THE MECHANISM OF VOMITING

ROBERT A. HATCHER

Department of Pharmacology of Ctwnell Vniversity Medical College,

New York City

1. THE RGLE OF THE STOMACH. Magnus (35) in 1903, reviewed

the literature relating to vomiting but many of his conclusions are
no longer tenable and many papers that have appeared since then
show the influence of prevalent errors. It is necessary, therefore,
to reconsider some of the older literature, and some of the more
recent may be disregarded or treated briefly.
The starfish (asterias) empties the stomach through the mouth,
the only opening into the stomach, and thus rids itself of useless,
inert and indigestible material (52); the frog is capable of vomiting
by the action of the stomach alone without aid from other structures
(37) ; man vomits through the coordination of numerous reflexes.
There are many gradations from the process in asterias to the type-
of vomiting usually seen in man, and much of the misunderstanding
concerning the actions of emetics and the mechanism of vomiting
has resulted from a too broad application of the results observed
in a single species or in a few closely related species.
The mechanism of vomiting varies somewhat with the character
of the gastric contents. Fluid may be expelled from the mammalian
stomach by a primitive mechanism, but the expulsion of a pasty
mass requires the aid of the accessory muscles. Obviously, the
mechanism of vomiting varies with experimental conditions, and
no single experiment suffices to explain the role of the various structures
concerned. The present discussion is concerned mainly with the
mechanism of vomiting in the cat, dog and man.
Cannon (3) states that Wepfer (56) in 1679, observed the closure
of the pylorus and contraction of the pyloric part of the stomach
during vomiting in cats, dogs and wolves but the question whether
the stomach participates in the act of vomiting, and if so, to what
degree, has been the subject of continued dispute. Magendie (34)
contributed to the confusion on this subject by his celebrated experi-
479

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

480 ROBERT A. HATCHER

ment in which he replaced a dog’s stomach with a pig’s bladder after

which he induced vomiting. That experiment throws little light on
the mechanism of vomiting because the cardia was removed with the
stomach.
One of the members of the committee (Cuvier, Pine1 and Humboldt)
that reported on Magendie’s memoir stated that he had seen a soldier
vomit after partial recovery from an injury in which a bullet had
severed the abdominal muscles so that they could take no part in
vomiting at a time when the stomach was in full view. Doctor
Gold in this laboratory (results not published) caused vomiting
in a cat in which the abdominal muscles had been severed longitudinally
and transversely, so that neither they nor the diaphragm could
compress the stomach. Vomiting was difficult, it did not empty
the stomach completely, and it did not occur in every instance. It
is evident that vomiting may occur under experimental conditions
in a different way from that in which it occurs under normal conditions.
Mellinger’s work is frequently mentioned in the literature, and
few have contributed more to our knowledge of the comparative
mechanism of vomiting in different species than has this student in
Hermann’s laboratory; therefore, his paper must be drawn upon
in the preparation of this review, because many subsequent investi-
gators have failed to consider certain facts that he states clearly.
Mellinger observed the stomach during vomiting in the frog, and
saw increasing antiperistalsis from the pylorus toward the cardia
with the expulsion of the vomitus from the mouth. The entire
expulsive force came from the stomach in experiments in which
the abdominal muscles were excised, and emesis was induced by
tartar emetic. Apomorphin induced gastric antiperistalsis, as
did a subcutaneous injection of tartar emetic. There was no local
action in the latter experiment, and Mellinger concluded that gastric
antiperistalsis is a function of the vomiting act in the frog.
Schueta (48) injected various emetics subcutaneously or intra-
venously, excised the stomach, and transferred it to a moist chamber.
He observed active and atypical movements of the stomach after
the administration of emetin, tartar emetic and apomorphin. The
present writer believes that these results throw no light on the mecha-
nism of emesis in mammals, at least not in the case of those which
act elsewhere than on the stomach, and they are mentioned here
merely because Magnus attached importance to them in his review
of the literature of vomiting.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 481

Openchowski (45) says that under the action of emetics (copper

sulphate and apomorphin) the normal gastric peristalsis is altered
so that with marked unrest of the intestine, the pylorus is closed,
the portion near the cardia remains at rest and expands like an in-
verted pear, after which contractions begin in the antrum and in-
crease in force, driving the gastric contents into the dilated fundus,
from which they are forced into the esophagus with the aid of
the pressure of the abdominal muscles and the diaphragm. He
believed that the dilatation of the fundus inaugurated the contraction
of the pylorus and the lower part of the stomach.
Mellinger states that the exposed empty stomach of a dog which
had been given an emetic showed antiperistaltic movements if the
stomach had been previously quiescent, but that if the stomach
had shown peristalsis this was usually increased in force.
Cannon (3) observed the movements of the stomach after the
hypodermic administration of apomorphin to cats which had re-
ceived food containing bismuth. He says the circular muscles
of the upper part became flaccid so that the slightest movements
of the abdomen changed the shape of the fundus, irregular twitchings
of the fundus wall occurred and soon a deep constriction started
about 3 cm. below the cardia and moved toward the pylorus with
growing intensity. It held fast at the transverse band and a wave
swept over the antrum. Another wave followed and the transverse
band relaxed slightly, but tightened with the next. Soon a firm
contraction, beginning at the antrurn, divided the stomach cavity,
and while waves passed over the antrum the fundus relaxed. There
were then a flattening of the diaphragm, a quick jerk of the abdominal
muscles, accompanied by opening of the cardia, and the gastric
contents were forced from the fundus through the cardia into the
esophagus. As spasmodic contractions of the abdominal muscles
were repeated, the gastric walls tightened about the contents. Cannon
saw anti-peristalsis in the stomach only once, then while the cat
was retching a constriction started at the pylorus and ran back
over the antrum, obliterating the antral cavity completely. Nearly
similar observations were made by Roux and Balthazard (47) on
dogs.
Hesse (22) observed the vomiting act in young dogs that had
received food mixed with bismuth. He says that the fundus dilated
and remained flaccid; the pyloric part then contracted, probably
in toto, whereupon the cardia dilated regularly. All other move-

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

482 ROBERT A. HATCHER

ments, such as the peristaltic waves from the middle toward the
pylorus , were inconstant, and antiperistalsis did not occur. The
food collected in the dilated fundus at the beginning of nausea,
about one minute before the dog actually vomited in at least one
case.
This question of antiperistalsis in the stomach during vomit-
ing has been the subject of controversy. There is no reason to
doubt that it occurs in primitive forms, as in the frog, in which that
organ plays the dominant role in emesis, and that it is usually absent
in vomiting in the higher animals in which the stomach usually
plays an important role, but not the dominant one. It is well known,
of course, that various highly developed functions are occasionally
performed in a primitive way, and it is not astonishing that the
higher animals sometimes employ the more primitive mechanism,
such as that seen in the frog. It seems probable that what we term
regurgitation in the case of the over-distended infant’s stomach is
a partial reversion to primitive methods, but it certainly partakes
of the nature of vomiting, for if the glottis were not closed and respira-
tion inhibited reflexly, the vomitus would be aspirated by the infant.
Levy-Dorn and Muehlfelder (32) observed patients with the
Roentgen ray during vomiting. They state that their observations
do not support the description of emesis given in the leading cur-
rent textbooks of physiology, several of which they quote. Follow-
ing these observations on patients, Muehlfelder, who was able to
vomit at will, swallowed a mixture of bismuth and syrup after fast-
ing. Observations during vomiting were made from behind and
from right to left by Levy-Dorn. During emesis the lower pole
of the stomach moved slightly up and down, and later to a greater
extent, equal to the width of several fingers. They state that the
stomach normally rises with each contraction of the abdominal
muscles, but it does not recede from the abdominal wall as it does in
emesis.
The lifting of the lower pole of the stomach by the contraction
of the abdominal muscles facilitates the twisting of the lower part
toward the front, and violent twisting may lead to volvulus of the
part between the cardia and the pylorus. It is probable that the
gastric movements in the voluntary vomiting of liquid, described
by Levy-Dorn and Muehlfelder, are much like those in reflex vomiting,
but we lack positive knowledge on that question.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 483

v. Czyhlarz and Selka (9) observed patients by means of the

Roentgen ray during vomiting, after the administration of bismuth
Svnchronous with the contraction of the abdominal muscles, the
shadow of the stomach rose, the antrum being fully contracted,
the cardiac part of the stomach being of normal width. They were
uncertain whether antiperistalsis occurred in the fundus. The
gastric movements ceased with the rhythmic contractions, suggesting
that the apparent antiperistalsis was only a passive movement.
The antrum remained fully contracted during vomiting.
It is commonly said that vomiting is attended with danger of
rupture of the stomach after a corrosive poison has been swallowed.
The only part of the stomach that contracts firmly, the pyloric, is
empty during vomiting, and it is obviously impossible to rupture a
stomach through any pressure exerted by the accessory muscles.
2. THE CARDIA AND ESOPHAGUS. The cardia is developed in
widely different degrees in different species, and its behavior during
vomiting has been the subject of dispute. As late as 1922 Carlson,
Boyd and Pearcy (5) stated that, while the literature concerning
the efferent nerve paths in the vagi to the cardia and the lower part
of the esophagus is fairly consistent, that relating to the sympathetic
innervation of these structures is contradictory.
Tantini (50) found that the cardia failed to dilate when he re-
placed the greater part of a dog’s stomach with a pig’s bladder which
was attached to the fundus, while the cardia and the adjacent part,
of the stomach were retained.
Openchowski (45) described ganglia in the serous coat of the
cardia which control its automaticity, and which are in connection
with a center in the region of the quadrigeminate bodies for contracting
the cardia, and with various centers in the cord and cerebrum for
inhibiting it. He administered apomorphin to a dog, in which the
thoracic aorta had been tied, and observed active vomiting move-
ments without actual emesis. The abdominal muscles compressed
the stomach, but the cardia failed to open, probably because the nec-
essary reflex from the stomach was then lacking.
Since the cardia must open during emesis, Openchowski sought
to explain the action of emetics largely with reference to these facts.
He supposed that apomorphin stimulates the dilator, and depresses
the constrictor fibers of the cardia. He found that the cardia opened
reflexly by irritation of the kidneys, bladder, uterus and intestines,
from all of which vomiting can be induced reflexly. We must ac-

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

484 ROBERT A, HATCHER

cept Openchowski’s observations on the innervation of the cardia

and the effects of many reflexes and, while it is possible that enormous
doses of apomorphin may have the actions which Openchowski
supposed, it is quite certain that apomorphin is capable of inducing
emesis without any direct action on any peripheral structure.
Cannon (4) found that the cardia of the cat remained in tonic
contraction for many hours following section of the vagi, so that
food then stagnated in the esophagus. These observations are
in harmony with those of earlier observers, including Krehl (28), whose
results are sometimes cited as being opposed to those just stated. Krehl
found the cardia of the dog patulous at death, some days after vagot-
omy, and that a sound could be passed up through a gastric fistula
into the esophagus without resistance, but he does not state how long
after vagotomy this was observed, and he says that food stagnated in the
esophagus or was regurgitated at first. Moss0 (43) easily drew
a wooden ball through the cardia by means of a thread. These
observations are in harmony with the observation of Carlson, Boyd
and Pearcy that the cardia may be inhibited reflexly from the stomach.
Carlson and Luckhardt state that Sinnhuber found the cardia
atonic when the vagi were cut just above the diaphragm, but the
tonicity was increased if the vagi were cut in the neck. I found
it impossible to pass a soft rubber catheter through the esophagus
into the stomach in a dog immediately after the vagi had been cut
in the neck, but Shaklee (personal communication) says he experi-
enced no such difficulty.
Carlson, Boyd and Pearcy confirmed Openchowski’s observation
on the automaticity of the cardia and extended his studies on the
innervation of the esophagus, cardia and stomach. They observed
both motor and inhibitory fibers in the vagus and splanchnic for
the cardia and lower part of the esophagus in the cat, but they were
unable to demonstrate inhibitory fibers for the cardia and esophagus in
the splanchnic in the dog. They find that reflex inhibition or con-
traction, dependent on the preexisting state of the cardia and lower
esophagus, can be initiated reflexly by the stimulation of any sensory
nerve, skeletal or visceral, in the decerebrated cat and dog, and those
under light anesthesia, but that the nose, mouth and alimentary
tract are in the closest reflex relation with the cardia.
Valenti and Hesse have measured the pressure in the esophagus
required to force liquid through the cardia into the stomach, and
Magendie observed that great pressure on the fundus of the dog’s

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 485

stomach is required to force liquids through the cardia into the

esophagus, but these observations have no obvious bearing on the
mechanism of normal vomiting.
Carlson, Boyd and Pearcy (5) say: “During the experiments on
dogs and cats, vomiting movements may also force the cardia guard
into the espohagus, because of the excessive abdominal pressure,
with dilatation of the cardia.” The cardia opens reflexly during
vomiting and there is no evidence that its contraction is overcome
by the excessive force exerted by the abdominal muscles and diaphragm
upon the stomach during ordinary emesis. Violent straining at
stool, with compression by the hands, does not “force ” the cardia,
even when the stomach is filled with food or liquid. On the other
hand, after the abdominal muscles have been cut, the relatively
feeble contractions of the cat’s stomach alone may suffice during
vomiting to empty part of the stomach contents into the esophagus,
since the cardia is then dilated. Hatcher and Weiss (20) saw two
cats, in which the vagi had been cut, vomit white mucus from the
esophagus alone, while the stomach contained blue fluid, showing
that the cardia was not forced open. Both vomited the gastric
contents later, when the cardia had dilated.
Hesse states that active dilatation of the cardia occurs before
vomiting, and this is in harmony with the observations of numerous
investigators. Hesse observed active dilatation of the cardia by
means of the Roentgen ray before vomiting. He stated that the
cardia remained open during the entire time that vomiting movements
were in progress in one instance, permitting the food to return to the
stomach.
Valenti (53) suggested that waves induced by muscle contraction
in the first part of the vomiting act pass up the esophagus and set
up the reflex in the upper part of the esophagus and pharynx for
opening the cardia. He found that the application of cocain to
the vagus, the glosso-pharyngeal trunk, or the pharynx and upper
part of the esophagus in the dog paralyzes the inhibitory mechanism
of the cardia and increases its tone. He concluded that for the
development of complete emesis the cardia must open through reflex
inhibition due to stimulation of a circumscribed area lying between
the pharynx and the uppermost part of the esophagus; that the
afferent path of the reflex is through the glosso-pharyngeus and
vagus, and the efferent path is through the vagus mainly, partly
through the ninth nerve.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

486 ROBERT A. HATCHER

The existence of this reflex has been confirmed by Carlson, Boyd

and Pearcy (6), but they found that the intravenous injection of 1
mgm. of cocain in cats and 1 or 2 mgm. in dogs weighing 15 kgm.
caused hypertonus of the cardia and they say it is evident that this
hypertonus following concainization of the pharynx is not due solely
to the elimination of afferent impulses by way of the glosso-pharyngeal
nerve, since cocain also stimulates the motor apparatus. It is a matter
of daily observation that moderate irritation of the throat induces
vomiting, but Magnus (36) disproved Valenti’s contention that this
reflex is essential for vomiting. Magnus tied a cannula into the lower
part of the esophagus of each of two decerebrated dogs, after rem .oval
of all of the upper part of the esophagus, pharynx, soft palate and
tongue, leaving the mucous membrane of the hard palate and naso-
pharynx, after which the administration of apomorphin caused emesis
promptly.
If a reciprocal relation exists between the cardia and the pylorus,
it is in abeyance during vomiting, since the pylorus is closed, but the
opening and closing of the cardia depend on the stage of the vomiting
act. Rossbach (46) states that cutting the vagi in the dog causes the
pylorus to open even when the stomach is filled with food. This,
obviously, does not occur during vomiting, however, nor does it occur
in the cat in which the vagi have been cut when the stomach contains
an emetic dose of solution of copper sulphate (20). The fact that the
cardia may fail to open for some time after vomiting movements have
been inaugurated in an animal in which the vagi have been cut seems
to suggest that impulses pass by the vagi from the centers for inhibiting
the cardia and that cutting the vagi disturbs the reflex for a time. It
seems probable, however, that the same mechanism that causes the
fundus to dilate during vomiting also causes the cardia to relax, since
Carlson, Boyd and Pearcy found that the cardia may be inhibited by
the stimulation of nerves in the mucosa of the stomach. The question
requires further investigation.
The participation of the esophagus in the act of vomiting has been
the subject of many investigations. Mellinger argued that the strong
inspiratory movements with the glottis closed that occur during retch-
ing cause a negative pressure in the chest and esophagus thereby
aiding in the passage of the gastric contents into the esophagus, and
v. Mikulicz (38) observed a negative pressure in the esophagus, equal
to -20 cm. of water during forced inspiration.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF’ VOMITING 487

Mellinger observed the exposed esophagus of the dog during vomiting;

he says that bubbles of gas were carried to and fro in the esophagus
before vomiting occurred, but there was no antiperistalsis. When
vomiting movements were induced by the intravenous injection of an
emetic in a dog with an empty stomach, the esophagus dilated during
vomiting and contracted to one-half its previous diameter in the in-
tervals between the paroxysms. When vomiting was induced in
dogs with food in the stomach, the esophagus was so stretched that
it seemed that it would burst. With the expulsion of the vomitus
through the mouth, the esophagus contracted in toto like the urinary
bladder during urination, and not with any antiperistaltic movement.
A small amount of the mass remained in the esophagus after vomit-
ing, and this was always carried to the stomach by peristalsis. Mel-
linger states that vomiting would be impossible without the partici-
pation of the esophagus when the stomach contains only a small amount
of food. Valenti (54) inserted a glass tube to various depths in the
esophagus in the dog and observed that it did not interfere with vomit-
ing. This may seem to imply that the esophagus does not actively
participate in the vomiting act, but the esophagus could dilate about
the tube, and the experiment, therefore, throws no light on the function
of the esophagus in normal vomiting.
Eggleston and Hatcher (10) saw typical vomiting movements with
the expulsion of a considerable volume of frothy mucus from the
esophagus, after the administration of apomorphin to dogs, following
the extirpation of the stomach and intestines, with the cardia tied; and
Hatcher and Weiss (20) saw cats vomit mucus from the esophagus
alone, after vagotomy.
Hesse said as late at 1913, that little is known concerning the condi-
tion of the esophagus in vomiting. He states that the rapidity of
the vomiting movements did not permit him to observe the movements
of the stomach and esophagus with the necessary sharpness. He,
therefo re, took serial pictures at intervals of half a second by means of
apparatus which was constructed in the Veifa-Werke laboratory for
Roentgen cinematographic work. His description of this stage of
vomiting, based on serial pictures, confirms that of Mellinger essentially,
but he was unable to state whether antiperistalsis occurred or the
esophagus contracted as Mellinger stated. Occasionally food passed
at once from the stomach out through the mouth, but the food mass
usually remained in the esophagus for variable periods in different
experiments, during which time the cardia sometimes remained open.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

488 ROBERT A. HATCHER

The food was carried up and down in the dilated esophagus as many
as eleven times during retching, and was expelled with sudden expira-
tion and closed glottis. Sometimes a part or all of the food returned
from the esophagus to the stomach instead of being vomited out of
the mouth. The esophagus did not expel the entire mass through
the mouth during vomiting and the residue returned to the stomach
after varying intervals of time, or was vomited during the succeeding
paroxysm.
Buch (2) reviewed the literature relating to the mechanism of vomit-
ing and described the vomiting act. He concluded that anti-peri-
stalsis in the esophagus is precluded by the rapidity with which the
vomitus passes through the cardia and out the mouth. This is proba-
bly true in the case of projectile vomiting, but vomitus does not always
pass from the stomach out the mouth so rapidly.
Cats and dogs vomit with difficulty or not at all while lying on the
back, and the failure to recognize this fact has lead to erroneous con-
clusions in many experiments. Hesse states that the difficulty is
concerned only with the expulsion of the mass from the esophagus,
since the food passes into the esophagus as usual while the animal lies
on its back, but he offers no explanation of this phenomenon.
This distribution of the vagus has been studied by L. R. Mueller (44) ;
the literature relating to the innervation and reflex control of the
cardia and the esophagus has been reviewed recently by Carlson, and
Luckhardt (7) and by Carlson, Boyd and Pearcy (5), (6).
3. THE ACCESSORY MUSCLES. It requires no special training or
knowledge of physiology to perceive that the abdominal muscles take
an active part in the vomiting in man, the cat and the dog, and, on the
other hand, rodents, which are incapable of vomiting, have nearly
similar abdominal muscles. Mellinger observed that fish vomit by
gastric contraction alone, since they have no diaphragm, and the
abdominal muscles because of their arrangement exert no influence
on the stomach during vomiting. It is unnecessary to consider further
the question of the participation of the diaphragm and the abdominal
muscles in the act of vomiting, but plain facts have been disregarded
in many of the discussions of this subject.
The esophagus in the cat is about 20 cm. in length, that in man
about 30, and considerable force is required to propel the vomitus-
especially when it consists of large pieces of food or a pasty mass-
through the esophagus rapidly. The cardia remains open in some
cases and the accessory muscles assist in emptying the esophagus by

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 489

-their pressure on the stomach, but it is probable that they do not

usually assist directly in expelling the vomitus from the esophagus in
this way, though they do assist indirectly by forcing the stomach up
against the diaphragm and thus diminishing the capacity of the thoracic
cavity. Moritz (42) found that the abdominal muscles in man exerted
a, pressure upon the stomach equal to that of a column of water 3 m.
in height.
Levy-Darn and Muehlfelder (32) state that the diaphragm con-
tracted at the beginning of emesis, but that it probably passed into
the expiratory position just before the vomitus was expelled from the
mouth. The observations were interpreted as indicating active parti-
cipation of the stomach, diaphragm and abdominal muscles in the
vomiting act, but they do not support the view that the diaphragm
is rigidly fixed in the inspiratory position throughout the whole period
of vomiting. v. Czyhlarz and Selka state that the diaphragm moved
up and down rapidly during vomiting in patients, but not to the extent
it does in deep respiration.
Miller (N), (40) studied the vomiting movements in cats under
highly artificial conditions. He says the muscles of respiration are
those which are concerned with vomiting, hence there is alwavs great
ventilation of the lungs during vomiting movements. This statement
is opposed to the commonly accepted view. His papers cannot be
discussed satisfactorily in a review such as the present one. Hesse
also described the retching as respiration. During the course of an
investigation which is now in progress in this laboratory we found no
lung ventilation during the vomiting movements induced by apornor-
phin in a dog in the natural position. The glottis closed after moderate
expiration, apparently.
The cat or dog usually vomits from an overfilled stomach with ease
after a few retching movements; later, when the stomach is nea*rly
empty, the paroxysms of vomiting are more prolonged and violent,
apparently because of the difficulty in compressing the stomach ef-
fectively. In a case in which a cat’s stomach was widely distended
with solid and liquid, the vomitus began to issue from the mouth after
the third retching movement, and continued to issue for two seconds
after the abdominal muscles had become fixed following the fifth
retching movement. Subsequently, when the stomach was nearly
empty, this cat retched twenty-five times and expelled very little
vomitus.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

490 ROBERT A. HATCHER

Brooks and Luckhardt (1) describe two types of vomiting, the pro-
jectile, in which the gastro-intestinal tract is mainly concerned, and
the slow labored type, “apparently, more of the skeletal muscles taking
part in the process which consists of prolonged violent retching move-
ments.” It seems probable, however, that the projectile vomiting
occurs when the contractions of the diaphragm and the abdominal
muscles suffice to drive liquid contents of the stomach [Link] through
the mouth. Obviously, this cannot occur with a paste-like gastric
content, and, on the other hand, liquid may be vomited only after
violent retching movements.
There has been no recent investigation, apparently, which adds
anything essential to our knowledge of the role of the accessory muscles
(those concerned with forced respiration) in the vomiting act.
4. THE VOMITING CENTER. .Giannuzzi (11) was the first to make
serious efforts to determine what part of the central nervous system
is concerned with vomiting induced by tartar emetic. Grimm (12)
administered tartar emetic to dogs and conducted artificial respiration
for the purpose of inducing apnea. Vomiting was prevented, but he
was unable to induce apnea. He concluded that tartar emetic acts
on the vomiting center and that it is identical with the respiratory
center. Harnack (15) disproved the latter contention, and the con-
clusions of Giannuzzi and those of Grimm now have only historical
interest, but they stimulated further investigation.
Openchowski (45) and his pupil, Hlasko (23), state that there are
independent centers for various acts concerned with vomiting, with
separate paths from each center. For example, Openchowski states
that there are centers for the contraction of the cardia in the region
of the posterior quadrigeminate bodies only, from which fibers run
mainly in the left vagus, partly in the right, while a few pass by way
of the splanchnics, and that there are centers for inhibiting the cardia
in the cord, in the sulcus cruciatus, and where the anterior inferior
end of the caudate nucleus unites with the lenticular nucleus, from
which fibers run by way of the vagus. He was unable to induce
vomiting by apomorphin after destruction of the quadrigeminate
bodies, but he could still induce it by copper sulphate, from which he
concluded that there are at least two centers for vomiting. Open-
chowski’s contention that there is a vomiting center in the quadrigem-
inate bodies finds apparent support in the results observed by Suke-
taka Morita (41), who found that removal of the hemispheres inhibits
the gnawing movements induced in rabbits by apomorphin. It is

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 491

probable that the vomiting center in the medulla was depressed by

the operative procedures, and that apomorphin then failed to stimulate
it. It is impossible to say why Openchowski failed to induce vomiting
with apomorphin after destruction of the quadrigeminate bodies, but
it was certainly because of some error of technic, and Magnus (36)
induced vomiting in decerebrated dogs with apomorphin.
Openchowski expressed doubt that any single center could co-
ordinate the many impulses essential for vomiting. Many of his con-
clusions rest upon faulty observations, and his work is reviewed here
at some length partly because of the extraordinary amount of work
that he did and partly because certain erroneous conclusions of his
were accepted by Magnus in his review of vomiting in 1903, while the
work of Thumas was not mentioned. Magnus (35) accepted the
view that there are at least two centers for vomiting, one of which is
automatic and on which apomorphin acts directly, the other a reflex
center.
Thumas (51) located the vomiting center within narrow anatomical
limits in the dog. He found a very small area in the posterior part
of the rhomboid fossa which is more sensitive than any other to the
emetic action of apomorphin. When this area was destroyed, apo-
morphin failed to cause vomiting, though the respiration was markedly
increased and the expiration was more forcible. Thumas concluded
from the results of these experiments that the vomiting center in the
dog lies in the deeper layers within a region 2 mm. in width, occupying
both sides of the midline, and extending from a point 2 mm. anterior
to the calamus scriptorius posteriorly a distance of 5 mm. through the
obex.
Hatcher and Weiss (20) accepted the conclusions of Thumas that
the vomiting center lies within the area described, or very near to it,
but they undertook to establish its physiologic identity, rather than
its exact anatomical location, since the area does not correspond to
any physiological unit. It is evident that the vagi are in close relation
with the vomiting center, and it seemed probable that it lies in the sen-
sory nucleus of the vagi in the ala cinerea, which lies very near to the
area described by Thumas. There is uncertainty concerning the exact
situation and extent of the ala, but according to the illustration given
by Landois and Rosemann (29), it immediately adjoins the area de-
scribed by Thumas.
Hatcher and Weiss induced emesis with mercuric chlorid in cats
and dogs after the destruction of the area described by Thumas, when

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

492 ROBERT A. HATCHER

care was taken to avoid, so far as possible, all injury to the sensory
nuclei of the vagi, but they were unable to induce emesis by any meas-
ures after the destruction of the sensory nuclei of both vagi, and even
the destruction of a small part of the nucleus on either side prevented
emesis in cats after the administration of mercuric chlorid, the most
certain emetic with which they experimented. They conclude that
certain cells situated in the sensory nuclei of the vagi are concerned
with the coordination of the various reflexes that are essential for the
vomiting act. Thumas states that the destruction of the ala cinerea
on one side does not interfere with emesis in the dog.
The area described by Thumas is so small, and it lies so close to the
sensory nucleus of the vagus, that Hatcher and Weiss made no attempt
to apply solutions of emetics to one area exclusively, but they found
t,hat the application of as little as [Link],l mgm. of apomorphin hydro-
chlorid per kilogram of body weight to this region caused emesis
promptly in the dog.
The fact that apomorphin does not induce vomiting after the de-
struction of the area described by Thumas is probably to be attributed
to depression of the true vomiting center due to operative procedure,
for mercuric chlorid then induces vomiting, as it does when the vomit-
ing center ia depressed by slight overdosage of morphin or apomorphin,
which also inhibits the emetic action of apomorphin. This view is
supported by the fact that after Thumas had made a cut along the
midline with a very sharp knife, apomorphin still induced vomiting.
Openchowski suggested that the area in the rhomboid fossa de-
scribed by Thumas serves merely as a path for emetic impulses, and
that apomorphin stimulates the structures concerned. It is possible
that this is true and that the destruction of this area involves the
path of afferent emetic impulses which are concerned with vomiting
induced by apomorphin, but that the impulses induced in the heart
or stomach by irritants take a slightly different path in the medulla
to the true vomiting center in the sensory nuclei of the vagi.
Massive subcutaneous doses of apomorphin almost invariably
cause emesis in the dog and corresponding doses of morphin usually do,
but large intravenous doses of morphin (13) and of apomorphin (10)
frequently do not. Leake (31) induced depression of the vomiting
center without causing emesis in dogs by repeatedly injecting small
doses of morphin subcutaneously. Hatcher and Weiss induced vomit-
ing, followed by depression of the vomiting center, in the dog by the
application of approximately 0.01 mgm. of morphin sulphate per

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 493

kilogram of weight to the floor of the fourth ventricle, after which a

subcutaneous injection of apomorphin was ineffective, but vomiting
was then induced reflexly by introducing a solution of mercuric chlorid
into the stomach.
Harnack (16), (17), (18) maintained that apomorphin stimulates
the medullary and higher centers and that even very large doses do not
depress them. Guinard (13) suggested that large intravenous doses
of morphin cause narcosis of the center before excitation can occur.
Cohen and McGuigan (8) suggested that morphin abolishes afferent
impulses to the respiratory center by depression of some peripheral
structure. Leake (31) concluded that fatigue of the center is not a
prominent factor in the anti-emetic action of morphin.
There is no satisfactory explanation of the phenomenon, however,
for in one experiment Hatcher and Weiss induced emesis in ten seconds
by the application of apomorphin to the vomiting center and when a
moderately large dose is injected intravenously, there is obviously an
interval of much more than ten seconds following the arrival at the
center of an effective dose, and before a depressant one reaches it. It
is possible that the center is stimulated only when there is a certain
concentration of these poisons in the blood, and depressed when that
concentration is exceeded, or that large doses of morphin and apo-
morphin depress some peripheral structure concerned with the normal
afferent emetic impulses. Doctor Gold has recently undertaken to
investigate this problem, but his results are not conclusive. It is
conceivable that large doses of morphin and apomorphin depress the
paths in the medulla for normal afferent emetic impulses, more readily
than those for impulses induced in the heart and stomach by irritants.
Harnack has called attention repeatedly to the depressant muscular
effects of apomorphin and other emetics, when administered system-
ically. Hatcher and Weiss (20) observed great muscular depression
in two dogs which were unable to stand after the direct application to
the area in the floor of the fourth ventricle described by Thumas of
about [Link],l mgm. of apomorphin hydrochlorid per kilogram of
weight. In these experiments there can be no question of peripheral
muscular action, such as Harnack described (16,) (17), (18) .I
Many observers have discussed the influence of blood supply on the
function of the brain centers. Hatcher and Weiss (20) induced vomit-
ing in cats after tying the carotid and vertebral arteries. The blood
1 The protocols of these two experiments were not published.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

494 ROBERT A. HATCHER

supply to the vomiting center was greatly diminished, and little bleed-
ing occurred when the carotid arteries and jugular veins were cut,
but the blood was arterial, and the vomiting center retained its normal
activity, for the injection of digitalis bodies then induced vomiting
as usual.
Valenti (55) concluded that cutting the vagi abolishes the emetic
action of apomorphin and of peripherally acting emetics, and that
apomorphin causes hyperexcitability of the vomiting center so that
impulses due to swallowing of saliva and the gastro-esophageal move-
ments then set up the vomiting reflex. The paper in which Valenti
announced this theory appeared in an Italian publication and no
mention of it is made in his paper which appeared in Schmiedeberg’s
Archives (54) in this same year, and Hatcher and Weiss were not
aware of his theory concerning the action of apomorphin on the center
when their paper was published in 1923. Valenti’s views concerning
the role of the vagi are obviously erroneous.
Miller (40) compared the vomiting reflex with the scratch reflex
with which it has much in common, allowance being made for the
fact that while the latter is a single rhythmic movement, the vomiting
reflex is one of rhythmic periods with groups of movements. Miller
apparently inclines to the view that the respiratory center is the center
for the vomiting reflex.
Hatcher and Weiss observed the sudden inhibition of the vomiting
reflex by the scratch (19) and defecation reflexes (20) and they con-
cluded that the vomiting center is purely reflex in character-never
automatic-as Thumas supposed, and that its reflex excitability is
increased by apomorphin in a manner analogous to that in which the
reflex excitability of the cord is increased by strychnin. They con-
cluded that normal impulses pass to the center from many parts of the
body, but being below the threshhold value are commonly ineffective,
but when the center is rendered hyperexcitable normal impulses in-
duce emesis. They adminktered a dose of apomorphin which was
insufficient to cause vomiting, after which forcing food into the mouth
induced emesi, at once, though only slightly increased normal impulses
were set up.
They concluded that the normal afferent emetic impulses are in-
hibited by ergotoxin, which acts peripherally, after which apomorphin
fails to induce emesis, but mercuric chlorid then induces vomiting
reflexly from the stomach, because ergotoxin does not paralyze all
the afferent endings on which mercuric chlorid acts. They did not

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 495

attempt to determine whether apomorphin acts upon a path of emetic

impulses within the area described by Thumas or directly upon the
cells in the sensory nuclei of the vagi.
Thumas attributed the inability of various mammals to vomit to
the absence of a vomiting center or to its rudimentary development.
Hatcher and Weiss suggest that rodents have probably lost the capacity
to vomit through the development of better means of protection
against poisons, and they point out that the vomiting center affords
only slight protection in the cat, dog and man against many systemic
poisons because a nearly fatal dose must usually be absorbed into the
circulation before vomiting is induced. Valenti (54) caused rabbits
to regurgitate liquid from the stomach after the injection of curare
to lower the resistance of the cardia, and stimulation of the vagua or
cervical sympathetic, but the conditions of the experiments were so
wholly abnormal that the results throw no light on the problem.
Many physiologists accept with reserve the arguments that have
been advanced in support of the view that a special center exists for
vomiting, and no detailed discussion of this broad problem will be
attempted here.
Hatcher and Weiss (20) say that no single movement and no in-
hibition concerned with emesis is peculiar to that phenomenon, and
that probably no single cell or structure is concerned solely with emesis,
but that that part of the central nervous system which alone is es-
sential for vomiting, which receives afferent emetic impulses, discharges
efferent, and coordinates in orderly sequence the various movements
and inhibitions necessary for vomiting-closure of the pylorus, con-
traction of part of the stomach, relaxation of the rest, inhibition of the
cardia and esophagus, closure of the glottis with inhibition of respira-
tion, stimulation and inhibition of the diaphragm, contraction of the
abdominal muscles, -is conveniently termed the vomiting center,
because it behaves as a unit when discharging this function. Its
activity can be increased by the local application of extraordinarily
small amounts of certain poisons, and when it is disorganized by
mechanical injury vomiting cannot be induced in the dog or cat.
Vomiting embraces several subsidiary reflexes which are coordinated
in the center; it may be completed within a few seconds, or it may be
checked at the stage of nausea, and the process held in abeyance while
the center maintains its control over certain of the structures involved
(pylorus and stomach) while others (respiratory) are released, and
after an interval of seconds or minutes the process be completed.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

496 ROBERT A. HATCHER

Retching movements with the expulsion of vomitus from the mouth

constitute a subsidiary reflex, which may be inhibited but which when
begun proceeds independently of the stimulus (Miller).
Mellinger (37) induced emesis by means of the purely local reflex
mechanism in a frog in which the cerebra-spinal organ had been ex-
tirpated completely. One cannot say that rudimentary vomiting
movements, or even actual emesis, never occurs in any of the higher
mammals without the participation of the central nervous system,
as the experiments of Schuetz may seem to suggest, but there is no
evidence that this occurs in adults, and all the evidence available
indicates that ordinary vomiting in the cat, dog and man is the result
of a reflex involving the medullary center.
5. THE PATHS OF EMETIC IMPULSES. Openchowski is the only one
who has studied the paths concerned with emesis comprehensively,
and his views are still widely held by physiologists, though certain of
his conclusions are based on the results of faulty experiments.
Thumas and Openchowski were both convinced of the automaticity
of the vomiting center (or centers) and believed that when apomorphin
failed to induce emesis in the dog, it indicated that the center was
paralyzed or the path was blocked. The center is especially prone to
become depressed after operations in its vicinity and this probably
explains why Openchowski failed to induce emesis with apomorphin
after destruction of the quadrigeminate bodies and after section of the
cord at the level of the fifth thoracic vertebra. His failure to induce
emesis with copper sulphate after vagotomy in the dog led him to the
general conclusion that the vagi alone carry afferent emetic impulses
from the stomach in various animals, and this view is still held by
many, though numerous observers have seen vomiting induced by
irritants in the stomach after vagotomy. Hatcher and Weiss induced
vomiting with copper sulphate in four cats in which the vagi had been
cut, and Dr. A. 0. Shaklee (personal communication) induced vomiting
with zinc sulphate in the stomach of four dogs in which the vagi had
been cut.
Miller (40) found that the emetic action of mustard is inhibited by
cutting the vagi in the anesthetized cat. He was unable to show that
the splanchnics carry sensory impulses from the gastric mucosa, but
such experiments can have no weight in view of the positive results
reported by many others.
It has long been known that vomiting may be induced reflexly by
irritation of the uterus, kidney, bladder (Openschowki), brain and

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 497

semicircular canals, as well as from the gastro-intestinal tract, but

much confusion has resulted from the commonly accepted view that a
drug which induces vomiting after its intravenous injection, acts
either on the vomiting center directly or reflexly from the gastro-
intestinal tract. Kleimann and Simonowitsch (26) injected tartar
emetic intravenously and found traces of it in the vomitus, from which.
they concluded that emesis was then induced reflexly from the gastro-
intestinal tract, but Weiss and Hatcher (20) found that tartar emetic
causes vomiting by its action on the heart after its intravenous injection.
Very small amounts of emetin and nicotin increase the reflex excitability
of the vomiting center to which they are applied directly, but they
cause emesis by their peripheral (cardiac?) action after their intra-
venous injection.
Emesis is induced by the direct application to the vomiting center
of minute amounts of certain normal constituents of the body (cholin,
epinephrin and histamine) and it is possible that when abnormal
amounts of histamine and cholin are present they may cause nausea
and vomiting in disease. It is highly improbable that epinephrin is
ever present in the circulation in amounts sufficient to exert any
perceptible effect on the vomiting center, but it is possible that its
peripheral effects may be concerned with vomiting in disease.
Hatcher and Weiss found that extirpation of the stellate ganglia
or section of the cord abolishes the emetic action of certain substances
(ouabain) in the cat, but not that of others (tartar emetic), and section
of the vagi abolishes the emetic action of tartar emetic but not that of
ouabain, hence each path carries both afferent and efferent emetic
impulses. Efferent impulses pass from the vomiting center by way of
the vagi and the spinal cord with equal facility regardless of the ex-
citing cause. They also found that atropin and ergotoxin each para-
lyze some of the afferent emetic paths in both the vagus and the
sympathetic, but neither paralyzes all paths in either.
If a given substance stimulates afferent endings about equally in
both the sympathetic and the vagus, the action of a nearly minimal,
emetic dose is inhibited by cutting one afferent path, but section of
one path does not appear to influence the action of a nearly minimal
emetic dose of a poison if its afferent path is uninjured. Afferent
emetic impulses evidently pass from many parts of the body through
many paths, but the essential efferent impulses probably leave the
medulla by the phrenics and one path in the vagus or one in the cord,
one of the latter paths being chosen and the ot,her inhibited in any
given case.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

498 ROBERT A. HATCHER

Cutting the cord does not influence the emetic action of small doses
of tartar emetic because the emetic impulses induced by it in the heart
and in the stomach pass up by the vagi almost exclusively. Mercuric
chlorid acts nearly equally on the afferent endings of the vagus and the
sympathetic in the stomach, and if one administers an oral dose which
would just suffice to cause emesis in the normal animal to one in which
the vagi have been cut or to one in which the cord has been severed,
the animal fails to vomit, because the impulses which reach the center
are insufficient to set up the vomiting reflex.
Miller (40) observed summation of emetic stimuli induced by mus-
tard in the stomach of the cat and faradic stimulation of the centra,l
end of the gastric vagus. Hatcher and Weiss (20) found a slight, but
distinct, synergism between the emetic action of apomorphin on the
center and that of ouabain on the heart, but they observed no synergism
between the emetic action of ouabain on the heart and that of mercuric
chlorid in the stomach. They say: “It seems probable that if all the
normal afferent impulses that arise throughout the body were capable
of summation the center would of necessity be very resistant, or that
it could not respond to abnormal stimuli arising from a single small
organ unless these were of great intensity. On the other hand the
synergistic action of the center and the peripheral organs in their
reaction toward a harmful substance is necessary for the fulfilment of
the function of such a protective mechanism.”
Sherrington (49) says : “The preclusion of the motor neurone from
one reflex while it is still left open to it to respond to other reflexes
appears to be one of the services of inhibition to the organism.” Its
seems probable that this applies also to neurones concerned with
afferent emetic impulses, for Hatcher and Weiss frequently observed
the abolition of the rapidly developing vomiting reflex, induced by the
action of one of the digitalis bodies on the heart, by the onset of the
defecation reflex induced by the same substance, presumably through
the same afferent paths, for the efferent paths for defecation are not
the same as those for emesis.
Knaut (27) states that only the anterior columns carry fibers con-
cerned in vomiting, induced in the cat by apomorphin. Knaut evi-
dently refers to efferent fibers, as he supposed that apomorphin stim-
ulated an automatic center. Hatcher and Weiss (20) found that
cutting the posterior columns of the cord does not interfere with the
vomiting reflex induced by the actio 1 of ouabain on the heart, in the
cat, but that compkte section of the cord usually does, from which it

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 499

seems probable that afferent emetic impulses from the heart pass up
by way of the antero-lateral column, since that is probably the only
sensory path from the viscera in the anterior part of the cord.
Langley (SO) includes only afferent fibers in the autonomic nervous
system, but Lewandowski (33) implies that it includes afferent fibers
from the viscera. Hatcher and WSss classified the afferent fibers
concerned with emesis as sympathetic and parasympathetic, but they
found that the afferent fibers of the vagus and those of the sympathetic
form numerous gradations, and they cannot be separated sharply
into the two classes, however convenient this may be for discussion
with reference to function or pharmacologic behavior.
This is illustrated by the fact that ergotoxin and atropin both inhibit
the emetic action of pilocarpin in the heart (?), that neither inhibits
its purgative action in the walls of the intestine; that ouabain and
digitalis exhibit differences in the paths of their cardiac emetic impulses,
both being inhibited by ergotoxin, neither by atropin. Afferent
emetic impulses induced by some chemical agents traverse the sympa-
thetic and the parasympathetic, but normal afferent emetic impulses
appear to pass up the sympathetic type only in the vagus and in the
sympathetic trunk. Langley calls attention to the fact that the
sympathetic is distributed to the entire body, whereas the parasym-
pathetic is distributed to certain parts only.
The paths for certain normal afferent emetic impulses are paralyzed
by ergotoxin, but whether it abolishes all normal afferent emetic im-
pulses is not known, and it is possible that it does not abolish the emetic
action of all drugs which increase the excitability of the vomiting center
without depressing it, as large doses of apomorphin do, supposing that
such substances exist. This subject requires investigation.
Hatcher and Weiss found the following types of afferent paths for
emetic impulses based on the pharmacologic behavior of their sensory
endings. Further investigation will almost certainly increase the sub-
division, and it is probable that some which are here placed in dif-
ferent groups may be found to have only minor quantitative differences
in behavior. Those marked with an asterisk are apparently common
to the vagus and sympathetic.
A. Sympathetic trunk, stellate ganglion and cord
l.* From the stomach, stimulated by small doses of mercuric chlorid, para-
lyzed by ergotoxin, not by atropin.
2. From the stomach, stimulated by large doses of mercuric chlorid, not para-
lyzed by ergotoxin or atropin. (Possibly the same as no. 1.)

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

500 ROBERT A. HATCHER

3.*From the heart, stimulated by digitalis, paralyzed by ergotoxin, not by

atropin.
4.” From the heart, stimulated by digitalis, paralyzed by ergotoxin and
atropin.
5.* From various parts of the body, conducts normal impulses that cause
emesis only when the center is hyperexcitable; paralyzed by ergotoxin, not by
atropin.
B. Vagus trunk
l.* From the stomach, like no. 1 in the sympathetic.
2. From the stomach, stimulated by tartar emetic, paralyzed by atropin,
not by ergotoxin.
3. From the duodenum, stimulated by tartar emetic, not paralyzed by atropin.
4. From the heart, stimulated by tartar emetic, paralyzed by atropin. (May
be like no. 2.)
5.* From the heart, stimulated by digitalis, paralyzed by ergotoxin, not by
atropin. (May be like no. 3 in the sympathetic.)
6.* From the heart, stimulated by pilocarpin, paralyzed by atropin and
ergotoxin. (Probably like no. 4 in the sympathetic.)
7.” From various parts of the body, conducts normal impulses that cause
emesis only when the center is hyperexcitable; paralyzed by ergotoxin, not by
atropin. (Probably similar to no. 5 in the sympathetic.)

THE VOMITING ACT. Many have described the vomiting act in

the cat, dog and man, but we have no ac curate and complete account
of it. Hesse states that vomiting takes place in three stages in the
dog. 1. Contraction of the pyloric part and passage of the gastric
conten ts into the fundic part with the onset of nausea. 2. The passage
of the vomitus into the esophagus. 3. The emptying of the esophagus
out through the mouth. These three stages are not sharply separated,
however.
The following description of typical vomiting in the dog is based
mainly on the facts reviewed in this paper. The subcutaneous in-
jection of apomorphin induces hyperexcitability of the vomiting center
after which normal afferent emetic impulses induce nausea and in-
augurate the vomiting reflex within a few minutes. If the animal lies
down after the injection, it gets up at the onset of nausea. Simul-
taneously, the pylorus and the pyloric part of the stomach contract
firmly, the antrum moderately, while the fundus relaxes, the walls
becoming flaccid, and the gastric contents are forced into the fund us.
There may be unimportant gastric antiperistalsis. The secretion of
saliva is markedly increased and the animal stands with the head
down and the nose nearly touching the floor. The respiration becomes
deep, rapid and irregular, and with increasing nausea the animal

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 501

makes chewing movements, licks the lips and swallows saliva repeatedly;
air is usually swallowed at this time. (Cats usually emit characteristic
cries of distress just before vomiting.) It is probable that the stomach
remains in the same position throughout the period of nausea, which
is followed bY dilatation of the cardia and characteristic vomiting
movements called retching (not common in man) during which the
heart rate is slowed and the blood pressure undergoes rapid oscillations.
Retching consists of simultaneous spasmodic contractions of the
muscles of respiration during which, however, the diaphragm contracts
with the muscles of expiration. The glottis closes (probably after
moderate expiration) , and remains closed until after the expulsion of
the vomitus has ceased. The neck is extended with each vomiting
movement until the vomitus begins to issue from the mouth. The
first few movements drive a variable portion of the gastric contents,
including air, into the (probably flaccid) esophagus and the air and
vomitus are carried up and down with the vomiting mov .ements until
near the last when the cardia (usually) closes, the diaphragm relaxes,
and the abdominal (and probably all other muscles concerned i.n forced
expiration) contract powerfully, forcing the stomach firmly against
the diaphragm. This results in changing the negative pressure in the
chest and esophagus into a positive pressure, while the esophagus con-
tracts in toto (or by antiperistalsis) and partially empties its contents
out through the mouth.
The larynx and the hyoid bone are pulled forward by the genio-
hyoid, Sterno-hyoid, Sterno-thyroid and thyro-hyoid muscles while
the neck is stretched forward and held rigidly, causing a widening of
the throat during the expulsion of the vomitus. The posterior pillars
of the fauces raise the palate and uvula and shut off the nasal chamber
from the throat.
The vomitus remaining in the mouth is swallowed, and that in the
throat and esophagus passes down by normal peristalsis and lies before
the cardia for some seconds or minutes before passing in to the stom ach,
or being vomited during the succeeding paroxysm It passes into
the stomach at once in those cases in which the cardia remains open
during the whole period of expulsion of the vomitus from the mouth.

It is remarkable that comparatively little attention has been paid

to the clinical significance of vomiting other than that of special types,
such as cyclic, or that which is of well-known significance, beyond the
mere notice of its presence or absence, and it is often deemed too in-

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

502 ROBERT A. HATCHER

significant to be mentioned in clinical reports. Hammer (14) protests

against “the oft-repeated assertion that the symptom of vomiting has
little clinical worth, and for all practical purposes may be quite dis-
regarded.” He says : “We observe vomiting as merely an insignifi-
cant symptom at one end of the scale, at the other end, this ominous
factor is a fore-runner of death.” The very language in which it is
sometimes discussed by clinicians is hardly intelligible to physiologists;
for example, Hurst (24) says: “By reflex vomiting is meant vomiting
which originates by a reflex from some organ other than the stomach.”
Janowsky (25) discussed the physiology and general pathology of
vomiting which he classified according to cause as central and reflex.
He includes psychic influences, tumor and inflammatory conditions
of the brain among the central causes, but vomiting due to affections
of the cerebrum is also reflex in character.
Emetics may be classified according to the nature of their action,
as those which increase the excitability of the center (apomorphin,
morphin, picrotoxin), and those which initiate afferent emetic impulses
through irritation of one or more structures in various organs or tissues
(copper sulphate, mercuric chlorid). A few (aconitin, emetin) act
in both ways.
It is probable that the greater number of systemic poisons which
cause vomiting do so through their action on the heart, and it is more
than probable that the heart is frequently the seat of the vomiting reflex
induced by disease. The frequency of vomiting in exophthalmic
goiter, in which the symptoms are mainly sympathomimetic, is sug-
gestive. The physician should never regard vomiting as a mere in-
cident without importance, more especially when it occurs in adults,
for it is probably the result of toxic substances in the circulation except
in those cases where it is induced by inflammatory processes, as in
gastritis.
Herrick (21) says the epigastric reference of pain with the nausea
and vomiting that frequently occur in patients having thrombosis of
the coronary artery often suggests to the patient and physician some
abdominal accident such as acute pancreatitis, perforation of the
gall bladder, or a gastric or duodenal ulcer.
The fact that [Link] is always induced reflexly by peripheral ir-
ritation or by increasing the excitability of the center, that the heart is
a common seat of the vomiting reflex induced by poisons; that the cen-
ter may be rendered hyperexcitable by certain normal constituents
of the body, that vomiting is one of the most common symptoms of

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

 THE MECHANISM OF VOMITING 503

disease suggests numerous problems for investigation. The selective

action of vakous poisons (ergotoxin, atropin) on afferent endings affords
a convenient method of investigating the afferent emetic paths in the
sympathetic and parasympathetic nerves to different organs.

BIBLIOGRAPHY
(1) BROOKS, C. AND A. B. LUCKHARDT. Amer. Journ. Physiol., 1915, xxxvi, 104.
(2) BUCH, 1111. Finska Lak. Hand., 1908, 1, 641.
(3) CANNON, W. B. Amer. Journ. Physiol., 1898, i, 359.
(4) CANNON, W. B. Amer. Journ. Physiol., 1907, xix, 436.
(5) CARLSON, A. J., T. E. BOYD AND J. F. PEARCY. Amer. Journ. Physiol.,
1922, lxi, 14.
(6) CARLSON, A. J., T. E. BOYD AND J. F. PEARCY. Arch. Int. Med., 1922,
xxx, 409.
(7) CARLSON, A. J. AND A. B. LUCKHARDT. Amer. Journ. Physiol., 1921, lvii,
299.
(8) COHEN, S. J. AND H. MCGUIGAN. Journ. Pharm. Exper. Therap., 1924,
xxiii, 145.
(9) v. CZYHLARZ, E. AND A. SELKA. Wien. klin. Wochenschr., 1913, xxvi,
842.
(10) EGGLESTON, C. AND R. A. HATCHER. Journ. Pharm. Exper. Therap., 1912,
iii, 551.
(11) GIANNUZZI, G. Cent. f. d. Med. Wissensch., 1865, iii, 1; 129.
(12) GRIMM, A. Pfltiger’s Arch., 1871, iv, 205.
(13) GUINARD, L. Lyon Medical, 1895, lxxx, 37.
(14) HAMMER, A. W. N. Y. Med. Journ., 1920, cxii, 64.
(15) HARNACK, E. Arch. exper. Path. u. Pharm., 1874, ii. 254.
(16) HARNACK, E. Arch. exper. Path. u. Pharm., 1875, iii, 44.
(17) HARNACK, E. Munch. med. Wochenschr., 19082, 1869.
(18) HARNACK, E. AND H. HILDEBRANDT. Munch. med. Wochenschr., 1910, 20.
(19) HATCHER, R. A. AND S. WEISS. Arch. Int. Med., 1922, xxix, 690.
(20) HATCHER, R. A. AND S. WEISS. Journ. Pharm. Exper. Therap., 1923, xxii,
139.
(21) HERRICK, J. B. Journ. Amer. Med. Assoc., 1919, lxxii, 387.
(22) HESSE, 0. Pfliiger’s Arch., 1913, clii, 1.
(23) HLASKO, B. Diss., Dorpat. Beitrage zur Beziehung des Gehirms zum
Magen, 1887.
(24) HURST, A. T. N. Y. Med. Journ., 1920, cxi, 45.
(25) JANOWSKY, W. Volkmann’s Sammlung klin. Vortrsge., no. 333, Inn. Med.,
no. 99, 1902, 745.
(26) KLEIMANN, A. AND R. SIMONOWITSCH. Pfltiger’s Arch., 1872, v, 280.
(27) KNAUT, A. Diss. Dorpat. Innervation des Magens seitens des Rucken-
marks in Hinsicht auf den Brechact. 1886.
(28) KREHL, L. Arch. f. anat. u. Phys.; Phys. Abt., 1892, Sup. 278.
(29) LANDOLS, L. AND R. ROSEMANN. Lehr. d. Physiol., 16 Auflage, 629.
(30) LANGLEY, J. N. The autonomic nervous system, part i, 1921, 1.

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.

504 ROBERT A. HATCHER

(31) LEAKE, C. D. Journ. Pharm. Exper. Therap., 1923, xx, 359.

(32) LEVY-D• RN, M. AND S. MUEHLFELDER. Berl. klin. Wochenschr., 1910,
xlvii, 388.
(33) LEWANDOWSKI, M. Praktische Neurolgie fur Arzte, iii Auflage, 69.
(34) MAGENDIE. Memoire sur le Vomissement, 1813.
(35) MAGNUS, R. Ergebn. d. Physiologie, II Jahrg. Abt. II (Asher-Spiro),
1903, 641.
(36) MAGNUS, R. Pfluger’s Arch., 1914, clvii, 75.
(37) MELLINGER, C. Pfluger’s Arch., 1881, xxiv, 232.
(38) v. MIKULICZ, J. Mitt. a. d. Grenz., 1903, xii, 596.
(39) MILLER, F. R. Journ. Physiol., 1910, xli, 409.
(40) MILLER, F. R. Pfltiger’s Arch., 1911, cxlii, 1.
(41) MORITA, S. Arch. exper. Path. u. Pharm., 1915, lxxvii, 188.
(42) MORITZ. Zeitschr. f. Biol., 1895, Neue Folge, xiv, 313.
(43) Mosso. Untersuch. z. Naturl. d. Mensch. u. d. Thiere, 1876, xi, 347. (Cited
by W. B. CANNON. The mechanical factors of digestion, 1911, 3.)
(44) MUELLER, L. R. Deutsch. Arch. f. klin. Med., 1911, ci, 421.
(45) OPENCHOWSKI, T. Cent. f. Physiol., 1889, iii, 1.
(46) ROSSBACH, M. J. Deutsch. Arch. f. klin. Med., 1890, xlvi, 296.
(47) Roux, J. C. AND V. BALTHAZARD. Seances et Mem. de la Sot. de Biol.,
10 series, 1897, iv, 704; also Arch. de Physiologie normale et path.,
1898, ser. 5, x, 85.
(48) SCHUETZ, E. Arch. exper. Path. u. Pharm., 1886, xxi, 341.
(49) SHERRINGTON, C. S. The integrative action of the nervous system. Lon-
don, 1906.
(50) TANTINI. In ORMODEI, Ann. Univ. de Med., 31, 1824. (Cited by MAGNUS
from S. MAYER in Hermann’s Handbuch.)
(51) THUMAS, L. J. Arch. f. Anat. u. Physiol., 1891, cxxiii, 44.
(52) TIEDEMANN, F. A systemic treatise on comparative physiology (Eng.
Transl.) London, 1834.
(53) VALENTI, A. Cent. f. Physiol., 1906, xx, 449.
(54) VALENTI, A. Arch. exper. Path. u. Pharm., 1910, lxiii, 118.
(55) VALENTI, A. Arch. di Farm. Sper. e Scienze Affini, 1910, ix, 299.
(56) WEPFER, J. J. Historia Cicutae Aquatica, Basil, 1679. (Cited by CANNON.)

Downloaded from [Link]/journal/physrev ([Link]) on August 7, 2020.