COLUMBIA LIBRARIES OFFSITE

HX641 00090
QP121 .C78 The reialKxi ol the

RECAP

GOOT.'BS

Ri^LATI'J.:

n-"

^Ij2d.

CzL

College o{ ^fjpstcians anb g>urgeon«

Hitjrarp

"^r*^^ ^ ff cr/^r/7 v.

THE RELATION OP THE DORSAL ROOTS OF THE
SPINAL NERVES AND THE MESENCEPHALON TO
THE CONTROL OF THE RESPIRATORY MOVEMENTS

BY

HELEN COPELAND COOMBS

A DISSERTATION

Submitted in Partial Fulfillment of the Requirements for the Degree

of Doctor of Philosophy in the Faculty of Pure Science,

Columbia University, New York City

Reprinted from The American Journal of Physiology, Vol. XLVI

July, 1918

THE RELATION OF THE DORSAL ROOTS OF THE
SPINAL NERVES AND THE MESENCEPHALON TO
THE CONTROL OF 'i HE RESPIRATORY MOVEMENTS

BY

HELEN COPELAND COOMBS

A DISSERTATION

Submitted in Partial Fulfillment of the Requirements for the Degree

of Doctor of Philosophy in the Faculty of Pure Science,

Columbia University, New York City

Reprinted from The American Journal of Physiology, Vol. XLVI

•July, 1918

Reprinted from The American Journal of Phtbiology, N'ol. 46, No. 4,

July. l5l8

THE RELATION OF THE DORvSAL ROOTS OF THE SPIXAL
NERVES AND THE MESENCEPHALON TO THE CON-
TROL OF THE RESPIRATORY MOVEMENTS

HELEN C. COOMBS

From the Department of Physiologij, Columbia University

Received for publication May 7, 1918

The general history of the work upon the nervous mechanism of
respiration begins with the experiments of LeGallois and Flourens in
the early part of the nineteenth century. In 1811 LeGallois (1) demon-
strated that after transection below the medulla, all respiratory move-
ments of the body cease with the exception of movements of the mouth,
which cease also after section of the medulla. Flourens (2) amplified
and confirmed this work (1842-1851). His experimental procedures
began by removing the cerebrum, then the cerebellum, then the corpora
quadrigemina of an animal (rabbits and pigeons were mostly used).
Respiration persisted until sections of the medulla were made, then it
failed. Reversing the operation and beginning with the lumbar spinal
cord, making successive sections upward he found that

In destroying the costal spinal cord, the rise and fall of the sides diminished
gradually and when he had finished, had entirely disappeared.

As he continued to make sections upward, respiration was main-
tained, although with difficulty, by the diaphragm until the origin of
the nerves of the diaphragm was reached when with their section and
consequent cessation of the movements of the diaphragm, all effective
respiration stopped, for the yawnings of the mouth and glottis which
alone survived had no effect. He then proceeded in a reverse manner,
removing the medulla by transverse sections from front to rear. The
yawning movements disappeared first; then the dilation of the nostrils;
the inspiratory movements of the trunk alone survived and finally
these failed also. These experiments appear to indicate that the spinal
respiratory nerves are unable of themselves to maintain rhythmic
respiratory movements but are dependent on the action of a central
coordinating mechanism situated somewhere above the lower end of
the medulla oblongata.

459

460 HELEN C. COOMBS

Volkmann (1842), Longet (1847) and SchifY (1858) (3) showed that
the central respirator}- mechanism is a double organ which can be
divided by a median longitudinal section without causing the death
of the animal; and Longet, and more particularly Schiff, endeavored
to show that this central mechanism is located in the nucleus of the
grej' matter in the alae cinereae in the lower part of the bulb, on each
side beneath the floor of the fourth ventricle, and that the paths by
which the impulse is conducted thence to the spinal cord run in the
lateral bundles,

Unilateral section of which at the lower level of the bulb, or at the level of
the second or third cervical vertebrae, suffices to produce respiratory paralysis
on the same side.

In opposition to this view, Brown-Sequard (4) enunciated his doc-
trine of "inhibitory centers." He showed in 1860 that if young ani-
mals were kept alive by artificial respiration for some time after section
of the spinal cord below the medulla, when the artificial respiration
was stopped, coordinated movements of the thorax and diaphragm
might still be observed for a time. He therefore concluded that the
center for respiration was not sharply localized in the medulla but
extended throughout the spinal cord. The effects of section of the
spinal cord below the medulla, he attributed to "inhibition" of these
centers following the lesion of the cord, and he compared the phenomena
with those of spinal shock.

This doctrine, including that of spinal respiration, was later pre-
sented in fuller form by Langendorff (1887) and Wertheimer (5) who
observed that the respiratory muscles of the trunk could sometimes
be made to contract after separation of the cord from the bulb in
animals poisoned with strychnine, in animals with the cord artificially
cooled or subjected to prolonged artificial respiration. Wertheimer
declared that such contractions showed the power of the spinal cord
to originate respiratory impulses.

Such an hypothesis has, however, been too often refuted to be at all
acceptable at the present time. Schiff, in his early exposition of
respiratory hemiplegia, demonstrated that section of the spinal cord
at the level of the second and third cervical vertebrae paralyzes the
respiratory mechanism.

Porter (6), in his study of the innervation of the diaphragm, fol-
lowed a similar line of reasoning. He showed that since hemisections
of the spinal cord above the phrenic nuclei do not inhibit the diaphragm

NERVOUS REGULATION OF RESPIRATION 461

on the same side, it follows that two hemisections altogether separating
the phrenic nuclei from the bulb do not inhibit the diaphragm on their
respective sides. In other words, the arrest of thoracic and diaphrag-
matic breathing in consequence of the separation of the phrenic nuclei
from the bulb is not an inhibition. But one explanation of the arrest
is then possible; the phrenic nuclei effect no respiratory discharge after
their separation from the bulb because they receive no impulses and
cannot originate them. Hence the cells for the discharge of respira-
tory impulses are situated above the calamus scriptorius and not in
the spinal cord.

In addition to this, StarHng (7) has pointed out that cells from which
the nerve fibers go to the respiratory muscles can, like the motor cells
of other parts of the organism, be affected by impulses reaching them
along various paths. Their normal activity in respiration depends
upon impulses from the medulla but they c^n also be affected along
other tracts derived ultimately from the posterior roots, at the same
or higher levels of the cord.

Hering (8) has concluded that after division of all the dorsal roots of
the frog, the motor cells cannot discharge when removed from periph-
eral stimulation, and in the case of respiration, he is convinced that

The normal rhythm of respiration is bound up with the integrity of the ac-
companying centripetal nerves.

Of late years, however, the relative importance of the dorsal roots of
the spinal nerves in the maintenance of respiration has been overlooked ;
perhaps the refutation of the ideas of ^'spinal respiratory centers" had
a discouraging effect, certain it is that little or no mention of the dorsal
roots and their connection with respiration is made in present-day
literature. The best pronouncement with which I am familiar has
been made by Luciani (9) who thus epitomizes their activity:

When the auto-regulation by means of the vagi is suppressed, an abnormal
type of respiratory rhythm appears which, although it provides for a degree of
pulmonary ventilation sufficient to maintain life, must yet be termed dyspneic
since it is not obtained without useless expenditure of energy. Under these
conditions it seems to us probable that a self -regulation comes into play due to
the rythmical and alternate excitation of the sensory paths of the inspiratory
and expiratory muscles.

The question has often arisen as to whether there is a mechanism
for the integration of the respiratory movements higher than the
medulla. The opinions of the various authors who have written upon
this subject appear to be somewhat divided.

462 HELEN C. COOMBS

Starling (7), in citing the work of Rosenthal and Marckwald, states,

In the rabbit, section through the upper part of the medulla oblongata, sepa-
rating the respiratory center from the higher parts of the brain, is equally with
out effect on the depth and rhythm of the respiratory movements. A great
change is observed, however, if the vagi are subsequently divided under these
conditions.

Nikolaides (10) says that in rabbits, isolation of the medulla ob-
longata from above causes almost the same effect as double vagotomy.
Luciani (9) states that

When the brain is extirpated to the level of a plane which passes along the
inferior limit of the pons, or when the section is made at the level of this plane, it
will be seen that after temporary disturbance, the animal continues to breathe
in a regular, perfectly coordinated manner.

H. Newell Martin (11) found that

On stimulation of the mid-brain of the rabbit, close to the iter and beneath
the corpora quadrigemina, there is a respiratory regulating center similar to
that of the corpora bigemina of the frog.

Marckwald (5) found that on blocking off the respiratory center
from the brain above bj the injection of parafl&n into the common
carotid, if these higher paths are cut off, the respiration remains regular,
although deep, and perhaps in the course of time tends to resume its
original type; but if the vagi are also sectioned, the respiration is
entirely changed; periods of rapid breathing alternate with periods of
complete cessation until the animal dies.

From the literature here quoted it will be seen that division of the
vagi, in connection with section at the level of the corpora quadri-
gemina has been considered. The possible relationship of the dorsal
roots of the spinal nerves has, however, been given no attention. I
have therefore performed a series of experiments with a view to deter-
mining whether there is a possibility of such a relationship.

These experiments have extended over more than a year, and include
results upon about forty cats. These animals were first etherized and
then tracheotomized and ether was given by means of a tracheal can-
nula. Tracings of both costal and abdominal respiration were taken
by means of Crile stethographs attached to Verdin recording tambours.
I will consider the results first of section of the dorsal roots of the spinal
nerves alone, then at the level of the posterior corpora quadrigemina
alone and finally the effects of the two operations together.

NERVOUS JiEGULATION OF RESPIRATION 463

After a control tracing of normal respiration (under ether) was taken,
laminectomy was done and the dorsal roots of the spinal nerves were
then sectioned, sometimes in both thoracic and cervical regions and
sometimes in the cervical region only. If the dorsal spinal roots are
cut in the thoracic region alone there is a diminution of costal respira-
tion although abdominal respiration remains unaltered and the rate is
very little changed; if the cervical dorsal roots also are involved, inde-
pendent costal respiration disappears, such costal respiration as is
present being passive and induced by the abdominal respiration as the
tracing of March 2, 1918 (fig. 1) shows. Such respiration is slower
than normal but the general character of the respiratory curve is not
altered. When the dorsal roots are cut in the cervical region alone,
thoracic respiration is not greatly changed. An animal whose dorsal
spinal roots have been divided aseptically may be kept ahve for an
indefinite period. Such an operation, indeed, is analogous to the con-
dition found in some cases of tabes dorsahs in which the functions
necessary for the maintenance of life may be performed adequately
enough although precision of movement is lacking.

It is of interest to observe in connection with the experimental work
the remarkable compensatory power of the individual dorsal roots. If,
for example, in sectioning the dorsal spinr / roots in the thoracic and
cervical regions, a single root on either side be left intact, costal respira-
tion remains much better than the general severity of the operation
and the number of roots cut would lead one to suppose possible. A
study of the nervous system impresses one more and more with its
remarkable adaptive faciHty in the rearrangement of channels for the
conduction of nervous impulses when the normal ones are cut off, and
this is particularly exemplified in the conduct of the dorsal spinal roots
of the thoracic region of the cord.

Following is a protocol of an experiment in which the dorsal roots
were divided.

March 2, 1918. Male cat (fig. 1).

Ether, tracheotomy.

Laminectomy.

Control tracing of respiration taken (part 1),

Dorsal spinal roots cut from third cervical to lower thoracic.

Respiration tracing taken (part 2).

In this experiment, the significant factor is the complete cessation of
an active form of costal respiration, such shght passive movements as

464

HELEN C. COOMBS

are present being the- results of the active diaphragmatic respiration.
The rate, however, is not greatly altered.

It is evident that in the maintenance of respiration a central integrat-
ing mechanism is of first importance. We are well aware of the neces-
sity of the integrity of the respiratory center in the medulla for the
initiation of respiratory movements, but is there no mechanism for the
integration of nervous impulses concerned with respiration higher than

imiumlimiiL

^mpmrn''

mwmvmmf'^jm^]

hWNMwmtmrnwNmmm

-:^,MMmmtmK'"&r''!MiS!

Fig. 1

Part 1. Respiration after laminectomy has been performed.

Part 2. Respiration after section of the dorsal roots of the spinal nerves
from the third cervical to the lower thoracic. Upper tracing represents costal
lower abdominal respiration.

the medulla? In other words, if all portions of the brain above the
medulla were removed, would respiration proceed in the same manner
as before?

In the technique for the operation of section of the brain stem above
the medulla, the carotids were first tied off to prevent excessive hemor-
rhage and the animal was then either decerebrated by removing the
hemispheres from the cranial cavity or a trephine opening was made

NERVOUS REGULATION OF RESPIRATION 465

over the occipital ridge and the corpora quadrigemina were sectioned
through it.

I have found that sections in front of the corpora quadrigemina and
between the anterior and posterior corpora produce no effect upon
respiration; when, however, the section cuts into or behind the pos-
terior corpora quadrigemina, there is a change in the character of the
respiration. It appears to become slower and less regular than the
normal type; still, it is hardly of a gasping character and maintains a
very fair type of ventilation.

The difference in conduct between the results of this operation and
that of complete section of the dorsal roots in the thoracic and cervical
regions is one of degree rather than kind. In the latter case, a few
channels for sensory impulses may remain above and below the sec-
tions— and we have mentioned the compensatory power of the dorsal
roots in this respect — while the conduct in the former case implies a
total lack of these sensory impulses.

In this connection, we cite the protocol of

March 16, 1918. Male cat (fig. 2).

Ether, tracheotomy.

2.00 p.m. Normal respiration (part 1).

2.45 p.m. Carotids tied off, then section behind corpora quadrigemina.

3.30 p.m. Good respiration, see tracing (part 2).

Such respiration as this gives no indication of the dyspnea which
some authors have found and does, to some extent, resemble the slow-
ing obtained after double vagotomy. I have observed at various
times, however, that if after section is made there occurs a hemorrhage
into the fourth ventricle, which causes a clot producing pressure upon
its floor, then dyspnea always occurs. But if no such hemorrhage
occurs, dyspnea is not present except in a very sHght degree.

As I have previously indicated, the spinal cord has not been regarded
as an important factor in respiration, during late years, and even when
the possibility of spinal respiratory centers was under consideration
few authors ever expressed the idea of a relationship between the dorsal
roots and these spinal centers. It has been shown in a previous paper
(15) that the dorsal roots undoubtedly play an important part in the
sensory mechanism of costal respiration and that fibers concerned with
afferent impulses pass up the spinal cord; our present work has served
to confirm these findings and to extend them. Moreover, section of the
brain stem at the level of the posterior corpora quadrigemina produces

466

HELEN C. COOMBS

immediate and lasting effects upon the respiration. Since the mesen-
cephalon contains afferent and efferent fibers from the spinal cord, the
question presents itself as to possible relationships between the dorsal
roots of the intercostals and the mesencephalon as shown by the effect
of section at the level of the posterior corpora quadrigemina and the
dorsal roots of the spinal nerves in the cervical-thoracic region. Fol-
lowing is a protocol of such an experiment.

Fig. 2

Part 1. Normal respiration.

Part 2. Respiration after section behind the posterior corpora quadrigemina.
Upper tracing represents costal, lower abdominal respiration.

July IS, 1917. Female cat (fig. 3).

Ether, tracheotomy, laminectomy (part 1).

2.40 p.m. Carotids tied off.

2.45 p.m. Decerebration.

2.55 p.m. Section behind the corpora quadrigemina (part 2).

3.20 p.m. Dorsal roots in cervical and upper thoracic regions cut (part Z).

NERVOUS REGULu\TION OF RESPIRATIOX

467

From this experiment an interesting phenomenon may be observed,
namely, that after section of the posterior corpora quadrigemina, subse-
quent section of the dorsal roots is followed by no additional effects. Such
a finding leads one to conclude that certain of the sensory impulses at
least, if not all connected with respiration from the dorsal roots of the

Fig. 3
Part 1. Normal respiration.

Part 2. Respiration after section behind the posterior corpora quadrigemina
Part 3, Respiration after section of the dorsal spinal roots in the cervica

and upper thoracic regions. Upper tracing represents costal, lower abdomina

respiration.

intercostals, pass through the posterior corpora quadrigemina since the
difference in severity between the results of the two operations is
about the difference that might be expected between total and partial
elimination of the afferent impulses from the dorsal roots. ^Moreover,
the fact that section of the dorsal roots after section of the corpora
quadrigemina produces no change in the character of respiration shows
that the entire effect was obtained bv the first division.

468 HELEN- C. COOMBS

Section of the dorsal roots before the corpora quadrigemina are
sectioned does leave some additional effect to be gained by the latter
operation, as the following protocol shows.

March 5. 1918. Male cat (fig. 4).

Ether, tracheotomy, laminectomy.

2.45 p.m. Control tracing ipart 1).

3.05 p.m. After section of the dorsal roots in the thoracic and lower cervical
regions (part 2).

3.30 p.m. After section of the posterior corpora quadrigemina. Note the
Cheyne-Stokes respiration (part 3).

From this experiment it is evident that there are still some afferent
intercostal impulses going through until the posterior corpora are
divided — not until then are all intercostal impulses cut off.

Such corroborative detail points strongly to the probabiUty of the
existence at the level of the posterior corpora quadrigemina of some
station closely related to the integration of the afferent impulses from
the respiratory ''cage.''

While certain of the motor impulses concerned in the skilled move-
ments of respiration must originate in the motor areas of the cerebrum
it is hardly likely that these are called into play during normal respi-
ration or during anaesthesia; and on the other hand, a purely medul-
lary type of respiration due to the movements of the diaphragm alone
is not normal either. I beheve, therefore, that sensor}' fibers from the
dorsal roots of the spinal nerves from the intercostals travel up the
brain stem as high as the level of the posterior corpora quadrigemina,
where some connection with the descending motor fibers is effected.
In other words, the dorsal spinal nerves and a region for the integration
of respiratory impulses at the level of the posterior corpora belong to
the same system. The fact that vagi and mesencephalon are unrelated
in this manner is what makes section of the vagi in this connection so
much more fatal than section of the dorsal roots — a relation which
will be discussed in a subsequent paper.

The time element concerned in section of the dorsal roots and the
corpora quadrigemina may also be considered. It is well known from
chnical e^-idence that people in whom accident or disease has destroyed
the dorsal spinal nerve roots are able to support life verj- adequately.
Stewart (12) has cited the case of a man in whom all the ribs became
completely immovable from disease of the spine in the lower cer\'ical
region. He was able to lead an active life and cany on his business
although he breathed entireh' by means of the diaphragm and abdom-

NERVOUS REGULATION OF RESPIRATION

469

inal muscles. Whether an animal in which both the dorsal roots and
the corpora quadrigemina were destroyed could maintain life very
long, I am not prepared to state; but experimentally, under anaesthesia,
good respiration may be maintained for several hours subsequent to
these operations.

Fig. 4

Part 1. Respiration after laminectomy.

Part 2. Respiration after section of the dorsal roots in the thoracic and lower
cervical regions.

Part 3. Respiration after section behind the posterior corpora quadrigemina.
Note the Cheyne-Stokes respiration. Upper tracing represents costal, lower
abdominal respiration.

Sherrington (13), in his work on decerebrate rigidity, describes the
persistent tonic spasm which occurs in certain groups of muscles after
section of the brain stem in front of the corpora quadrigemina. The
groups of muscles which are contracted are the retractors of the head

470 HELEX C. COOMBS

and neck, the muscles of the tail, the extensors of the elbow, knee,
shoulder and ankle — the antigravity muscles. The spasm depends
on the integrity of the dorsal spinal roots and appears not at all, or
only imperfectly, in the limbs of which the corresponding dorsal nerve
roots are divided. Section at that level of the corpora quadrigemina
also does away with decerebrate rigidity, a fact which offers further
confirmation that certain fibers of the dorsal spinal nerve roots have
end stations at this level.

coxcLrsioxs

In summarizing the effects upon the respiratory movements of
section of the dorsal roots of the spinal nerves and at the level of the
posterior corpora quadrigemina, my findings are:

1. Section of the dorsal roots of the thoracic and cervical spinal
nerves results in a diminution or cessation of active costal respiration.
The effect of section of both thoracic and cervical nerves is a more
marked diminution of costal respiration than after section of the
thoracic roots alone. After section of the thoracic roots, abdominal
respiration remains unchanged and there is no marked alteration in
the respirator^' rate.

2. Section of the brain stem below the anterior corpora quadrigemina
results in a slower, deeper form of respiration than normal somewhat
similar to the most severe effects which follow double vagotomy.
Abdominal respiration is more prominent than costal.

3. Section of the dorsal roots of the spinal nerves after section into
or behind the posterior corpora ciuadrigemina produces no more severe
effect than section of the posterior corpora quadrigemina alone.

4. Section of the posterior corpora quadrigemina subsequent to
section of the dorsal roots of the spinal nerves produces an effect on
respiration somewhat greater than when the dorsal roots alone are
sectioned.

o. The general relationship of afferent to efferent spinal nerve roots
which Sherrington (14) describes obtains also in the afferent and
efferent intercostal roots.

I wish to express my thanks to Professor F. H. Pike of this department
for his valuable suggestions and criticisms of this work.

NERVOUS REGULATION OF RESPIRATION 471

BIBLIOGRAPHY

(1) LeGallois: Experiments sur la principe de la vie, 160. 1824.

(2) Flouren's: Recherches experiment ales sur les proprieees et les fonctions
du systeme nerveux, 2d ed., 173, 1848.

(3) Rosenthal: Hermann's Handb. Phj'siol., iv, Leipzig. 1882.

(4) Browx-Sequard: Journ. d. Physiol., 1858, i.

(5) Marckwald: Zeitschr. f. Biol., 1887.

(6) Porter: Journ. Physiol., xvii, 455.

(7) Starlixg: Schiifer's Textbook of physiology, ii, 288.

(8) Herixg: Arch. f. Physiol., 1893, Iviii, 614.

(9) LuciAXi: Human physiology, i.

LO) NiKOLAiDEs: Arch. f. Phj'siol., 1905, 465.

11) Martix: Journ. Physiol., i, 370.

12) Stewart: Manual of physiology, 6th ed., 223.

L3) Sherrixgtox: The integrative action of the nervous system.
L4) Sherrixgtox: Schiifer's Textbook of physiology, ii, 802.
Pike axd Coombs: This Journal, 1917, xlix, 395.

VITA

Helen Copeland Coombs was born in St. Joseph, Missouri, July 25,
1891. She graduated from the Yonkers High School and entered
Barnard College in 1907. She received the degree of Bachelor of
Arts from Barnard College in 1911; the degree of Bachelor of Science
from Teachers College in 1914; and of Master of Arts from Columbia
University in 1915. Since that time she has carried on research work
in experimental physiology in the Department of Physiology of
Columbia University. She is now Research Assistant in that
department.

Her publications are:

The Postural Activity of the Abdominal Muscles of the Cat. By F. H. Pike

and Helen C. Coombs. American J. Physiol., vol. xlii, p. 395, 1917.
The Relation of Low Blood Pressure to a Fatal Termination in Traumatic Shock.

By F. H. Pike and Helen C. Coombs. Journ. Am. Med. Assoc, vol.

Ixviii, p. 1892, 1917.
The Role of Afferent Impulses in the Control of Respiratory Movements. By

Helen C. Coombs and F. H. Pike. Proc. of the Soc. for Exp. Biol, and

Medicine., vol. xv, p. 55, 1918.

THE WAVERUY PRESS

BALTIMORS. U. ■. A.

Date

Due

j 1

1

I

1 '

1

' i

1

j 1

1

______'_ i 1

: i

' ■

!

! 1
1