1900 B 3 135 fll? LIBRARY OF THE UNIVERSITY OF CALIFORNIA. CTVvxAxO I T (^-iVh-A^^LO , BIOLOGY A PLETHYSMOGRAPHIC STUDY OF THE VASCULAR CONDITIONS DURING HYPNOTIC SLEEP. A DISSERTATION SUBMITTED TO THE BOARD OF UNI VERS1TY STUDIES OF THE JOHNS HOPKINS UNIVERSITY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY. 1900. BY ELISHA CHISHOLM WALDEN. VV3 BIOLOGY LIBRARY G A PLETHYSMOGRAPHIC STUDY OF THE VASCULAR CONDITIONS DURING HYPNOTIC SLEEP. BY E. C. WALDEN. [From the Laboratory of Physiology in the Johns Hopkins University^ CONTEXTS. Page Introduction 124 Description of the apparatus 125 Description of the plethysmographic curves 132 a. Normal curves 133 b. Exceptional curves 137 c. Effect of suggestion 139 Description of other phenomena 141 a. Blood-pressure curves 141 b. Pulse rate 144 c. Respiration 146 d. Temperature 147 1. Rectal temperature 147 2. Surface temperature 148 Discussion of results 149 a. Suggested explanation of the meaning of the plethysmographic curves 149 b. Probable explanation of the changes in the blood-pressure .... 155 Summary and general conclusions 158 INTRODUCTION. THE principal object of these experiments has been to deter- mine the changes occurring in the volume of the arm as a consequence of hypnotic sleep and suggestion, and to compare the results so obtained with the observations which have been made by the same methods on normal sleep. It has been shown by Mosso, 1 Howell, 2 and other investigators, by means of the water plethysmo- graph, that the volume of the arm is increased during normal sleep. The same authors have also shown that mental and muscular activity cause a constriction of the arm. These changes are assumed by How- ell to be due to vasomotor changes in the cutaneous blood vessels 1 Mosso: Ueber den Kreislauf des Blutes im menschlichen Gehirn, Berlin, 1881 ; Die Temperatur des Gehirns, Berlin, 1894. 3 HOWELL : Journal of experimental medicine, 1897, ii, p. 313. L31179 Study of Vascular Conditions during Hypnotic Sleep. 125 and a consequent alteration in the amount of blood flowing through the peripheral vessels. The present experiments have been extended so as to include not only the plethysmographic records, but records of the blood-pressure, pulse, respiration, and temperature, as well. DESCRIPTION OF THE APPARATUS. The plethysmograph used was similar to that previously described by Howell. 1 It consisted of a glass cylinder of sufficient size to allow the hand and a portion of the fore-arm to be inserted in it. One end of the cylinder was drawn out and w,as connected to one arm of a three-way stop-cock by stiff rubber tubing. On one side of the cylinder there was a small neck, into which was fitted a piece of glass tubing provided with a stop-cock. This opening served as an escape for the air while the apparatus was being filled with water. The two remaining arms of the three-way stop-cock were joined, one to the recording apparatus and one to the reservoir, which contained water for filling the apparatus. By turning the cock, the plethysmograph could be connected with the recording apparatus, the reservoir, or with both. The recorder used was the form devised by Bowditch. 2 It con- sisted of a test-tube swung on a spiral spring in such a manner that the height of the water always remained constant, the test-tube being pulled up by the tension of the spring as water was withdrawn from the tube, and the spring, in turn, being stretched out as water was poured into the test-tube. The spiral spring was fastened to a short vertical rod, and this was attached to another longer vertical rod by means of a universal joint. The longer vertical rod was firmly fast- ened to a table. By means of the universal joint the height of the test-tube and the tension of the spring were very easily regulated. A pen of thin paper was attached to the test-tube and wrote against the blackened surface of a drum kymographion, which revolved once in six hours. Two other pens were arranged to write in the same vertical line. One of these pens was connected with an electric signal, in circuit with a clock, and marked intervals of one minute. The other pen was attached to a lever and was used to record the application of any stimulus that was given to the subject, or any change that was noticed in his condition. 1 HOWELL: Journal of experimental medicine, 1897, ii, p. 313. 2 BOWDITCH : Proceedings of the American Academy, May 14, 1896. 126 E. C. Walden. To keep the arm immovable in the plethysmograph, the device described by Howell 1 and Shields 2 was used. This consisted of a hinged collar of hard rubber, which fitted around the thumb between the first and second phalangeal articulations. This collar was rigidly attached by a brass rod to another collar of hard rubber which fitted loosely over the fore-arm. The outer circumference of this collar was of such a size that it fitted snugly into the mouth of the glass cylin- der. The object of this device was to prevent the arm from slipping farther into the plethysmograph. To prevent the arm slipping out of the plethysmograph, the device described by Shields 3 was used. This consisted of two hinged hard rubber rings. The larger one of these rings fitted around the end of the glass cylinder, the other ring, of just sufficient size to allow the fore-arm to pass through it, was connected to the first by screw clamps, so that the smaller collar could be pressed up against the end of the cylinder, and in this way prevented the collar within the cylinder from being pulled out. The plethysmograph was swung from the ceiling, and was so arranged that it could be adjusted to any desired level. The elbow was supported by means of a sling, which was fastened to the chain holding the plethysmograph. The greatest difficulty encountered in all plethysmographic experi- ments has been to secure some device whereby the arm could be en- closed within the plethysmograph in such a manner as to prevent leakage from the cylinder, and at the same time to avoid compression of the arm. The errors which occur in either case spoil the records obtained. The original device employed by Mosso, 4 consisting of a rubber sleeve, was not entirely satisfactory when used alone, the chief objection being the difficulty in adjusting the sleeve to the size of the arm of the subject for each experiment, without causing undue compression of the arm. The rubber sleeve only serves this purpose when it is very carefully adjusted to the arm. A device has been used in these experiments which can readily be adjusted to any arm without danger of either a leakage from the instrument or a compression of the arm. A piece of heavy rubber band tubing, fifteen centimetres in length and of sufficient diameter to allow it to fit loosely around 1 HOWELL: Journal of experimental medicine, 1897, ii, p. 313. 2 SHIELDS : Journal of experimental medicine, 1896, i, p. 74. 3 SHIELDS. Journal of experimental medicine, 1896, i, p. 74. 4 Mosso : Ueber den Kreislauf des Blutes im menschlichen Gehirn, Berlin, 1 88 1 ; Die Temperatur des Gehirns, Berlin, 1894. Study of Vascular Conditions during Hypnotic Sleep. 1 2 7 the fore-arm, was drawn over the hand and fore-arm, the upper end of the sleeve reaching the elbow. A surgeon's glove, of thin rubber and provided with a long sleeve, was then drawn over the hand and the heavy rubber sleeve. A second piece of band tubing, similar in every respect to the piece first slipped over the fore-arm, was then pulled over the sleeve of the rubber glove, in such a manner that the thin rubber sleeve was sandwiched between the two pieces of heavy rubber tubing. The thumb was next secured in the holder used to pre- vent the arm from slipping too far into the cylinder, and the hand was then thrust into the plethysmograph. The upper ends of the heavy rubber tubing, between which lay the thin rubber sleeve, were then inverted over the mouth of the cylinder, and were securely tied. The hard rubber collar was next adjusted on the arm in such a manner that the rubber sleeves were tightly clamped between the inner collar and the smaller one of the collars on the outside of the cylinder, and this collar was then firmly fastened to the hard rubber ring encircling the end of the cylinder. The rubber glove completely closed the open end of the glass cylinder, so that there was no possibility of a leak. The object of the pieces of heavy rubber band tubing, one on each side of the thin rubber sleeve, was to reinforce the thin rubber at the mouth of the plethysmograph. Were it not for this protection the water within the cylinder would affect the thin sleeve, pushing it out, and in this way the accuracy of the instrument would be destroyed. Since the rubber sleeves used did not bind the arm, there was no danger of compression, the arm being under the same pressure it would have been were it enclosed in the same volume of water with- out the sleeve intervening. The hand and about nine centimetres of the fore-arm were enclosed in the thin rubber sleeve ; this sleeve was forced snugly against the skin by the water within the instrument, the water at the same time forcing any air out that might have been imprisoned between the glove and the arm. With this arrangement the thin rubber glove acted as a second skin, allowing the arm and hand to increase or decrease readily in volume, and these changes in the volume were promptly recorded by the corresponding outflow or inflow of water from the plethysmograph to the hanging test-tube. The adhesion of the thin glove to the arm entirely prevented any chance of air forcing its way between the glove and the skin. When the subject was ready for an experiment, the glove was drawn on and the arm secured in the plethysmograph. The three- way stop-cock was turned so that the plethysmograph was placed in p p * OP ( UNIVERSITY x. i~ OF 77 /** 7T7 7^ >N *&i-/FORJikt^ 128 ^. C . Walden. connection with the reservoir. As the instrument filled with water, the air within the plethysmograph was forced out through the small opening in the top of the cylinder. As soon as the instrument was filled with water, this opening was closed. The water was still forced into the plethysmograph from the reservoir until the hand was under considerable pressure. This pressure was sufficient to force out ajiy air that might have remained between the sleeve and the arm, and it was also effective in fitting the glove closely to the arm and hand. When the glove had been pressed down against the hand, the stop- cock was turned so that the plethysmograph was placed in connec- tion with the recorder, the water supply from the reservoir being shut off at the same time. Under these conditions water flowed from the plethysmograph to the recording test-tube until the pressure within the cylinder, and consequently the pressure exerted against the arm, was equal to the level of the column of water in the test-tube. If the water level in the test-tube was higher than the level of the cylinder, the arm in the plethysmograph was subjected to positive pressure. If, on the contrary, the cylinder was higher than the level of the water in the test-tube, the arm was under negative pressure. This must be avoided, for, as was shown by Shields, 1 a positive pressure on the arm may cause a marked constriction, while, on the other hand, negative pressure causes the arm to dilate. If the test-tube was so arranged that the level of the water within it was at the height of the middle of the cylinder, then the arm within the cylinder was half of it under a slight negative, and half of it under a slight positive pressure. Records of the blood-pressure were taken by a modification of Mosso's sphygmomanometer. 2 The apparatus consisted of two glass tubes, one above the other, and of sufficient size to allow the fingers to be easily inserted. The tubes, which were connected with each other, were filled with water, which placed the fingers under a counter pressure. To prevent leakage, the fingers were inserted into thin rubber or membrane fingers which were securely fastened to the cylinders. Besides these thin fingers, thin leather collars were pulled over the rubber fingers to reinforce them and prevent the possibility of bulging of the rubber when the fingers were subjected to great pressure. The hand was held in position by a hard rubber collar which fitted around the wrist and which was secured to the base sup- 1 SHIELDS: Journal of experimental medicine. 1896, i, p. 74. 2 Mosso : Archives italiennes de biologic, 1895, xxiii, p. 177. Study of Vascular Conditions during Hypnotic Sleep. 129 porting the glass tubes. This prevented the fingers from slipping out of the apparatus when pressure was applied. The pressure was regulated by means of a pressure flask swung from the ceiling, and so arranged that it could be raised or lowered as was desired. The pressure was registered by a mercury manometer arranged to take graphic records. The method of determining the blood-pressure by this instrument was as follows : the fingers were first inserted into the tubes, and the hand secured in its position. The pressure on the fingers was in- creased gradually by raising the pressure bottle. As the pressure was increased, the amplitude of the pulsations increased until a cer- tain pressure was reached ; any increase in the pressure beyond this point caused the pulse to diminish in amplitude, the pulse being en- tirely obliterated if the pressure was raised to a sufficient height. This point having been reached, any decrease in the pressure was followed by a return of the pulse and an increase in its amplitude, the maximal amplitude on decreasing the pressure being observed at about the same pressure as with the increasing pressure. The varia- tion in the maximal amplitude of the pulsations on the rising and descending scale never amounted to more than five millimetres of mercury. If before the pressure records were taken, the pressure was rapidly raised and then lowered, the readings secured were al- most the same with both ascending and descending variations in the pressure. It was found that the temperature of the water used in the instrument had a great effect upon the amplitude of the pulsa- tions. If the water was cold, the amplitude of the pulse was very small, and it was with difficulty that the maximal pressure could be distinguished. When the water used was- of a higher temperature than that of the fingers, the pulsations were markedly increased in amplitude, and the differences in the amplitude when the fingers were under different pressures were easily recognized. In order to test the accuracy of the principle of the Mosso sphyg- momanometer, experiments were made upon dogs. A small mem- brane tube, of sufficient size to allow the carotid artery of a dog to slip through it easily, was fastened securely to one end of a glass tube 8 cm. long and I cm. in diameter. The membrane tube was pushed into the glass tube and was prevented from slipping out by a cork fastened into the end of the tube. A small hole was bored through the centre of the cork, large enough to allow the artery to be passed through it. The opposite end of the glass tube was closed 9 3 o E. C. Walden. by a cork through which a small glass tube entered the cylinder. This tube was connected by a glass " T" piece to an ordinary mer- cury manometer and to a pressure bottle. The carotid artery to be experimented upon was exposed and carefully dissected out from the tissues for a distance of about eight centimetres. It was then ligated and cut through. To the peripheral stump a strong thread was attached, and this was pulled through the cork and 'the membranous tube, until at least five centimetres of the artery were enclosed within the tube. The free end of the artery and of the membranous tube were then tied securely together, in such a manner that there was no leak when pressure was applied. The artery and membrane tube enclosing it were held in position in the glass tube by a thread tied to the free end of the artery and passed out through the cork at the distal end of the glass tube. The appa- ratus was next filled with water from the pressure flask. Even at zero pressure pulsations were visible in the mercury manometer, and, as the pressure was raised, these oscillations of the mercury became more pronounced until a certain pressure was reached, at which point the amplitude of the pulsations was maximal. Any increase or diminution in the pressure from this point caused a diminution in the amplitude of the pulsations. Two experiments were made with this apparatus, in one of which the right and in the other the left carotid artery was fastened in the glass tube of the sphygmoma- nometer. The carotid artery of the opposite side was connected in the usual way with an ordinary mercury manometer and served as a control to the pressure observations made with the sphygmomano- meter. The results obtained in these experiments are given below. Experiment 1. Length of time during which the observations were taken, 48 minutes. Sphygmomanometer on the right carotid. Right Carotid. Left Carotid. 142 mm. Hg. 150 mm. Hg. 140 " " 148 " " 150 " " 152 " " 141 " " 149 " " 136 " " 144 " " 134 " " 144 " " Average 140.5 mm. Hg. Average 147.8 mm. Hg. In this experiment the pressure registered in the left carotid is a few millimetres higher than the pressure registered on the right side. In the following experiment the Sphygmomanometer was placed on Study of Vascular Conditions during Hypnotic Sleep. 131 the left carotid and the mercury manometer was connected with the right carotid artery. The results of this experiment are shown in the following table : Experiment ^. Period during which the observations were made, 1 hour and 20 minutes. Sphygmomanometer on the left carotid. Right Carotid. Left Carotid. 153 mm. Hg. 160 mm. Hg. 144 " " 144 " " 139 " " 143 " " 130 " " 138 " " 137 " " 143 " " 138 " " 140 " 145 " " 150 " " 142 " " 148 " " 128 " " 132 " " 230 " " 134 " " Average 138.6 mm. Hg. Average 143.2 mm. Hg. In this experiment the greatest pressure was also recorded by the instrument in the left carotid. The temperature was registered by standard thermometers, read- ings being made every fifteen minutes. The rectal temperature was obtained by thrusting a thermometer up the rectum about five centimetres. The temperature was also taken of both the arms. A small mat of cotton was fastened loosely to the arm, and the ther- mometer was thrust between the cotton and the skin. The cotton prevented the slight variations in the room temperature from affect- ing the readings of the thermometer. The pulse was taken from the radial artery, the number of pulsations counted in one-half minute being doubled and the result taken as the number for one minute. The respiratory rate was obtained by simply counting the number of respirations in one minute. The subject for the experiment was placed upon a bed. He was allowed to rest from fifteen minutes to one-half hour before any records were taken. The subject was kept as quiet as possible, and all unnecessary noise on the part of the observers was avoided. When the subject had rested a sufficient time, the readings were begun, two or three readings being taken before any suggestion was given to the subject. 132 E. C. Walden. DESCRIPTION OF THE PLETHYSMOGRAPHIC CURVES. Over twenty-five experiments have been made on hypnotic sleep in the present investigation. The plethysmographic records ob- tained in the first few experiments were not so satisfactory as were the tracings obtained from the later experiments, owing to the diffi- culty encountered in so adjusting the rubber sleeve to the arm that it neither compressed the skin veins nor allowed leakage from the instrument. Although the first experiments were inaccurate in regard to the volume changes in the arm, yet the general course of the curves was the same as in the later experiments, and hence these tracings may be taken as confirmatory of the curves obtained later, in which the volume changes in the arm were more accurately registered by the recorder. The curves shown in Figs. I, 2, and 3 are the plethysmographic tracings of three experiments in this series, and they show the most characteristic changes noticed in a subject when in perfectly quiet hypnotic sleep. Many investigators, among whom are Mosso, 1 Howell, 2 Shields, 3 and Kiesow, 4 have shown that, in normal physiological conditions, the arm constantly undergoes changes in its volume, and that this is true no matter what may be the position of the body. These changes are usually small, but vary greatly in amplitude, and are attributable to mental and sensory stimuli acting upon the vaso- motor centres. The course of the plethysmographic curve from an individual in the normal waking state, care being taken that all muscular movement is absent, is in a general horizontal direction. On this curve there may appear rhythmic variations due to the respiratory movements and other longer, irregular, wave-like varia- tions which probably depend upon rhythmic changes in the vaso- motor centres. Besides these, the irregular variations due to sensory and mental stimulation, already referred to, are more or less abun- dant. With continued sensory stimulation or mental activity, the curve mounts up above the normal level, showing that there has been a decrease in the volume of the arm. In normal sleep there is 1 Mosso: Ueber den Kreislauf des Blutes im menschlichen Gehirn, Berlin, 1 88 1 ; Die Temperatur des Gehirns; Berlin, 1894. 2 HOWELL: Journal of experimental medicine, 1897, ii, p. 313. 3 SHIELDS: Journal of experimental medicine, 1896, i, p. 74. 4 KIESOW: Archives italiennes de biologic, 1895, xxiii, p. 198. Study of Vascular Conditions during Hypnotic Sleep. 133 a fall in the curve which lasts with variations as long as the sleep continues ; this fall in the curve denotes an increase in the volume of the arm. In normal physiological con- ditions, therefore, the course of the plethysmographic curve varies under different conditions, a rise in the curve following psychical activity, while normal sleep causes the curve to sink. With this summary of the changes observed under normal physiological conditions, the phenomena observed in hyp- nosis may now be described in detail. Normal Curves. Before the hypnotic suggestion was given to the subject, the curve was allowed to estab- lish its normal level for the recumbent position. This level having been established, suggestion was begun. The instant the suggestion of hyp- notic sleep was given to the subject, there was a pro- nounced rise in the curve, corresponding to a constric- tion of the arm, the curve mounting upward for from one to ten minutes. The change in the volume of the arm during this period was by no means the same in the different experiments, vary- ing, in round numbers, from ^ .S * .SP-c or? _* LO Jr S JS -M g v rt e g> S $ -0 '" J$ > > 4> C o ^ c S S *J3! E- u +j * o -5 M C W S S B o ' s 3 1 H to O H -M ja pQ !tifi f o-^-g CL ^ > S w J cr - ^ 1 S "2 S e S ,7> -M J=l Si o S rt C !s-3 S KH g O, K rt * r l!! I ^5 ^ w 2, it 134 E. C. Walden. two to eight cubic centimetres for the hand and that portion of the fore-arm within the plethysmograph. At just what point during the suggestion the subject fell into hypnotic sleep it is impossible to determine. That hypnosis does occur during this rise is probable, since the constriction of the arm corresponded, in every case, to the period during which the suggestion was given, and as soon as the sug- gestion was ended, the subject being in hypnotic sleep, there was a fall in the curve. In some of the experiments this fall was so small that it was easily overlooked, while in other experiments there was a marked fall indicating a change in the volume of the hand and fore-arm of at least ten cubic centimetres. This change, corresponding to a vascular dilatation of the arm, was never so rapid as the previous rise. The time during which the fall continued varied greatly in the several experiments. In some cases it lasted but one or two minutes, while in other experi- ments the curve continued to sink for more than two hours. The fall in the curve differed from the previous rise in another particular. The rise was as a rule continuous, while the fall was broken and often concealed for some minutes by Study of Vascular Conditions ditring Hypnotic Sleep. 135 irregular oscillations which occurred in the path of the tracing. These variations correspond in every par- ticular to the sharp vaso- motor variations noticed in all plethysmograp hie tracings in normal physio- logical conditions, and hence they are probably due to the same causes. After having reached its lowest point, the curve usually began to show a steady rise, which was quite gradual and continued as long as the hypnotic sleep lasted, that is, from two to five hours. The curve during this rise not only reached the normal waking level, but in every case it mounted above this level, the hand and fore-arm constricting in some cases to an extent equal to a diminution in volume of as much as thirty cubic centi- metres. In a few experi- ments the rise in the curve was much sharper, lasting about one hour, the change in the volume of the arm being about the same as in those experiments in which the curve rose more slowly. When the curve rose in this more rapid manner, having reached its maximal height for the rapid rise, the tracing continued 1 36 E. C. Walden. to rise much more gradually, with slight oscillations, until the end of the hypnotic sleep. At the instant hypnotic sleep was ended at the suggestion of the experimenter, there was a sharp rise in the curve, lasting about one minute, similar to, and, in many cases, as great in extent as the rise which occurred at the beginning of the hypnotic suggestion. In some cases the diminution in the volume of the hand and fore-arm during this rise amounted to as much as eight centimetres. During the rise the subject usually opened his eyes and made some movements ; these movements, however, did not per- manently affect the curve. After reaching its maximal level, the curve began to drop, and continued to fall, broken by sharp oscilla- tions, until the tracing had reached about the level it had at the beginning of the experiment previous to the suggestion of hypnotic sleep. The course of the curve from this time until the end of the experiment was in a general horizontal direction and corresponded in every way to the tracings of the normal waking curve. The general course of the curve of hypnotic sleep may then be described as follows : A sharp rise, lasting from one to ten minutes, is followed by a slower fall very variable in duration, usually comparatively brief, but, in exceptional cases, lasting for two hours. This fall is in turn followed by a gradual, long-lasting rise, continuing until the end of the hypnotic sleep, after which the curve sinks to the level it had previous to the suggestion. The general tendency of the curve during hypnotic sleep is upward, and this change in the record corresponds to a constriction of the hand and that portion of the fore-arm within the plethysmograph. Besides these general changes, other secondary variations were observed. These have already been alluded to in reference to the changes noticed in normal physiological conditions, and probably depend upon the activity of the vasomotor centre. These variations were by no means uniform ; in some cases a rapid rise of a few milli- metres was followed immediately by an equally rapid fall, while in other cases, after a fall or a rise, the curve continued for several minutes in a horizontal direction and then gradually returned to about the former level. These changes cannot be due to external stimulation, since the subject was quiet, making no movements what- ever, save those due to respiration. All noises and other forms of sensory stimuli were also excluded as carefully as possible. These oscillations are due, then, to some internal stimulus acting upon the vasomotor centre. In some of the experiments, rhythmic, wave- Study of Vascular Conditions ditring Hypnotic Sleep. 137 like variations, such as were noticed by Howell 1 in normal sleep, have been observed ; these variations were never so pronounced in the curves of hypnotic sleep as they are in the curves of normal sleep, and they were not found in all of the records of hypnotic sleep, and hence they must be considered as exceptional, depending upon the condition of the subject at the time of the experiment. Exceptional curves. In four of the experiments of the present series there were variations which have not been described in the plethysmographic curves of other investigators. These changes, which were of two different types, are shown in Figs. 4 and 5. In one case, Fig. 4, after gradually falling for fifty minutes, the curve suddenly mounted upward. In one such experiment there was a diminution in the volume of the hand and that por- tion of the fore-arm within the instrument of 16.9 cubic centimetres in three minutes, and a further 1 HOWELL: Journal of ex- perimental medicine, 1897, ii, P- 313. 138 E. C. Walden. \ \ shrinkage of 5.6 cubic centimetres during the following fifteen min- utes. This surprising constriction of the arm could not be traced to any external stimulus, nor to an alteration in the position of the hand within the plethysmo- graph. No difference could be noticed in the rate of the pulse or of the respiration. After this sudden rise, the tracing maintained the high level until the mo- ment of waking, when after the usual rise which occurs on waking, the curve sank to about the level it had at the be- ginning of the experi- ment. The sudden vari- ations already described, due t'o vasomotor changes, were much sharper and more nu- merous after this sudden rise than they were in that portion of the trac- ing preceding it. In Fig. 5, a curve of an opposite character to the one just described is reproduced. After a gradual rise, lasting ninety minutes, which in this case corresponds to a diminution in the Study of Vascular Conditions during 'Hypnotic Sleep. 139 volume of the arm of 7.6 cubic centimetres, the curve fell abruptly, the arm increasing in volume 20 cubic centimetres in seven minutes. After this sudden fall, the curve gradually rose for two hours and five minutes, rising beyond the waking level. The subject was then awakened. On waking, the tracing sank to about the level it had previous to the suggestion of hypnotic sleep. Here, again, the sharp variations of vasomotor origin differed, those preceding the fall being much sharper than the oscillations which followed it. Unfortunately the sudden fall occurred at a time when the subject was not under close observation, and the possibility of some external stimulus being one of the causes for the fall cannot be excluded. The position of the hand in the instrument could not have been altered, for had such a change occurred, the tracing would not have returned to so nearly its former level at the end of the experiment. A point of some interest in this connection is that these variations never occurred in the tracing when the subject was in the best condition for an experi- ment. He complained on each of these occasions of not having had sufficient sleep the night before, or at the time of the experiment he was suffering with headache. Effect of suggestion. A series of experiments was undertaken to determine the effect of various external stimuli upon subjects in dif- ferent hypnotic conditions. It was suggested to the subject that his arm had grown smaller or that it had increased in size, or he was told that he had forgotten his name or could not open his eyes. In other cases it was suggested that the subject would hear music dur- ing his sleep, and at certain intervals a music box was set into action. The effect of suggestion during the normal waking state was always to give a sharp rise in the curve (Fig. 6) which lasted until the suggestion ended, when the tracing gradually sank to its former level. On suggesting hypnotic sleep, the phenomena observed were the same as those already described for the hypnotic sleep curves. During hypnotic sleep, the curve having risen above the normal wak- ing level, each suggestion caused a sudden rise in the tracing of from one to five millimetres ; this rise was followed by a sharp fall in the curve. The fall was closely related to the suggestion as it invariably ended with the suggestion, the curve then gradually mounted upward until it had reached about the level it had before the suggestion was given. In suggestion during hypnotic sleep there was always a fall in the curve, and in suggestion during the normal waking state there was always a rise in the tracing, no matter what the character of the 140 E. C. Walden. I- I I I "I- OHIJKLMNOPOR I 23 45 6.7 a 9.10 FIGURE 6. Plethysmographic record of the hand and lower fore-arm, on the right side, taken December 7, 1899. This experiment was to determine the effect of suggestion during the normal waking and during the hypnotic sleep. Mr. Steele operator, and C. D. H. subject. A fall of 1 mm. in the original curve corresponds to an increase of 0-282 c.c. in the volume of the arm. The curve here presented has been reduced 57 per cent. The letters and figures show the points at which suggestions were given, and correspond to the following : A. Suggested that the subject could not open his eyes. B. Suggested that the subject could open his eyes. C. Suggested that the right arm of the subject was swelling. D. Suggested that the arm had returned to its normal condition. E. At this point the subject was left alone in the room in order to see if the curve would sink to the level it had originally. F. Came back into the room. G. Suggested that the subject could not open his eyes. H. Suggested that the subject could open his eyes. I. Suggested that the right arm of the subject was swelling. The subject at this time made a great many movements. J. Suggested that the arm was normal. K. Suggested that his right arm had diminished in size. L. Suggested that his arm was normal. M. Suggested that there was an increased flow of blood to the arm. N. Suggested that the arm was normal. O. Suggested that the arm was constricting. P. Suggested that the arm was normal. Q. Suggested that the arm was dilating. R. Suggested that the arm was normal. 1. Quiet hypnotic sleep was suggested. 2. Suggested that the arm was swelling. 3. Suggested that the arm was normal, this suggestion was followed by the sugges- tion of deeper hypnotic sleep. 4. Suggested that the arm was constricting. 5. Suggested that the arm was normal. 6. At this time the subject had a coughing fit. 7. Suggested deeper hypnotic sleep. 8. Suggested that the arm was swelling. 9. Suggested that the arm was normal. 10. The subject was awakened. This occurred at 2.45 P. M. Study of Vascular Conditions during Hypnotic Sleep. 141 suggestion given may have been. When a suggestion was given to a subject in hypnotic sleep, he became restless and acted as he would have done if the suggestion had been given to him while he was awake. The small but sharp rise followed by the fall in the curve was very similar to the rise and subsequent fall which always occurred on waking a subject from hypnotic sleep. The slight movements made by him cannot account for the fall in the curve, since the same movements were noticed when the suggestion was given during the waking state, in which case there was a rise in the curve. This fall can be more easily explained by assuming that in hypnotic sleep the voice of the operator partially awakened the subject. The small rise which preceded the longer fall also adds weight to this assump- tion. The rise which followed, the suggestion having ceased, would correspond to a return of the deeper hypnotic sleep. DESCRIPTION OF OTHER PHENOMENA. Blood-pressure curves. Mosso, 1 Frangois-Franck, 2 Gley, 3 Hill,* Colombo, 5 and others have demonstrated by various methods that muscular and mental activity, the position of the body, the time of day, and barometric conditions may all have an effect upon the blood- pressure. It is known that during normal sleep the blood-pressure is lower than it is under similar conditions when the individual is awake. As the vasomotor phenomena accompanying hypnotic sleep are the reverse of those found in normal sleep, it was decided to determine whether there was as marked a difference in the blood- pressure. It would be but natural to expect a rise in arterial pres- sure as an accompanying phenomenon to the vaso-constriction of the cutaneous blood vessels, but such is not the case. In fact, it is impossible to compare the curves representing the volume changes in the arm, taken by means of the plethysmograph, with the curves registered by the sphygmomanometer. While the plethysmographic 1 Mosso: Archives italiennes de biologic, 1895, xxiii, p. 177. 2 FRANCOIS-FRANCK : Travaux du laboratoire de M. Marey, 1877, p. 273. 8 GLEY : Expose' des donne'es experimentales sur les correlations fonctionnelles chez les animaux, Paris, 1897; fitude expe>imentale sur I'e'tat du pouls caroti- dien pendant le travail intellectuel, Paris, 1881. 4 HILL : Journal of physiology, 1895, xviii, p. 15 ; 1897-98, xxii, p. xxvi; 1898- 99, xxiii, p. iv. 6 COLOMBO : Archives italiennes de biologic, 1899, xxxi, p. 345. 142 E. C. Walden. curves are all of them more or less similar, the blood-pressure curves differ greatly from each other, and hence it is difficult to determine what the normal pressure curve is during hypnotic sleep. The characteristic changes in the blood-pressure can only be recognized by taking the observations secured in a number of experiments, and from them constructing a mean curve. The pressure observations taken in five experiments have been tabulated in the table given below. The observations were made at intervals of fifteen minutes during the time the experiment lasted, and they include the pressure TABLE I. Experiment. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 IS Dec. 13. 60 50 50 55 60 58 56 54 50 48 50 50 54 56 56 56 58 76 Dec. 15. 80 76 72 69 77 90 93 88 90 91 93 97 96 98 95 88 87102 Dec. 21. 93 80 100 94 94 69 91 81 75 72 79 76 % 97 96 101 102 103 Jan. 11. 96 87 93 98 98 95 97 98 99 98 95 99 105 Jan. 16. 70 72 77 77 71 66 68 70 72 66 68 70 72 66 91 97 99 99 Average. . . 80 73 78 78 80 76 81 78 77 71 84 84 86 S3 82 82 82 97 observations taken just before the suggestion was given, and those secured after the suggestion had been finished, as well as the obser- vations made during the period of hypnotic sleep. In the first column to the left are given the pressure readings taken just before the suggestion was given ; the following sixteen columns contain the observations taken at stated intervals during the suggestion ; and the last column contains the readings taken just after the subject awakened. In order to secure the general blood-pressure curve from these experiments; the average height of the pressure at each interval was taken, and from these figures a curve was plotted. This curve is given below. The average blood-pressure in the fingers in a horizontal position was found to be eighty millimetres of mercury. Mental and muscular activity, or sensory stimulation of any kind, increased the pressure, this increase varying from five to thirty-five millimetres of mercury according to the nature of the stimulus. In order to avoid as far as Study of Vascular Conditions during Hypnotic Sleep. 143 possible all external stimuli, the subject was kept quiet, and the observers made no unnecessary noise or movements. After one or two experiments, the subject became so accustomed to the apparatus that on lying down, he soon began to show signs of sleepiness. The pressure readings were begun when the subject had reached this condition, and the height of the mercury was assumed to be equal to the arterial pressure of the individual without marked mental or sensory stimulation. Following the hypnotic suggestion, there was a fall in the arterial pressure, which, on the average, amounted to seven millimetres of mercury. After this fall, the pressure fluttered around the normal level ; the tendency during the first part of the too 95 90 as 80 7S 70 ? 3 4 5 6 7 8 9 1 1 1 1 ? / 3 1 4- / 5 / 6 1 7 t 8 1 1 ^^^x 1 ^^^ 1 / / i ^s "- --. 1 \ / ^ \ / ^ \. / \ f \ / FIGURE 7. General curve of the blood-pressure during hypnotic sleep. The figures along the abscissa represent consecutive periods of fifteen minutes. The figures along the ordinate represent the blood-pressure in the fingers in millimetres of mercury. experiment was for the pressure to remain below the normal level, while during the remainder of the experiment the curve was above the normal level. When the suggestion of hypnotic sleep was ended, the pressure rose rapidly, the average increase in the pressure being seventeen millimetres above the pressure at the beginning of the experiment and fifteen millimetres above the pressure just before the end of the hypnotic sleep. It was never possible to continue the pressure observations for any length of time after the subject was awakened. On this account it was not determined how soon the pressure returned to the normal level after the suggestion was ended. That it does not return to this level as rapidly as the plethysmographic curve returns to its base line is 144 E. C. Walden. or evident, since the experiment was never stopped until the plethysmo- graphic curve had reached about the level it had at the beginning of the experiment, and in every case the pressure curve was still much above this level. The marked rise at the end of the curve was coin- cident with slight muscular movements made by the subject at this time, and an increased heart rate which, as will be shown, was also very striking. In hypnosis, therefore, the general course of the blood-pressure curve is in a horizontal direction, the pressure varying but little from the level it had previous to the suggestion. On waking there is a sudden rise in the curve, which is in all probability due to the effect of increased mental and muscular activity on the heart rate, as de- scribed in the next section. The pulse rate. The general pulse rate curve was determined by the same method that was used to obtain the general blood-pressure curve during hypnotic sleep. The pulse rate curves in the different experiments did not show the great differences among themselves that were noticed in the blood-pressure curves. The general char- acter of these curves was the same in all of the experiments, and only in the minor variations were there any differences. The table given below contains the observations taken in the same experiments from which the observations of the blood-pressure were obtained. TABLE II. Experiment. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Dec. 12. 58 56 57 52 58 62 58 57 54 54 54 56 56 57 57 58 62 74 Dec. 15. 64 62 62 62 60 62 60 61 64 66 65 60 64 60 54 56 61 68 Dec. 21. 62 58 55 60 64 63 64 62 58 60 58 66 67 57 69 72 60 70 Jan. 11. 68 60 60 63 57 56 52 58 60 60 60 62 60 64 Jan. 16. 56 52 60 52 52 56 52 52 60 60 60 58 60 62 Average . . 62 58 59 58 58 60 57 58 59 60 59 60 61 58 60 62 61 67 From the table it will be seen that before hypnotic sleep was sug- gested, the pulse rate was sixty-two per minute ; after suggestion there was a fall in the rate to fifty-eight per minute. During the entire period of hypnotic sleep, the rate varied about this figure, the aver- Study of Vascular Conditions during Hypnotic Sleep. 145 age rate for the entire period of hypnotic sleep being fifty-nine pulsa- tions per minute. The tendency of the pulse curve was to rise gradually, the heart beating less frequently at the onset of hypnotic sleep, than it did after the subject had been asleep for several hours. After hypnotic sleep had lasted for about three hours, the pulse rate was nearly as high as it was just before the suggestion was given. On waking the subject, the pulse increased in frequency, the average rate for the observation just after the hypnotic sleep had ended being sixty-seven pulsations per minute. In the accompanying figure is given the plotted curve constructed from the averages given in the preceding table. 9 I O /i 13 /6 67 66 65 62 <5/ 69 68 67 FIGURE 8. General curve of the pulse rate during hypnotic sleep. The figures along the abscissa represent consecutive periods of fifteen minutes, those along the ordinate represent the number of heart-beats per minute. The table and figure both show that the heart rate was slower dur- ing hypnotic sleep than during the waking state, they also show that this effect upon the heart rate was greater at the beginning of the hypnosis than it was after the hypnotic sleep had continued for a considerable time, since the rate gradually increased until at the end of four hours the heart had about the same rapidity as at the beginning of the experiment. On waking, the pulsations were more frequent than either during or before the hypnotic sleep. After waking, the rate of the pulsations did not return rapidly to the 10 146 E. C. Walden. rate previous to the suggestion, as it was always faster when the last observations were made, about fifteen minutes after the subject had been awakened, than it was at the beginning of the experiment. Respiration. It is well known that during periods of mental and muscular rest, as well as in sleep, the respiration is slower than dur- ing periods of activity. Since in hypnotic sleep, there are but few movements, it is but natural to expect that the respirations would be less frequent, and such is the case. Hoover and Sallman * have observed the changes in the respiration in a hypnotic subject who was in a hypnotic state for one week. During this period, the res- piration was slower than normal, except at those times when the subject made muscular movements. At such times there was an increase in the frequency of the respirations which these authors attribute entirely to the movements. In the present experiments, the hypnotic state was much shorter than in the experiment referred to, and but few movements were noticed. In the table given below are tabulated the observations of the respiratory rate and the average rate at each reading taken in the five experiments from which the pulse and pressure observations, already described, were obtained. TABLE III. Experiment. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Dec. 12. 16 18 18 16 16 16 17 16 16 16 16 18 17 17 17 16 17 22 Dec. 15. 19 16 19 16 16 16 16 18 20 16 12 15 16 16 20 16 16 20 Dec. 21. 22 16 16 16 16 17 16 16 16 17 15 16 16 17 12 18 16 20 Jan. 11. 18 18 18 21 18 IS 20 16 17 16 16 17 20 20 Jan. 16. 16 15 16 16 15 15 14 15 16 18 18 19 16 18 Average . . 18 17 17 17 16 16 17 16 17 17 15 17 17 17 16 17 16 20 It will be noticed that while in some of the experiments the variations were well marked, in others there was but little difference in the respiratory rate during the whole experiment. This can only be accounted for by assuming that the subject was less restless in those experiments in which the variations were but little marked than in those cases where the differences between the wakine and the HOOVER and SALLMAN : Journal of experimental medicine, il o r THE UNIVERSITY or Study of Vascular Conditions during Hypnotic Sleep. 147 hypnotic conditions were more prominent. The average figures for the experiments show that there was a fall in the respiratory rate during the period of hypnotic sleep. In the accompanying figure is shown the general course of the curve of the respiratory rate during hypnotic sleep. 5 6 8 9 /O / tf> J- \ FIGURE 9. General curve of the respiratory rate during hypnotic sleep. The figures along the abscissa represent consecutive periods of fifteen minutes, those along the ordinate represent the number of respirations per minute. At the moment of hypnotic suggestion there was a fall in the curve ; during the entire period of hypnotic sleep it ran a horizontal course with slight variations, and on waking it rapidly mounted upward, passing the level it had at the beginning of the experiment. Temperature. It has only been possible to obtain observations of the rectal temperature in two experiments. The course of the curves constructed from these observations was the same in the two experiments, and the general curve constructed from the obser- vations may be safely assumed to represent the general course of the rectal temperature during hypnotic sleep. Table IV. contains the temperature observations taken in the two experiments and also the averages obtained from these observations, and Fig. 10 repre- sents the curve plotted from the mean figures. These observations and the plotted curve of the general course of the curve of the rectal temperature show that there is but little change in the rectal temperature. The rectal temperature was always higher at the beginning of the experiment than it was at any time during the rest of the experiment. On suggesting hypnotic sleep, there was a slight fall in the tem- perature, and during the period of hypnotic sleep the temperature slowly and gradually fell. When the subject was awakened the temperature rose, but it did not reach the height it had before the suggestion was given. 148 . C. Walden. Average P '-n 1 O I Experiment. 2 3 B ~ g \c o N) s b bo - b CO 5^ CO | b C^J C^i Vl b ; H- \c SO b vc - O ^J CN co i-^ vC bo '^ ^ ^ vC s Osi b 5 g b bo ' bo \C CO to ^\ ^ bo | b $ Co 1 b 5 ^r j Sa ^J II '-n s bo bo s bo bo bo " 1 1 1 CO It may be said that the general effect of hypnotic sleep is to slightly lower the rectal temperature. Surface Temperature. The effect of hypnotic sleep upon the temperature of the skin was different from the effect upon the rectal temperature. The variations in the temperature were not the same in the two arms, the changes being less marked in the arm surrounded by the water in the plethysmograph than in the other arm. Table V. contains the temperature obser- vations taken in five experiments and Fig. 1 1 the curves plotted from the average observations of the two arms. The general course of the temperature curves in the two arms was the same. The general course of the curve, as shown in the figure, was as follows : On suggest- ing hypnotic sleep, the temperature rose rapidly, as determined by observations at intervals of fifteen minutes, this rise con- tinuing for about one hour, after which the curve slowly and steadily sank until the end of hypnotic sleep ; on awakening the subject, the curve rose rapidly, reaching a higher level than it had at any previous time during the experiment. It will be seen that the general course of the temperature curves of the skin and rectum is not the same. While the rectal temperature remains nearly uniform, sink- ing slowly though slightly during hyp- nosis, and showing but little rise on waking, the skin temperature follows in general a reverse course to that of the plethysmographic record, rising during the first part of hypnosis, sinking during the latter portion of the sleep, and rising suddenly and markedly upon awakening. Study of Vascular Conditions during Hypnotic Sleep. 149 It should be noted, however, that the general temperature of the skin during the hypnosis was above that observed at the time the sugges- tion was given. DISCUSSION OF RESULTS. Suggested explanation of the meaning of the plethysmographic curves. The changes which occur in the arm as the result of mental and muscular rest and activity have been shown to be due, in all probability, to vasomotor dilatation and constriction of the cutaneous blood vessels. Mosso 1 has clearly shown, in individuals with defective skulls, that there is a diminution in the volume of 37.1 37. 36.9 368 367 / 2 3 4- S 6 7 8 9 / / 1 / 2 / 3 / # / 5 / 6 t ? / e \ \ z \ \ / \ / \ / FIGURE 10. Plotted curve of the rectal temperature observations in hypnotic sleep. Average before, 37.1; during, 36.9 ; after hypnotic sleep, 36.9. The figures along the abscissa represent consecutive periods of fifteen minutes, those along the ordinate, the temperature in tenths of a degree centigrade. the brain during normal sleep, and he attributes this shrinkage in the volume of the brain to a diminished amount of blood in the blood vessels of the brain. TarchanofF 2 has shown that in sleeping dogs there is a fall in the blood-pressure, and that on waking the pressure again mounts upward. Since the vasomotor changes in the cuta- neous vessels are not local, but general, this fall in the blood-pressure is easily accounted for by the diminution in peripheral resistance, by the diminished heart rate and respiratory rate. Howell 3 has observed that there is a marked increase in the size of the arm and 1 Mosso: Ueber den Kreislauf des Blutes im menschlichen Gehirn, Berlin, 1881; Die Temperatur des Gehirns, Berlin, 1894. 2 TARCHANOFF : Archives italiennes de biologic, 1894, xxi, p. 318. 8 HOWELL: Journal of experimental medicine, 1897, ii, p. 313. E. C. Walden. \ sr c s s 2 c o> a n o r^ o o o < p r $ is > " 3 QJ \^' ^^ ^^ ^ ? s B P 5 P S2 K 5 is W X 1 s' OQ ? eg eg eg oo . g S vs ^8 P GO CO CC CO P vS S v* g a> 3 s --? OO OJ OJ OJ do OJ C^J C^J C^J O to b : -n * OJ CO ON OJ OJ UJ OJ Co Co CO to CO t/J b '^n CO CO Co - Co co b C^J OJ C^J OJ OJ 4^ U) C^> OJ O b 4^ I" C> '^n OJ CO o Co Co Co Co U) CO CO C/J CO OJ C\ 'In '^i 4^ Co to Co 5* C^O CA> C^ C^J C^J CM CM wi OJ p CO In ^ Cn <> u> CO vO Co Co Co Co Co CO CO C^J CO O-> O CO 'Ln CC 4* CN W f ' Co Co Co Co Oo Co CO 4^ CO Co b - Oo OJ C^J Oo Oo U> OJ -u 4- O ^ co b i^j > CM Oo CM CM CM CM CM Co Co 4^ CO Co b b b ^ co ^-n 3 to Co CM CM O4 OJ CM CM CM 4*- O ^ 00 CO U> 4^ S3 f i CM CM CM CM CM CM CM CM to CM CO O CO t-' Co ON CO U> -fi- CM 4* O ^ >1 3 is vO Co Co Co Co Co to CM 4^. co co vi Co b i' in 00 > Co OJ CM CM CM CM CM U) -^ >^> 4> C u> oj ^i u> b g Co Co Co Co Co CO Co 4>- CO CO in 4.. O H- 1 CO vO g b C^J UJ Oo t*J tO U) 4^ OJ 4- \O b '>-" ^ c^j co OJ CO ^ Co Co Co Co Co CO Co Co CO CO jv O *-4 CO CO co to vO Os> OJ OO C^J to OJ 4^- OJ 4- vC to In 4^ LM b is <-n CO CO - CO CO OJ U> U> OJ tO C^> 44. CA> 4* GO C/J ^ Co Co Co Co Co CO Co Co CO CO CO w wJ CO Co to vj OJ C/J Oo tO -^ U) -^ CO I CO CO CM In OJ CO vO Co Co Co Co Co Co CO CO I vO '-n d/J CO Co Oo to CO -^ U) -^ CO I CO b 4^ CN CM CO \C Co Co Co Co 4^ OJ CO CO I b u> 4^ co ( i 4^ CM t* -I Co oj u> ro 4- to 4^ CO I Co - CO Oo 4^ Co CO CO I O CO 4*- H- 1 i i t^i CO to CN OO U OJ CO 4^ CO 4^ CO ' vO In CM ^ U> S OJ Co C/J OJ -^ co co co 1 b b co co C> &s Cn OJ Oo Oo CO ; yi co yo po O '-n CO CO JS \o Co Co Co Co -fi- CO CO CO ! co b H-" co i "