vol. 14.

6 Jickeli. Morph. Jahrb. vol. VTIL
1 Lendenfeld. Zcol. Anz, No. 181.

8 Chun. Nervensystem dee Siphonophores, Zool. Anz. No. 77.

434 S. W. CONN AND K 6. BEYER.

This paper describes a system of ganglion cells in the ectoderm
of Vellela, scattered quite abundantly over nearly all parts of the
animal. No central system or nerve ring each as appears in
most Coelenterates was seen. This observation, as far as I am
aware, stands alone, bat as Chnn is a very careful workman
there is no donbt as to its truth. Some work which has been
done in the Biological laboratory dnring the present year, upon
Porpita, shows that here also is fonnd a similar system of nerve
ganglion cells. The observations were made without a previons
knowledge of Chan's paper, and are therefore more valuable as
confirming his statement as to the existence of a nervous system
among Siphonophora, as well as in extending oar knowledge of
the relation and distribution of the same.

Oar specimens of Porpita were collected at Beaufort, N. C, and
were preserved by cemic acid. The animals were placed alive
in a very weak solution of osmic acid and allowed to stain for a
few minutes. Then after washing they were hardened in alco-
hol, at first in a weak solution, 50 per cent, then in 70 per cent.,
95 per cent, and absolute alcohol. This preserved the tisanes
in beautiful condition for histological work, staining the cell
nuclei and the nerve cells slightly. It was hardly necessary to
use any farther staining reagents, although to bring out the
nuclei of the nerve ceils it is best to stain the specimen with
hematoxylin.

To make the arrangement and distribution of the nervous sys-
tem intelligible, a few words upon the rough anatomy and his-
tology of Forpita will be necessary. Forpita is a small button-
shaped siphonophore, with a diameter varying from half an inch
to an inch and a half, and with a thickness, in large specimens,
somewhat over a quarter of an inch. Their color is a beautiful
greenish blue, and when floating on the water with their long
tentacles spread out, they are as handsome a specimen as one
wishes to find. At sea they are usually seen floating on the sur-
face of the water in large schools, appearing as a greenish band,

cuparativcly narrow bat very long, extending in a straight line

■ miles. They possess some power of locomotion, but this
r is slight, and they float largely at the mercy of the winds

i waves,
ha upper surface of Forpita is a plain, nearly flat, circle,

THE XERVOUS SYSTEM OF POBPITA. 435

which is perforated by numerous openings lending into a series of
air chambers lying directly beneath. The under surface is more
curved in outline, and is covered by large numbers of zooids,
nutritive, generative and tentacular. The general anatomy can
be seen from Fig. 1? Plate XXXY, which is a perpendicular radial
section through one-half the animal, t. e. from the centre to the
edge of the disk. The upper half of the disk can be seen to be
occupied by a series of air chambers AC, arranged in concentric
circles around the centre, each circle being separated from the
others by circular partitions of chitin, and being further divided
by radial partitions into many smaller chambers. Each cham-
ber communicates with the exterior in two ways. First, by an
opening through the upper surface, Fig. 1 0, leading directly to
the exterior, and second, by means of a large number of tubular
filaments, the pneumatic filaments, Fig. 1 pf. These pneu-
matic filaments arise from the lower side of the air chambers,
and can be traced from these through the lower half of the disk,
pursuing a more or less complex course. They finally make
their appearance on the under side of the animal, and can be
seen as long tubular threads, which in great abundance are
wound around the nutritive zooids, Fig. 1 pf.

The number of these concentric rings of air chambers varies
very much, but they never reach the edge of the disk. Outside
the outermost air chamber the disk is prolonged into a thin
flexible velum, fig. 1 F. This velum is filled by a gelatinous
tissue, and is traversed by numerous branching canals. It is
very abundantly supplied with circular ectodermal muscles, thus
forming a movable membrane extending around the animal and
giving it some power of motion.

Upon the lower surface of the animal are found the various
forms of zooids. These consist of three kinds. (1) One very
large central zooid, Fig. 1 CZ, the primary nutritive organ. (2)
A very great number of smaller nutritive zooids, Fig. 1 iT2T,
varying much in size from minute buds to large organs, nearly
the size of the central zooid. They fill the space from the cen-
tral zooid to the base of the tentacles, occupying thus a large
part of the under surface of the animal. Most of the feeding of the
Porpita is done by these zooids, and they serve also as the origin of
the generative organs, the medusae appearing as buds around their

436 H. W. CONN AND H. O. BEYER.

bases, Fig. 1 GZ. (3) External to the feeding zooids are three
or four rows of tentacles, Fig. 1 T. Most of these tentacles are
very long, even surpassing in length the diameter of the disk ;
and when the animal is floating on the water they are stretched
out as a deep fringe around it. The outer rows are younger
and much shorter, not even reaching the edge of the velum.
They are all movable and highly sensitive, and are armed with
quantities of thread cells, many of which are collected in numer-
ous knob-like batteries, Fig 1 B.

The most external layer of cells.over the whole of the animal
is an ectodermal epithelial layer. The cells of this layer vary
considerably in different regions. Upon the upper surface of the
disk they are high columnar cells, Figs. 1 and 5 E, many of
which, especially near the edge of the velum, are epithelio-mus-
cular cells, Fig. 7. Upon the under side of the velum the cells
are smaller and by no means as high, Fig. 5. The nutritive
zooids are covered with a still smaller layer of cells, and upon
the central zooid they become quite flat. The tentacles finally
reach the extreme, and are covered by a layer of large but thin
scale-like cells, Fig. 1 T and Fig. 3. Immediately beneath the
epithelial cells is found a layer of ectodermal muscle fibres. In
the tentacles and the nutritive zooids the ectodermal muscles are
longitudinal. The ectodermal muscles found in the velum, how-
ever, are circular muscles. This system of muscles is much more
highly developed than the endodermal muscles which are found in
the nutritive zooids, and to it seems to be due most of the move-
ments of the animal. In all parts of the body there is devel-
oped just beneath the ectodermal muscle layer a supporting
membrane, Figs. 4 and 5 SI. The thickness of this supporting
layer varies much, it being thin in the tentacles, but very thick
in the central zooid and in the upper part of the disk. Succeed*
ing the supporting layer, as we go toward the interior, are found
in some regions a system of endodermal muscles. Neither the
tentacles nor the velum, where the ectodermal muscles are so
powerful, possess endodermal muscles; but the nutritive zooids,
and particularly the central zooid, have an abundant supply.
They form in all cases a circular system. The innermost layer
of cells is the endoderm, which presents many varieties, according
to the region of the body where it is found and the function it

THE NERVOUS SYSTEM OF PORPITA. 437

performs. A very peculiar endoderm cell is found in consider-
able numbers in the tentacles, of which Fig. 8 is a representation.
Each has a quite large body, very clear and perfectly transparent.
Toward the interior of the tentacle the cell is prolonged as a
highly granular columnar process, ending in a knob in which is
contained a large nucleus. Toward the exterior the transparent
body is continued as a long seemingly tubular process which
reaches to the supporting membrane. These cells are found
among other endoderm cells of ordinary form which nearly fill
the interior of the tentacle. Elsewhere in the animal the endo-
derm has cells usually characteristic of this layer.

There are in Porpita two distinct structures which are prob-
ably nervous in junction. The first consists of scattered ganglion
cells widely distributed and quite abundant. The second is a
large number of organs around the edge of the velum, which
seem to be sensory organs of some kind.

Nerve Ganglion Cells.

If a bit of the tentacle of an osmic acid specimen of Porpita
be teased out in glycerine, in such a manner as to flatten the
ectoderm without pulling it to pieces, quite a number of different
ectodermal structures will be seen. Most prominent will be the
longitudinal muscle fibres, which section shows are entirely out-
side the supporting membrane, and therefore ectodermal. Lying
among the muscle fibres and sometimes seen to be connected
with them are numerous thread cells. The outlines of the ecto-
dermal cells are also plainly seen, showing them to be large flat cells,
each of which contains a prominent nucleus. Careful observa-
tion will show another structure much less prominent than those
mentioned, faintly stained with osmic acid or picro-carmine or
more deeply with haematoxylin. These cells are, as far as can
be judged from their histology, true ganglion cells. Fig. 3 is a
camera drawing of such a preparation. Muscle fibres and thread
cells are omitted, to avoid confusion.

The body of these ganglion cells is very small, smaller indeed
than the nuclei of the ectodermal cells. They are only about
Tfjfos of an inch in diameter in ordinary specimens, though some-
times somewhat larger. Once seen, however, they can be readily

438 H. W. CONN AND H. G. BEYER.

found in large numbers. Each cell consists of a small cell body
with several long processes, Figs. 2 and 3. In a majority of
cases the cell bodies are triangular, Fig. 2 a, with a long fibre
given off from each angle. Bipolar cells are also frequently seen,
though they are much less frequent than the tripolar cells; in
these cells the body approaches an oval form, Fig. 2 5. In still
other cases cells with four processes are seen, Fig. 2 c. Occa-
sionally multipolar cells with more than four processes are found,
though they are extremely rare. The tripolar cells with a trian-
gular body are much the most common.

The body of the cell at first sight seems to be completely homo-
geneous, and it is with difficulty that a nucleus can be distin-
guished. Careful examination of favorable specimens, however,
particularly those stained with haematoxylin, shows what is
represented in Fig. 2. There is present in each cell a large but
faint nucleus, nearly filling the body of the cell, and within this
a small bright point, the nucleolus. The cell is very slightly
granular and usually appears as a clear, almost hyaline mass, in
which can be seen the nucleus as a somewhat dark area, and the
nucleolus as a small bright spot, Fig. 2.

The fibres which arise from these cells are, as above stated,
usually three in number, though there may be two or four, or
occasionally more, given off from each cell. Very thin delicate
fibres they are, pursuing a tolerably straight course closely
applied to the muscular layer. They all divide more or less into
finer branches, and thus the processes from each cell cover quite
a considerable area. They are remarkable for their extreme
length, and can in favorable preparations be traced as delicate
branching fibres for a long distance before they finally disappear
in the muscular layer. How much farther they may be con-
tinued within this layer it is of course impossible to say. Fre-
quently the fibres from one cell unite with those of other cells,
as in Fig. 3, thus putting the different nerve ganglia into com-
munication with each other, and forming to a certain extent a
continuous nerve plexus. Many of the fibres, however, do not
present any such connection with other fibres, but after branch-
ing in a complex manner, finally appear to enter the muscular
layer lying beneath them and thus disappear from view. They
do not seem to have any connection with the thread cells, which

THE NERVOUS SYSTEM OF PORPITA. 439

are found abundantly scattered in the ectoderm, although they
are found in the ectoderm of the stalks which bear the thread
cell batteries, Fig. 1 B. They have, indeed, connection with no
structures except the muscles.

These cells are entirely ectodermal structures, as is abundantly
proved by section. A cross section of the tentacle, Fig. 4, will
indicate this relation. The ganglion cells G are seen to lie
within the ectodermal cells. Beneath them are the ectodermal
muscles M, and still further toward the interior is seen the sup-
porting layer, Fig. 4 Sl9 which separates ectoderm from endo-
derm. This system of ganglion cells therefore lies in the outer-
most layer of the ectoderm, even exterior to the ectodermal
muscles. The same can be seen in sections from other parts of
the animal. Fig. 5 is a section through the edge of the velum,
and shows the ganglion cells G lying among the ectodermal
epithelial cells and outside the ectodermal muscles. All the
ganglion cells are ectodermal, therefore, and the endoderm does
not seem to possess nervous elements, as is the case in some
hydroids1 and in ctenophorae.'

Though the cells are most readily seen in the ectoderm of the
tentacles, owing to the thinness of the ectodermal cells in this
region, after they are once recognized they can be found in
various other parts of the animal, and are indeed quite widely
distributed. They are always found in connection with the
muscular system, and are most abundant when this system is
most highly developed. All of the tentacles are well supplied
with them. The velum, which is highly muscular, is particu-
larly rich in its nerve supply, both upon its upper and its under
surfaces. The nerve cells are found here more abundantly than
elsewhere. Upon the upper surface of the animal, as we
approach the centre, the muscular system becomes less and less
noticeable, and parallel with its decreasing importance the nerve
cells become less abundant. They can be found, however,
scattered here and there over the entire dorsal surface of the
disk. Upon the nutritive zooids we have been unable to find a
single ganglion cell, although we have searched patiently for
them. Neither can they be found in the ectoderm of the central

1 Lendenfeld, Zool. Anz. No. 131.

* Hertwig, Ctenophora, Jenaisches Zeit. Vol. XTV.

440 H. W. CONN AND H. O. BEYER.

zooid, although here, owing to the thinness of the ectoderm cells,
they would be easily seen if present. Chun, in his paper on
Yellela, states that the nerve cells are to be found here as well as
elsewhere. This is certainly not the case in Porpita, for in no
case, either by teasing or by section, have we been able to
discover a single nerve cell in any of the nutritive zooids.

The distribution of the ganglion cells then is as follows : They
lie wholly in the ectoderm, and their fibres, after running for a
considerable distance beneath the outer ectoderm cells and imme-
diately upon the muscle layer, finally penetrate this layer and
are lost. The whole of the upper surface of the animal is sup-
plied with them, somewhat sparsely toward the centre, but much
more abundantly toward the edge and especially in the velum.
The under surface of the velum has also a rich supply, and the
tentacles which come next in order contain large numbers. Be-
yond the base of the inner row of tentacles, toward the centre of
the lower surface, they are no longer to be seen either in the
secondary nor the central zooid.

The numbers of ganglion cells in these different regions differ
very much, but everywhere, even where they are the most abun-.
dant, their relatively small number is quite surprising if they are
to be considered as forming a nervous system. In the tentacles,
where the ectodermal cells are large, there is found on an average
about one nerve cell to a dozen ectodermal cells. In the upper
surface of the velum they are somewhat more abundant, but
owing to the fact that the ectodermal cells are smaller their rela-
tive number is much less ; while in the centre of the upper sur-
face not more than a single nerve cell is found to 200 or 300
ectodermal cells. They are not distributed with any regularity.
Quite a number may be found lying very near together, Fig. 3,
in adjacent or even in the same cell, and then there will be seen
a large tract which does not seem to be at all supplied with them.
Nothing like a central system can be made out. No union of
the cells into a nerve ring, such as has been made out by the
Hertwigs1 in Medusae and by Lendenfeld3 in Eudenduum, eeems
to exist.

There is still perhaps some doubt as to whether the structures
here described are really what they have been considered;

1Hertwig. Loceit. * Lendenfeld. Loc.cit.

THE NERVOUS SYSTEM OF PORPITA. 441

whether they may not be some form of connective tissue cor- '
puscle without any nervous function. They are, as we have seen,
very few in numbers as compared with any organs which they
are supposed to enervate; they are connected with no central
system, and simply form a more or less connected plexus of scat-
tered cells. If they are true nerve elements they are only to be
considered as what may be the beginning of a nervous system.
It can hardly be possible that they play any important function
as nervous organs. Always associated as they are with the mus-
cular system, they are to be regarded as muscular rather than
sensory cells ; but the relatively small number of even their fibres,
as compared with the number of muscular fibres which each
must be supposed to control, certainly indicates that the mus-
cular system cannot be to any great extent dependent upon
them for its stimulation. The cells here described and those
described by Chun in Vellela are, however, undoubtedly similar
structures to those found by various observers in other OoBlen-
terates, and in many cases, as in Medusae and Actinia, they are
connected with a central nervous system. In these cases there
can be little doubt as to their nervous functions. The fact of
the great resemblance of the cells here found to those of the
peripheral nervous system of other coelenterates, shows therefore
that we are probably correct in viewing them as nervous struc-
tures, and as forming a very primitive nervous system, but one in
which the nervous function is probably very slightly manifested.

Sensory Organs.

Under this head are included a group of organs, hitherto
undescribed, whose nature is somewhat problematical, but which
from their structure seem to be organs of sense of some kind.

If the velum of Porpita be examined from the upper surface
with a lens, it will be seen that its edge is not a plain circle, but
is marked by serration, and looks somewhat like the rim of a
wheel studded irregularly with small cogs. A close examination
shows that this is due to the presence of a series of organs, many
hundreds in number, which, side by side, are arranged around
the edge of the velum. Each organ is a small ectodermal
pocket, and is separated from its neighbor by a small space,
equal in width perhaps to that of the pockets themselves. They

442 H. W. CONN AND H. 6. BEYER.

thus form a sensory ring extending around the edge of the disk
and composed of hundreds of entirely separate organs.

The minute structure of these organs can only be made ont
from sections and teased specimens. They are best seen in
radial sections through the edge of the velum. Such a section
is shown in Fig. 1 S> and much more highly magnified in Fig. 5.
Such sections show at a glance the nature of the organs. They
are nothing more than little invaginations of the ectoderm, form-
ing a little pocket filled with peculiar cells. The supporting
membrane, separating the ectoderm from the cndoderm, can be
traced along the velum to its edge Sly and there bending in-
creased to form the inner lining of each pocket ST. Beneath
the supporting membrane, in the interior of the velum, is seen
a gelatinous tissue perforated by numerous endodermal canals,
Fig. 5 C. Outside this membrane, upon the upper and under
surface of the velum, lie the ordinary ectodermal epithelial
cells, and outside the same membrane, but within the pocket
formed by its invagination, lie a large number of cells* still
ectodermal cells but highly modified.

The cells which fill these pockets are large and highly special-
ized, but they are nevertheless only modified ectodermal cells.
This is readily proved by examination of many sections which
show a complete gradation from the ordinary ectodermal
epithelial cells to the large peculiar cells in the interior of the
pocket, Fig. 5. Toward the edge of the pocket the ectodermal
cells of the velum are seen to elongate, and thus, even at the
deepest part of the organ, while the base of the cell is applied
closely to the supporting membrane, its free end is still upon a
level with the rest of the ectodermal cells*. The ectoderm cells
can thus be traced from the short columnar cells* by almost
insensible changes, to the peculiarly modified sense cells in the
interior of the sensory organs.

Each pocket of this row is thus seen to be filled with a large
number of long, quite large cells, with a broad base applied to
the supporting membrane, and with their narrower free ends
lying exposed to the exterior. Two distinct types of these eeUa
can be distinguished, although they usually graduate into each
other without an abrupt break. There are first in the middle
and deepest part of each pocket a number of large cells,

THE NERVOUS SYSTEM OF PORPITA. 443

highly granular, Fig. 5 and Fig. 6 a. Each of these cells is
somewhat conical in shape, with its apex, in most cases but not
in all, reaching the surface of the velum and thus exposed to
the water. At its base the cell shows a broad band more highly
granular than the rest, Fig. 6 a, in which is seen a very large and
very distinct spherical nucleus containing a prominent nucleolus,
Figs. 5 and 6. These cells fill the middle of each pocket. The
second type of cell is found around the edge of the organ, some-
times passing insensibly into the cells of the first type and some-
times ending more abruptly. They differ from the first type in
being much more slender, and in not being granular, but com-
posed of a clear hyaline substance which appears perfectly homo-
geneous. Each cell shows one or two swellings within which is
an oval mass of more dense material, which stains more deeply
than the rest of the cell. It is the nucleus, but it is seldom
definitely outlined, and in no case is it as prominent and distinct
a structure as is the nucleus of the central cells. No nucleolus
is discernible. These cells are much more abundant than those
of the first type, occurring in thick masses around the sides of
each pocket and enclosing the central cells in the middle. In
their natural position they remind one somewhat of the layers of
rods and cones in the retina of the eye. At the extreme edges
of the organs they of course become shorter and finally pass
into the ordinary ectoderm cells.

The functions of these organs it is impossible to tell with cer-
tainty without observations on living specimens, and as we have
only had alcoholic specimens to work upon, we cannot say con-
clusively what they are. From their histological appearance,
however, they would seem to be organs of touch. The presence
of such long delicate cells with free ends exposed to the sur-
rounding water would certainly point to such a function ; and
their position at the extreme edge of the velum would favor the
same view. They have no connection with the nerve ganglia
above described ; not a single nerve cell is to be found in them
or in any way connected with them. But this is not surprising,
for we have seen that the ganglion cells are associated with the
muscular system alone, and their absence in these bodies is to be
expected. Until further evidence can be obtained they may be
considered as organs of sense and probably organs of touch.

444 H. W. CONN AND H. O. BEYER.

These same organs degenerate with great readiness. In speci-
mens kept in aquaria for a few days, the whole of the central
cells, except the densely granular area at their base, fused into a
homogeneous mass, giving them the appearance of secreting
organs. In well-preserved specimens, however, the cells are dis-
tinct and have the above-described shape.

EXPLANATION OF PLATE XXXV.

Figure. 1. A diagrammatic perpendicular radial section of Porpita
from the centre of 'the animal to its circumference.

2?. Batteries of thread cells.

0. Opening of air chambers through the upper surface.

n Velum.
AC Air chambers.
CZ. Central nutritive aooid.
GZ. Generative sooids.
XZ. Secondary nutritive looids.
pf. Pneumatic filaments.

Figure 2. Specimens of nerve cells.
a. Tripolar cell,
ft. Bipolar cell.
c Qnadripolar cell

Figure 3. Teased preparation from tentacle, showing ectodermal
cells and ganglion cells.

(?. Ganglion cells.

F. Nerve processes from the cells.

Figure -L Cross action of tentacle.
(?. Nerve fibre.
M. Muscle fibres in section.
SL Supporting membrane.

Figure 5. Cross section through edge of velum showing sensory
bodv.

C. Endodermal canals of velum.
E. Ectodermal epithelial cells.

G. Ganglion cells.

SI, Supporting membrane.
ST. Supporting membrane lining the sensory organs.

THE XERVOUS SYSTEM OF PORPITA. 445

Figure 6. Sense cells from sensory organs,
a. One of the larger central cells.
£. Smaller prriphrral cells of the sense organs.

FiorBE ?. Epithelio-niaseniar oeiU from the upper surface of the
velum.

Fiorsz S. Peculiar endodermal cell found in the tentacles.
Figures drawn bv H. W. Conn.

ON THE PRESENCE OF CILIATED EPITHEL-
IUM IN THE HUMAN KIDNEY. By ALBERT H.
TUTTLE, Professor of Zoology in the Ohio State University ;
Fellow by Courtesy of the Johns Hopkins University. With
Plate XXXVI.

The presence of vibratile cilia in the renal organs of the cold-
blooded vertebrates was fully established many years ago : the
extent of the observations made in that direction toward the close of
the last half century is, however, not generally recognized. Those
of Bowman (Philosophical Transactions, 1842) upon the kidney
of the frog are most commonly referred to, and are frequently so
cited as to leave the impression that only the neck of the capsule
was known to be ciliated ; that author, in the paper referred to,
interested as he was in a far different question, that of the true
relation of the Malpighian corpuscle to the uriniferous tubule,
making mention only of the cilia observed in the neck of the cap-
sule and in that portion of the capsule itself which immediately
adjoins the opening into the tubule. The publication of this
important paper, which, as is well known, contained the first true
solution of the question with which it directly dealt, called the
attention of observers to the organs in question, which were, in
accordance with the usage of the day, very generally examined
in the fresh condition : the fact last mentioned gives the reason
why structures which have to a great extent escaped observation
in the hardened and stained preparations more common at the
present day were seen with the far less efficient instruments of
the earlier observers. Bowman speaks of the cilia as seen in
action, producing a current away from the capsule, beyond the
neck of which he did not follow them. Kolliker, however ( Mai-
lers ArchiVj 1845), describes cilia in action throughout the entire
extent of the tubules in the kidney of an embryo lizard ; and in
a note to Kolliker's paper Muller states that he has observed the
same phenomenon in the tubules of the kidney of a. skate. Ke-
mak {jFrariepJ8 Neue Notizen^ 1845) records the observation of

448 ALBERT H. TUTTLE.

cilia in action throughout the extent of the tubules in the kid-
neye of lizards and newts. G. Johnson, the author of the article
on the kidney in Todd's Cyclopedia of Anatomy (Vol. IV, 1848),
speaks of ciliary action as observed by him in all portions of the
tubule in the kidneys of two genera of newts, Triton and Lisso-
triton, in considerable portions of the tubules of the kidney of
the frog, and through a large extent of the tubules in the kidney
of a snake : he also predicts their eventual discovery in the kid-
neys of all vertebrates. In 1854 Kdlliker, in his Microscopische
Anatomiey mentions- the ciliation of the tubules in reptiles, am-
phibians, and fishes as a well-established fact, referring to the
observations cited above and others. This conclusion, while fully
recognized by those who have carefully examined the matter,
seems to have dropped out of the general literature of the histology
of the kidney ; the observation of Bowman upon the neck of the
capsule being, as I have already said, the only one generally cited.

As regards warm-blooded vertebrates our present knowledge is
far less extensive. Most of the papers above alluded to speak of
the impossibility of recognizing the cilia in the kidneys of the
animals under consideration after their characteristic action had
ceased : this doubtless takes place as an almost immediate con-
sequence of the change of temperature caused by the removal of
a portion of the kidney of a bird or mammal to the stage of the
microscope; and the best microscopes of that day, and indeed of a
much later period, were wholly inadequate to the detection and
resolution of such delicate and thickly set cilia as are really pres-
ent, when in a state of rest. Gerlach, however, as quoted by
Kolliker (Micr. Anat\ saw what he believed to be ciliary action
in the kidney of the common fowl, and Hassall (Microscopic Ana-
tomy, London, 1852) described it as witnessed by him in the kid-
neys of the sheep, the horse, and the rabbit.

The first person to recognize the presence of ciliated epithelium
in the hardened and stained mammalian kidney was Elein, who
published in the Quarterly Journal of Microscopic Science for
April, 1881, a notice of their detection in the kidney of the mouse.
He found them in the neck of the capsule, but makes no men-
tion of having seen them in any other portion of the tubule. The
object of the present communication is to call attention not only
to their presence in the human kidney, but also to their extensive

CILIA IN TEE HUMAN KIDNEY. 449

distribution ; and to record similar observations made upon the
kidney of the cat.

The human kidneys which I have examined in this connection
were obtained from a series of autopsies made during the month
of February last at the small-pox hospital by Dr. W. T. Council-
man (who was then lecturing on pathological histology in this
laboratory), under very favorable circumstances as regards their
perfectly fresh condition : they were carefully hardened in alcohol,
being intended originally for the demonstration of micrococci.
Their exceptionally fine state of preservation led me to study
them carefully with high powers, with the result (among others)
of the detection of the cilia in question in all that were not ex-
tensively diseased, viz. in sixteen out of nineteen kidneys ex-
amined. '

The sections made use of were from .01 to .03 mm. in thick-
ness, were chiefly stained with Bismarck brown and mounted
in glycerin, though some were examined unstained or stained
with other reagents, and some were mounted in balsam. It was
while studying the structure of the nuclei with a Zeiss one-twelfth
oil-immersion objective that I came, to my surprise, upon fine,
closely set cilia projecting freely into the lumen of the tubule,
which is considerably enlarged in the small-pox kidney. Al-
though they were seen in numerous places in the section under
examination, my first impression was that each place under con-
sideration must be in close structural proximity to the classical
neck of the capsule of its respective tubule, until after several
days9 examination of the same section, when I came upon the
region represented in Fig. 1, Plate X&XVI, which is plainly the
place where the lower part of the convoluted tubule ( " spiral
portion " of Schachowa) passes into the descending limb of
Henle's loop. The subsequent examination of a large number
of sections from the whole series of kidneys in my hands has
convinced me that the convoluted tubule is very extensively if not
generally ciliated. Fig. 2 represents a cross section, and Fig. 3
a longitudinal section of such a tubule. (Figs. 1, 2 and 3 are
drawn from different kidneys.)

It is somewhat remarkable that while I have examined an in-
definite number of capsules lying in the planes of the sections
that I have studied most carefully, I have not happened to come

450 ALBERT H. TUTTLE.

upon a single one in which the plane of the section coincided
with the neck of the capsule. I am therefore as yet unable to
say from observation whether or no the cilia exist at that historic
point. In the case of the cat, however, I have met with some-
thing approximating success in this direction, as I shall presently
state.

The question of the relation of the cilia to the rod-like bodies
(or stabcheri) of Heidenhain readily presented itself. As the al-
cohol-hardened human kidneys did not reveal these structures, I
determined to make a comparison of the two kidneys of some
mammal, one hardened with alcohol and the other with some
chromium compound. A kitten three or four days old was
therefore killed and the kidneys immediately removed, one being
divided and placed in strong alcohol, and the other treated in a
similar way with Muller's fluid, a solution of ammonium chromate
not being on hand.

The kidney of the kitten at this age presents a very interesting
functional " waking up " (if I may so term it) from within out-
wards ; the more central of the glomeruli and tubules being fully
developed and evidently active, while the more peripheral are
still quite embryonic. I hope to consider this further at an early
date.

The alcohol-hardened kidney was first examined for cilia :
these were readily found in the more active portions of the kid-
ney where the lumen of the tubule was sufficiently large : the
small i] ess or absence of the lumen in the more distally situated
tubules made a satisfactory examination impossible. Fig. 4
represents a somewhat longitudinal section of a convoluted tubule
from this kidney, the plane of section cutting the lumen of the
tubule at two or three adjacent points in the course of the latter.
Fig. 5, to which I desire to call particular attention, represents a
section passing through a Malpighian corpuscle situated in the
zone between the more active and more embryonic portions of
the kidney. As I have endeavored to represent, the plane of
section passed a little above the neck of the capsule, though
nearly parallel to it, a bit of the capsule thus overhanging the
opening into the neck. The capsule is lined throughout the
greater portion of its extent with the flattened epithelium usually
described as characteristic of its whole surface, but as this ap-

CILIA IN THE HUMAN KIDNEY. 451

proaches the neck it passes rather abruptly into a cuboidal epithe-
lium, which in the portion outlying the overhanging part of the
capsule above referred to is plainly seen to be ciliated. We have
here cilia within the capsule, the situation in the mammal re-
calling that figured by Bowman (loc. cit.) in the frog's kidney,
and by Ecker (Icones Physiologicm^ 1851-9) in that of a snake
(Tropidonotus).

The kidney hardened in lluller's fluid showed the presence of
the rod-like bodies of Heidenhain distinctly, though not conspi-
cuously ; and also, though not as clearly as in the case of that
hardened in alcohol, the cilia, situated in some cases upon cells
in which the former structure could be detected, in others upon
those in which it was not demonstrated. I am not prepared to
state any definite conclusions as to the relation between the two.

The cilia in the human kidney are from 3.5 to 5 /i. in length,
in the kitten somewhat less : they are exceedingly fine and very
numerous and closely set ; hence the great difficulty of their re-
solution. I am of the opinion that they will eventually be
demonstrated in the kidneys of mammals generally. Where
present they may be seen, I think, without difficulty under the
following conditions : first, the material should be perfectly
fresh ; the kidneys should be taken from the body of the animal
in question immediately after killing (in the case of the human
subject within a very few hours after death) and speedily harden-
ed— preferably, I think, with alcohol — at a low temperature ;
second, the sections employed must be quite thin ; third, they
should be lightly stained, if at all, and high-colored staining-
fluids, such as carmine and hematoxylin, should be avoided;
fourth, they should be mounted in glycerin ; after one is familiar
with the appearance of the cilia they can be recognized in balsam
preparations, but with considerable difficulty ; finally, the ex-
amination of the sections should be made with objectives of high
aperture: high amplification is not so important. My own
examinations have been chiefly made with a Zeiss one-twelfth,
but in part also with a Oundlach one-eighth and a Tollcs one-
sixth, all so-called homogeneous-immersion objectives ; after be-
coming familiar with my sections I could recognize the presence
of cilia with water-immersion objectives of various makers, by the
detection of what appeared to be a striated layer over the granu-
lar cells of the epithelium ; no dry objective that I have used has

452 ALBERT H. TUTTLE.

proved able to resolve this " layer " even into distinct striation,
though I can generally recognize its nature by the characteristic
diffraction color that is produced.

I have gone at length into the conditions which I believe to be

important for the successful observation of cilia in mammalian

kidneys, partly with the hope that others may be interested in

• taking up the search in this direction, and partly for the purpose

" of throwing light upon observations already made. In this latter

connection I would mention a paper in Virchow's Archiv for

Feb. 2, 1883, by S. A. Lebedeff (Zur Kenntniss der feineren

Verdtiderungen der Nieren bet der HaemogloHnavsscheidung) ;

the " striated border," figured and described by that author in

connection with the epithelium, of the convoluted tubule in the

kidney of the dog, presents an appearance exceedingly similar to

that seen when a layer of cilia (clearly shown as such under a

homogeneous-immersion objective) is examined with a good

• water-immersion objective of moderate aperture.

The general distribution of ciliated epithelium throughout the
convoluted tubules of warm- and cold-blooded vertebrates alike,
if established, would indicate a corresponding functional im-
portance. The suggestion that the cilia play a considerable part
in the propulsion of the urine toward the pelvis of the kidney, is
probably the most reasonable.

The figures in the plate wore all drawn in outline with the
camera lucida upon the same scale, and the details afterwards
added. They represent, as nearly as it is in my power to do bo,
the appearances observed ; my want of skill as a draughtsman
and my lack of familiarity with the peculiar mode of drawing
required by the process of reproduction employed must divide
the responsibility for all obvious defects. The cilia are perhaps
rendered too conspicuous in all the figures ; this is certainly the
case in Fig. 1 .

DESCRIPTION OF PLATE XXXVI.

Fig. 1. Union of convoluted tubule (spiral portion of Schachowa)
with the descending limb of Henle's loop. Man.

Fig. 2. Cross section of convoluted tubule. Man.

Fig. 3. Longitudinal section of convoluted tubule. Man.

Fig. 4. Longitudinal section of convoluted tubule. Kitten.

Fig. 5. Malpighian corpuscle, showing ciliated epithelium within
the capsule. Kitten.

ON THE EFFECT OF VARIATIONS OF ARTE-
RIAL PRESSURE ON THE DURATION OF
THE SYSTOLE AND THE DIASTOLE OF THE
HEART-BEAT. By WM. H. HOWELL, A. B„ Fellow
in Biology, and J. S. ELY, Ph. B. With Plate XXXVll.

That variations of arterial pressure have no' direct influence on
the rate of beat of the isolated mammalian heart has been clearly
demonstrated by the investigations of ProfessorMartin (x). It is
possible, however, that although the pulse-rate in any given time
may remain unchanged, still the duration of the systole or of the
diastole in each individual heart-beat may be altered, according
as the arterial tension is increased or diminished. A shortening
of the systole, for instance, might be compensated by an increase
in the length of the diastole, or vice versa, so that the total num-
ber of beats in a given period would be unaffected; just as in
electrical stimulation of the heart, when a systole is provoked
before the completion of the previous diastole there is a compen-
satory increase in the following diastole, the pulse* rate in a given
time remaining the same ('). Since the rate of beat of the heart
is not directly affected by variations of arterial pressure, within
limits; it follows that any change in the duration of the systole
consequent upon a change in arterial pressure must go hand in
hand with an inverse change in the duration of the diastole. The
same holds true, of course, for any change in the length of the
diastole.

In view of the fact that alterations in the time relations of the
heart-beat, as the direct result of changes in arterial pressure, might
take place, although the pulse-rate remained the same, it seemed
well to submit the question to investigation, especially as positive
statements with regard to the influence of greater or less arterial
resistance upon the time of the systole or diastole are not un-
frequently met with in physiological works. Marey (') considers
that it is principally the diastolic phase of the heart-beat which
is affected. According to him, when an increased resistance is

454 WM. H. HOWELL AND J. 8. ELY.

opposed to the heart, although the length of the systole itself may
not be altered, yet the following diastole will be of greater dura-
tion in order that the heart may recover from the excessive effort
it has made. Talma (4), on the contrary, in a recent article makes
the statement that the " duration of a ventricular systole increases
as the resistance increases." It is. possible that in a heart still in
connection with the rest of the body, and especially the central
nervous system, the duration of the systole may be indirectly
influenced by changes in resistance, but we hope to show that in
a heart completely isolated from extraneous nervous influences
and cut oft* from all other organs of the body, except the lungs,
variations of arterial pressure alone, within wide limits, have no
direct effect upon the systole and diastole with regard to their
time relations.

Our experiments were all made upon thd isolated heart of the
dog, kept alive by feeding with defibrinated calf's blood. The
method of isolating the heart has been described by Professor Mar-
tin in former numbers of this journal (Vol. II, Nos. 1 and 2). The
method used by us is the same in principle, although very much
altered in many of its details. To briefly repeat the essential
points of the operation : the animal, tied down upon a dog board,
is anaesthetized by means of a mixture of chloroform and ether,
both carotids are ligated and the vagi cut; the top of the sternum
is removed and the internal mammary arteries ligated ; artificial
respiration is, of course, used after this point has been reached.
As quickly as possible the sides of the thorax are cut away, a
cannula placed in the left subclavian artery, the right subclavian
ligated below the origin of the vertebral, and the superior vena
cava and azygos vein tied. A large cannula is then placed in
the aorta and fastened by a stout ligature just below the origin
of the left subclavian ; through this cannula the heart pumps out
its blood after being removed to the warm case. A large glass
cannula is now introduced into the inferior vena cava below the
diaphragm. This cannula is connected by rubber tubing with a
Mariotte flask filled with defibrinated and filtered calf s blood
heated to 37° C. The air in the tubing and cannula, it is scarcely
necessary to say, is replaced by blood before the latter is placed in
the vein. The warm blood is now allowed to run into the heart
from the flask while the clamp is removed from the left subclavian

ARTERIAL PRESSURE AND THE HEART-BEAT 455

artery, and the heart permitted to pump out all coagulable blood
through a tube connected with the cannula. Care must be taken
at this part of the operation to keep up a good arterial tension by
partially clamping the outflow tubing. The coagulable blood is
also removed from the aorta through the cannula connected
with it. When all of the dog's own blood has been washed out of
the heart and lungs, the animal is transferred to the warm case.
The arrangements here can be scarcely understood without the
aid of a diagram. In papers shortly to be published, embodying
the results of some previous work by Professor Martin and others
under him, the details of the apparatus with an accompanying
diagram will be given. It is sufficient to say that within the case
are two large Mariotte flasks, each capable of holding several
litres of blood, and so arranged that they can be used alternately,
the heart, when receiving blood from one flask, pumps it out
through the cannula in the aorta and the long rubber tubing
which is now connected with it, back into the other flask, so that
when one is empty the other is ready to be used. The tubing
connected with the aort$ extends above the top of the case, and
the arterial pressure against which the heart works can easily be
varied to any desired extent by increasing or diminishing the
height of the end of this tube above the heart. The exact varia-
tions in arterial pressure thus produced are given by a mercury
manometer connected with the cannula in the left subclavian
artery. The pen of this manometer writes upon the roll of paper
of the kymograph, and from its tracings the pulse-rate is also
obtained.

The essential point in our experiments was to register accurate-
ly the duration of the systole and the diastole of the heart isolated
in this way and exposed to varying arterial pressures. It would
have been a comparatively easy matter to have taken tracings of
the heart-beat directly by means of levers, after the method em-
ployed by Hoffa and Ludwig (5), or by the application of the more
simple device used by Baxt (6). But it seemed questionable to us
whether such methods possess sufficient accuracy. Outside of the '
complications arising from the possible changes in position of the
lever on the heart's surface, or from changes in position of the
heart as a whole, it appears very uncertain whether or not
the very beginning of the diastolic relaxation will be promptly

456 WM. H. HOWELL AND J. S. ELY.

registered by such instruments. Owing to the smallness of the
dog's ventricle, on the other hand, it is scarcely practicable to
introduce an ampulla into the heart in the way employed by
Chauveau and Marey (7) for the horse.

The method determined upon, and which, it seems to us,
leaves but little to be desired in the way of accuracy, was as
follows. After the operation of isolating the heart was finished,
and the dog had been transferred to the case, a catheter with
terminal and side openings was passed down the superior cava
and right auricle into the right ventricle, and fastened firmly in
position by a ligature around the superior cava. The catheter
was filled beforehand with defibrinated blood. Its free end was
connected by means of lead tubing, as short as possible and filled
with 0.6 per cent NaCl solution, with an ordinary Fick spring
manometer. The arm of the manometer carrying the writing
point, had all vibrations of its own, arising from its inertia, damp-
ened in the usual way by a carrier immersed in oil. The tracings
were taken upon the blackened paper of a rapidly revolving
drum-kymograph, upon which, immediately under the manometer
pen, a tuning fork vibrating fifty times a second was likewise
made to write. The accuracy of the manometer in recording
rapid variations of pressure was tested before using by connecting
it with a small rubber bag, filled with liquid, which could be
compressed under an ordinary telegraph key, the beginning and
end of the stroke being registered by an electro-magnet. The
writing point of the manometer did not move in a straight line,
but described the arc of a large circle. When the height of
the curve was small, not exceeding ten or twelve millimetres,
this arc did not differ appreciably from a straight line. In most
cases, however, it was necessary to introduce a correction for this
error. The correction was made by simply allowing the pen to
describe its arc upon the drum when stationary, and then
measuring the displacement from the vertical for any given
height. The difference was added to or subtracted from the
recorded time of the systole, according as the displacement was
in the direction of the movement of the drum or opposed to it.

In every case but one the tracings were taken from the right
ventricle, owing to the fact that a catheter can be introduced into
this side of the heart with great ease and without causing any

ARTERIAL PRESSURE AND THE HEART-BEAT. 457

injury. We made several attempts to place a catheter in the left
ventricle, either through one of the pulmonary veins or through
a slit in the left auricular appendage. By the latter method it
is necessary to remove the pericardium and to expose the heart
to more or less handling. The consequence was that* it never
lived well for any length of time after the operation. By the
former method we succeeded in obtaining several series of ob-
servations, one of which is given in the following table (Experi-
ment VII). The results are in accord with those obtained from
the right heart. It can make but little difference from which of
the ventricles the tracings are taken, since the complete syn-
chronism of the two sides of the heart, when beating normally,
is a matter about which there can be no doubt.

In making an observation tracings were taken simultaneously
upon the drum and the large kymograph, beginning generally
with a mean arterial pressure. As soon as one tracing was fin-
ished the arterial pressure was quickly changed, and another
similar tracing taken. So that the heart was not exposed as a
rule to any given arterial pressure for more than one or two min-
utes before the tracing was taken. Three or four such tracings
at different arterial pressures, forming a series the members of
which were comparable amongst themselves, were taken upon
each drum.

Outside of the variations of arterial pressure the only condition
which was liable to change during a series was the pulse-rate.
Any change in pulse-rate would produce an alteration in the re-
lations of systole and diastole, and destroy the value of the series.
As a matter of fact many series were rejected on this account.
Since, however, the tracings of any one series were always taken
from the same flask of blood, the temperature and therefore the
pulse-rate remained constant in the majority of cases. To obtain
the duration of the systole and the diastole at each arterial
pressure, vertical lines were drawn from the tuning-fork curve to
the beginning and end of the heart-beat, for ten successive beats.
The time of each systole and diastole was then counted out,
the average taken, and the necessary correction made for the
arc described by the pen.

Very soon after the commencement of our work a difficulty
presented itself in determining at what point to reckon the be-

458 WM. H. HOWELL AND J. S. ELY.

ginning of a systole. In some heart-beats, especially those in
which there was a slow pulse-rate, the ascending limb of the curve
was of the character shown in Fig. 3. The curve as seen in
this figure, does not rise from the base line with uniform rapidity;
there is 'at the beginning of the wave a slow rise, which later
suddenly increases in steepness. It seemed to us that the pre-
liminary rise was merely the indication of the auricular beat,
and that the systole proper of the ventricle began at the com-
mencement of the steep and sudden ascent of the wave. So long
as the pulse-rate remains the same, as it does in each single series
of observations, and the arterial pressure is not lowered below
the limit at which the heart is well nourished, it really makes no
difference whether the systole is counted from the bottom of the
wave or from the beginning of the steep rise, as far as the effect
of arterial pressure upon the time relations of the phases of the
heart-beat is concerned, since the difference would only affect the
absolute length of the systole and not its comparative relation to
the length of the diastole at different arterial pressures. If we
wish to make a comparison between the times of the systole with
different pulse rates, then it becomes necessary to settle this
point. We had it in mind to go on to the effect of changes of
temperature on the time relations of systole and diastole, and
therefore carried out several experiments for the purpose of
determining which point of the curve indicates the actual begin-
ning of the ventricular systole. The result at which we arrived is
that the first shallow rise is really caused by the auricular con-
traction, and in counting out our tracings we always began to
reckon the systole from the beginning of the steep rise. So that
our figures indicate, for the given pulse rates, the absolute length
of the systole and the diastole in the dog's heart.

The experiment which we made to determine this point was
to take simultaneous tracings, in the way described, from both
auricle and ventricle. A catheter was introduced into the right
auricle through the superior vena cava, and into the left ventricle
through a slit made in the auricular appendage; each catheter
was connected with a Fick manometer. It was necessary to make
the auricular catheter larger and to connect it with its manometer
by means of wide lead tubing, in orcter to obtain distinct auricular
waves. This had the disadvantage that distinct oscillations of the

ARTERIAL PRESSURE AND THE HEART-BEAT. 459

large column of water took place, and were evident in the
tracings, though this was of little consequence for the question
in hand. The1 exact position of the catheters in the heart in
these, as in all the other experiments, was determined by post-
mortem examinations. It was not possible to make any series of
observations at different arterial pressures with catheters in both
auricle and ventricle. The heart was usually injured to such an
extent by the operation that it soon became too weak to pump
the blood to any considerable height, and shortly died. Several
such experiments were made, however, in which the heart beat
normally for some length of time. Figure 4 gives a portion of a
tracing taken in this way. The two pens in this case were un-
fortunately not writing in the same vertical line ; the pen of the
manometer connected with the auricle, giving the lower of the two
tracings in the figure, was about a millimeter in advance of the
other. In the lower curve, given by the auricular manometer,
it is seen that both the auricular and ventricular contractions
are recorded. By comparing it with the upper curve, which was
given by the manometer connected with the left ventricle, it is
very evident that the short preliminary rise in the contraction
wave of the latter is synchronous with the auricular contraction
as given in the former.

From this, and other simultaneous tracings in which the con-
traction wave wa6 of a different form, we were led to the conclu-
sion that the proper systole of the ventricle begins at the steep
rise, and in all our tracings, as we have said, we have reckoned
it from this point. When the pulse-rate is rapid, and there is no
appreciable pause after the diastolic expansion, " the auricular
wave does not appear in the ascending, systolic limb of the wave,
but at the end of the previous diastolic descent ; the systole in
such cases was counted from the beginning of the wave.

At the result of many series of observations, most of which
are given in the following table, we are able to state that varia-
tions of arterial pressure, between 50 and 160 mms. of mercury,
have no direct influence whatever upon the duration of the systole
or the diastole of the heartbeat in the dog.

When the blood pressure sinks so low that the proper nutrition
of the heart is prevented, there is a diminution in pulse rate and
a consequent change in the time relations of the systole and
diastole. (See Exp. II, Obs. A4.)

460 W3i. H. HOWELL AXD J. S. ELY.

Professor Martin, in his work on the effect of arterial pressure
on the pulse-rate, placed the limit to which arterial pressure could
be lowered without affecting pulse-rate at about 20 mm of
mercury. In some of the hearts need in our experiments this
effect was manifested when the arterial pressure fell to 30 mms.
of mercury, as in the case cited. On the other hand, we hare no
doubt that the arterial pressure might be raised to considerably
more than 160 mms. of mercury without affecting the time of
systole or diastole. In one observation, indeed, the arterial
pressure was increased to 180 mms. without causing any change;
the series as a whole in this case was not of a uniform pulse rate,
and hence is not given in the table. The numbers given, 50 to
160 mms. of mercury, can be fairly assumed as the limit of prob-
able variations of arterial pressure in living dogs of the size used
in the experiments

In the table given below the pulse rates for the different obser-
vations cannot be taken as absolutely correct. The pulse-rate
on the kymograph tracing was always estimated, as nearly as
possible, for the ten seconds during which the drum, upon which
the ventricular tracings were taken, was revolving. But owing
to the irregularity of the tracing of a mercury manometer, espe-
cially when the kymograph to which it is attached is going
rather rapidly and shaking the manometer more or less, errors of
half a beat or more may easily be made in counting out When-
ever the difference in pulse-rate was greater than one beat in ten
seconds the series of observations was rejected.

A consideration of the table will show that only in rare in-
stances do the* tiroes of the systoles or diastoles, in any one series
of observations, differ amongst themselves by as much as .01 of a
second. And of the cases in which a difference as great as this
occurs, it sometimes points to an increased length of systole with
increased arterial resistance, and sometimes the reverse, appear-
ing to indicate that the difference was probably owing to errors
of marking. In the method adopted by us it is not always possible
to mark with absolute precision the beginning or end of the
systole, and erron of .01 of a second might readily be made in
this way. Id some forms of waves no such difficulty occurred, and
the time of the systole or of the diastole for the ten waves counted

t remained practically identical. In other forms more serious

Table.

Number

Pulse Rate

Arterial

Duration of

Observation

Pressure in

Average

Average Di-

Experiment.

10 seoonds.

mm. of Hg.

Systole In seo.

astole in seo.

81.6

140

.121

.189

81.6

.121

.189

39.6

.138

.193

29.26

140

.121

.197

84.6

100

.120

.170

36.

152

.121

.166

86.

.118

.164

A«

81.6

.092

.226

38.

106

.127

.179

83.

149

.187

.170

82.

.188

.177

B«

82.6

106

.137

.174

30.6

101

.160

.170

in.

160

.162

.170

.159

.174

27.76

100

.112

.262

.117

.242

28.

187

.118

.239

A«

28.76

.115

.232

IV.

27.

100

.113

.256

27.

118

.119

.246

26.6

.119

.261

30.6

121

.156

.166

80.6

.157

.160

81.

145

.157

:i63

81.76

101

.146

.168

81.6

124

.146

.164

81.6

.147

.169

30.26

.167

.170

29.6

186

.149

.177

29.6

.146

.194

28.3

102

.253

.182

VI.

23.

186

.246

.188

23.

.260

.184

24.

.133

.251

Vil.

Catheter in )
Left Ventricle )

28.26

184

.146

.248

28.

.147

.244

31.6

104

.126

.200

30.6

.129

.193

30.

152

.132

.196

80.5

107

.132

.195

vin.

80.

160

.127

.208

80.26

.130

.203

80.6

100

.126

.194

80.76

.129

.196

80.6

151

.180

.198

80.

.129

.206

464 WM. H. HOWELL AND J. S. ELY.

traction of the right ventricle as well as of the right auricle is regis-
tered by the manometer. The upper curve is from the manometer
connected with the left ventricle. The tuning-fork curve marks
hundredths of a second.

REFERENCES.

1. Martin. Studies from the Biol. Lab., Johns Hopkins Univer-
sity. Vol. II, No. 1, p. 110, and No. 2, p. 213.

2. Marey. La Circulation du Sang. 1881, p. 347.

3. Marey. Ibid.

4. Talma. Beitrage zur Kenntniss des Einfiusses der Respiration
auf die Circulation des Blutes. P finger's Archiv, 29 Bd., S. 311,
1882.

5. Hoffa u. Ludwig. Einige neue Versuche uber Herzbewegung.
Ztschr. f. rat. Med., IX, S. 107.

6. Baxt. Die Verkiirzung der Systolenzeit durch den N. accelerates
Cordis. Archiv f. Anat. u. Physiol., 1878, S. 122.

7. Marey. Physiol, rn/d. d. I. circulation, p. 54, 1863.

ERRORS IN PLATE.

Fig. 1. Ci, 4th systole from the left, instead of .10 of a second
should be .11 of a second.

Cs, 1st and 2d systoles from the left, instead of .10 of a second
should be .115 of a second.

5th systole from the left, instead of .10 of a second should be .11
of a second.

Ci, 2d systole from the left, instead of .10 of a second should be
.11 of a second.

The plate was drawn from the original tracings, but two attempts
of the printer have failed to get the chronographic tracings correctly
transcribed.

NOTES ON THE MEDUSAE OF BEAUFORT, N. C.

Part II.1 By W. K. BROOKS, Associate Professor of Biology,
Johns Hopkins University.

Tukritopsis Nutrioula (McCrady).

Oceania nutricnla (McCrady). Modeeria multitentacula
(Fewkes). Modeeria nutricula (Fewkes). Turritopsis nutri-
cola (Haeckel).

This is one of the most abundant medusae at Beaufort during
the summer months, and I have been able to verify the extreme
accuracy of McCrady's graphic account of the structure and
habits of the adult. The larva is probably a deep-water form,
as it was found only once, notwithstanding the abundance of the
medusa.

The Larva. — The only colony of the hydra which I obtained
was scraped from the piles of the steamboat wharf at Morehead
City, seven or eight feet below low-tide mark. It lived for two
weeks in the house, and set free great numbers of hardy medusae,
which were reared without difficulty. The upright stems, from
one- third of an inch to half an inch high, bore large terminal
hydranths, as well as smaller ones scattered irregularly along
the stem on short stalks. The long fusiform body of the hydranth
carries from eighteen to twenty thick, short filiform tentacles,
which are arranged in three or more indefinite whorls. The
medusa buds grow around the stem just below the hydranth,
and are carried on short stems. The perisarc is not annulated,
and it forms a loose cylindrical sheath around the main stem
and the short branches which carry the lateral hydranths and the
young medusae, while the latter are closely invested by a thin
capsule of perisarc. The sheath on the stems is thick and crusted
with foreign matter. It terminates abruptly by a sharp collar
just below each hydranth. The young hydranths and medusae
are budded off at this point, but soon become entirely sheathed in

1 For part I, see this Journal, VoL II, p. 185.

466 W. K. BROOKS.

perisarc by the growth of the stem. The pale yellowish-red
hydranths are very similar to those of Tvhiclava (Allman.)

Metamorphosis of the Medusa. — The little medusa remains
attached to the stem for some time after the rupture of the
sheath of perisarc. At this time it is nearly spherical, and
covered with large conspicuous ectoderm cells. Its eight short
tentacles are thrown backwards in contact with the outer surface
of the bell, and their tips are hooked or bent upon themselves.
This position of the tentacle renders the bulb at the base, with its
ocellus, very prominent.

The medusa, when set free, has eight tentacles, a thin globular
bell, and a short simple proboscis. When swimming the tentacles
are bent into hooks and thrown back against the umbrella, which
is lengthened and emarginated during each contraction. When
at rest the height of the umbrella is about equal to its diameter,
and it forms a spherical segment almost equal to a sphere. The
tentacles are capable of extension to a length equal to about
twice the diameter of the umbrella, and when the animal is at
rest they are stretched out almost horizontally, and the distal
half is bent downward a little, forming an obtuse angle near the
middle of the tentacle. The interradial tentacles lie nearly in
the plane of the bell-margin, and the perradial tentacles a little
lower. The tips of the extended tentacles are slightly clavate,
with a spot of dark orange pigment. The length of the pro-
boscis is about two-thirds the height of the umbrella, and its
upper and lower ends are smaller than the middle. The
mouth is simple, and the endoderm of the oral end of the pro-
boscis is very thin, but just below the constriction at the aboral
end it becomes very thick ; the thickened area arching outwards
on to the subumbrellar surfaces of the radiating tubes.

This thickening of the endoderm cells of the aboral end of the
stomach is characteristic of Turritopsis ; and in a specimen a
week old, the whole upper half of the proboscis is filled by four
great masses of very large endoderm cells, which meet in the
central axis and run out for some distance into the radiating
tubes. The singular structure which is thus formed has been
described by various authors as a peduncle, but it is not at all
the same as the gelatinous projection from the substance of the
umbrella which, in many medusae, hangs down into the stomach.

MEDUSAE OF BEAUFORT, N. C. 467

As the medusa grows the proximal ends of the radiating tabes
are drawn down into the cavity of the umbrella, until, in speci-
mens two weeks old, the stomach is suspended some distance
below the sub-umbrella, by a transparent mass of large cells,
meeting in the central axis and perforated by the four tubes. In
the adult this body almost entirely fills the upper half of the
umbrella-cavity. In a medusa a week old the oral lobes have
appeared, and are fringed by the large projecting lasso-cells
which have been noticed by McOrady and others. At about
this time the reproductive organs make their appearance on the
proboscis at the lower ends of the masses of endoderm cells. The
tentacles are still only eight, and no more were developed in the
medusae which I reared from the larva, but I captured many
specimens in the same stage and at all the following stages up
to maturity.

In specimens from a week to two weeks old the lower surface
of the very wide velum is pushed out to form eight hemispherical
pouches ; four of them perradial and four interradial, in the planes
of the eight tentacles. These pouches project so that they are
visible in a profile view below the free edge of the umbrella.

Cunena Ootonabia (McCrady).

McCrady's remarkable discovery that the young of this species
exists as a parasite within the bell of Turritopsis, a medusa
belonging to a totally different group, is of so much interest that
I was well pleased to have an opportunity to verify it at Beaufort
during August and September, 1882. Since McCrady's paper
was published no one has succeeded in rediscovering these larvae,
and as both Cunina and Turritopsis occur at Beaufort, the latter
in considerable numbers, I had kept a sharp watch for them for
nearly three years before I found them. Near the end of July,
1882, I found a single specimen of Turritopsis filled with the
larvae, and from this time until the end of the season they could
be obtained in great abundance. I was therefore able to verify
McCrady's accurate account of the metamorphosis, and to add
a number of new points which I hope to publish soon in an
illustrated paper.

468 W. K. BROOKS.

Nemopsis Baohei (L. Agassiz).

Nemopsis Gibbesvi (McCrady).

This medusae is quite common at Beaufort during the spring
and early summer months, and specimens were found at all stages
of growth. There does not seem to bo any reason to doubt its
identity with the northern form, and Agassiz' specific name must
therefore be retained in place of McCrady's name.

The Larva. — Most writers upon the subject have questioned
the relationship between the floating hydroid found and described
by McCrady, and Nemopsis, and my observations show that the
medusa is derived from a fixed hydroid closely related to Bou-
ganviUeia and Endendrium.

On May 29th, 1882, the dredge brought up from about twelve
feet of water in Newport river, a piece of decayed wood covered
with a small Endendrium-like hydroid about an inch high.
Each main stem gave rise to three or four short alternating
branches, and these, as well as the main stem, ended in hydranths,
which were sharply separated from the stem by a fold or collar.
The thin transparent horny ectosarc extended almost but not
quite up to this fold, and there were two or three irregular
annulations on each side branch close to the main stem. The
hydranth carries twenty-four long slender tentacles, with their
proximal ends in a single circlet, but with their tips bent alter-
nately backwards and forwards, thus forming two circlets. The
very extensible funnel-shaped proboscis is sharply distinguished
from the body of the hydranth, and the hydra therefore resembles
Endendrium, as described by Allman, more than it does Bou-
ganvilleia in this particular.

The six or eight medusa buds are arranged in a ring around
the body of the hydranth, about midway between the bases of
the tentacles and the proximal end of the body. The various
medusae in this ring are in different stages of growth, and only
one is usually set free at a time. The terminal hydranths and
those near the end of the main stem have no medusa buds, as
these seem to be developed only upon the older hydranths.

The Metamorphosis of the Medusa. — The medusa is very
small when set free, and it is flattened and folded together so
that the proboscis projects out of the umbrella. In half an hour
or an hour it expands and begins to swim. It is then about two

MEDUSAE OF BEAUFORT, N. C.

469

one-hnndredths of an inch high, and the diameter is a little Less
than the height. The proboscis is short and simple, without oral
tentacles, and the umbrella is about as thick at its sides as it is
in the oral axis. Most of the specimens had four perradial
tentacles — one at the end of each radiating tube. In others
there were six tentacles, arranged in this way, and in these

3.2

2.3

the tentacles 1. 1. were much larger than those marked 2. 2.,
and these again larger than 3. 3. — the latter being very small
transparent buds in most specimens. The order of appearance
of the tentacles varies considerably. In one medusa, twenty-
five one- thousandths of an inch in diameter, they were like thi

3. 2. 4.

4.2.3

No. 1 being larger than No. 2, this again larger than No. 3,
while No. 4 was a very small bud.

In another specimen of the same size they had the arrange-
ment shown in this diagram — No. 1 being in each case larger

1.2.

2. 1.

1. 2.

2 1.

470 W. K. BROOKS.

than No. 2, and more directly in the line of the radiating tube.
In this specimen, twelve days later, after it had grown to a
diameter of four one-hundredths of an inch, the arrangement was
like this, with an ocellus in 1 and 2. At this stage all four ten-

4. 2. 1. 3.

3.
1.
2."
4.

4
2.
"l.
3.

3. 1. 2. 4

tacles in each cluster were of about the same size; but in a
younger specimen, which was taken with the tow-net on May 9th,
and which was three one-hundredths of an inch in diameter, No.
4 in each bunch was a small transparent bud. The oral ten-
tacles appear when the medusa is about three one-hundredths of
an inch in diameter. They are simple at first, but they soon be-
come forked at their tips, and each of these forks becomes forked
in the same way, and so on. It is hardly possible to give a clear
account of the changes in the shape and outline of the umbrella
without figures, but I am sure that/when my figures are pub-
lished, they will prove the specific identity of the northern with
the southern form.

Phortis Gibbosa (McCrady).

Eirene gibbosa (L. Agassiz). Irene gibbosa (Haeckel).

As all the other species of Irene have marginal cirri, the ab-
sence of these structures in this form seems to justify the retention
of McCrady's generic name. It is a very rare medusa, and
McCrady gives no figure of it, although I have in my possession
a sketch made by him from memory. The occurrence of the
medusa has never been noted by any one except McCrady.
Specimens were occasionally met with at Beaufort during the
summer months, and I had therefore been able to secure a
pretty complete series of the older stages, when, in September,
1882, 1 obtained the hydra stage in great abundance, and reared
from it hundreds of young medusae.

MEDUSAE OF BEAUFORT, N. C. 471

The Larva. — On September 19th, 1882, quantities of stems of
Aglaophenia were torn up by a gale and thrown upon the beach
at Fort Macon. Attached to these stems were specimens of a
peculiar campanularian hydroid. A long slender hydrorhiza
runs along the stem of Aglaophenia, and gives rise, at pretty
regular intervals, to short annulated branches, some of which
terminate in hydranths and others in reproductive calicles,
which do not differ very greatly from the hydrothecae either in
size or in shape. The hydrothecae are trumpet-shaped, slightly
curved, and they taper gradually from the base, which is no
larger than the short stem, to the wide, flaring, reflected open-
ing. The hydranth has a long slender body and about twelve
tentacles, with rings of lasso-cells.

The gonotheca is very similar to the hydrotheca in size and
shape, as well as in its position upon the stem. The chief differ-
ences are that the gonotheca tapers somewhat more gradually
towards the stem than the hydrotheca, the annulations run
up it for half its length or more, and its distal end is less
flaring. The blastostyle runs along one side of it, to terminate
in a club-shaped tip or manubrium, and medusa buds are placed
along one side of it. There are only three or four of these, in-
creasing in size from the base to the free end.

The Metamorphosis of the Medusa. — Only one medusa
escapes at a time — the largest one nearest the distal end of the
blastostyle — and, as soon as it is set free, it expands or unfolds
so as to become about as long as the entire gonophore.

As soon as the hydroids were captured each specimen was
placed, alone, in a tumbler of sea-water, and, when the labora-
tory was reached, each tumbler was found to contain hundreds
of swimming medusae. These were carefully picked out with a
dipping tube, and the hydroids were then placed in larger
aquaria, where they lived for nine days and continued to throw
off medusae, although the number set free daily was very much
less than the number set free within a few minutes after the
specimens were found. The rarity of the adult medusa stands in
marked contrast with the hardy and prolific nature of the larva,
and as the latter were found on this occasion in great abundance,
I conclude that its proper home must be at some distance from
the shore, and that the stems of Aglaophenia upon which they
were found had been torn up from deep water.

472 W. K. BROOKS.

When set free the ex-umbrella of the medusa is regularly
curved, and its height is a little greater than its diameter. The
sub-umbrella, on the other hand, is bent upon itself about half
way up, at an obtuse angle, and the lower or free half of the um-
brella is about twice as thick as the upper half. The proboscis
is very short and is divided into two portions — an upper flat
division which runs out along the radiating tubes fbr about one-
sixth of their length, and a pendant portion of about one-fifth
the height of the umbrella, and folded into four oral lobes.
There are no traces of marginal sense organs at this stage, and
the arrangement of the tentacles, in all my specimens, was some-
what peculiar, as shown in the diagram.

There are four perradial tentacles, of
which one (3) is very much smaller than
the others, and consists, in some specimens,
of a bulb only, the lash being undeveloped.
Opposite this is a somewhat larger tentacle
(2), with a short lash. The two remaining
perradial tentacles (1 1) are fully developed
and alike. Their lashes are very slender l

and delicate, and may be thrown out to two or three times the
diameter of the bulb. Between these four perradial tentacles four
interradial tentacles (4 4 4 4) are represented by bulbs without
lashes. In one of the quadrants there are two small protnber-
ances from the wall of the circular tube — the adradial tentacles
(5 5). They are placed midway between the perradial and inteiv
radial tentacles, and there are no traces of them in the other
three quadrants. At this time there are no otocysts, and lateral
cirri are totally absent at this stage, as they are at all later
stages.

At the end of the first eighteen hours the shape of the medusa
has changed completely. When contracted in swimming, its
height is nearly equal to its diameter ; but when it is at rest the
diameter is about twice the height, so that it is no longer
globular but saucer-shaped. The upper half of the umbrella has
begun to thicken to form the peduncle, and it is now about as
thick as the lower half, except at the angle in the sub-umbrella,
where it is still thin. The four perradial tentacles are all fur-
nished with lashes, but these are still absent in the interradial

fi 1

5
2

4 4

MEDUSAE OF BEAUFORT, N. C. 473

tentacles, although these have grown larger. There are now
two adradial thickenings of the circular tube in each quadrant.

In a specimen one-fourth of an inch in diameter there are six-
teen fully-developed tentacles and sixteen thickenings of the cir-
cular tube, without lashes. The four perradial tentacles and
the four interradials are equal and similar, while the eight
primary adradials are somewhat smaller. There is an otocyst
with one otolith on each side of the base of each tentacle, making
thirty-two in all.

In a specimen two-thirds of an inch in diameter there are forty
tentacles, and sixty in one an inch wide.

Amphinbma Apioatum (Haeckel).

Saphenia apicata (McCrady). Stomotoca apicata (L. Agas-
siz). Stomatoca apicata (Fewkes).

This medusa is not very common at Beaufort, although
specimens are found occasionally all through the summer. The
structure of the adult has been well described by McCrady and
Fewkes, and I have little to add td their accounts.

The Zarva.^-The hydra stage was found on three occasions at
Beaufort between July 5th and July 11th, 1882, on the lower
surface of the shell of the living LvrwuVm^ fastened to the sand-
tubes of Sabettaria. It is a Perigonomus very much like P.
ndnutus (Allman). The simple unbranched slender upright
stems are from eight one-hundredths to one-tenth of an inch
high, and their bases are encased in the sand-tubes of the anne-
lid so that the presence of a creeping stolon could not be ob-
served. The flexible stem is covered for one-half or two-thirds
its length by a delicate, closely adherent film of perisarc, to
which foreign particles are attached. The stomach occupies
about one-fourth or one-fifth the total length of the stem, from
which it is separated by a slight constriction. There were ten
tentacles in each of the thirty or forty specimens which 1
examined, and, when fully extended, they point alternately
backwards and forwards — those pointing forwards being a little
longer than the others. The medusae are attached by very short
peduncles along the stems, but as most of them were set free
before the specimens could be examined, the mode of attachment
could not be carefully studied.

474 W. K. BROOKS.

Each colony of larvae was placed by itself in a bottle of sea-
water as soon as it was found, and, when the laboratory was
reached, each bottle was found to contain hundreds of minute
but very active medusae. They proved to be quite hardy and
lived for more than a week in aquaria, although the great length
and delicacy of the tentacles caused great difficulty in rearing
them, as the tentacles became entangled with each other and
with the sides of the jar, so that the medusae could not be drawn
into a dipping tube without injury, and many were destroyed
each time the water was renewed.

Metamc^rphosis of the Medusa. — When the medusa is set free
there is no trace of the apical process, which is not a larval
structure, but an adult characteristic. The bulb is about twice
as high as wide — the height being about twenty-five thousandths
and the diameter about thirteen thousandths of an inch. The
wall of the umbrella is thin, and its surfaces are nearly concen-
tric and regularly curved. The proboscis hangs down to about
one-half the height of the umbrella cavity, and ends in a circular
mouth. The stomach is a little enlarged at its base, where it
joins the radiating tubes. There are two tentacles with large
bulbs, faintly tinged with pale orange. The long delicate lash
springs abruptly from the bulb, and its base is very little larger
than its tip. Immediately after the medusa is liberated the
length of the tentacle is four or five times the height of the um-
brella. Alternating with the two opposite tentacles there are
two small pigmented perradial bulbs without lashes.

In a medusa three days old and thirty-five one-thousandths of
an inch high, the apical process is present as a short, solid,
rounded projection from the aboral pole. The tentacles are from
ten to twenty times as long as the height of the bulb, and four
pigmented interradial enlargements of the wall of the circular
tube have appeared midway between the four perradial bulbs.
The length of the proboscis is now a little more than half the
height of the sub-umbrella.

When five days old the medusa begins to assume the adult
form. The apical process grows rapidly, and becomes pointed or
conical, the lower or free half of the umbrella becomes thicker
than the upper half upon which it is bent at an angle. The four
oral folds have appeared, and the upper end of the proboscis is

MEDUSAE OF BEAUFORT, N. 0. 475

slightly enlarged, probably by the growth of the sexual elements.
The tentacle tapers more gradually at the bulb, and the lashes
and marginal enlargements are relatively a little larger than
they were at an earlier stage.

In specimens eight days old the process is equal to or greater
than half the height of the umbrella, and the medusa has essen-
tially the adult form, except that the marginal enlargements are
much larger relatively and less numerous than they are in the
adult. I was not able to keep them longer, as the tentacles, fif-
teen or twenty times as long as the height of the umbrella,
became entangled with each other and attached to the sides of
the glass jar, so that I was not able to remove the animals to
change the water without injuring them.

Liriopk Soutigeka (McCrady).

This is one of the most abundant medusae at Beaufort, and
there is no difficulty in obtaining a supply of segmenting eggs
and young medusae. The eggs are very small and transparent,
and, as they develop with great rapidity, they are very favorable
subjects for embryological work. My results agree perfectly
with those of Metschinchoff, and there is no difficulty in witness-
ing the actual delamination of the inner ends of the cells of the
developing egg.

THE ACTION OP ETHYL ALCOHOL UPON THE
DOG'S HEART. By H. NEWELL MARTIN, M. A.,
M. D., D. So., Professor in the Johns Hopkins University, and
LEWIS T. STEVENS, B. A., Fellow of the same.

The physiological action of alcohol is a subject in connection
with which very much has been written. In the Index Cata-
logue of the Library of the Surgeon-General's office there are
more than one hundred and fifty separate references under the
title "Alcohol, physiological effects of." From this vast mass
of literature bearing on a subject which has been so often prom-
inent in social and political discussions, very much may, of course,
be at once eliminated as of no immediate interest to the physiolo-
gist or therapeutist in his capacity as such. It contains no orig-
inal experiments, and is mainly a rhetorical and uncritical
account of the work of Others, often also described with a mental
bias. After throwing aside these productions of the orators and
essayists, there still remain numerous articles professing to deal
with the physiological action of alcohol which can hardly be
accepted as so doing, for in many cases all sorts of alcohol-con-
taining drinks have been administered to men or the lower
animals, and the results, if any, set down as due to the alcohol
only. That this is not justifiable a moment's consideration will
make clear, for it is well known that in different wines and spirits
various substances are present which have potent action on the
system, and cause these drinks, quite apart from the percentage
of alcohol in them, to produce each its own characteristic effect,
not only immediately after consumption, but, when taken in excess,
remotely and permanently ; as illustrated by the different patho-
logical states to which they give rise or predispose. It is to this
cause undoubtedly that the very discordant statements of various
workers are mainly due; while there has also been a good deal of
careless experimenting, such as the injection of large doses of
90 per cent, alcohol into the alimentary canal and the ascription
of the consequences to absorbed alcohol, quite regardless of the
intense local irritation which must have been set up in the
stomach or rectum of the animal experimented upon. During

478 H. NEWELL MARTIN AND LEWIS T. STEVENS.

the last thirty years more careful work with reasonable doses
and dilution, and with attention to the kind of alcoholic liquid
used, has given better results. So far at least as the pulse is
concerned, it seems fairly settled that alcohol diluted with water
and in doses sufficient to produce transient disturbance of the
mental faculties, has no effect on the pulse-rate of healthy men
or other mammals, though even here there is not absolute agree-
ment. Zimmerberg,1 whose paper is the most satisfactory of all
those on this subject with which we are acquainted, found no
pulse alteration caused by alcohol in dogs and cats when the
animals were not tied down. Rabbits, on the contrary, showed
a quickened pulse, but this seemed due to scare, for the same
phenomenon was observed when a little water was injected
into the animal's stomach. He also could discover no pulse
quickening in man. Dr. Edward Smith,9 however, found his
own pulse quickened by alcohol, while that of Mr. Moul was un-
affected. As Dr. Smith makes no statement as to whether he
was accustomed to the daily use of alcohol, it seemed possible
that he was an habitual abstainer, and that the pulse-quickening
action of the alcohol in his case depended upon the fact that his
system was quite unaccustomed to it. As this point seemed of
interest and perhaps of practical importance, we asked a friend,
aged about twenty-six, and who had never, so far as he knew, drank
anything containing alcohol, to allow us to make an observation
upon him. He kindly consented, and we give here the result
before proceeding to the main series of our experiments. The
alcohol used in this case and throughout our researches was that
prepared by Squibb, and sold as " Absolute Alcohol " of sp. gr.
0.7850 at 25° 0., and warranted to contain not less than 99.75
per cent, of pure ethyl alcohol. Mr. J.'s last meal was taken
at 7 P. M. At 9 P. M. he lay down on a bed, and his pulse-
rate was noted at intervals for an hour. At 9h. 05m. it was
74 per minute, and varied between that and 71.5 until 9h.
30m.; he then became drowsy, and this and the recumbent
posture brought the pulse down to 68 at 9h. 58m. At lOh.
08m. he was roused ; at lOh. 10m. told he was to be given the
alcohol. The substance really administered was, however, only
some sugar and water — the object being to see what effect, if
any, the idea of taking the drug (which might well excite a per-
son accustomed to regard it somewhat in the light of a poison)

ETHYL ALCOHOL UPON THE DOG'S HEART. 479

would have on the pnlse. There was a transient quickening to
73 per minute, but this was probably merely duo to rising from
the recumbent position in order to drink. At lOh. 31m. P. M.,
when the pulse had fallen to 70, 15 cub. cent, of alcohol in 50 cub.
cent, of water were given. This caused no rise of the rate of heart-
beat greater than two beats in a minute, and this only lasting a
few minutes, and easily accounted for by the muscular effort in-
volved in changing the posture. At lOh. 52m. the pulse was
again 70 per minute, and thenceforth until the final counting, at
12h. 10m. A. M., its rate lay between 72 and 67 per minute —
on the whole slowing towards the close of the experiment.
This slowing can hardly have had any dependence on the alco-
hol, as it is well known that the pulse normally becomes less
frequent towards midnight, and especially in a person who has
lain for hours at rest. That the dose of alcohol was sufficiently
large was evidenced by the dizziness produced by it.

We here give in tabular form the results of the experiment
just described.

Hour.

P.M.
9h. 05m.

58
lOh. 08m.

52
llh. 00m.

48
12h. 00m.

A. .M.

12h. 10m.

Pulse-rate
per minute.

71.5

72.5

67.5

72
71
70

71
70
72
70
67
69
08
67
68
70
68

Notes.

Subject lay down on bed at 9 P. M.

Drowsy.

«<

Aroused.

45 cc. of water with sugar in solution administered
immediately before.

15 cc. alcohol in 50 cc. of water given.

Complains of slight dizziness.

480 H. NEWELL MARTIN AND LEWIS T STEVENS.

Combining this experiment on a teetotaller with those of pre-
vious workers, we think it tolerably certain that moderate quan-
tities of pure ethyl alcohol so diluted with water as to have no
local irritant action, exert no influence on the pulse- rate of
healthy men. Possibly the contrary result obtained by Dr. Ed-
ward Smith is to be explained by the fact that he was experi-
menting upon himself. Although practised in so doing, he may
not have always been able to suppress such an amount of inter-
est in the result as amounted to a nervous excitement sufficient
to influence his pulse. It is, perhaps, necessary here to defin-
itely state that the above conclusion applies only to ethyl
alcohol, and not to various wines and spirits. As regards
several of these, the evidence collected by Dr. Edward Smith
and others points the other way Some quicken the pulse,
and, so far as diseased persons are concerned, the clinical
evidence seems conclusive that, under certain conditions, some
alcoholic liquids will remarkably diminish the rate of heart-beat.
In the treatment of the sick, however, pure diluted ethyl alcohol
has rarely been used, and it may be that the influence observed
on the pulse-rate is a specific action of some of the other con-
stituents of the liquids administered.

When a substance acts upon so many different systems of the
body as alcohol does, it becomes no easy matter to get at its
immediate specific action upon any one organ ; yet a knowledge
of this may be of primary importance. A given substance, for
example, is known to raise arterial pressure ; perhaps it is often
a matter of no consequence whether it does this by increasing
the heart's work or by constricting the arterioles ; yet obviously
circumstances may arise, e. g. a greatly weakened heart, when
the administration of a drug constricting the arteries would per-
haps temporarily increase arterial pressure, but in so doing throw
so much extra work on the feeble heart as to lead to disastrous
results. To raise therapeutics from empiricism or guesswork it
is essential to know precisely the action of each drug on each organ
in the body, and then its action upon them when working together
in the living man. By the combination of careful observations at
the bedside, with experiments made in physiological laboratories
on the action of substances on healthy animals, and in laborato-
ries of experimental therapeutics on healthy and diseased, we

ETHYL ALCOHOL UPON THE D0GP3 HEART. 481

may hope in time to know, at least with tolerable exactness (for
there will always be individual idiosyncrasies to be met and com-
bated) exactly what any dose given to any patient is going to
effect in him. The educated physician does not now prescribe as
his predecessor would have done, a dose of salts for every case of
constipation ; he selects his purgative to suit the particular case
and in accordance with his diagnosis of the seat of the trouble
and his knowledge of the physiology of the alimentary organs
and the specific action of the drug. To clearly establish for
every substance used in medicine, first its special action upon
each organ when isolated, and then its action upon each organ
when that organ is in vital connection with all the rest, is a task
of almost appalling magnitude ; but in proportion as it is accom-
plished will medicine become a trustworthy art based on scientific
knowledge. Fortunately so much has been done of late years,
especially in physiological and pharmacological laboratories, as
to show that the task is not hopeless.

The investigation whose results are given in the following
pages was undertaken with the hope of contributing some little
to the attainment of the end above described, and also with the
view of testing the availability of the dog's heart, isolated from
all other organs of the body except the lungs, for therapeutical
research. The latter subject seemed well worth investigating, as
the hearts of frogs and reptiles, which have hitherto alone been
experimented upon as regards the direct action of drugs upon the
organ, differ in many fundamental points of anatomy, physiology,
and nervous supply from the heart of man, while the dog's heart
is practically identical with it in structure and working.

The animal having been narcotised by a large dose of acetate
of morphia subcutaneously injected, or by the inhalation of the
vapor of a mixture of ether and chloroform, the heart was
isolated essentially in the manner described in a previous number
of this journal (Vol. II, p. 213, plate XV). Certain modifica-
tions in the method, however, require mention.* Instead of
allowing the right carotid to pump out through the tube q (Plate
XV), and regulating the pressure in the aortic arch by opening

* The modifications here described are so inconsiderable and easily intelligible
that it has not seemed to us necessary to illustrate them by a new plate.

482 H. NEWELL MARTIN AND LEWIS T. STEVENS.

the stop-cock 22 more or less freely, the cannula inserted into the
artery was attached to a long rubber tube which was led through
the top of the warm chamber, in which the heart lay, to a height
of several feet, where it ended in an outflow orifice. By vary-
ing the height of the point of outflow any desired arterial pres-
sure could be easily obtained. We usually chose such a height
as gave a mean pressure of 100 to 140 mm. of mercury, measured
by a manometer connected with the left carotid, which recorded
upon the paper of a kymograph, and thus also enabled us to
count the pulse. Wo may at once dismiss the latter by saying
that the doses of alcohol given by us had no effect upon its rate,
thus confirming the results of the majority of recent observers.

In some cases the method was modified by tying up the right
carotid instead of the aorta, and inserting into the latter a can-
nula of thin braes, as large as it would admit. This cannula
was pushed up to the origin of the left subclavian and firmly
tied there. To its distal end was connected a wide rubber tube,
which led through the top of the warm chamber and ended in
an outflow tube which could be raised or lowered at will. This
modification was adopted to secure to the left ventricle a wide
outflow channel, and thus eliminate a possible source of error
due to its having only one carotid through which to empty itself.
As will be seen subsequently the result was the same whether
the left ventricle had only the carotid through which to force
its contents, or a tube of the full diameter of the thoracic
aorta. This might perhaps have been expected, as the height
to which the column of blood had to be pumped was, in both
cases, arranged with reference to the diameter of the tube through
which it was forced, so as to give about the same pressure in the
aortic arch ; in other words, to oppose the same resistance to the
systole of the left ventricle.

The nutrient liquid sent to the heart was supplied from four
Mariotte's bottles, either of which could at will be connected
with the organ. One of these flasks, at the commencement of
the experiment, contained two litres of fresh defibrinated dog's
blood, mixed with 500 cub. cent of 0.75 per cent, solution of
sodium chloride in distilled water. At the commencement of an
experiment this flask was put in connection with the superior
vena cava, and supplied the right auricle under a pressure equal

ETHYL ALCOHOL UPON THE DOG'S HEART. 483

to that of a column of the blood mixture fifteen centimetres in
height This supply-pressure was the same for all the four flasks,
as they stood on the same level, and, as repeated trials showed,
gave rise, when the cannula usually inserted into the superior
cava was disconnected from that vessel and allowed to pour into
a beaker, to a greater flow of blood than the left ventricle ever
pumped out in an equal time; so that the heart always had
opportunity to take up more blood than it accepted.

The blood received by the right auricle from the first Mar-
iotte's bottle having passed through the lungs, was finally sent
from the left ventriclo through the outflow tube connected either
with the right carotid or with the aorta. From the outflow tube
it poured into a funnel from which it passed back into bottle No.
2, where it collected ; this bottle being meanwhile in free commu-
nication with the atmosphere, but shut off from the heart.
When No. 1 was nearly empty and No. 2 full, by turning a
couple of stop-cocks, No. 2 was cut off from direct connection
with the outer air and converted into a Mariotte's flask, and at
the same time placed in communication with the superior cava.
No. 1 was, simultaneously, cut off from connection with the
heart and arranged to receive the blood pumped out by the left
ventricle and now supplied to the heart by No. 2.

One of us stood by the kymograph and looked after it ; the
other stood by the outflow tube. The former at intervals of a
few minutes gave the word " get ready," and a few seconds after-
wards " go." The other then immediately turned the outflow
tube connected with the left ventricle so that it emptied into a
beaker held in his hand. At the expiry of fifty-five seconds
from the word " go " the warning " get ready " was again given,
and at the end of a minute, upon a second utterance of the word
" go," the collection in the beaker was stopped. The blood col-
lected during this minute was measured and noted ; and soon after-
wards a new measurement of the quantity pumped out by the heart
in a minute made in like manner. When bottle No. 2 was nearly
empty and No. 1 full, the stop-cocks were reversed and the heart
fed from No. 1 ; and so on as often as necessary. The blood col-
lected for measurement was poured back through the funnel into
the bottle which happened to be the receiving one at the moment
When such measurements made five or six consecutive times

484 H. NEWELL MARTIN AND LEWIS T. STEVENS.

agreed within a few cubic centimetres, the heart was considered
fit for the examination of the action on it of alcohol -containing
blood. Bottle No. 3 contained when the experiment commenced
two litres of defibrinated dog's blood. As soon as it was ascer-
tained that the heart was working with fair uniformity, 500 cub.
cent, of 0.75 per cent, warmed sodium chloride solution to which
alcohol had been added were mixed with the contents of No. 3.
The quantity of alcohol used was such as to form either 0.25
or 0.5 per cent, of the whole; or, put in another way, 25
or 50 parts in 10,000. The total quantity of alcohol admin-
istered did not exceed in any case which wo here record (larger
quantities were given in other experiments with marked
pathological results) 10 cubic centimetres, an amount contained
in about § oz. of good brandy. It must, however, be borne in
mind that under the conditions T>f our experiments the only
organs concerned were the lungs and heart, and that when
alcohol is swallowed much of it may be held back in the liver or
eliminated by the kidneys. It is therefore probable that much
larger quantities of alcohol than those we employed might be ad-
ministered by the mouth and absorbed and removed from the whole
body without producing that influence upon the heart which our
experiments demonstrate. When the alcohol-containing Mariotte's
bottle was connected with the heart, the stop-cocks were so turned
that the blood pumped out flowed into bottle No. 4 ; and while
the heart was fed from No. 3, measurements of the blood pumped
out in a minute were made in the manner above described. After
the action of the alcohol had fully manifested itself, a bottle
(No. 1 or 2) containing no alcohol was connected with the heart ;
if no marked recovery took place the experiment was rejected,
as the diminished work might have been due to gradual death
of the isolated heart, independent of any specific action upon it
of the alcohol. When unmistakable recovery took place the
experiment was recorded as a satisfactory one, even though the
heart did not regain completely its original working power.

Care was of course taken to keep the blood supplied to the
heart of as uniform a temperature as possible. Its temperature
was observed by means of a thermometer inserted into the sup-
ply tube close to its attachment to the superior vena cava.

In a preliminary and general way our results may be stated as

ETHYL ALCOHOL UPON THE DOGPS HEART. 485

follows: When dejibrvnated blood containing £ of one per cent,
by volume of ethyl alcohol is supplied to an isolated dog's heart
which has been hitherto working with uniformity r, the invariable
result is a very rapid and marked diminution in the work done
{indicated by the quantity of the blood pumped out from the
left ventricle) by the heart in a given time. When the blood
contains only \ of one per cent of alcohol the result is, in most
cases, the same, but sometimes is little or none. After the action
of the alcohol has been fully manifested the heart can in many
cases be restored to its original working state if supplied with
defibrinated blood containing no alcohol. Blood containing but
one-eighth of one per cent, of alcohol exerts no influence upon
the work done by the heart, at least for several minutes.

As the heart was, under the conditions of the experiment,
isolated from all extrinsic nervous control, and supplied under
exactly the same pressure with blood of exactly the same compo-
sition, except that one sample contained a little alcohol and the
other did not, it was clear that in seeking an explanation of the
above results we were limited to two directions : our apparatus
might be imperfect, or the alcohol had a direct action upon the
living organs, heart or lungs, or both.

As regards the apparatus, it was possible that the bottles filled
with alcoholised blood flowed less freely than the others, and thus
cutting off the supply to the heart, gave it less to pump out.

Repeated and most careful examination quite precluded this
. explanation. In many cases before commencing an experiment
each of the four Mariotte's bottles was in turn connected with
the vena cava cannula and allowed to pour for a minute into a
beaker, with the invariable result that the quantity collected
from each one did not vary four per cent, from that obtained
from any of the other three. Wo had in fact taken such care
to have the connections and stop-cocks of each bottle so similar
that a different result could hardly have been possible. In
other cases bottle 1 was first used to feed the heart ; then alcohol-
ised blood supplied from bottle 3, with the usual result. The
heart was then recovered by good blood supplied from bottle 2,
and meanwhile bottle 1 emptied of good blood and filled with
alcoholised, its connections being left undisturbed. Then alco-
holised blood from bottle 1 being supplied to the heart, we found

486 H. NEWELL MARTIN AND LEWIS T. STEVENS.

invariably a marked diminution of work, although this bottle
had previously, when filled with good (i. c. non-alcoholised)
blood, kept the heart at full work; and it returned to this stand-
ard when subsequently supplied from bottle 3, which meanwhile
had had its contents syphoned off and replaced with good blood.
An absolutely incontrovertible proof that possible different rates
of supply from the bottles had nothing to do with the general
result will appear later when we describe the effect of removal
of the pericardium.

Once defects of the apparatus were eliminated we had to seek
the cause of the result obtained in the heart or lungs. It
seemed conceivable (a) that the alcoholised blood constricted
the pulmonary vessels or otherwise impeded the flow from right
ventricle to left auricle ; or (J) that it greatly dilated the coronary
vessels of the heart and allowed so much blood to be diverted
through them as to seriously diminish the proportion of the total
amount pumped into the root of the aorta, which was left over
to be pumped through the carotid or aortic cannula, with which
our outflow tube was connected; or (c) the alcoholised blood
might act injuriously on the ganglia and nerves of the heart ; or
(d) it might act injuriously upon the cardiac muscular tissue.

We were quite at a loss for a time in endeavoring to decide
between the above possibilities. At last it was observed that
when the heart was supplied with alcoholised blood and this
diminished the work done, the organ invariably was much dis-
tended, closely filling the pericardiac sac. In the latter a
minute hole was always cut as soon as the heart was placed in
the warm chamber, to prevent the accumulation of lymph within
it, which otherwise is apt to occur; probably because the efferent
lymphatic trunks have been tied or twisted in the operations of
isolating the heart and inserting the cannulas. After noticing
the expansion of the heart above mentioned, our next experi-
ment was modified by cutting away the pericardium before any ob-
servations were made. We then found that even blood containing
i of one per cent, of alcohol, which had never previously failed to
cause a marked diminution in the heart's work, was almost with-
out effect on it. In other cases the experiment was modified by first
leaving the pericardium intact and getting the usual alcohol re-
sult ; next, recovering the heart by supplying it with good blood ;

ETHYL ALCOHOL UPON THE D0CP8 HEART. 487

then cutting away the pericardium and supplying alcoholised
blood from the same flask as before. This now had no effect on
the work done by the heart in a minute; though, as will be more
precisely stated later, it had a noticeable influence on the bulk of
the heart.

Removing the pericardium could obviously have no influence
on the rate of supply from our bottles or on the calibre of the
pulmonary arterioles; so those possible causes of the general result
of the alcohol administration were definitely set aside. It also
seemed hardly conceivable that dilatation of the coronary vessels
caused the less outflow from the carotid artery or thoracic
aorta ; for compression of a distended heart by its surrounding
pericardium would oppose such dilatation, and the effect ought
therefore to be most marked after the removal of that sac, which
was exactly the reverse of what we found to occur. That the
contractile force of the heart was not directly affected seemed
also demonstrated by the very slight diminution of work, if any,
which occurred on the administration of alcohol after removal of
the pericardium. We thus seemed driven to seek for some
alteration in the physical condition of the organ which impeded
its action and diminished its work. This alteration was not far
to seek. The great swelling of the heart when under the influ-
ence of alcohol was obvious. At the height of each systole it
nearly filled the pericardiac cavity, and during the diastoles
had little opportunity to dilate and receive a fresh supply of
blood. Hence the quantity pumped out at each beat became
less and less in proportion as the heart swelled. As it seems
tolerably certain that the normal heart-beat is of such character
that, at the end of each systole, the ventricular cavities are en-
tirely emptied and obliterated, we may state our restilts as follows :
The action of alcohol administered in the manner and doses
above described is, without primarily altering the force of heart-
beat, to alter its character, so that the ventricular cavity is not
obliterated at the end of systole, and less so the longer the alcohol
has been administered. At first this incomplete systole is com-
pensated for by a more extensive diastole, so that the difference
between the capacity of the ventricle in complete diastole and
that in complete systole remains the same as when the organ was
normally beating. Consequently, the quantity of blood pumped

488 H. NEWELL MARTIN AND LEWIS T. STEVENS.

out at each beat remains as great as before. If the heart be con-
fined in the pericardium it soon, however, ceases to have room to
swell during diastole to a size sufficient to compensate for its in-
complete systole ; and thenceforth, as the swelling increases, the
difference between diastolic and systolic capacity becomes .less
and less. As the necessary result, the quantity of blood pumped
round by the organ is proportionately diminished. Removal of the
pericardium prevents this result, at least for a considerable time.

Probably the diastolic increase would ultimately, even wjth the
pericardium removed, gain a maximum before the systolic in-
crease of ventricular capacity had reached its limit, if alcohol
were administered a longer time, and there would then be a dimi-
nution in the blood pumped round ; but upon this point we are
not prepared at present to make a positive statement. When
hearts freed from the pericardium showed a distinct diminution
in the work done, we have never been able to obtain any satis-
factory recovery ; and, as above stated, we are unwilling to lay
stress on experiments in which no such recovery was obtained
when good blood was substituted for alcoholised.

Gaskell has shown* that the heart of the frog and toad can
have the extent of its systole or diastole controlled by the vagus
nerve. Hence it may be that the characteristic physical change
wrought in the muscle of the dog's heart by alcohol is indirectly
produced by a primary action of the drug on vagus nerve end-
ings in the organ. Gaskell, himself, however,4 and Roy,5 Ringer*
and others, have found that various substances supplied to the
apex of the frog's ventricle bring about a condition of imperfect
systole similar to that which we find produced in the dog's heart
by alcohol ; while other substances exert the reverse effect, bring-
ing the frog's apex into an almost tetanic state of systole. Hence,
reasoning from analogy, it is also possible that the alcohol acted
directly upon the cardiac muscle. At present we do not find
ourselves in a position to decide between these possibilities.*

* This paper was read before the Medical and Ghirurgical Faculty of Mary-
land on April 27, 1883, and an abstract of it published in the Medical News,
Philadelphia, May 5, 1883. Since the present article was put in type, a paper
by Ringer and Sainsbury has appeared in the Practitioner for June, 1888.
They experimented with various alcohols on the frog's ventricle, and found all
stopped the heart in diastole. Their work makes it probable that our results are
due to direct action of the ethyl alcohol on the muscular tissue of the dog's heart*

ETHYL ALCOHOL UPON THE DOG'S HEART 489

The therapeutical significance, if any, of the results obtained
by us we do not feel qualified to discuss ; but we may point out
that our work seems to show that alcohol should be used with
caution in cases of pericardiac effusion, where any increase
in the size of the organ, hampered as it is already by the
liquid around it, could only be harmful. We trust shortly to
investigate the action of other substances upon the isolated dog's
heart ; especially those substances which have been found to pro-
duce dilatation or contraction in the hearts of amphibia and
reptiles. If we can establish for the mammal the results which
others have obtained on the lower vertebrates, wo may perhaps
add some little to the knowledge available to the physician in
his treatment of the pathological conditions known as dilated
and contracted heart.

We append in tabular form the details of some of our experi-
ments. The only point which we think may need explanation is
the fact that in some cases arterial pressure is seen to fall while the
heart was still pumping some blood up to and out of the outflow
orifice, which was maintained at a uniform height. This is due to
the fact that the pressure recorded by the manometer depended
on two factors: one (the main one), the height of the exit of the
outflow tube above the level of the heart ; the other, an elastic
reaction of the aortic arch and the arterial stumps connected with
it, and of the elastic rubber outflow tube, due to the fact that
when in good working condition the heart kept them all slightly
on the stretch. When the heart pumped less blood this tension
diminished or disappeared, and the pressure in the stump of the
carotid with which the manometer was connected fell accord-
ingly.

The numbers in the column headed " outflow " give the num-
ber of cubic centimetres of blood pumped by the heart through
the outflow tube in the minute ending at the time stated in the
first column. The figures in the column headed " pressure "
indicate millimetres of mercury.

490 H. NEWELL MARTIN AND LEWIS T. STEVENS.

March 12, 1883. Animal under the influence of morphia
during the preliminary operation. Heart isolated at 2h. 05m.
P. M. Outflow through right carotid. Pressure measured in
left carotid.

Time— P. M.

2h. 23m,

47
2h. 47m. 80s.

56
2h. 56m. 15s.
8h.01m.

06
8h. 14m. 00s.

86
8h. 87m. 80s.

48
3h. 48m. 15s.

Pressure.

Outflow.

140

198

188

198

140

197

139

188

141

199

140

190

125

118

122

120

124

103

135

142

134

148

133

166

140

205

142

203

142

199

135

169

138

161

185

168

134

168

Notes.

Alcoholised blood, 0.25 per cent., turned on.

Good blood turned on instead of alcoholised.

Marked recovery.
Pericardium out away.

0.25 per cent, of alcoholised blood turned on.

Pulse slightly irregular.
Good blood turned on.

The heart now became very irregular and was
obviously dying. The experiment, however,
shows well enough the comparatively slight
action of the alcohol after the removal of
the pericardium.

ETHYL ALCOHOL UPON THE DOCPS HEART. 491

April 26, 1883. Very small dog; under morphia while heart
was being isolated. Heart isolated at 2h. 03m. P. M. Outflow
cannula in aorta. Pressure measured in left carotid.

Time— P. M. i Pressure.

2h. 21m. 00s.

30
2h. 31m. 00s.

44
2h. 44m. 20s.

59 !
3h. 00m. 00s. '

02 I

3h. 10m. 15s.

17
3h. 19m. 00s.

24
3h. 24m. 45s.

35
3h. 35m. 15s.

99
99
99
99
99

98
98
98
98
98.5

98.5

96
96
95

Outflow. •

140
142
140
145
145

121
116
100
98
100

129
125
123
122
120
126

28
8
0

100

98.5

121

116

135

133

97.5

120

97.5

110

103

105

137

127

128

Notes.

0.25 per cent, alcohoiised blood turned on.

Good blood turned on.

0.5 per cent, alcohoiised blood turned on.

Pressure rapidly falling as the blood sank in
the outflow tube ; not enough being pumped
out by the left ventricle to supply the coron-
ary arteries.

Good blood turned on ; the exact moment of
turning on the good blood has unfortunately
been omitted in the record of the experi-
ment. It was probably at the time here
stated, but may nave been just before 3h.
09m.

A few drops of blood pumped out of the out-
flow orifice.

Pericardium cut away.

Alcohoiised blood (0.5 per cent.) turned on.

Heart greatly swollen.

Good blood turned on.

This experiment shows well the much greater
effect produced by the blood containing J
of one per cent, of alcohol than that con-
taining J. Also the much less effect of the
alcohol in so far as quantity of blood pumped
around is concerned, after removal of the
pericardium.

492 H. NEWELL MARTIN AND LEWIS T. STEVENS.

May 31st, 1883. Medium sized dog, etherised while the heart
was being isolated. Heart isolated at 3h. 55m. P. M. Outflow
cannula in aorta. Pressure measured in left carotid.

Time.

Pressure.

Outflow.

Notes.

4h. 20m. 00s.

102.5

283

108

285

108 i

278

108

279

4h. 30m. 80s.

0.5 per cent, alcoholised blood turned on.

100

198

109

Heart much distended.

96.5

The pulse waves on the kymograph tracing
have become very feeble.

4h. 88m. 20s.

Good blood turned on.

102.5

264

102

266

102.5

270

4h. 48m. 00s.

A large slit cut in pericardium.

108

288

103

281

102

278

102

277

4h. 57m. 45s.

0.5 per cent, alcoholised blood turned on.

180

5h. 01m. 00s.

175

The poricardium was now completely removed,

101

260

as it was observed that although the ven-

100.5

245

tricles projected through the opening made
in it, the auricles, especially the right, were

100

241

compressed and impeded in their diastole.

5h. 12m. 80s.

Good blood turned on.

108

295

102.5

278

102.5

284

The experiment was stopped here, with the
heart and lungs still in good condition. On
the whole, it is one of the most satisfactory
in our series, as the lungs remained in good

order throughout, instead of becoming cede-

matous towards the end of the experiment,
as they usually do, impeding the blood flow

and more or less vitiating ' the result. This

gradually increasing pulmonary oedema is

one reason why wo have rejected all experi-

ments but those in which the heart showed

decided recovery after the removal of the

alcoholised blood ; it is, also, we feel sure,

mainly responsible for our failure in most

cases to get a complete recovery of the organ

as indicated by the outflow.

ETHYL ALCOHOL UPON THE DOG'S HEART. 493

February 26th, 1883. Animal under morphia while heart was
being isolated. Isolation completed at lh. 05m. P. M. Outflow
cannula in right carotid. Pressure measured in left carotid.

Time- P. M.

Pressure

Outflow.

Sot**

lh. 42m. 00s.

122

204

Perimnlmm removed before the first measure-

131

202

ment of outflow was made.

123

207

lit. 50m. 00b.

0.25 per pent. uliroholUt'd blood turned on.

119

200

117

184

117.5

183

lh. 57m. 00s.

Good blood turned on.

116

183

3h. 18m. 00s.

122

108

Three measureiTu'uls rmidc between lh. 59m.

123

203

and 2h. 13m. were thrown aside as useless.

122

300

■on ■WBBnt of the discovery of a bubble of

211

119

202

pis imprisoned in a Iwnd ot !he supply tube
of the Mariotte's lmttle. This greatly di-
minished the quantity nf blood reaching the
heart. Another bottle having been con-

nected with the heart, the gas was removed

and the experiments continued.

2h. 27m. 00s.

0.23 per cent, ak-oholiscd blood turned on.

118

197

120

198

2h. 33m. 00s.

Good, blooil turned on.

120

210

Mi, 38m. 00?.

0,5 per cent, nleoholised blind t timed on.

120

202

123

305

123

205

31i. 4«m. 00s.

Good blood turned on.

123

200

3h. 61m. 0&.

I pT cent. alcoholized blood turm.d km.

119

195

117

192

117

190

118

191

3h. Win. 00s.

Good blood turned ou.

121

203

123

308

Thouj-hout this experiment the limps kept in
good condition. It shows very well the
slight effect, of alcohol on the quantity of
blood pumped out by the heart when the
pericardium has iieeri removed. Even blood
eoiitaininu 1 percent, of alcohol had very
Jiltle influence in iliiuihi-hiiiirtlie outflow.

494 H. NEWELL MARTIN AND LEWIS T. STEVENS.

REFERENCES.

1. Zimmerberg, Heinrich. Unters. u. d. Einfluss d. Alkohols auf
d. Thdtigkeit des Herzens. Dissert Dorpat, 1869.

2. Smith, Dr. Edward. On the action of foods on the respira-
tion during the primary processes of digestion. Phil. Trans. Vol.
149, p. 731.

3. Gaskell, Dr. W. H. Proc. Roy. Soc December 21, 1881.

4. Gaskell, Dr. W. H. On the tonicity of the heart and blood-
vessels. Journ. of Physiology, Vol. Ill, p. 48.

5. Roy, Dr. G. S. On the influences which modify the work of
the heart. Journ. of Physiology, Vol. I, p. 452.

6. Ringer, Dr. Sydney. Concerning the influence exerted by each
of the constituents of the blood on the contraction of the ventricle.
Journ. of Physiology, Vol. Ill, p. 380.

THE DIRECT INFLUENCE OP GRADUAL VARI-
ATIONS OP TEMPERATURE UPON THE RATE
OP BEAT OP THE DOG'S HEART. By H. NEWELL
MARTIN, M. A., M. D., D. So., Professor in the Johns Hopkins

University.

(Abstract.)*

Iu the investigations described the method of experiment was
such as to completely isolate physiologically the heart of the
dog from all the rest of the body of the animal, lungs excepted.

This was accomplished by occluding the right and left carotid
and subclavian arteries, the aorta just beyond the origin of the
left subclavian, and ligaturing both venae cavse and the azygos
vein. In consequence the only fraction of the systemic circula-
tion left open was that through the coronary system of the
heart : no organ but the heart itself has any blood sent it, except
the lungs. Hence the cerebro-spinal nerve-centres and the sym-
pathetic ganglia very soon die, while the heart Remains alive, in
good working condition, for two hours or more. The right
auricle is supplied uniformly with defibrinated calf s blood, con-
veyed to the superior vena cava from Mariotte flasks. The
blood, after traversing the pulmonary circuit, is finally pumped
by the left ventricle into a cannula, which is tied into the aorta
just beyond the origin of the left subclavian artery. From the
distal end of the cannula a wide rubber tube carries the blood to
an exit cannula seven or eight feet above the level of the heart.
By raising or lowering this exit, and by raising or lowering the
level of the Mariotte flasks feeding the heart, arterial and
venous pressures could be changed at will, or maintained very
nearly constant.

Venous and arterial pressures being kept constant, the tem-
perature of the blood supplied to the heart was gradually
changed by raising or lowering the temperature of the water

* Reprint from Proc. Roy. Soc. No. 223, 1883. This paper will shortly be pub-
lished in full in the Philosophical Transactions, as the Crooiiian Lecture for the
year 1888.

496 H. NEWELL MARTIN.

contained in the vessels in which the feeding Mariotte flasks
were immersed.

The pulse-rate was recorded by a Fick's spring manometer, and
arterial pressure by a Marey's mean-pressure mercury mano-
meter, each being connected with the central stump of a carotid
artery. Temperatures were read by means of a thermometer tied
into the root of the left subclavian, so that its bulb projected into
the aortic arch.

Uniform artificial respiration was maintained.

As the result of many experiments it was found (1) that the
isolated dog's heart beats quicker when supplied with warm blood,
and slower when cold blood is supplied to it ; (2) that the rate of
beat depends much more upon the temperature of the blood in
the coronary arteries than on its temperature in the right auricle
or ventricle ; (3) that when deflbrinated calf's blood is used to
feed the heart, that organ cannot be kept alive as long as when
deflbrinated dog's blood is employed; (4) that no matter how
long an experiment lasts, the deflbrinated blood, circulated again
and again through heart and lungs, shows no tendency to clot ;
hence fibrinogen is not produced in those organs.

The question answered by the first of the above results was
the one for whose solution the research was undertaken. The
experiments show that, in spite of its highly- developed extrinsic
nervous apparatuses, the heart of the mammal does, so far as its
rhythm is concerned, in its own nervo-muscular tissues, respond
to temperature variations within wide limits (42° — 27* CO, just
as the frog's heart or that of the embryo chick does. To account
for the quick pulse of fever we, therefore, need not look beyond
the mammalian heart itself; we require no theoretical assump-
tion of any paralysis of inhibitory, or any excitation of accelera-
tor cardio-extrinsic nerve-centres.

Amblystoma punetatum.

Sm'l F. Ctorte, Oe(

Amblyatoma punctatum.

Sam't F. Clarke, Del.

Amblystoma punctatum.

Sam'l F. Clarke, Del.

->■:

■jv!

■ i

€1

7 5 r

&jji

Sturl.Hinl.LHbAbl.il

Fuji.

<m. ■ •

-.••••«,•# .• *•

••■ * t * • •
•••••••• » • •

*~"-'

- ■ >

•••V • * . ••

.«» ... ,,

-£ .■••'• "

PlaloV.

Fi<$ 2.

».'.»> y

■ ■■* xz

*'/,

V .

il c. b. ii. c y "V,^,

/ " /*♦» VA ,1,(1 V,'.;...

7 "^ ft' ■ ■ ,- 4 *

•'■V N

.'' .«■

lev*

S ..

f { j \H (i.

l'i«4 7. * i • * r^.*'

7 ' • - '

7 ' • . •

Vi 44 ^ t |» f .

(I (V I

"-. -v

;•-

;»--

''V* T

■5"

•i

Vv\.\\.

A ri->r-* ' . ". ,r,r- t«**t" B»l5iiww

Poreellana ocellata.

W. K. Brook* and E B. WiUon, Del.

Segmenting Vertebrate Ovum.

W. K. Brook*, Del.

B. N. Martin, Del.

B. It. Martin, Del.

STUDiESFROM BIOL LAB

W\. 11, - PLATE X

STUDIES FROM BIOL. LAB

VOL. I!, - PLATE X!

STUDIES FROM BIOL. LAB

VOL II, -PLATE XII

STUDIES FROM BIOL. LAB

VOL II, -I LATE XIII

-\rJ

' ■-..,■

STUDIES FROM BIOL. LAB.

VOL II. -PLATE XIV

0»O Ooj.O 0 *»*'?

STUDIES FROM BIOL. LAB.

VOL. II, - PLATE XV.

STUDIES FROM BIOL LAB

VOL. IT, -PLATE XVI.

' 0

= , __^

xzX^ "^

^^\ ~"^\

^bU^^

STUDIES FROM Bid. LAB

VOL. It. -PLATE XV

STUDIES FROM BIOL. LAB.

VOL. U, - PLATE XV.

riJDIF.S FROM BIDL LAB

VOL II. -PLATE XVI.

STUDIES FROM PIOI 1 Ab

VOi II. -RATE XX.

o.ae, m. m.

m jt

7 8. 9

13. I* 15

18 _ 19 ^*^ 2a 21.

STUDIES FROM BIOL LAB

VOL II. -PLATE Xll

.0666

STUDIES FROM BIDL UT?

VOL II. -PLATE XXII

q Q ■& 6: Q

8 <S 0 ® @

« 55 „ 71. 71.

STUDIES FROM BIOL. LAB.

VOL. 11,-PLATE.Xmi.

STUDIF.5 FROM BIOL. LAB

/Ill

N^i

/Ill,

STUDIES PROM BIOL. LAB.

VOL II.- PLATE XXVm,

STUDIES FROM BIOL LAB

VOL II- PLATE XXX

\ I

— . a

•, f i -I... ■••

mj' /

* V'r ->.

1 «*•

.__ — WW , t <

\T t MX'

■■■•"<.; '".■•■

fcVi /•• ../ '>

v- *

j >-.d .«! ."■

Si V

m«f

, V

> ^ ■»

i
i

•..■»•

••"v'-..-i-:iv '

7? • -

•

./'

a \"

.>'-*

i rmikiaKkWAui uthmmp

STUDIES FROM 3I0L LAB.

VOL II -PLATE XXXI

.-.' .» ■•

iS . •'

'. ■ < . .

" \> '

f %

/•..n

: i

■» .. ■_• ■

I I-*

\ I

■ r

1 >■ J
i

1 * *

^ I.-."

« » * i

j *

~'X*\

*■

■i:

i ■ ■ *

■ i

"&&*' J- )

_, **

<# )
■ '-r'

. i

• i
i

/ 1

1 !

L i/ ..
/. i.

\ .'

)

\
\

./'

. lUIIHk.MN^IillVnWlLItt

STUDIES FROM BIOL LAB,

VOL II -PLATE XXXII

a "v

A V

x>' ,

•/'

md / ■ . •

a. «,

%,-

.ix.

•' md

y i

iC
to

tf>

'$*

\ »

iff

.•^

i
^

.- #

f "v

v.v

/.-.

.a>

^T '-4

^Jl ^

y ,r v

•I'l*'

% V \

.2?

c/f

>; .

• /^---""

^,.-

t nuKwunukiJt urtn»*)UT%

i c

I IL

STUDIES FROM BIOL LAB.

VOL II- PLATE XXXIII

. i ;:

i'vi;.i

HJ.

'."")

S*8i^.

'v-

v..

1 \

■V

-*"*to^

-V

/■-#

■ J

-j^^*5^^

f 1

^//7

-7

^T^

' V

V^-^

\ t ■

. V

ft \\

-• *- / -

J ;

,v;

j j -

\ • -
i -

te mp'Y

' V\

v-0
f.

i rMiMmemf At* u-m m

STUDIES FROM BIOL LAB

VOL n- PLATE XXXIV

STUDIES FROM BIOL LAB

VOL II-PLATE XXXV

, «

r ;

STUDIES FROMBTOLLAB.

VD1 II- EL ATE XXXVI

lOOJi MLl lilmlilililihl

STUDIES FROM BIOL LAB.

VOL n- PLATE XXXVII

) )

■>

:» >'

» »

y y>

< - 5> *

J )

)

> i>

> J»

/ >

J >

> »

» j

1 >

J J>

y ■

) >

> i

>
>

> >

» > ' ■

> >

> i ■
i >

J> 1

0 >

* •

•

1) >

% \

> )

)

B»

■ > .

1 ) >

.3*

) j > i

' K 1

,1 ) ) )

ID i 1 )

1 1

1 1 )

) )

> >

> )

) )
) I

. 1) 1 ■>

>
"> 1

•>
>

j
)

)

> )

1) > J )

J> J

1
3

i l

s >

'1 > )

i .) .'

) '

>> ) 5

1> > J )

J >

) D

15 )

>J } )

> ;

. > >

B > ,

> <> J J

a i

i >

» '

) i

"> ) J

> ) )

)

O )

> >>

i >

i »
>

> > >

> 1 J
> > >

3 5

) 1

5 >

)

1 "> )

j ■>

§
>

) >

»5

i >

)■ >
>> >

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "STUDIES FROM THE BIOLOGICAL LABORATORY"

See other formats