COLUMBIA LIBRARIES OFFSITE HX64101681 QP391 .C81 A new clinical lest RECAP. QF33/ C8I Columbia ^nitiersfitp intI)p(lCilpofJlrtnfork College of ^fjpsicians anb feurseons Hifararp Digitized by tine Internet Arciiive in 2010 with funding from Columbia University Libraries http://www.archive.org/details/newclinicaltestfOOcorn A NEW CLINICAL TEST FOR TEMPERATURE SENSITIVITY A CONTRIBUTION TO THE STUDY OF TEMPERATURE REACTION IN NERVOUS DISEASES BASED ON THE REACTION TO SIMULTANEOUS COLD AND HOT STIMULATION An two parts^ ETHEL LETITIA CORNELL, A.B. DEPARTMENTS OF PSYCHOLOGY AND NEUROLOGY COLUMBIA UNIVERSITY NEW YORK CITY SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN THE FACULTY OF PHILOSOPHY COLUMBIA UNIVERSITY Colunrfjia ^HntbErgitp intteCitpof^etogorfe CoUege of ^bpSitianS anb g>urgeong Reference Hibrarp A NEW CLINICAL TEST FOR TEMPERATURE SENSITIVITY A CONTRIBUTION TO THE STUDY OF TEMPERATURE REACTION IN NERVOUS DISEASES BASED ON THE REACTION TO SIMULTANEOUS COLD AND HOT STIMULATION (in TWO PARTS BY ETHEL LETITIA CORNELL, A. B. DEPARTMENTS OF PSYCHOLOGY AND NEUROLOGY COLUMBIA UNIVERSITY NEW YORK CITY SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN THE FACULTY OF PHILOSOPHY COLUMBIA UNIVERSITY REPRINTED FROM VOL. I, NO. J, PAGES II9-I58, AND VOL. I, NO. 9, PAGES 335-372 THE NEUROLOGICAL BULLETIN PUBLISHED MONTHLY BY PAUL B. HOEBER, NEW YORK COPYRIGHT, 1918 Q"PSt)l / L A NEW CLINICAL TEST FOR TEMPERATURE ^' SENSITIVITY J^ BASED ON THE REACTION TO COLD AND HOT STIMULI SIMULTANEOUSLY APPLIED By ETHEL L. CORNELL, A.B. ~-^ Departments of Neurology and Psychology, Columbia University -' ^ NEW YORK CITY GENERAL OUTLINE Historical Experimental Statement of problem Apparatus Material and Method Subjects, Areas tested, Arrangement ol Series, Temperatures used Procedure Conditions aflfecting results Limitations of method Changes in temperature of skin or stimulus Topography of warm and cold spots Results General character of reactions to double stimuli Proportion of responses giving cold first Proportion of responses giving only one sen- sation Reaction to single stimuli Overlapping of reaction-time to cold and hot Responses giving both sensations Detailed consideration of responses to double stimuli Responses giving hot first — comparison with those giving cold first Responses giving cold first General distribution Effect of double stimulus on reaction-time to hot Summary of individual cases Conclusions HISTORICAL A review of the literature on the tempera- ture senses is a task of rather large dimensions. There are so many factors, however, influenc- ing any investigation in this field, and there is so little in the literature that bears specifically on the present problem, that I shall attempt to give a somewhat sketchy survey of the whole field. The problems attacked have been many, the methods of approach various, and the con- clusions of all possible degrees of diversity; so that it is not always easy to give a very clear presentation. Indeed, one fervently reechoes the words of Trotter and Davies: "The study of thermal sensitivity is a peculiarly difficult one, not only in regard to technical detail, but especially in regard to obtaining clear conceptions and expounding them without confusion." ^ 'J./. Psychol, u. Neurol.. 20, 1913. p. E 106. There seem to be four general classes of problems that have been investigated in regard to the temperature senses — four general methods of approach, which while not mu- tually exclusive, are sufficiently distinct to give a certain coherence to the discussion. These four classes are: 1 . The psychophysical — under which I have classed the various problems of sensitivity that occupied so large a part of the earlier litera- ture — problems of the threshold values of sen- sation, of discrimination, and of the limiting values where sensations of temperature pass over into those of pain; and the question of topography — the relative sensitivity of vari- ous parts of the body. 2. The physiological — which took its im- petus from the question of the specific energies of the nerves, because of the apparent con- tradiction that cutaneous sensations ofTered to this theory. Under this class will fall the Reprinted from the March, IQ18, issue of the Neurological Bulletin, Vol. I, No. j, pages Jig-158. Copyright, igi8 119 'IViiiix-raturc Sciisiti\ ity question of the nature and distribution of the end-organs (closely allied to the topography), the question of the nature of the arousal of the sensation and the consequent explanations of adaptation, and the question of the con- stitution of the nerves which was raised by the recent work of Head. 3. The psychological — which is, to be sure, bound up with all the rest, but under which I have chosen to consider the questions related chiefly to qualities of sensation and reaction- time. 4. The clinical — which is, to a great extent, merely the application of results obtained by other methods to abnormal cases. It has, however, set many of the problems under the other headings, and has, besides, its peculiar problems of sensory dissociation to treat. PSYCHOPHYSICAL INVESTIGATIONS It is quite fitting to begin from the psycho- physical point of view, since the very first recorded observations are those of Weber (24), that a metal cylinder of 15° C. felt colder on some parts of the body than on others, and similarly, one of 40° felt warmer in some places than in others. Fechner, also, did not omit to test temperature sensitivity to corroborate Weber's law; but these experiments we may pass over, to consider the more specific attempts to measure temperature sensitivity. Threshold values may here, as in other cases, be of three sorts, although our authors have not always distinguished them very clearly. The stimulus limen (R. L.) would correspond to the least difference from the skin temperature that could be sensed; the difference limen (D. L.) would correspond to the least noticeable difference between two stimuli at other than skin temperature; the terminal limen (T. R.) would be the limiting value where sensations of temperature ceased to be felt as such. The first experiment of this sort was that of Nothnagel (16), in 1866. He used two small cylindrical vessels filled with water, each fitted with a thermometer dipping into the water. The water was heated or cooled to the desired temperature by inserting into it a large key previously heated or cooled. The two vessels were applied successively to the same spot, and the discriminati%e interval determined. He made out discrimination tables for various parts of the skin surface, as a starting point for the investigation of pathological conditions, in which he was primarily interested. Discrimination, which he found to be best between 27° and 33° C, varied from .2° on the forearm, upper arm, and cheek — the most sensitive parts — to 1.2° in the middle of the back. At that time, of course, the separation of thermal sensations into two separate series had not been thought of; consequently, Nothnagel's values may be sometimes true difference thresholds and sometimes may represent the absolute thresh- old, or least noticeable sensation for cold or warm. After the distinction had been made, Eulenberg made an attempt to determine the least differences above and below the skin temperature that could be perceived as warm and cold respectively (7). Eulenberg's ap- paratus was somewhat more refined. He used two thermometers with flattened bases, one of which had its base wound with a platinum wire through which a current could be sent to raise the temperature of the base. The heat was controlled by a rheostat. One ther- mometer was placed on the skin and left until it registered the skin temperature; the other was then heated by the current to a point slightly above or below this and set down in the same place. Minimal differences perceived as cold and warm respectiv^ely were thus determined for \'arious areas. The difference R. L. FOR Cold and Warm (Eulenberg) Minimal Minimal Cold Warm Difference Difference Sternum .2° .2° Fingers — dorsal .2 — •3 .2 Ball of thumb ■:•,— •4 •4 Back of hand ■3— •4 •4 Forearm — dorsal •3— •4 .2 Forearm — volar •4 •3— 4 Cheek •4 •4 Upper arm •5 •5 Top of foot •5 .5- -6 Leg — upi>er part ■5 — .6 .5- .6 Leg — middle part •7 •7 Thigh — ventral .6 .5- .6 Back — lateral •9 .9 — 1.0 Back — median I.O I.I Temperature Sensitivity for each sense was found to be practically identical for the same area of the skin, and to vary, for different areas, from .2° C. in the most sensitive parts (stern urp, back of fingers) to l° or i.i° on the back. The third sort of threshold — the pain threshold for thermic stimuli — was first inves- tigated by Donath, whose work was published in 1884 (6). He contrived two instruments, one for investigating cold-pain, which he designated kn,'algimeter (xp^o^ — cold, oKyos — pain), the other for warm-pain, which he called thermalgimeter. The kryalgimeter was a thermometer ending in a spiral mercury vessel which fitted exactly into an opening in a piece of heavy flannel. This was placed on the skin and cooled by spraying it with ether — the flannel protecting the surrounding skin — until a sensation of pain was felt. The thermalgimeter was a similar thermometer, ending in an elongated bulb which was wound with a platinum wire through which a current could be sent. This was placed on the skin, and the temperature increased until pain was reported. He found noteworthy individual differences among his six normal subjects and wide variations, according to the part of the skin stimulated, from — 11.4° to 2.8° C. for the appearance of cold-pain, and from 36.3° to 52.6° for warm-pain. In general, ventral surfaces were more sensitive than dorsal, upper extremities than lower, left side than right; the ends of the fingers were the least sensitive. He concluded that the pain aroused by extreme degrees of temperature is a much grosser sensation than that from the middle intensities, and lacks the fineness of discrim- ination of the latter. The values that Donath obtained for cold-pain seem extraordinarily low, and suggest that his method may have overlooked the effect of adaptation: he does not state how long the contact with the skin continued. He does, however, say that cold- pain is much more difhcult to recognize and define, and arises more slowly than warm. Neumann, in 1903, investigated the same question, in regard to the warm sense only, in order to compare the interval between the sensation of warmth and that of pain (15). He used a metal cylinder partly filled with water that could be heated by an electric cur- rent. A thermometer was inserted into the water and attached to the handle; the handle also had a little plunger by which the water was kept constantly circulating, in order to assure an even temperature throughout. Neumann found that while the threshold for warm might vary as much as two degrees C. for the same individual in the same place at different times, the threshold for pain was fairly constant for the same place and varied according to the area stimulated from 35° to 52° — values practically identical with Do- nath's. Both Donath and Neumann were interested chiefly in pathological conditions, and their work will be referred to later. The other question pertaining to sensitivity is that of the relative sensitivity of different parts of the body. Goldscheider was the first to make a systematic attempt to experi- ment upon this question. His earliest experi- ments were published in 1887 (9a). He ex- plored the skin with a small metal surface that had been heated or cooled, and divided sensitivity into twelve degrees of intensity for cold stimulation and into eight for warm. It is clear, of course, that the actual tempera- ture of the skin in this case is of great impor- tance, and also the proportion of warm and cold spots in the various areas. Alrutz re- peated topographical experiments after adapt- ing each area tested to a temperature of 32° (2a). He was able to differentiate only three degrees of sensitivity; but he used a stimulus temperature of 34.5°, in order to have it below the point at which the cold nerves are para- doxically stimulated by high temperatures, while Goldscheider had used a temperature of 45° — 49°.- Such radically different modes of procedure could not, of course, arrive at the same results. The truth probably is that these differences of sensitivity are not pecu- liarly areal differences, but are caused by differences in skin temperature and by the presence of more or less strongly reacting spots in certain areas. Whether these spots have any constant topographical distribu- tion is doubtful; in any case, it could only be - The difference is more tiian one of metiiod. It is due to the divergence in the views of Goldscheider and Alrutz as to the nature of the sensation of heat — the former main- taining it to be just an intensified warm, the latter a fusion of warm and paradoxical cold. Temperature Sensitivity determined by a systematic exploration, point by point. PHYSIOLOGICAL INVESTIGATIONS It was originally supposed that the same nerves conducted all cutaneous impressions. It was not until 1883 that Blix, who was in- terested to determine how well-founded was the contradiction that cutaneous sensations offered to the theory of the specific energies of the nerves, discovered that sensations from various kinds of cutaneous stimuli were lim- ited to permanent spots — among the sensa- tions, those of cold and heat (3). The same discovery was made independently at about the same time by Goldscheider (gb), and by Donaldson (5). Blix used an induced current with a pin for the stimulating electrode, and found that distinctly localized spots responded by sensations of pain, pressure, cold, and heat, and that these spots were never superimposed. Other methods of stimulation confirmed the results, and it was further observed that a cold spot never responded to a warm stimulus, nor a warm spot to a cold stimulus. Blix con- cluded that there were separate cutaneous nerves with separate terminations in the skin. He mapped out various regions of his own body for temperature spots and found them arranged irregularly in groups, distributed fairly uniformly over the body, most closely on the dorsal side of the hand and fingers, most scattered on the sacral region of the back, with cold spots always more numerous than warm. Donaldson's observations con- firmed Blix's. Goldscheider found a much more numerous count, but his method was probably at fault, since later work has cor- roborated Blix. The latest and most com- plete observations were those of Sommer (19), reported in 1900. He found (for the benefit of anybody who is statistically minded) a total of half a million cold spots and thirty thousand warm — an average of 12-13 cold spots and 1-2 warm spots per square centimeter. Donaldson made the earliest attempt to discover the end-organs associated with the sensations of warm and cold. He made a histological examination of a piece of skin that had previously been explored and marked and then excised, but his findings were entirely negative. He evidently did not cut deep enough, for he found that the recovering area was perfectly sensitive. Von Frey, whose work was published about ten years later, has done the most constructive work in the dis- covery of the nerve-endings (23). It is now generally supposed that the end-bulbs of Kraus mediate cold sensations, and the Ruffini corpuscles those of warmth. It is generally assumed that the warm end- organs lie deeper than the cold, because of the greater difficulty in localizing warm sensa- tions and because of their longer reaction- time. Thunberg pointed out that these assumptions were not sufificient proof ; both con- ditions might be due to the nature of the end- organ — Goldscheider indeed had assumed that the warm end-organs had a longer latent time. Therefore Thunberg undertook an investiga- tion of the question by an indirect method {20b). He applied silver plates of the same area but of varying degrees of thickness to the skin. The principle, if I understand it correctly, was as follows: Two plates of the same area and temperature, but of different thickness, will add warmth to, or withdraw it from the skin in different amounts, in a measurable proportion to the thickness. Since the skin-temperature is a balance between the temperature of the blood and the outer tem- perature in contact with the skin, the amount of warmth added to — or withdrawn from — the skin until the different layers of the skin reach a new temporary balance will be a measure of the depth at which the balance is attained. Consequently, the application of a very intensive stimulus of less thickness will have a greater effect on the more superficial layers than on the deeper-lying laj'ers of the skin; while on the other hand, a stimulus of less intensity but greater thickness will give a relatively favorable opportunity for the stimulation of the deeper layers. Since then, as Thunberg found, the more intense, thinner plate provided the optimum conditions for the arousal of paradoxical cold sensations, while a stimulus of less intensity but of greater mass offered the best conditions for obtaining a strong warm sensation, he concluded that the cold nerve-endings lie more superficially, the warm nerve-endings deeper. Temperature Sensitivity 123 The question of the depth at which the end- organs lie is closely connected with the ques- tion, somewhat theoretical but nevertheless important in its practical bearing, of the nature of the arousal of thermal sensations. Does the end-organ change its temperature or does it remain constant? If it changes, do the changes take place in the same direction as the changes in the outer skin-temperature or in the opposite direction? It is generally considered that, since the skin-temperature is found to be differently tempered for chang- ing conditions of temperature without arous- ing any sensation, there is a zone of indifference — usually called the "physiological zero" — which is capable of shifting up and down the scale within rather wide limits. But Thunberg points out (20c) that we have no proof that the temperature of the end-organs is actually the same that w^e obtain when we measure skin-temperature, or that it necessarily changes in the same direction. What we get by measur- ing the skin-temperature is a balance between the different layers of the skin, which are normally cooled from without by the air and warmed from within by the blood. But when the outer layers are cooled by contact with the cold, the deeper layers, according to Thunberg, have their temperature raised by an increased flow of blood to restore the bal- ance; so a fall of temperature toward the sur- face takes place. Somewhere between, there may be a layer that maintains its own temperature. Since it is not known in what layer the end-organs are, it cannot be said what is the nature of the change. There are two views current: that of Weber, that tem- perature sensation is due to change of skin- temperature (or, as amended by Thunberg, to change in the layer of the end-organ) and lasts as long as the change lasts; and that of Hering, that temperature sensation is due to the actual temperature of the end-organ and lasts until the physiological zero reaches a new location — the temporary temperature of the end-organ. According to Hering, the shifting physiological zero — the adaptation to any actual temperature of the end-organ — is what accounts for the absence of sensation, though just what the physiological zero is is not clear. According to Weber, any relatively stable temperature following a period of change is the cause of the absence of sensation. The difference in the two views is due partly to differences in their findings as to the duration of sensation. An experiment to determine the duration of a sensation with a constant stimulus was made in 1903 by Holm (10). He found that the duration of the sensation decreased as the temperature of the stimulus approached that of the skin, and he thought this suggested that the duration of the sensation probably agreed with that of the temperature change, since prolongation of the stimulus did not pro- long the sensation. He found no after-images with stimuli of middle intensities, but extreme cold gave rise to an after-image which ap- peared some moments after the removal of the stimulus — due. Holm thought, to paradoxical cold sensations occasioned by a quick warming of the cooled cold-organs by the blood. The absence of after-images he thinks is a further support of Weber's theory; because if the absolute temperature were effective, one would expect after-images with weak as well as with strong sensations. Duration of Sensation (Holm) Cold Temperature Average Duration 30° 31 sec. 25 47 20 72 15 112 10 165 5 Warm 210 •peratui 'e Average Duration 40° 126 sec. 45 152 There is a recent study on the effects of adaptation by Edwina Abbot, which suggests a somewhat different conclusion (i). She made four groups of experiments for four con- ditions of adaptation. In the first group her subjects had their hands adapted to a tem- perature of 40°, by immersion in warm water. Then the D. L. was determined by dipping the two hands into separate jars, for 37.5°, 35°, 32.5°, 30°, 27.5°, 25°, 22.5°, 17.5°. She found 124 Temperature Sensitivity a "consistent drop" in the D. L. at 32.5°. In the second group, she determined the D. L. for 40°, 32.5°, 27.5°, 25°, after adaptation to various temperatures. She found that the limen for 32.5° varied hardly at all under the varying conditions of adaptation. This seemed to point to 32.5° as a condition of adaptation suitable for a norm by which to compare the limena for different temperatures under other conditions of adaptation. In the third group, to test this, she compared the D. L.'s for various temperatures after adapta- tion to 40°, 32.5°, 17.5°, and found that the D. L.'s for 32.5° under the three conditions of adaptation were within .5° of each other, while for other temperatures the D. L. rose after adaptation to 40°, and fell after adaptation to 17.5°. The fourth group was undertaken to test the possibility of a double adaptation — an adaptation to both extremes at once. In this group, she compared the effect on the D. L.'s for various temperatures of adaptation to 40°, with the efi'ect of a double adaptation, to 40° immediately followed by adaptation to 17.5°. The values for the D. L. were practi- cally alike in this experiment for extreme tem- peratures, but between 35° and 27.5° the D. L. was higher after double adaptation, except again, at 32.5°, where the value was about the same for both conditions. This was considered to confirm the hypothesis. All of Dr. Abbot's results suggested that varying conditions of adaptation had practically no effect on sensitivity at normal skin-tempera- ture, and she concludes that "persistence of skin-temperature as a more or less constant basis for appreciating variation of temperature is a much more possible conception than the . physiological zero," "since it does not fit the facts . . . and since the facts it at- tempted to explain may be explained without it." Dr. Abbot's observations were not very numerous, however, and her results are sug- gestive rather than conclusive. There is one other question to be considered under the physiological caption — the question of what constitutes an adequate stimulus for warm and cold spots. Blix had first used an electrical stimulus, and found both warm and cold spots sensitive. Kicsow (11) found tem- perature spots likewise sensitive to electrical stimulation, anil to mechanical stimulation — if the observations were carefully made, if the observer was sufficiently practised, and if conditions were otherwise favorable. That is, he obtained responses to mechanical stim- ulation in nine out of thirty tested cold spots, in twenty-one out of forty-si.x trials; in ten out of thirty trials with warm spots for one observer, five out of fifteen for another, and seven out of ten for a third. The percentage of successes is not startling, but is enough, Kiesow thinks, to prove mechanical stimula- tion possible. Goldscheider (9c), Alrutz (2c) and others have confirmed the possibility of mechanical stimulation, though Head (18) finds no evidence of it. The earlier investi- gators had found that cold spots never re- sponded to warm stimuli, nor warm spots to cold. Kiesow tested the spots by means of the "opposite stimulus" and got a rather curious result. He never got a cold sensation by cold stimulation of a warm spot, but, on the other hand, he found scarcely a cold spot that would not respond to temperatures of 47° to 50° by a sensation of warmth. This, Titchener says, is a misreading of the facts: "In reality, radia- tion occurred, and the process observed was not that of warmth but of heat." ' However, paradoxical sensations of cold — i. e., cold sensations from a warm stimulus applied to a cold spot — were discovered by von Frey (23), and confirmed by Thunberg (20a), Alrutz (2d), and are admitted by all. Para- doxical warm sensations have been described by Goldscheider, but no other investigator has been able to discover them. PSYCHOLOGICAL INVESTIG.\TIONS It is perhaps somewhat arbitrary to attempt to separate the psychological questions from the physiological, but I have reserved this designation for the investigations relating chiefly to the question of the nature and qual- ity of the sensation . of heat — the question that has loomed largest on the psycholog- ical horizon and that brings about a trian- gular debate between Thunberg, Alrutz, and Goldscheider — and for the determination of the reaction-time to cold, warm, and hot stimuli. 'Titchener. 'Quantitative Instructor's! Manual.' p. 86. Temperature Sensitivity 125 Paradoxical cold sensations are the crux of this discussion. While everybody admits their existence, there is considerable diversity of opinion as to the exact manner of their arousal and the interpretation of their significance. Goldscheider, who originally believed they were due to mechanical stimulation, now ad- mits that they are truly paradoxical, but diffi- cult to arouse except from very sensitive cold spots (gd). He says they never appear with a stimulus of less than 38°, and are not a frequent phenomenon with temperatures below 4i°-45°. Alrutz and Thunberg, how-. ever, find paradoxical cc5ld sensations a con- stant phenomenon from temperatures of 36°- 38° and up (2b, 20a). Goldscheider also re- ports paradoxical warm sensations from cold stimuli (of 2i°-27.5°); consequently, the significance of paradoxical sensations for Goldscheider is simply as another means of confirming the specific nature of the end- organs. In addition to paradoxical cold sen- sations by point stimulation, Alrutz and Thunberg showed that they could be obtained also by areal stimulation with an object at a fairly high temperature, if the area was first adapted to moderate warmth for two minutes. These facts, and the absence of paradoxical warm sensations, in their observations, led these tAvo investigators to the belief that par- adoxical cold sensations were an essential element in the sensation of heat. Alrutz found two facts in his earlier work that con- firmed this: first, that on areas of the skin where cold spots are lacking, stimulation by any degree of heat produces warm sensations and pain sensations, but never heat; second, that on areas where cold sensitivity was strongly developed but warm sensitivity weak, sensations of heat could be experienced but never strong warm sensations. In a later paper (2a), Alrutz added a third proof that the hot sensation involved the stimulation of cold nerves, from experiments in reaction-time of warm, hot, and paradoxical cold sensations. Alrutz found that hot sensations had a char- acteristic quality, which he called Anschwel- lung — a sort of crescendo effect from the first perception of warm to the full perception of hot. This meant that the reaction to warm was quicker than that to hot. The average reaction-time for warm at a temperature of 42.5°, is 3850-; for hot at about the same temperature, 7950-; for warm at 43-44.5°, 2670-; for hot at 43-46.5°, 6240-. This longer time for the perception of hot Alrutz finds to be parallel with the reaction for paradoxical cold sensation — 7450- for a stimulus of 42°, 7500- for a stimulus of 51°. Alrutz's figures are based upon rather few observations, but he compares them with those that Goldscheider had previously obtained, and finds them quite similar. Goldscheider had found a quality in the sensations he calls warm, similar to the Atischwellung of Alrutz's hot sensations — that is, a gradual rise from the "minimum" to the "maximum" of intensity. Reaction-Time to Hot (Goldscheider, Alrutz) Goldscheider Average Temperature Minimum Maximum 35° 7 SO'' 39 371 1,105'^ 40 345 777 44 282 585 49 283 456 Alrutz Temperature Average Time 36.5° 805' of the constitu- tion of the nerves; but Head, too, while he separates the mechanism for warm and hot impressions, postulates a simple sensory "heat spot." Cf. p. 128. Temperature Sensiti\ity 127 w arm Reaction Mini muti Maximum •750 sec. •371 1. 105 sec. •345 •Ill .282 ■585 .283 •456 Stimulus-Temperature 35° 39 40 44 4Q.5 It may be noted that for cold there is a marked drop in the reaction-time at 15°, while for -warm the most marked drop occurs at 39°. The latter point is very much nearer the skin- temperature than the former — a result that, if constant, is a little puzzling. CLINICAL IXVESTIGATIOXS Following Fechner, interest was much stim- ulated to the investigation of thermal sensa- tions. Nothnagel pointed out in 1866 that temperature sensibility was the first to be disturbed in diseases of the brain and spinal cord and in neuroses. He established norms of discrimination for various parts of the body, as has already been explained, with which clinical cases might be compared. He found hypo-esthesia to be the general rule — a raising of the threshold for discrimination. Often, however, in pathological cases, an apparent hyperesthesia occurs — an increased effect of a given temperature, apparently greater sen- sitivity — while discrimination remains normal or is reduced. This method did not prove very valuable, being too laborious and unneces- sarily fine. Donath next devised a method, on the as- sumption that the important thing clinically was to discover where temperature sensations passed over into pain. The method has been described. He applied it to the examination of thirteen cases of tabes, but curiously, he did not examine the same cases for both the warm-pain-threshold and the cold-pain-thresh- old. He found that warm-pain usually ap- peared at a higher temperature than normally, and cold-pain usually (though not always) at a lower. He concluded that the pain aroused by cold and warm was a much grosser sensa- tion than the perception of moderate degrees of temperature and lacked the possibilities of discrimination of the latter. Neumann also investigated the pain threshold for warm stimuli. In four cases of tabes he found the thresholds for warm and for pain higher than normal; in four cases of spondylitis he found two hypoesthetic in the affected region; in six cases of hysteria he found three normal, two hypoesthetic, and one hyperesthetic. Goldscheider thought that the sensitivity of different areas would be a good clinical test (9a). He made a topographical map of the body on the basis of the intensity of sensation from the same stimulus, and then, in cases of disease, compared parts that normally had a constant relation to each other, to see if the sensitivity was affected. Obviously, the lack of any control for the introspective com- parisons of clinical subjects would make this method of slight value. The next clinical attempt was Leegaard's, who took as his point of departure the fact of the indifference zone — the region of the scale at and near the skin-temperature which is not felt as either cold or warm (13). He used little copper cylinders which were applied to the same spot for five seconds each, begin- ning at a temperature of 20°-25° C, and ascending by steps of .5° until the lower and upper limits of the indifference zone were reached. He found that normally the indif- ference zone was not over 1°, but in cases of brain and cord lesions and peripheral affec- tions that he investigated, it might vary to any extent. In one case of hemiplegia, for instance, he found an indifference zone of 26°, in a case of transverse myelitis, an inditference zone of 14° in the affected region. These are rather interesting, because, while Leegaard did not differentiate two senses, and while his procedure was somewhat questionable, his experiments in a way anticipate those of Head. The possibility of using tests of cutaneous sensitivity as a means of differential diagnosis, in the case of lesions of the nervous centers, is partly' effect and partly cause of the more recently increased knowledge of conduction paths. Laehr did some important early work in distinguishing different types of partial sensory dissociation (12). In particular he pointed out the so-called syringomyelic type of dissociation, which is characterized by a dis- turbance of pain and temperature and the pres- ervation of contact and pressure sensations. By 1905, four distinct types of dissociation 128 Temperature Sensitivity had been described. Marinesco (14) classifies them as follows: (i) Deterioration of all forms of cutaneous sensi- bility. (2) Abolition of pressure and vibratory senses (or one of these with conservation of all other forms). (3) Abolition of thermic and pain senses with con- servation of all others (syringomyelic type). (4) Impairment of thermic, tactile, and pain sen.ses with conservation of pressure and vibratory sensations. An entirely new direction of work in the sphere of sensory dissociation has been sug- gested within the last decade by the work of Head and his colleagues. Head had the radial nerve in his arm divided, a small portion excised, and the ends sutured. Then he studied the return of sensation to the area of distribution of the nerve. His most complete report (18) was published in 1908. His find- ings were that sensitivity returned gradually: first, to pressure, cutaneous pain, and ex- tremes of heat and cold, accompanied by the phenomena of intensification and peripheral reference of the sensation; and after a con- siderable interval (160 to 400 days) sensitivity to light touch, to moderate degrees of tem- perature, and the discrimination of two points simultaneously applied. His conclusions are familiar: that the skin is supplied by two systems of nerves — one providing for a more fundamental, reflex, diffused response to pain- ful stimuli and the extremes of heat and cold ("protopathic sensibility") ; the other produc- ing localized responses, discrimination of light touches, of two points simultaneously applied, and of intermediate degrees of temperature ("epicritic sensibility"). More specifically, in regard to temperature he found that on the recovering area of the hand that was in a "protopathic" condition, punctate stimuli of about 17° C. were the first to be sensed; a little later, a stinging sensation in response to punctate stimulus of 50°, and then a sensation of heat from a stimulus of 45°. As sensitivity increased, the limits approached slightly nearer, but at no time while the hand was in a protopathic condition {i. e., while it was in- sensitive to cotton wool and unable to dis- criminate two compass points), did any cold spot react to a temperature above 26° (e.xcept paradoxically to a hot stimulus), nor did any "heat spot" react to any temperature under 37°. Only particularly sensitive spots reacted to these temperatures: the usual limits were about 20° and 40°. But simultaneously with the restoration of sensitivity to cotton wool, an areal stimulation (by a surface of 1.25 cm. diameter) of 36° to 34° gave a warm, well- localized sensation from parts where no "heat spots" had been discovered. Still, however, punctate stimulation of the "heat spots" gave no sensation if the temperature was under 38°. With cold, the experience was not quite so clear but seemed to Head analogous. With the appearance of the finer discrimination, no increase in the number of heat spots was ob- served; on the contrary, the separate spots lost their increased vividness. The recovered sensitivity to moderate stimuli. Head thinks, cannot be due to the regeneration of the nerve, because he finds that a similar condition ob- tains on the normaj skin. If he marked out all heat and cold spots in a normal area, and then applied a stimulus between 26° and 37° to a small area where no spots had been found, such area would nevertheless respond with the appr&priate sensation. He concludes there- fore, that there are two mechanisms, one punc- tate, which responds to teinperatures below 26° and above 37°; the other not punctate, which responds to areal stimulation between these two points. The facts of adaptation Head finds offer further proof of different mechanisms. That epicritic sensibility depends on the tempera- ture to which the skin is adapted, while protopathic sensibility does not, he concluded from the following observation: If both hands were adapted to water at 50°, a copper block at 29° felt cold to the normal hand, but gave no sensation on the protopathic area; simi- larly, after adaptation to 0°, it felt warm to the normal hand, but indifferent to the proto- pathic area. This is interpreted as meaning that the protopathic condition does not undergo the shifting of the threshold. When both hands were adapted to a cold of 10°, a paradoxical condition arose where 22° pro- duced no sensation on the normal hand but gave cold sensation on the affected area. This is another evidence, Head says, of the fact Temperature Sensitivity 129 that the protopathic parts are incapable of adaptation, and so allow the cold spots to react to 22°. This experiment has been repeated by sev- eral investigators. Franz (8) supported Head's contention that different sets of nerves pro- vide for the reception of extreme and medium stimuli, but he regards the difference in sensation as due to an overlapping of the ner\'e supply — a possibility that Head rejects. Trotter and Davies repeated the experiment imder elaborate and stringent experimental conditions (21). The results they obtained led them to conclude that all the observed phenomena were phenomena of hypoesthesia due to the regeneration of the nerve fibers and that all forms of sensitivity tended to re-appear at about the same time in a hypo- esthetic form, except sensitivity to hot, which was somewhat delayed. Their observations in regard to sensations of temperature were prac- tically identical with Head's, but their inter- pretation is quite different. Normally they say there is a zone of the temperature scale in the neighborhood of the skin temperature which is indifferent and extends over about five degrees of the scale. '" This indifference zon« can be shifted up or down by warming or cooling the skin. As the distance between the skin-temperature and that of the stimulus is increased, the intensity of the sensation is increased, and it is generally necessary to have a difference of fifteen degrees before the maximum sensation is elicited. Now if there is a general reduction in the thermal sensi- tivity of the skin, this will produce an increase in the range of the intermediate insensibility, and a diminution in the" intensity of the sensa- tion produced by various stimulus tempera- tures. As an illustration: Suppose on the normal skin a stimulus of 32° is indifferent, 35° slightly warm, 38° warm, and 50° hot; if the skin is hypoesthetic, then 32° and 35° will both be indifferent, 38° faintly warm, 50° warm, while hot perhaps cannot be elicited at all. This is what really happens, according to Trptter and Davies, in the hypo-esthetic zone of the recovering area, and partly ac- counts for the delayed reaction to hot. Nor- ' This is a much wider zone tlian the results of any of the earlier investigators imply. maily, a stimulus of 50° produces a sensation of hot, accompanied by a stinging quality; in the hypoesthetic area, before any thermal quality is recognizable, a stinging sensation is produced by a stimulus of 50°. Trotter and Davies' experiment was very carefully done and seems adequate to the facts it explains, but it fails to account for the cases reported by Head where epicritic sensibility remained after the protopathic type was destroyed. There is some evidence also to support Head's theory that the two sorts of sensibility are conducted by different spinal paths. Ranson has attempted to show histologically that the fibers that serve epi- critic impulses are myelinated, enter the dorsal columns of the cord from the larger ganglia of the dorsal roots, and ascend on the same side; while those that serve protopathic im- pulses (pain, and extremes of temperature) are unmyelinated both peripherally and centrally, enter the cord from smaller ganglia, cross im- mediately to the other side of the cord and ascend in the lateral columns. Temperature impulses, however, are an exception to the uncrossed epicritic tracts, and, to provide for them, a few of the myelinated fibers were found to cross (17). The literature leaves us then, in a rather inconclusive state as regards the nature of thermal sensations and the physiological con- ditions of their arousal. Authorities differ, on the one hand, as to whether hot is a simple sensation or a fusion involving both warm and paradoxical cold sensations; and, if a fusion, whether it can be analyzed or not. On the other hand, they are not agreed as to whether the facts of dissociation are to be explained by assuming different sensory mechanisms for extreme and medium stimuli, or by an overlapping nerve supply, or by a greatly diminished sensitivity. More clinical evidence of a searching character is needed before positive conclusions can be drawn. EXPERIMENTAL STATEMENT OF PROBLEM The problem that I here present in its pre- liminary results is that of devising a clinical test of temperature sensitivity that may be I30 Temperature Sensiti\ity of some ditignostic \alue in certain diseases of the nervous system. The test used is a test of reaction-time to two distinctly different stimuli applied simultaneously. It was sug- gested by Dr. Tilney that a disturbance in the response to simultaneous stimuli of cold and hot might be apparent at an earlier stage than would appear from tests of either sensa- tion alone, and might consequently be of early diagnostic value. This part of the experiment is concerned only with the responses of nor- mal subjects, and the problem specifically stated becomes: \\hat is the nature of the normal indixidual's reaction to two stimuli, one cold, one hot, applied simultaneously to immediately adjacent areas of the skin, in various regions of the skin surface? APPARATUS The stimulus was applied by means of little blocks of soft iron, attached to wooden han- dles, bearing a letter or number for ease of identification. These blocks were of two sizes, to suit the areas tested — the larger ones were two inches long, one and one-half inches wide and one-quarter inch thick; the smaller, one inch by three-quarters by one-quarter. There were ten of each size. The blocks to be heated were kept in a bed of sand in a copper drying-oven, into the roof of which a thermometer was inserted — held by a cork — which reached to the bed of sand and recorded its temperature. The oven was heated over a Bunsen burner, and could be maintained at an accuracy of tw-o degrees Centigrade range, which was considered suffi- ciently accurate for the purposes of the experi- ment, at least in its preliminary stages. It required a good deal of attention, howe\er, to keep it at the desired temperature. The blocks to be cooled were placed likewise on a bed of sand, in a hea\"y zinc bo.x which rested on a wooden rack, beneath which broken-up ice was placed. The box was also fitted with a cork into which a thermometer was inserted, reaching to the floor of sand. The accuracy of this device likewise varied within two de- grees or less. Before the testing began, a rough attempt was made to determine how long, under the conditions of the experiment, a block would retain its heat. A hole was made in one of the blocks so that the bulb of a thermometer would fit into it, and some tests were made with this block. It was found that in a half- minute (which was as long as a block was ever kept out of the oven), of which time it was in contact with the skin from five to ten seconds, the block lost less than half a degree Centigrade, but that after returning the block to the oven it was four or five minutes before the block regained the oven-temperature. Enough blocks were provided so that no block Was used oftener than once in five minutes, and only rarely oftener than once in seven or eight minutes. It was desired for the purposes of the experi- ment to time the reactions in some simple fashion that would be suitable to use for clin- ical purposes. A stop watch could not be used, since both the experimenter's hands were occupied with the blocks; therefore a Thomas poKgraph was used, so arranged that the experimenter could record the reactions with her foot. A tambour was so fixed that a light tap upon it with the experimenter's toe was transmitted to the writing lever of the instru- ment and made a mark upon the record paper. The polygraph was provided with a Jacquet chronograph, which marked fifths of a second. Writing pens were used, thus obviating the necessity of smoked paper, and the record could therefore be marked permanently- as it was made. MATERIAL AND METHOD Five women acted as subjects for the experi- ment. They included a janitress of a flat and a college instructor; they ranged in intelligence from mediocre to superior ability, and in age from twenty-one to fifty years. They had no known nervous disturbances — except a slight neuritis reported once by one subject — al- though one subject was regarded as of a nervous temperament. These five subjects were tested on forty test spots previously determined, each area being tested four times with the double stimu- lus — both cold and hot. An enumeration of the spots follows: Area i. — Fingers of right hand, dorsal sur- face. Area covered by two small blocks (held Temperature Sensitivity 131 with the long sides juxtaposed and the long axes running in a radio-ulnar direction) on the index and middle fingers, closely peripheral to the metacarpal bones. Area 2. — Fingers of left hand, dorsal sur- face. Same area. Area ^. — Right hand, dorsal surface. Area covered by two small blocks (held in same position) immediately proximal to second and third metacarpal bones. Area 4. — Left hand, dorsal surface. Same area. Areas. — Right forearm, dorsal surface. Area covered by two large blocks (held with the short sides in juxtaposition and the long axes running in a peripherocentrad direction) which were set down proximally to a line drawn in a radio-ulnar direction one inch above the protuberance of the ulna, the long axis of the blocks coinciding with a line inter- secting the first at right angles, one inch radiad to the protuberance of the ulna. Area 6. — Left forearm, dorsal surface. Same area. Area 7. — Right arm, lateral surface. Area covered by two large blocks (held with the long sides juxtaposed and the long axis at right angles to the axis of the arm) which were set down proximally to a line drawn two inches above the olecranon process of the ulna, the short axis of the blocks coinciding with a line joining the olecranon process of the ulna and the acromion process of the scapula. Area 8. — Left arm, lateral surface. Same area. Area 11. — Fingers of right hand, ventral surface. Area corresponding to that of Area i . Area is. — Fingers of left hand, ventral sur- face. Same area. Area 13. — Right hand, palm. Area corre- sponding to that of Area 3. Area 14. — Left hand, palm. Same area. Area iS- — Right forearm, ventral surface. Area corresponding to that of Area 5, but measured from protuberance of radius. Area 16. — Left forearm, ventral surface. Same area. Area 17. — Right arm, mesial surface. Area covered by two large blocks (held as for Area 7) which were set down proximally to a transverse line one inch above the crease made by bending the elbow, the short axis corre- sponding with the line of the biceps. The lat- ter line was measured from the inner condyle of the humerus, and the distance varied with the different subjects but remained the same for any one subject throughout the experi- ment. Area iS. — Left arm, ventral surface. Same area. Area ig. — Face, middle of forehead. Area covered by two small blocks (held with the long sides juxtaposed and the short axis in a cephalocaudad direction), set down so as just to clear the eyebrows. Area 20. — Dorsal surface of neck. Area covered by two small blocks (held as for Area 19), set down in the median line, just below the hair. Area 21. — Face, right cheek. Area covered by two small blocks (held as for Area 19), set down between the hair and the point of at- tachment of the lobe of the ear, just in front of the tragus. Area 22. — Face, left cheek. Same area. Area 2j. — Right leg, lateral surface. Area covered by two large blocks (held with the short sides in juxtaposition and the long axis in a cephalocaudad direction) which were set down proximally to a line drawn transversely three inches above the external malleolus, the long axis of the blocks coinciding with a line intersecting the first at right angles, drawn upward from the same point. Area 24. — Left leg, lateral surface. Same area. Area 25. — Right leg, mesial surface. Area corresponding to that of Area 24, measured from the internal malleolus. Area 26. — Left leg, mesial surface. Same area. Area 27. — Right thigh, lateral surface. Area covered by two large blocks (held with the long sides juxtaposed and the short axis in a cephalocaudad direction), which were set down proximally to a line drawn transversely three inches above the outer condyle of the femur, the short axis of the blocks coinciding with a line drawn upward longitudinally from the same point. Area 28. — Left thigh, lateral surface. Same 132 Temperature Sensitivity Area2Q. — Right thigh, mesial surface. Area corresponding to that of Area 27, measured from two inches above the inner cond\le of the femur. Area 30. — Left thigh, mesial surface. Same area. Area 31. — Right foot, sole. Area covered by t\vo small blocks (held with the long sides juxtaposed and the long axis running trans- versely), set down at the intersection of two lines — a longitudinal line from the middle of the cur\-e of the heel to the junction of the second and third toes, and a transverse line bisecting this. Area 32. — Left foot, sole. Same area. Area 31. i. — Right foot, dorsal surface. Area covered by two small blocks (held as for Area 31), set down proximally to the third and fourth toes. Area 32.1. — Left foot, dorsal surface. Same area. Area 3^. — Abdomen, median line, caudad to navel. Area covered by two large blocks (held with the long sides juxtaposed and the long axis in a transverse direction), set down immediately caudad to the navel. Area 34. — Abdomen, median line, immedi- ately cephalad to navel. Area 35. — Abdomen, right side. Area covered by two large blocks (held as for Area 33), set down at the intersection of a transverse line at the level of the navel and a longitudinal line three inches to the right of navel. Area 36. — Abdomen, left side. Same area. Area 37.— Back, median line. Area covered by two large blocks (held with the long sides juxtaposed and the long axis running trans- versely), set down caudad to a transverse line at the level of the lowest rib, the short axis of the blocks coinciding with the line of the vertebrae. Area 38. — Back, median line. Area covered by two large blocks (held as for Area 37), set down cephalad to a transverse line four inches cephalad to the level of the lowest rib. (In the case of Subject G six inches was used instead of four.) Area 39. — Back, right side. Area covered by two large blocks (held as for Area 37), set down at the intersection of the line at the k\c'l of the lowest rib and a longitudinal line four inches lateral to the s[)incs of the v'er- tebra\ Area 40. — Back, left side. Same area. These forty areas were tested four times for each individual, twice with the hot block above the cold {i. c, proximal or cephalad to it), and t%vice with the hot block below the cold {i. e., distal or caudad to it). The two positions are referred to as Positions A and B respectively. Twenty-six of the areas were also stimulated with two cold and two hot blocks, separately. The experiment was divided into two series, each consisting of sixty-six stimulations, forty with the double stimulus, thirteen with hot and thirteen with cold. The series were ar- ranged in as nearly a haphazard order as was possible. Certain limitations were necessary, howe\er, in order to make the experiment go smoothly. It was necessary, for example, to restrict the region tested in any group of stim- ulations to a part of the body that could be reached when the subject was lying in one position; it was not permitted to test the same area twice without a lapse of at least three stimulations between (to avoid disturbance from after-images); the hot and cold, large and small blocks had to be so distributed that no block was used oftener than once in five minutes, etc. The series finally took form as follows: SERIES I Double Stimltlations given in Position A Group I. Right Arm and Hand !ST AREA BLOCKS I II cold S 1,2 2 5 — La, I 3 3 hot S a, b 4 17 cold L2,3 5 11 — Sc, 3 6 15 hot Lb,c 7 13 — Sd,4 8 3 — Se, 5 9 17 — Ld,4 [Q 5 hot Le, a II I — S a, I [2 15 — Lb, 5 Temperature Sensitivity 133 the latter on the first and third. Practice tests were given to each subject before each group, on the areas included in the group, and the arrangement of the series was such as to distribute as evenly as possible the effects of practice. Group 5. Left Arm and Hand TEST ARE.\ BLOCKS 55 16 cold L 4. 5 56 i8 — L a, I 57 4 — S a, 2 58 6 cold L 2, 3 59 2 — S b, 3 60 12 — S c, 4 61 14 — S d, 5 62 18 hot L b, c 63 4 hot S e, a 64 16 — L d, 4 65 12 cold S I, 2 66 6 — L e, 5 Double Stimulations given in Position B Group I. Left Arm and Hand test area blocks 1 6 — L a, I 2 18 — Lb, 2 3 16 hot L c, d 4 12 — S a, I 5 2 — S b, 2 6 4 cold S 3, 4 7 14 — S c, 5 8 18 cold L 3, 4 9 12 hot S d, e 10 16 — L e, 5 11 6 hot L a, b 12 4 — S a, I Group 2. Back, Neck, Lateral Surfaces of Arms TEST AREA BLOCKS 13 40 hot L c, d 14 8 — L e, I 15 38 — La, 2 16 7 cold L 3, 4 17 39 — L b, 5 18 20 hot S b, c 19 8 hot L c, d 20 40 — L e, I Series I was given to three subjects on the 21 37 — L a, 2 first and third daj's, and to two on the second 22 39 cold L 3, 4 and fourth; while Series II was given to the 23 7 — L b, 5 former on the second and fourth days ^nd to 24 20 — S d, 2 G roup 2. i ■Ibdomen and Face TEST AREA BLOCKS 13 35 — L 1,2 14 34 — Lc, 3 15 21 — S b, 2 16 36 — Ld,e 17 19 — Sc, 3 18 22 hot S d,e 19 33 — L a, 4 20 35 — Lb, 5 21 21 hot S a, b 22 19 cold S4. 5 23 36 — Lc, I 24 22 — S c, I Group J . Legs and Feet TEST AREA BLOCKS 25 25 — Ld,2 26 28 — Le,3 27 30 — La, 4 28 24 — Lb, 5 29 31 hot Sd,e 30 32.1 — S a, 2 31 23 — Lc, I 32 32 — Sb,3 33 29 cold L2,3 34 26 — Ld,4 35 27 — Le,5 ■ 36 3I-I — S c, 4 37 32 cold S5, I 38 31 — S d, 2 39 24 hot La, b 40 23 cold L I, 2 41 29 — Lc,3 42 30 hot Ld,e Group 4. Back, Neck, Lateral Surface of Arms TEST AREA BLOCKS 43 39 — L a, 4 44 40 cold L5, I 45 8 — Lb, 2 46 38 — Lc, 3 47 7 hot Ld,e 48 20 — S d,4 49 37 — L a, 4 50 8 cold L5, 1 51 40 — Lb, 2 52 7 — Lc, 3 53 39 hot Ld,e 54 20 cold S5-I 134 Temperature Sensitivity Group 3 . Legs and Feet TEST AREA BLOCKS 25 29 — Lc. 1 26 23 — Ld, 2 27 31 cold S3. 4 28 32 hot S e, a 29 30 cold L3.4 30 26 — Le. 5 31 29 hot La, b 32 32.1 — S b, 5 33 27 — Lc. I 34 25 — Ld,2 35 3I-I — S c, I 36 31 — S d,2 37 24 cold L3.4 38 32 — S e, 3 39 2S — Le, 5 40 23 hot La, b 41 30 — Lc, I 42 24 — Ld, 2 Group 4. Abdomen and Face TEST AREA BLOCKS 45 35 hot Le, a 44 19 — Se, 3 45 33 — Lb. 3 46 21 — Sa.4 47 36 — Lc,4 48 22 cold S 5. I 49 34 — Ld. 5 50 19 hot S b, c 51 35 — Le, I 52 21 cold S2, 3 53 36 cold L2,3 54 22 — Sd,4 Group j. Right Arm and Hand TEST AREA BLOCKS 55 13 — Se, 5 56 15 cold L4.5 57 3 — S a, I 58 17 hot La. b 59 I — S b, 2 60 II — Sc, 3 61 15 — Lc, I 62 5 — Ld. 2 63 3 cold S4. 5 64 17 — Le, 3 65 II hot S d, e 66 5 cold L4,5 The temperature of the hot stimulus was 40°-42° C; that of the cold stimulus was i8°-20°. A few variations beyond these limits are noted in the individual records of Table l. The room temperature varie.J from 21° to 24°. This temperature, somewhat higher than ordinary- room temperature, wis chosen because it was desired to keep the part of the body tested exposed. However, even this temperature was not comfortable for most subjects, and the covered parts of the body were not generally exposed for more than five minutes before they were tested. They were kept uncovered throughout the test of any region. PROCEDURE The subject lay on a cot in such a position that the experimenter, standing beside her with one foot over the tambour used to record the reaction, could, with very little movement, take the blocks from their resting places and set them down on the part to be tested. The subject was instructed somewhat as follows: "I am going to perform an experiment to see how you respond to hot and cold. \\'hen I put this little block down on your hand, I want you to tell me just as quickly as you can what you feel. Tr>' it now" . . . (Se\-eral trials, either stimulus.) "Xow I'll put both of them on you, and you tell me which you feel first, just as soon as you feel it; and then tell me just as soon as you feel the other. Close your eyes. Ready, on your arm!" Several trials were gi%"en, until S understood that she was to respxmd as soon as she felt anything. Some practice was required before S was able to respond with certainty' on the first day, but thereafter there was rarely any difficulty^ The directions were summed up, after S felt fairly sure of her replies: "Xow you see what I am going to do. Sometimes you will feel both, and sometimes you will fee! only one or the other, so you must pay close atten- tion and answer what you feel just as soon as you feel anything." S then closed her eyes, the area to be tested was named; E removed the blocks from their boxes, started the poly- graph, gave a ready signal, and set the blocks down, simultaneously tapping with her foot upon the tambour which operated the writing lever of the polygraph. S's responses were likewise recorded by a tap of E's foot, the blocks were returned to their bo.xes, the number of the test and S's verbal response (whether "cold-hot," "hot-cold," "cold," or Temperature Sensitivity 135 "hot") were noted beside the drum record of the reaction-time, and the second area was then tested in the same way. A sample of the record paper is reproduced below. k' ?f^ c/^T ff If the double stimulus was given and both cold and hot were reported, the stimulus was immediately removed. If both were given and only one reported, the stimulus was left in contact about fi\e seconds and then removed. This same procedure was followed when only one stimulus was given, so that S's replies could not be controlled by knowledge of her errors. If for any reason the record for any test was unsatisfactory — if the ink flowed too freely and blurred, if the tambour failed to record, if E was conscious of having bungled and given the stimulus badly, as occasion- ally happened — the test was repeated at the end of the group in which it occurred, with- out S's knowledge of its being a repeti- tion. When the record was completed, it was scored by counting the time marks between the moment of stimulation and the first response, and between the latter and the second response. To do this, lines were ruled from the rise of the recording mark to the time line by means of a straight edge. Read- ings were taken to the nearest tenth of a second. CONDITIONS AFFECTING RESULTS The reaction-time to cold has generally been found to be quicker than that to hot — a fact that corresponds to the probably deeper situa- tion of the warm end-organs below the skin surface. It would therefore be assumed that if both stimuli were applied at the same time, other things being equal, the reaction to cold would precede that to hot. My results show that in most cases this is what happens {v. Table i), but the majority is in some cases so slight that it is important to consider the "other things." Limitations of Method. — In the first place, the limitations of the method must be taken into account. The method of measuring the reaction itself has a considerable possibility of error. The simultaneity with which E ac- tually sets down the blocks and taps with her toe upon the recording tambour may be ques- tioned. The error might be quite large in this case, for it is a fairly complicated procedure; nevertheless, I think the general consistency of each individual's responses (which may be seen from a glance at Figure 4) justifies the assumption that as a matter of fact this error was not large enough to afifect the results materially. Then the recording of the sub- ject's responses by E is also subject to a slight variable error; but this would undoubtedly be less — since E was practised before she began the tests — than the error of unpractised sub- jects in recording their own responses by unfamiliar means. The method of reading the record is also susceptible of a certain amount of error, which would not be greater in any case than one-tenth of a second. A more serious source of error, however, in the judg- ment of the experimenter, is the possibility of a slight inaccuracy in setting the blocks down exactly together and with equal pres- sure. An attempt was made to control this error by providing a tambour to be attached to the blocks, which would trace on the record the moment at which each block made con- tact with the skin. This device could not, however, be made to work satisfactorily. Con- sequently when the experimenter recognized that she had made a distinctly bad stimula- tion, she sometimes repeated it to see what the result would be. Sometimes also when the response was given in the order "hot-cold," the experimenter repeated the stimulus even though she thought it had been given ac- curately. This was not done systematically because it would have lengthened the experi- ment too much; but the following is a sum- mary of the total number of times a test was repeated which first gave the response "hot-cold." 136 Temperature Sensitivity First Response "Hot-Cold" 1st Badly Given ist Not Badly Given 2ND RESPONSE 2ND RESPONSE Subject C— H H— C C— H H— C G. 7 ■ I — Co.* I I 13 8 K. — — — — H.t I — 4 I CI. — 2 6 5 * One response was "cold" only. In one test that gave the response H — C, E thought that the cold block touched first, but this test was not repeated. t One response was "hot" only. Of twelve tests badly given and repeated, nine gave the response C-H, and three H-C on the repetition. Of thirty-eight tests re- peated, though not consciously badly given, twent>'-four gave the response C-H, and fourteen H-C on the repetition. This would seem to indicate that while some of the responses that occurred in the order H-C may have been due to inaccuracies of method of stimulation, some of them probably were not. This source of error, however, remains un- determined and probably considerable. Changes in Temperature of Skin or Stimulus.— In the second place the relation of the skin-temperature to the stimulus-tem- perature, and also the actual stimulus-tem- perature, may have affected the reaction-time in a variable way. Although the room tem- perature was fairly constant, the skin-tem- perature of different individuals on different days, on different regions of the body, prob- ably varied rather widely. This is very likely one reason for the greater constancy of the reactions on the abdomen and back, which is shown by every figure — a less scattered dis- tribution on these areas for each individual, a smaller variation among different indi- viduals, and a higher percentage of responses in the order cold-hot. For, of course, the skin- temperature of the usually covered areas of the body varies less than that of the extremi- ties and uncovered areas. The thighs and .the back of the neck also show a greater constancy of response than the hands and feet. Variability of skin-temperature, then, will account for some of the variability of reaction. The variations in the temperature of the stimulus constitute a possible source of error more difficult to determine. The hot stimulus varied for the most part between 40° and 42° C. On three days, a few tests were made with the oven temperature at 42.5°, one day it rose to 43°, and one day fell to 38°; but if the individual records for those days are examined, it does not appear that the general character of the results was affected (v. Table i). The cold temperature varied between 18° and 20°, except on two occasions — once when it was between 18° and 16°, and once between 17° and 19°. The hot stimulus temperature was chosen because it was desired to have a temperature distinctly hot, yet below the threshold for burning. This point of course varies; Neu- mann places it on the average at 43.5°, Head at about 50°. Temperatures below 40° fre- quently do not elicit a hot sensation. There- fore the limits were so set. In a similar way, a cold, distinctly cold but not painful, was desired. Crawford places the limit for great intensity of cold at 15°, and a temperature above 20° is not distinctly cold to a part of the skin whose temperature is 27° — as that of hands and feet often is. Therefore a tem- perature of i8°-i9° was considered best. The reaction-time depends, to some extent, on the temperature of the stimulus. Gold- scheider found that for a limited area of the forearm, the mean reaction-time to cold for a stimulus of 20°-i9° was .377 seconds; and for I5.5°-I5°, .278 seconds — a difference of a tenth of a second for a range of four degrees. He also found that for a stimulus of 40°, the mean time was .777 seconds; and for 44°, .585 seconds — a difference slightly less than two-tenths of a second for a range of four degrees. Thus the variations in the stimulus temperature may have affected the results by a few hundredths of a second, but not significantly. TOPOGR-A.PHY OF WAR.M AND COLD SfOTS.^ In the third place, the number of warm and cold spots in the areas tested may have been unequal, and have given a different emphasis to one sensation or the other. This effect should have been neutralized by the procedure of reversing the relative position of the blocks in Temperature Sensitivity 137 alternate experiments. Nevertheless, slight discrepancies in the exact location of the spots might perhaps have thrown a strongly reacting spot into one-half or the other of any area, or one-half or the other may have contained a particularly sensitive spot of one kind. This possible source of error could not be corrected without a laborious exploration of the skin of every subject, which would not be worth while. It might perhaps be inferred that some idio- syncrasy of warm or cold spots is responsible for those occasional instances when one area gave a hot-cold response in each of the four tests, or only one reaction to each test. Examples of this may be seen in Table i : Subject Co., Areas 11, 14, 16; Subject H, Area 31. It may be noted that Area 13 in three subjects gave three responses in the order hot-cold. These, then, are the factors that are or may be present in the results and should be borne in mind. G. Co. I DI N ^I ^J 1 1 III IV Y VII nil III I I I © K t, I I III! I U ¥ i) 1 II 3 .5 4 24 12 8 142 Temperature Sensitivity TABLE I [continued) Reaction-Time for Double Stimulations Subject a. Position of stimuli: Pos. A, Experiment 2 and 4; Pes. B, Experiment i and 3 Temperature: H— Ex. i, 40.5°-42° C; Ex. 2, 40.5-42; Ex. 3, 40-42.5; Ex. 4, 40-42 C — Ex. I, I7.8°-I9°; Ex. 2, 18-20; Ex. 3. 18-20; Ex. 4, 18-20 Room — Ex. I, 22''-23°; Ex. 2. 24-25; Ex. 3. 21-22; Ex. 4, 21-22 Dates of experiment: March 26, April 23, May 23 (o. m.). May 23 (/>. m.), 1917 Order C-II Order II-C Area 1st Intv. 3nd Intv. 1st Intv. 2nd Intv. Single Resp. Ex. I 2 3 4 / 2 3 4 1234 I 2 3 4 12 3 4 I 9 7 II 4 3 3 II 7 3 9 9 5 3 12 9 24 4 5 7 7" 5 3" 9' 9 8 5' 4 3 7 9 8 5 7 6 II C7 II 12 7" 10 3" 7' 12 10* 7' 4 i.-^ 14 8 9" 4 4 4" t' 10* t' 4* 15 6 9 8 4 4 3 H8 17 8 6 8 8 4 3 4 3 2 11" 8 4" 3 10' 8* 4' 8* H8 4 10 6 9 9* 8 6 4* 6 6 8 3 89 8 5 6 6 8 7 9 8 7 4 4 4 6 12 8 8 4 5 8 4 C8 14 10 10 II 6 8 4 9* 4* 16 9 9 4 4 8 8 5 6 18 10 8 8 4 4 4 ID 5 19 10 8 II 4 10 7 4 4 21 8 9 10 9 9 3 4 4 22 9 7 8 6 C8 C12 20 9 8 9 8 5 6 5 4 23 12 8" 9 8 5 8" 6 15 8' 5' 24 10 8 10 10 16 10 8 6 27 9 8 9 6 4 5 5 9 29 8 8 8 8 4 5 4 4 31 9 11 9 9 9 II C12 3I-I 8 8 10 15 12 14 16 4 24 7 9 16 7 9 4 C8 26 9 9 14 10 Cio C8 28 10 7 8 8 5 6 6 4 30 10 6 9 8 4 4 6 4 32 9 12 12 II 16 II 15 14 32.1 9 8 9 10 II 21 28 12 33 8 10 7 3 4 3 9 4 34 9 8 10 8 5 4 4 4 35 8 8 9 8 4 3 4 4 36 8 8 10 8 4 4 6 4 37 8 9 8 8 4 4 7 5 38 8 8 7 6 12 5 8 4 39 7 8 9 8 4 4 5 5 5 40 8 8 9 4 9 3 9 t Record could not be read, but response was in Order H — C. * Stimulus was repeated and gave reaction in the same order. Temperature Sensitivity H3 Explanation: Meaning of "Position of Stimulus" is explained under Method. Under "Order C-H" are grouped those reactions that gave response to cold first; under "Order H-C," those that gave response to hot first; under "Single Resp." those that gave either cold or hot alone, each figure designated by the appropriate letter, "ist Intv." means time from stimulus to first response; "2nd Intv.," time from first response to second response. Twice-accented figures indicate repetitions of the stimulus, to which the first response is shown by once- accented figures. Conditions under which repetitions were made are described under Method. TABLE 2 Reaction-Time (in Tenths of a Second) for Single Stimulations Cold Subject G. Co. A'. H. CI. Av. M. V. Area Seconds 3 II 10 13 12 10 19 9 12 8 9 1.04 .2 1 4 10 10 10 10 II 12 C14 H8 II 9 10 1.03 •07 II 8 9 8 II II 9 12 12 8 8 .96 •15 12 10 9 II II 12 10 12 13 8 8 1.04 •14 5 12 8 14 '3 10 15 12 II 8 9 1. 12 .20 6 12 10 14 II 12 24 12 14 10 7 1.28 .26 7 8 8 10 10 12 II 9 13 7 8 .96 .16 8 10 10 10 12 13 8 10 12 8 8 1. 01 ■13 15 10 9 10 ID 9 10 ID 12 8 9 •97 .08 16 II 9 17 13 10 12 10 15 8 8 I-I3 •24 17 10 10 II 10 10 9 II 13 6 6 .96 .16 18 9 8 10 II 9 7 12 10 8 9 •93 .12 19 II 10 19 II 10 8 12 13 8 10 1. 12 .20 21 10 9 11 10 10 II 14 13 II 10 1.09 .11 22 II 8 17 II 13 12 12 12 10 9 I-I5 .16 20 9 12 12 14 II 9 II 13 10 9 1. 10 •14 23 n II 15 Cii H16 14 II 13 13 8 8 115 .20 24 Cii H4 ID 12 14 C27 H6 C12 H9 12 13 10 9 1. 14 •15 29 Cii H6 C8 H5 12 9 10 8 12 14 6 8 .98 .21 30 10 10 9 II 10 10 14 12 7 8 1. 01 •13 31 12 10 13 13 12 II 12 13 8 8 1.12 .16 32 12 C9 H4 14 15 18 15 12 13 10 10 1.32 .20 35 ID 9 S 10 10 9 9 12 8 7 .92 .11 36 9 C8 H4 9 9 9 9 8 13 10 8 •94 .10 39 10 9 8 10 7 10 10 II 6 7 .88 .14 40 12 9 8 9 14 17 13 10 7 9 1.08 .26 144 Temperature Sensitivity TABLE 2 (continued) Reaction-Time for Single Stimulations Hoi Subject Area Co. K. H. Av. M. V. Seconds .OS .18 .06 •24 .10 .26 .07 •24 .11 •19 .11 •25 19 23 10 24 12 29 10 30 10 31 10 32 12 35 10 36 9 39 HS C4 40 14 4 9 10 II II 13 12 17 I 9 10 12 10 10 9 15 2 9 9 9 9 12 II 12 5 9 II 15 12 12 II 10 6 9 8 II 10 10 ID 22 7 10 9 13 II 9 10 II 8 II 8 II II 16 II 13 5 12 9 II 13 14 8 II 6 8 8 8 9 10 10 H13 C6 14 16 21 9 14 12 10 17 13 10 12 12 16 25 15 14 8 10 10 II H12 9 12 C8 H13 9 12 C6 8 7 6 8 10 10 16 10 9 9 9 8 10 II 9 II 9 8 8 6 8 9 Iio 9 C8 44 .18 07 .20 94 •19 97 .11 07 .22 01 -13 94 .09 18 •25 03 .22 25 .28 37 .28 14 .18 •23 99 .09 99 ■19 05 •15 08 .16 TABLE 3 Proportion of Responses in Order, C-H and H-C Cold— Hot Subject G. Co. K. H. a. Xo. Xo. No. No. No. ■ M.V. Right arm Left arm Head Right leg Left leg Abdomen Back Average 28 94 13 34 14 45 83 15 38 15 47 75 10 59 II 62 88 19 80 13 52 96 12 48 14 59 100 IX 65 14 88 81 14 82 13 81 88.1 55-3 62 19 15 3S 21 57 53-6 17.6 56 18 50 54-8 11.8 75 II 69 68.0 6.0 46 22 88 70.8 17.4 71 19 80 70.8 14.0 64 15 94 82.2 14.6 83 14 87 82.8 1.8 61.9 75 69 Tern peratur TABLE 3 Hot- e Scnsith continued) -Cold ity 145 Subject G. Co. K. H. CI. No. % No. % No. % No. % No. % Av. %M.V. Right arm I 3 22 58 13 42 12 38 14 38 34.8 13.0 Left arm 6 17 23 59 12 38 9 27 16 44 37-0 12.0 Head I 6 6 35 4 25 3 19 3 19 20.8 7.4 Right leg I 4 I 4 7 28 3 13 2 8 11-4 7-3 Left leg I 4 7 28 7 29 I 4 2 8 14.6 II. I Abdomen 6 35 2 12 3 18 I 6 14.2 9.8 Back 3 19 3 i8 2 12 3 17 2 13 15.8 2.6 Average 7.6 33- 8 26.6 19.4 19.4 21.4 Totals TABLE 4 Single Responses to Double Stimuli Subject G. Co. K. H. CI. Totals C II C H C H C H C H C H Fingers I — — I 4 I 1 2 6 Hand — — 3 I — — 2 I — — 5 2 Arm — — — — I — — 8 I I 2 9 Face 3 I — — — — I 2 — 6 I Back of neck — — — — — — — — — — — — Leg I — 2 — — — — 2 3 — 6 2 Thigh — — — — I — 3 2 — — 4 2 Foot — I 7 I — 2 7 2 I — 15 6 Abdomen — — — — — — — 3 — — — 3 Back — — — — — — — — — — 13 2 2 3 12 23 40 31 TABLE 5 Median Reaction-Time for Single Stimuli — Grouped by Subjects Cold Subject G. No. of Co. No. of A'. No. of H. No. of a. No. of Region Med. Cases P.E: Aled. Cases P.E. Med. Cases P.E. Med. Cases P.E. Med. Cases P.E. Right arm •9 II .10 1.05 12 .10 I.O 12 ■05 1.2 12 •05 .8 12 .00 Left arm I.O 12 •05 I.I 12 .10 I.I5 12 .10 1.2 n .10 8 II .05 Head I.O 8 .10 I-I5 8 •15 1.05 8 •05 1-25 8 •05 I 8 .05 Right leg I.I 4 .00 1-3 5 •05 I.I 6 .10 1-3 6 •05 8 6 .00 Left leg 1.0 4 .00 1-3 6 ■15 125 4 •25 1.25 6 •05 I 5 -05 Abdomen •9 4 .00 •9 4 .00 •9 4 .00 1.05 4 •15 8 4 .00 Back ■95 4 •05 •85 4 •05 •85 4 •15 1.05 4 •05 7 4 .00 146 Temperature Sensiti\ it\- TABLE 5 {conlinued) Hot Subject G. Co. K. H. CI. No. of No. of No. of No. of No. of Region Med. Cases P.E. Med. Cases P.E. Med. Cases P.E. Med. Cases P.E. Med. Cases P.E. Right arm I.O 12 .05 I I 8 .10 1.0 12 •15 I 2 12 .10 8 12 .00 Left arm •9 12 .10 I 12 .10 1.05 12 .10 I 5 II ■45 8 12 .00 Head •95 8 .10 I I 8 .10 •95 8 •05 I 2 8 .10 8 8 .00 Right leg 1.0 6 .05 I 3 6 .10 i^i5 6 .10 I 7 6 .10 I 6 •05 Left leg 1.0 6 .10 I 35 6 .10 1-35 6 ■25 1 4 6 .20 95 6 •15 Abdomen •95 4 .03 I 4 •05 1.0 3 .00 I 2 4 .00 8 4 .00 Back 1.0 3 .00 I I 4 .00 I.I 3 .20 I 25 4 .05 9 3 .00 TABLE 6 Median Reaction -Time for Single Stimuli — Groiped by Regions Med. Cold No. of Cases P.E. Hot Med. No. of Cases P.E. Hand — dorsal (Areas 3, 4) Hand — ventral (Areas 11, 12) Arm — dorsal (Areas 5, 6, 7, 8) Arm — ventral (Areas 15, 16, 17, 18) Face (Areas 19, 21, 22) Neck — dorsal (Area 20) Leg (Areas 23, 24) Thigh (Areas 29, 30) Abdomen (Areas 35, 36) Back (Areas 39, 40) Note: The lack of a complete total for any region indicates that the remaining responses were "cold-hot" or "hot-cold." * P.E. in these tables is really Q, the semi-interquartile range, or the distance above and below the median that will 1.0 19 05 I.I 20 •«5 1.0 20 15 •95 20 ■15 1.0 35 15 1.0 40 .10 1.0 40 10 1.0 39 .15 I.I 30 10 1.0 30 .20 I.I 10 15 1.0 10 .20 I-I5 16 15 1.2 20 .20 1.0 17 15 1.05 20 .15 •9 19 10 1.0 19 .10 •9 20 10 I.I 17 ••15 include fifty per cent, of the cases. TABLE 7 Median Reaction -Time of H-C Responses Subject G. Co. A-. H. CI. 1st 2nd No. of 1st 2nd No. of isi 2nd No. of 1st 2nd No. of 1st 2nd .Vo. of Region Int. Int. Cases Int. Int. Cases Int. Int. Cases Int. Int. Cases Int. Int. Cases Right arm — — — 1.2 •4 19 1.2 4 13 14 ■95 12 ■9 •4 II Left arm 1.0 •35 6 I.I .5 20 i^25 4 12 1.6 1.0 9 .8 ■4 13 Head — — — I.I ■45 6 •95 -4 4 14 .8 3 .8 •4 5 Right leg — — — — — — 1.2 .4 7 1.6 •9 3 — — — Left leg — — — 1-3 •75 7 14 .4 7 ■— — — — — — Abdomen — — — 1^3 •5 6 — — — 1.6 ■5 3 — — — Back .8 •4 3 I.I .6 3 — — — 1.2 •5 3 — — — Temperature Sensitivity TABLE 8 Median Reaction-Time for C-H Responses — Grouped by Subjects First Interval H7 Co. K. H. a. Subject G. No. of No. o_f No. of No. of No. of Cases Med. P.E. Cases Med. P.E. Cases Med. P.E. Cases Med. P.E. Cases Med. P.E. Right arm 31 1.0 Left arm 29 I.O Head 12 I.O Right leg 21 1.0 Left leg 2.1 I.O Abdomen 16 .9 Back 13 I.O 14 .20 .20 19 1.2 .20 13 I.I I-,^ •15 14 I.I I.O ■15 14 I.O I.O .10 13 I.O Second Interval .8 •05 18 85 .10 II 9 .05 22 8 •05 19 9 .10 15 8 •05 14 8 .00 Co. K. H. CI. Subject G. No. of No. of No. of No. of No. of Cases Med. P.E. Cases Med. P.E. Cases Med. P.E. Cases Med. P.E. Cases Med. P.E. Right arm Si •4 .00 13 •4 .00 14 •4 .10 12 •65 .40 21 4 •05 Left arm 29 •4 •05 15 •4 .00 15 •4 .00 19 •7 .20 18 4 •05 Head 12 •4 .10 10 ■45 ■05 II •4 •05 12 •7 .10 II 5 .10 Right leg 21 •4 .10 19 .6 .20 13 •4 •05 II .6 .20 22 8 •30 Left leg 23 •4 •05 12 •4 .00 14 •55 •15 17 .6 .20 19 I •45 .■\bdomen 16 ■4 .00 II •4 .10 14 •4 •05 II .5 .10 15 •4 .00 Back- 13 ■4 ■05 14 •45 .10 13 •4 .00 15 •5 .10 14 5 .10 table 9 Median Reaction-Time for C-H Responses — Grouped by Regions Region First htterval No. of Cases Med. P.E. Second Interval Med. P.E. Hand — dorsal (Areas 1-4) Hand — ventral (Areas 11-14) Arm — dorsal (Areas 5-8) Arm — ventral (Areas 15-18) Face (Areas 19, 21, 22) Neck — dorsal (Area 20) Leg (Areas 23-26) Thigh (Areas 27-30) Foot (Areas 31-32. i) .\bdomen (Areas 33-36) Back (Areas 37-40) RESULTS With these precautions concerning the re- liabiHty of the method we may consider the results. Table i presents the individual responses in detail to double stimulations for the forty areas. Table" 2 presents the 50 I.I 15 •4 .10 41 I.I 15 •4 •05 50 1.0 15 •4 .05 46 •95 iS •4 .00 40 I.O 10 •4 .10 16 I-I5 20 ■5 .10 53 I.I 25 .6 .20 71 I.O 20 ■4 •05 49 I.I 20 .8 •35 67 I.O 10 •4 •05 69 I.O 15 •4 •05 detailed responses to single stimulations. The first thing that may be noticed in a rapid inspection of Table i is the considerable number of responses that were given in the order hot-cold. Table 3 shows the proportion of responses in each order for each individual. Temperature Sensitivity TABLE 10 Comparison of Reaction to Hot in Single and Double Stimuli Double Stimulus C-H Resp. Single (i. e., Total Time) Stimulus Med. P.E. Med. Difference Hand — dorsal Hand— ventral Arm — dorsal Arm — ventral Face Neck — dorsal Leg Thigh Foot Abdomen Back 1-5 •25 1. 1 4 1-5 •25 •95 55 1-5 • 15 I.O 5 1-3 • 15 I.O 3 1-5 • 15 I.O 5 1-7 •25 I.O 7 2.0 •25 1.2 8 1-4 • 15 1.05 35 2.0 •30 1.2 8 14 .20 I.O 4 1.4 •15 I.I 3 The averages show that on the arms the percentage of C-H responses is only slightly greater than half — 53.6 per cent, and 54.8 per cent., while on the abdomen and back it rises to 82.2 per cent, and 82.8 per cent, re- spectively. These relations do not hold exactly among individuals, but a tendency toward greater constancy for the latter regions is shown by the fact that on these regions, in only two cases does the per cent, fall below 80, while on the arms (with the exception of Subject G., whose per cents, are all high), no case rises above 57 per cent. As regards the H-C responses, the highest percentages fall on the arms, on the average, and for all sub- jects except G.; but the percentages on abdomen and back are not significantly smaller than on legs. Individuals vary greatly — from G., who has no higher than 19 per cent, of H-C responses for any region, and no less than 81 per cent, of C-H responses, to Co., who has as high as 59 per cent, of H-C responses for one region and only 34 per cent, of C-H responses for another. With these responses may be compared the cases in which only one response was made to the double stimulus (Table 4). It will be noticed from Table i that one subject, K, sometimes reported that both sensations were perceived simultaneously. This fact will be considered later. Excluding these responses of K., we find that there are 71 instances (out of about 800 responses) when only one of the two stimuli is reported {v. Table 4). Of these, 40 are cold and 31 hot — a nearly equal dis- tribution, but this is produced by the responses of one individual, H., who gave 23 hot re- sponses: otherwise, the balance is in favor of the cold. With the exception of H.'s responses (which will be taken up again), nothing of par- - ticular significance appears from this table. These responses are probably to be explained by lapses of attention or slight inequalities in the stimulus. The facts of Table 2 are more easily under- stood from Tables 5 and 6, and Figures i and 2, which attempt to present graphically the distribution of the reaction-times to separate cold and hot stimuli. The most striking thing about the distribution is the amount of over- lapping in the reaction-times of cold and hot. The range in both cases is about the same, with a slightly greater variability for hot. The medians show no consistent difl'erence for the two sorts of stimuli. In 20 cases out of 35 {cf. Table 5), the median cold reaction is shorter. In 14 of these, it is shorter by as much as .1 second, and in 9, by .2 second or more. In 10 cases the median for hot is shorter, and in the remaining 5 cases the medians are the same. This may be due to the rather rough method of measurement, since other observers have found a clearer difference; on the other hand, if the measure- ment adequately represents the fact, it ex- plains, perhaps, the rather large number of Temperature Sensitivity 149 responses to double stimuli that occurred in the order, hot-cold. The averages for the individual areas {cf. Table 2) show the reac- tion-time for hot to be greater than for cold in 17 out of 26 cases — a slightly greater percentage of cases than appears from the medians of the grouped areas (Table 6), although in only 8 of these instances was the difference in time as much as .1 second. It is of interest in passing to compare the median reaction-time to cold and hot sep- arately (Table 5) with the percentages of C-H and H-C responses (Table 3), though it must be remembered that the medians of Table 5 are computed on too few cases to be very reliable. Still it is rather curious that Sub- ject G., who showed fewest cases of H-C responses, nevertheless has more areas that give a shorter median reaction to hot than any other subject — three cases out of seven. A certain tendency toward consistency is, however, suggested by a comparison of Tables 3 and 6. Table 3 shows the greatest number of H-C responses, as well as the least number of C— H responses to occur on the arms. Table 6 shows that, with the exception of the back of the hand, the other areas on the arms have less than .1 second difference in the medians for cold and hot, one being greater for hot than for cold, one less, and one equal to cold. The abdomen and back, on the other hand, show .1 and .2 second difference respectively, the median for cold being shorter; and these areas likewise have the greatest number of C-H responses. This seems to indicate that the distribution of C-H and H-C responses is a real one and is not due exclusively to faults or inaccuracies in the procedure. One other thing may be noticed about Table 6 — the instances of double responses to the single stimulus. Some regions show an incomplete total for the group of areas desig- nated, and the remaining number of responses were either "cold-hot" or "hot-cold." Fifteen such responses were to cold, five to hot — that is, cold was more often felt as cold-hot than was hot felt as hot-cold. (H-C responses to cold and C-H responses to hot did not occur.) This is probably another illustration of the fact, often mentioned, that hot is a more dififi- cult, less certain sensation than cold; so that any change or imagined change after cold is apt to be reported as hot. Subject CI. so explained to herself several cases in which she reported the cold stimulus as "cold-hot" and then immediately corrected herself. (These cases, if included in the count, would further increase the difference.) Possibly the slight change in sensation was due to inexactness in setting down the blocks simultaneously and with equal pressure. The greater number of cold stimuli perceived as cold-hot may also, perhaps, be in part the effect of suggestion — more responses to double stimuli occurred with cold first; therefore greater likelihood existed that hot would follow cold than vice versa. When we turn to the double responses, the hot-cold responses to double stimuli are not numerous enough to form a very accurate basis of comparison with the cold-hot re- sponses. Table 7 shows the medians for those regions that have as many as three such responses. Of course most of these have a high degree of unreliability, but no particular differences from the distribution of cold-hot responses appear. These responses may also be compared with the C-H responses in regard to the length of the interval between the first and second re- sponses. In order so to compare them, the ratio of the second response to the first was determined for each response. The distribu- tion of these ratios is presented in Figure 3. The medians for both sorts of responses are indicated by circles and crosses respectively. In twenty-two cases there were at least three H-C responses for a given region. Of these, five are equal to the median ratio for C-H responses for that region, five are less, and twelve are greater. That is, when the response occurred in the order hot-cold, there was a slight tendency for the second interval to be longer in relation to the first than when the response occurred in the order cold-hot. This is rather unexpected, and may perhaps be considered an intimation that in such cases the blocks were not applied quite simultane- ously. We may turn now to a study of the cases in which the response was given in the order 5.=.© - © '^- - = =.fL = = >• . _. be:: o — =©;=; :^:=i . ^ -.= _# = ~ _. !;e.i - _ =§ ?oiTe\j Ia 3 'S S if? 2. i ; 0--r--roJ c^ro •0'j:**t''^'0^ -^ t~ r-;- oo 150 oo (r> o^ c5._-_c^ oJ tJ i S = _ i © - s :r. - =e = = =© - ■ - e s _ :©:== = OS 151 152 Temperature Sensitivity cold-hot {v. Tables 8 and 9, Figures 4 and 5). Figure 4 shows the distribution for each indi- vidual and brings out such individual differ- ences as there are. Subjects G., Co., and K. are quite similar as to median times and distribution; H. has somewhat longer times and more extreme reactions, but a similar difference in the median time of the two inter- vals; CI. has a curious distribution, in that the median time for the second interval is relatively long — longer even than for the first interval, in one case. Four subjects show less variability for the second interval than for the first; that is, the second interval follows the first at a fairly constant interval of about .4 second, whatever the length of the first inter- val. In the case of CI., however, the reverse is true. This accords with the fact shown by the distribution of ratios (cf. Figure 3): the ratio of the second interval to the first varies more than the second interval. Individuals vary least in reaction-time on abdomen and back (range of medians for these regions, .8- 1.2 seconds — cf. Table 8), and most on legs (range of medians, .8-1.5 seconds). The double stimulus, when cold is perceived first, does not appear much to affect the reaction-time to cold. Comparing the reac- tion-time to cold in the case of single and double stimulations {cf. Table 6 and Table 9), we find that three are shorter for the double than for the single stimulus, five are longer, and three are equal in the two cases. The effect on the reaction-time to hot is. naturally, to increase it. This may be seen by comparing the medians for the total reac- tion-time of C-H responses {i. e., the time for the perception of hot in the double stimula- tions) with the medians for the reaction to hot separately. Table 10 presents the com- parison. The difference is about .4 second, on the average, but it is noteworthy that on the foot, leg and neck, the time is considerably in- creased for the double stimulus. This increase is due partly to a delay in the perception of hot for those regions, as the table shows. But this is not enough to explain the amount of difference in the reaction of these regions to double stimulations; the greater delay in this case must be in some way dependent on the double stimulation. It is possible that this is occasioned by the phenomenon referred to by Alrutz and Goldscheider as the Anschwellung of the sensation — the increase in intensity of the sensation immediately after the first per- ception of it — which is certainly most notice- able on the feet and legs. At this stage of the experiment, however, nothing definite can be said on this point. Since the purpose of this paper is chiefly to present the facts as they appeared from an examination of normal subjects, the paper will close with a summary of the individual cases. The five individuals chosen were presumably normal. They were not given a neurological examination, but they were all personally known to the experimenter. TABLE II Summary of Individual Differences Total No. of Resp. Reactions Giving Response C- -H Ratio of 2nd Reaction to to Double St im. Reaclions Range of Areas Total Intv. to 1st Sing. Stim. Greater Medians of Having Time Single Ratios Medians Med. for than Different Regions less than Greater <.2S >i.o >.44 H. less 1.4 Sec. 2 Resp. than 2.S Sec. than C C-H H-C Sing. No. ofCasesist Inlv. 2nd Intv. No. of Ca.'ies No. of Cases No. of Cases No. of Cases G. 144 13 5 4 . 1 sec. sec. 230 3 Co. 94 68 15 II .3 .2 10 2 3 5 I 2 K. 92 47 5* II .2 .2 10 3 14 2 I 2 H. 97 34 35 37 -3 -25 9 17 4 15 6 I CI. 120 39 10 7 .1 .6 3 4 23 6 2 ' Thirteen cases were reported as simultaneous pQjc. © i ©:=: = '^>i. 3&tw . > - + -. i \ -X ^^.^ \ '.^^ '■•. \ < 1^ ^ ---'""' \^..-' Fig. 6. General Distribution of Responses TO Double Stimuli; (a) Percentage of C-H Responses; (6) Percentage of H-C Responses; (c) Percentage of Single Responses to Double Stimuli Subject G. Subject K. Subject Co. -.-. Subject H. ^■^^^"" Subject CI. 154 Temperature Sensitivity 155 Subject G. — A young woman of about twenty-one years of age, a student of domestic science, without any known nerv- ous or temperamental abnormalities. Compared with other subjects, G. showed a greater constancy of reaction in every re- spect. All the figures of distribution show a close grouping of her reaction-times — only four instances in the C-H responses when the reaction was greater than 1.4 seconds, only eighteen cases when the hot reaction preceded the cold, five reactions in which only one of the two stimuli was perceived, scarcely any variation in the median time for different areas for either first or second interval — but, on the other hand, three areas where the median for the separate hot stimulus was shorter than the median for cold; a close grouping of ratios, and no individual area which did not give at least two reactions in the order cold-hot. On the whole, G. is rather an ideal illustration of the norm. Subject Co. — A woman of about fifty years of age, without any known abnormalities. The distribution of Co.'s responses is more variable but shows the same general charac- teristics as G.'s, except in the number of H-C responses. Seventy cases occurred, most of them on the arm, hand, or foot. It is to be noted (c/. Table i) that the H-C responses that were repeated gave the same kind of response only on arm, hand, and foot — never on other regions. It is also noticeable that Area 1 1 gave no C-H responses but six in the order H-C; Area 14, no C-H responses but five H-C; Areas 3, 31, and 32.1 gave three single responses, and in the case of Area 31, all three were "cold." Co. shows a somewhat greater variability for different regions of the body than does G., with shortest median times on abdomen and back. Subject K. — A young woman of about thirty years of age, a graduate student in psychol- ogy — no abnormalities known. K. also shows the same general form of dis- tribution, with an intermediate number of forty-seven H-C responses. K. however, reported thirteen times that the sensation was simultaneous. This is rather curious. since it occurred in no other subject, and it was probably the effect of suggestion. K. had heard a little about the experiment before act- ing as subject, and while she did not know the procedure, she evidently speculated a little about it. Toward the end of the second day she remarked that she supposed the blocks were put down simultaneously and she felt one before the other, because she did not per- ceive two contacts! It will be noticed from Table i that on the first day no cases of this sort appear. Nevertheless K. appeared to have some hesitancy in determining which sensation came first, and it is probably in consequence of this that she has fourteen ratios (compared with a small number in other subjects) less than .25, which indicates a relatively long first interval. Subject H. — A woman of about forty-five years of age, janitress of a flat, generally re- garded as of a rather nervous temperament. H. deviates somewhat from the general distribution of the first three subjects. Her reaction-time is longer, and she has more extreme reactions — thirty-seven of her C-H responses are greater than 1.4 seconds. Her total times are in seventeen instances greater than 2.5 seconds (compared with eight, the maximum for any other subject). Her ratios are also longer, fifteen being greater than i.o, and six medians being more than .40-.44, the most frequent ratio. H. has thirty-five cases of a single response to a double stimulus — • three areas giving a single response as many as three times out of four. Twenty-three of these were hot and twelve cold, reversing the proportions of every other subject but K., in whom the two sorts were practically equal. In contrast to this, it will be noticed from Table i, that not infrequently H. responded by "warm" instead of "hot," although she was not asked to differentiate. It is true that for this subject on the first day the temperature of the hot stimulus fell to 38°; however, most of the "warm" responses did not occur on this day, and on other days the temperature was not below 40°. Subject Cl. — A young woman of about thirty years of age, a graduate student in psychol- ogy. On the second day Cl. was recovering 156 Temperature Sensitivity TABLE 12 Summary of Reactions Cold- Separate Hot- Separate Cold— Hot Hot— Cold Interval 27id Interval isl Interval 2nd Interval II III IV V VI VII VIII IX X XI XII Fingers 10 8-II 9 9-12 Hand 10 10-12 II 9-12 Forearm 10 9-12 10 9-12 Arm 10 S-ll 10 8-11 Face II 10-12 10 8-12 Neck — dorsal II 9-12 10 8-12 Leg II IO-I3 12 10-14 Thigh 10 9-12 10 10-12 Foot 12 IO-I3 12 11-15 Abdomen 9 S-io 10 9-1 1 Back 9 S-io II 9-12 * Total range: group included only three cases. from an attack of neuritis in her left arm. This does not seem to have changed the character of her responses for th is day or for her left arm in general. The general character of Cl.'s responses shows quite individual characteristics — a markedly short reaction-time to hot and cold separately and to cold in the case of the double stimuli, but a delay in the reaction to hot on the legs, which made the median for hot, in the C-H responses, as long or longer than the median for cold. She has few ex- treme reactions, but all of them are to hot stimuli. Of the ten instances of single reaction to double stimuli, only two were "hot." In Figure 2 extreme reactions to hot separately may be noted on the left arm and right leg, while in Figure 4 nearly all the reactions that are extreme occur on the right and left legs and not at all on the left arm. Her ratios tend to be longer than the others; six medians are over .40-.44, and twenty-three separate ratios are greater than i.o. CONCLUSIONS In so far as these five subjects may be rep- resentative of the normal reaction to cold and hot stimulation under the conditions of this experiment, we may summarize the results as follows: I. More than half of the total number of ) 10-12 4 3- 5 12 10-13 54-7 ) 10-14 4 4- 6 12 10-14 44-6 ) 9-12 4 3-4 II 9-14 5 4- 6 ) 8-11 44-5 12 10-14 54-7 ) 9-11 4 4-6 II 10-14 54-6 > 9-1 1 4 4-6 14 10-16* 4 4-12* 9-14 6 4- 8 12 10-16 4 4 ) 8-11 4 4- 3 12 12-15 54-9 10-14 8 5-12 16 15-16 8 5-14 > 9-11 44-5 13 10-16 54-6 I 8-11 44-6 10 9-11 5 4- 6 reactions to simultaneous cold and hot stim- ulation elicit first cold, then hot. The back and abdomen have a higher percentage than other regions of such responses. 2. The proportion of responses to double stimulation that does not include both re- sponses is small — 8 per cent, on the average, with a minimum of 3 per cent, and a maximum of 20 per cent. 3. The reaction-time to cold and hot stim- uli separately does not show any distinct difference, as measured in this way, except perhaps on the abdomen and back. 4. The length of the first interval, when "cold" is the first response to the double stim- ulation, is about 1.0 second; the length of the second interval, about .4 second, on the average. A second interval of more than i.o second is rare. 5. The order in which the response occurs — whether "cold-hot" or "hot-cold" — is prob- ably dependent in part on the simultaneity of application and in part on physiological conditions. 6. Individual differences in all of these re- spects are so considerable that a wide devia- tion from the average must be admitted as normal. 7. The method, though not accurate within very fine limits, gives a clinical picture suffi- ciently precise to show individual differ- ences among normal subjects, and should Temperature Sensitivity 157 TABLE 12 Summary of Reactions % C-H % H-C % Sing. %? XIII XIV XV XVI Fingers 55.6 34-6 8.8 I.I Hand 47-1 41-5 7.8 3-3 Forearm 50.6 36.1 6.0 3.6 Arm 58.2 32.6 7-4 1-3 Face 65.5 23-1 11.4 — Neck — dorsal 84.2 15-8 — — Leg 64.6 18.3 97 7.2 Thigh 83.1 9.6 7.2 — Foot 61.2 II. 2 25.0 2.4 Abdomen 81.7 14.6 37 — Back 83.1 157 — 1.2 therefore be useful in dealing with abnormal cases. For convenience of reference, a table sum- marizing the data is appended. The first two columns of the table give, respectively, the median reaction-time in tenths of a second, and the limits between which 50 per cent, of the cases are included, for responses to sep- arate coW stimuli. Columns I II and IV give the same facts for separate hot stimuli. Columns V-XII contain similar data for responses to double stimuli, first for Cold-Hot, then for Hot-Cold reactions, and for each interval. Columns XIII-XVI show the proportion of double stimuli that gave responses of each kind: C-H, H-C, Cold orHot alone, and Doubt- ful (where the priority of either could not be decided). If individuals are to be compared with this table, it must be remembered that a subject's median reaction-time may lie anywhere within the limits of the probable error or even slightly outside of these limits and still be normal. The percentage of stimuli giving any par- ticular sort of response is even more subject to individual variation. All that it would be safe to say about the normal reaction in this respect is that a majority of the total responses of an individual should be Cold-Hot, though on any region except the face, thigh, abdomen, and back the percentage may fall below fifty; and further, that a very small proportion of the total will fail to give both responses, and that the neck, back, and abdomen are least likely to present this type. The graphs in Figure 6 will make this clear. BIBLIOGRAPHY 1. Abbot, Edwina. 19 14, Effect of Adaptation on the Temperature Sense, Psych. Rev. Monog., xvi. No. 2. 2. Alrutz, Sydney. (a) 1908, Untersuchungen iiber die Tempera- tursinne, Ztsch.f. Psychol, u. Physiol, d. Sinnesorg., 47, p. 161. {b) 1900, Skandin. Arch.f. Physiol., x, p. 340. (c) 1897, Notes on the Temperature Senses, Mind, vi, p. 445. {d) 1898, The Sensation Hot, Ibid., vii, p. 141. 3. Blix, Magnus. 1884, Experimentelle Beitrage ziir Losung der Frage iiber die specifische Energie der Hautnerven, Ztsch. f. Biol., N. S. ii, p. 141 and N. S. iii, p. 145. 4. Crawford, J. F. 1898, A Study of the Temper- ature Sense, Psych. Rev., 5, p. 63. 5. Donaldson, H. H. 1885, On the Temperature Sense, Mind, 10, pp. 399-416. 6. DoNATH, Julius. 1884, Uber die Grenzen des Temperatursinnes in gesundem und krankem Zustande, Arch. f. Psychiat., xv, p. 695. 7. Eulenberg, a. 1885, Zur Temperatursinnes- messung, Monatsch. f. prakt. Derm., 4, p. i. 8. Franz, S. I. 1909, Sensations following Nerve Division, /. Contp. Neurol, and Psych., 19, pp. 107, 215. 158 Temperature Sensitivity 9. GOLDSCHEIDER, A. (a) 1887, Eine neue Methode der Temperatur- sinnesprufung, Arch. /. Psychiat. u. Nervenkrankh., 18, p. 659. {b) 1884, Die specifische Energie der Gefuhls- nerv'en der Haut, Monatsch. f. prakt. Dermal., 3, Nos. 7, 9, 10. (c) 1898, Physiologie der Hautsinnesnerven, Gesammelte Abhandlungen, vol. i. {d) 1912, Beitrage zur Lehre der Hautsen- sibilitat, Zlsch. /. klin. Med., 75, pp. 1-14. (e) 1912, Revision der Lehre vom Tempe- ratursinn, Ber. u. d. V. Kong. f. exper. Psychol., Leipzig. 10. Holm. 1903, Die Dauer der Temperaturemp- findung bei konstanter Reiztemperatur, Skan- din. Arch. f. Physiol., 14, p. 242; cited 1905, in Nagel's Handbuch, iii. 11. KiESOW, Friedrich. 1895, Untersuchungen iiber Temperaturempfindungen, Philos. Stnd., xi, P- 135- 12. L.\EHR. 1895-6, tJber Storungen der Schmerz- u. Temperaturempfindung in Folge von Erkranken des Riickenmarks, Arch. J. Psy- chiat., 28, pp. 773-874. 13. Leegaard, Chr. 1891, ijber eine Methode zur Bestimmung des Temperatursinns am Kran- kenbett, Deutsch. Arch. f. klin. Med., 48, p. 207. 14. Marinesco, G. 1905, Les troubles de la bares- thesie et leur coexistence avec panesthesie vibratoire, Semaine med., 25, p. 565. 15. Neumann, Fritz. 1903, Beitrage zur Klinik des Warmesinnes, Deutsch. Arch.f. klin. Med., 76, p. 106. 16. NoTHNAGEL. 1866-7, Beitrage zur Physiologie und Pathologie des Temperatursinns, Deutsck- es Arch.f. klin. Med., 2, p. 285. 17. Ranson, S. W. 191 5, Unmyelinated nerve- fibrgs as Conductors of Protopathic Sensation, Brain, 38, pp. 381-389. 18. Rivers, W. H. R., and Head, Henry. 1908. A Human E.Kperiment in Nerve Division, Brain, 31, p. 323. 19. SoMMER. Zahl der Temperaturpunkten der ausseren Haut, Ber. d. Physik.-med. Gesellsch. ZH Wlirzburg, Sitz. von 22 Nov., 1900. 20. Thunberg. (a) 1896, Article in Upsala Ldkaref. Forh.; cited, 1898, by Alrutz, Mind, p. 143. (6) 1901, Untersuchungen iiber die relative Tiefenlage der Kalte-, Warme-, und Schmerzpercipierenden Nervenenden in der Haut, Skandin. Arch. f. Physiol., .\i, p. 382. (c) 1905, Article in Nagel's Handbuch der Physiologie, iii, p. 680. 21. Trotter and Davies. 1913, The Peculiarities Found in Cutaneous Areas Supplied by Re- generating Nerves, Jour. f. Psych, ic. Neurol., 20, p. E102. 22. Veress, Elmer. 1902, Beitriige zur Kenntniss der Topographic der Warme-Empfindlichkeit, Arch.f. Physiol., 89, p. i. 23. VON Frey, M. Beitrage zur Physiologie des Schmerzsinns, Gesellsch. d. Wissensck. z. Leip- zig, 1894, 46; Pt. I, p. 185; Pt. 2, p. 283; ibid., 1895, 47, Pt. 3, p. 166. 24. Weber. 1851, Tastsinn und Gemeingefiihle. Paul B. Hoeber, 67-69 East S9Iu Street, New York A CONTRIBUTION TO THE STUDY OF TEMPERATURE REACTION IN NERVOUS DISEASES* THE REACTION TO SIMULTANEOUS COLD AND HOT STIMULATION IN SYRINGOMYELIA, TABES DORSALIS, MULTIPLE SCLEROSIS, PERIPHERAL NEURITIS, AND A CASE PRESENTING A CAPSULO-THALAM IC SYNDROME By ETHEL L. CORNELL, A. B. Departments of Neurology and Psychology, Columbia University NEW YORK CITY OUTLINE Introduction — Summary of the normal reaction to the double thermal stimulus Consideration of individuals' reaction in the various phases of the test Comparison of the syndromes studied General tendencies observable in the reaction to double stimulation — tentative interpretation Conclusions The present study is a continuation of in- vestigations of the reaction to a double ther- mal stimulus, preliminary results of which were published in the Neurological Bulletin for March, 1918. The former paper dealt with the reaction of normal subjects. The present paper is a study of twenty-two patients suf- fering from organic nervous diseases. The general method of the experiment was to provide several obser\'ations in each of which the reaction-time was measured for a series of stimuli combining simultaneous and separate cold and hot stimulation. For de- tails of the method, description of the appara- tus, plan of the series, etc., the reader is referred to the previous article.^ The patients ' The experiment as originally planned covered four days, on each of which the double stimulus was given on forty areas of the body, and a single stimulus on twenty-six of these areas — the latter being equally divided between cold and hot, so that in four experiments, two measures for cold and hot respectively were obtained for each of the twenty-six areas, and four measures of double stimulation were obtained for forty areas. Owing to the difficulty of included six cases of syringomyelia, five of tabes dorsalis, six of multiple sclerosis, four of peripheral neuritis, and one of a thalamic tumor. With the exception of one case of syringomyelia obtained from Dr. Tilney's pri- vate practice, they were patients from the Neurological Department of the Vanderbilt Clinic, and the diagnosis of each case was confirmed by the Chief of Clinic, Dr. McKen- dree. The experiments were made during 1917-1918, in the forenoons, in a room of the clinic, free from distraction, the temperature of which was kept between 22° and 25°C. As a rule, rather more preliminary practice was required with these patients than with normal getting patients to return so often, the plan was varied so that three observations should give the same number of tests. On the first two days forty double stimulations were given as before, but the twenty-six areas were tested for both hot and cold. On the third day. two series of double stimulation were made, single stimuli being inserted only often enough to serve as a check. This was not a source of suggestion, however, because, as will be seen, all patients gave a considerable proportion of responses to double stimuh in which only one stimulus was perceived. ♦ Submitted in partial fulfillment of requirements for the degree of Doctor of Philosophy in the Faculty of Philosophy, Columbia University Reprinted from the Neurologic.u, Bulletin. Vol. i. No. 9, pp. 335-372- Copyright 1918. 335 336 Temperature Sensitivity subjects; enough was given to put the patient thoroughly at ease, and to give the experi- menter an idea of the general type of response to be expected and a reasonable certainty that the patient understood the directions. Analysis of the cases will be made, on the basis of the tables and figures to be found at the end of the paper, the cases of each syndrome preceded b\- a summary' of the findings for that syndrome. The various syndromes will then be compared, the general nature of the disturbance will be discussed, and the conclusions that seem justified will be drawn. As a preliminary', however, it will be con- venient to summarize the results of normal reactions in a form easily comparable with that in which abnormal cases will be presented. I. Gross- Disturbance. — While normal subjects show many individual differences in the general nature of their responses, nothing that may be called a gross disturbance appears. By this term is meant a striking disturbance shown by failure to react; or by perverted response which is a hot-cold response to a cold stimulus or a cold-hot response to a hot stimulus, as well as a hot response to a cold stimulus or a cold response to a hot stimulus. II. Gexer.\l Char.\cter of Reaction. Type of Response to Double Stimulus. — In spite of considerable fluctuation, normal sub- jects all show a majority of C-H responses, and a ver\- small number of single responses. - Range of Medians for Various Regions of the Body. — Among normals, no subject shows a range of more than .6 second for the median reaction-times of all regions, in any t\'pe of response. That is, a subject's reaction-time is fairly constant for all regions of the body.' Extreme Reactions. — Since the measure used is the median and not the average time, it is not much afTected by a few extremeh' long reactions. This is desirable, of course, in measuring the normal time, but in comparing abnormal cases with normal, it might cause the neglect of a number of extreme reactions, which might be of some importance. They 'Neurological Bulletin I. p. 154. • Ibid.. Tables 5-8 and 1 1. p. 145 el seq. can, therefore, be considered by themselves. Reactions of 3.0 seconds for separate stimuli or for the first inter\-al of double responses do not occur among normals.* Reactions of i.o second or more as the second interval of double responses are rare, generally, though they occur in one subject in twenty-five per cent, of the cases.* Consistency. — As it might be possible for a disturbance to be confined to a few areas, but at the same time to cause a large probable error in the measurement of the region includ- ing those areas, or a large fluctuation in the type of response for that region, the con- sistency of response was measured in the case of abnormal subjects (Infra, Table 8). For single stimuli, the measure used was the per- centage of areas stimulated giNnng two cor- rect responses — this is called consistency in type, and varies, for different normal sub- jects, from 85 to 100 per cent, for cold, and from 92 to 100 per cent, for hot. For double responses, the same measure was used — the percentage of areas stimulated gi'ving two C-H responses, etc.; but this allowed more varia- tion, as each area was stimulated four times. Consistency of C-H responses was high among normals, ranging for different sub- jects from 75 to 100 per cent. The fluctuation of single areas in time might have been measured by the de\iation of each area from its own a^-erage if there had been a sufficient number of cases, but it seemed best under the circiJmstances to take as a measure the range of each area having two similar responses. Thus, for normals, it was found that the percentage of areas which responded at least twice by C-H that had a very close range — less than .3 second — varied from 43 to 73 per cent, for the first interval, and from 45 to 80 per cent, for the second interval, for different individuals; while the percentage having a large range — i.O second or more — was not more than 3 per cent, for the first internal, and varied from o to 29 per cent, for the second. In reaction-time, separate stimuli gave somewhat more con- sistent responses — 68 to 96 per cent, of the areas having a small range for cold, 50 to 'Ibid., Figs. I, 2 and 4. ' Ihid.. Table i, Subj. H. p. 141. Temperature Sensitivity 337 92 per cent, for hot, while the largest per- centage having a range of i.O second or more was 7.6 per cent. Diminution of Intensity of Sensation. — While subjects were not asked to differentiate degrees of intensity, there were a few instances among normals, and many among abnormals, when the sensation was reported as 'warm', instead of the usual 'hot'. Isolated instances of this sort might be due to sporadic fluctua- tions in the temperature of the blocks used for stimulus; but any considerable number of such cases would have to depend on sub- jective factors. Of course, if all responses should be 'warm', it might be explained as a language-habit of the patient; but if any distinction is made, it is fair to assume that it represents a subjective difference. Among normal subjects, no responses to a separate hot stimulus were of this sort; in the case of double stimulation, the highest percentage of warm responses was 7 per cent. There was no suggestion of any corresponding tendency to report 'cool' instead of 'cold'. Intensification of Sensation. — It would be desirable to have a measure of this phenome- non also, but as any questioning by the experi- menter would have been a distraction from the main purpose of the experiment and would in any case have been too open to sug- gestion to be very valuable, it is necessary to resort to collateral evidence. There is much less tendency to report increase in intensity, verbally, than decrease; partly, perhaps, be- cause a very intense stimulus causes a reflex withdrawal which is itself an evidence of in- crease. This reflex is about the only measure we have of intensification, and it is of ques- tionable value both because of large indi- vidual differences and because no record was kept of its appearance in normal subjects. It occurred, on abdomen or back in at least two subjects, but it was generally inhibited in the course of the preliminary practice. SYRINGOMYELIA Six cases were studied. In three, the dis- turbance was confined to one or a few regions ; while in the other three there is a general dis- turbance, although in two of the latter, it is more marked on a few areas. The distur- bance is shown by a marked lengthening of the reaction-time to single stimuli; by failure to perceive the stimulus at times; by a tendency to give perverted responses; by a higher percentage of single responses than normal to double stimuli and a long reaction- time in every type of response; by a large number of extreme reactions; by a more than normal inconsistency of response; and by a tendency to report the sensation as weak in intensity. Analyses of the cases follow, with references to the tables and figures at the end of the paper. Case I. — P. K. ; age 16; student; intelli- gence and cooperation good. Onset. — Spinal curvature eleven years be- fore test. Sensation. — Not described in history. I. Gross Disturbance. — Tables 6 and 7. No failure. Perversions occur on hand and arm to both cold and hot, but more often to cold. II. General Character. Type of Response.— Fig. I. Cur\'e is practically normal for face and legs; but on arms, abdomen and back single re- sponses to double stimuli are numerous: — H on arms; C on abdomen and back. Median Reaction-tiine. — Table I. In single stimuli C tends to be somewhat lengthened, except on face; while H is about normal except on arms and back. In double stimuli there is a general tendency toward lengthened reaction-time in the first interval of double responses; a marked similar tendency in the second interval on arms and back, and in the case of single responses. Extreme Reactions. — Table I, Fig. 6. Single re- sponses show no more tendency to extreme reactions in double than in single stimuli, and are in both cases confined to cold. The tendency is more marked in double responses, and is more frequent for the second interval— slightly more frequent for H-C than for C-H type. Consistency. — Table 8. In single stimuli there is a marked inconsistency in type of response for cold which is most evident on the right arm. The fluctu- ation in time, while greater than normal, is not very pronounced. In double stimuli no type of re- sponse occurs twice on half the areas, though C-H is the most frequent type. The fluctuation in time is greater than normal, and is particularly marked for the second interval of H-C responses. H er- sponses fluctuate more than C. 338 Temperature Sensitivity Diminulion oj Sensation. — Table 9. Hot is di- minished on all regions in a considerable percentage of cases — more in double than in single stimuli. Cold is somewhat diminished on arms. Intensification. — Table 10. There is no reflex e\i- dence, but patient reported that when sensation was 'sharp' distinction was diflicult. He thought that it was cold that sent a shooting pain through region and that it 'got worse and worse' while warm 'stayed warm all the time'. Once he said: "I guess it's warm, because if it were cold I'd feel it." Per- haps this explains the number of times that C was reported as warm. It also implies an intensification of cold — though there is evidently a hypesthesia. Single stimuli show cold to be more disturbed than hot : by its longer reaction-time ; more extreme reactions; greater inconsistency in type and in time; and greater likelihood of perversion. But the results of double stimulation do not confirm this. The C-H type has at least a pluralitN- of all responses; reaction to hot tends to be longer in double than in single stimuli, while to cold it remains about the same; and H responses are less consistent than C. The disturbance is moderate in degree and is general rather than local, although face and legs escape. Case II. — E. Wi. (colored); age 25; dental student; intelligence and cooperation good. OxsET. — About eighteen months before test. Sensation. — Described in history, July, 1917: "Mild degree of disturbance to hot and cold." Frequently patient burned himself be- cause "sensation was so delayed he did not feel pain in time." I. Gross Disturb.\nce. — Failure to react was consistent on left hand. Perversion occurred on right hand to hot, and on left foot to cold. II. General Character. Type of Response. — Curve is altogether comparable with normal, except for the failures on left arm and the per cent, of single hot responses on neck. Median Reaction-time. — In single stimuli the only lengthening is for hot, on left arm and back. In double stimuli, C-H responses are about normal, but H-C and H responses show some lengthening, par- ticularly on left arm and abdomen. Extreme Reactions. — The highest percentage of extreme reactions occurred for single responses to double stimuli, thus indicating a real disturbance for this type, both in the failure to perceive the double stimulus and in the slowing up of the response that did occur. In double responses the second interval is extreme more often than the first; but H is more frequently extreme than C, whether it is the first or second response. Consistency. In single stimuli, H is less than normally consistent in type. The fluctuation in time is not excessive — though greater than normal — for either reaction. Both are very consistent (1. e., they have a range of less than .3 sec.) in about half the areas. In double stimuli, no type occurs twice on fifty per cent, of the areas, though C-H approxi- mates this. The fluctuation in time is greater than normal; and the deviation from the normal is greater in the C-H type than in the H-C. Diminution of Sensation. Hot is diminished on all regions in a considerable number of cases — more in double than in single stimulation. The left arm, however, which showed the only gross disturbance, had relatively more reactions that were hot than had other regions. An initial perception of 'warm' fol- lowed by 'hot', from .3-2.0 seconds later, occurred a number of times but not with any regularity. 'Warm' was reported with temperatures even of 45° to 50°C. No weakening of the cold sensation was reported. Intensification. No evidence, either reflex or verbal, was present. Both single and double stimulations show more disturbance of the hot sensation than of the cold — more inconsistency in type; greater delay; more extreme reactions (at least in double responses) ; greater tendency to inhibition in double stimulation; and more tendency to diminished intensity. Disturbance is marked locally, affecting, chiefly the left hand. Case III. — A. C; age 31; tailor; intelli- gence and cooperation fair. Onset. — About nine years before test. Sensation. — Described in history, June, 1917: "Normal to touch and apparently, also to pin-point. Thermal sense absent down to shoulder joints." I. Gross Disturbanxe. — Failures were frequent or usual on arms and neck, and occurred on ab- domen. Sensation was entirely lacking on back, except on shoulder blades. Perversion of cold was the rule on arms and occurred on abdomen; per- version of hot occurred on left arm and abdomen. II. Gener.\l Character. Type of Response to Double Slitnuli. — Curve shows almost complete absence of double response; C type present only on face, neck, and abdomen; all other responses H. Temperature Sensitivity 339 Median Reaction-time. — In single stimuli, reactions are all very prolonged — hot more so than cold. In double stimuli the H type only occurs with any fre- quency, and is even slower than in separate stimuli. Extreme Reactions. — About half the single re- sponses to double stimuli and single stimuli are extreme. Double responses are very infrequent, but evidently tend to be relatively longer for the second inter\-al than for the first. Consistency. — As patient came only twice, while the e.xperiment was planned for four days, there were not enough stimulations on each area to provide a measure for consistency. But consistency for single stimuli could not be very high for either stimulus, be- cause of the number of failures and perversions, while for double stimuli, forty-two and five-tenths per cent, of the forty areas gave two hot responses, and fifty-three per cent, of these had a range of more than l.o second. Diminution of Sensation. — Diminution of hot occurred to a considerable extent on arms and abdomen, and to some extent elsewhere. 'Warm' was reported with temperatures as high as 47°C. Diminution of cold occurred on face. Intensification. — On legs, the double stimulus and the cold stimulus were several times accompanied by a marked reflex withdrawal. In one such case — to cold — patient said: /'Cold or hot — something strong." Both sensations are much disturbed. Cold is more often perverted and is more apt to cause a reflex movement; but although hot is more delayed and apparently more diminished, it takes most marked precedence in double stimulation. The disturbance is marked generally over all re- gions, but is profound on arms and back. Case IV. — L. A.; age 49; paper-hanger; intelligence and cooperation fair. Onset. — About ten months before test. Sensation. — Described in history, June, 1917: "Normal to pain and touch; cold felt as cold, but heat at times retarded and at times mistaken for cold on small finger and up ulnar side of forearm." I. Gross Disturb.\nce. — Failure to react oc- curred, but not persistently, on sole of each foot. Per\'ersion of cold occurred on right forearm, right thigh, and right side of abdomen. II. General Character. Type of Response. — Curve shows pfcponderance of double responses except on arms and face, with H-C type more fre- quent than normal. Single responses are frequent on arms and face and are present on legs; but do not occur on neck, abdomen or back. Median Reaction-time. — Owing to the fact that patient came only twice, instead of four times, medians are not altogether reliable. For single stimuli, C seems to be a little slower than H — espe- cially on face, left leg and abdomen, but times are not far from normal. For double stimuli, C-H re- sponses are nearly normal, although face and legs seem to give rather long first interval and face a long second interval; H-C type tends to have more lengthened second intervals and generally longer first intervals; single responses are long. Extreme Reactions. — The highest percentage of extreme reactions occurs for single responses to dou- ble stimuli, indicating, in connection with the long medians for this type, that these responses imply the most disturbance. Consistency. — As in Case 3, consistency cannot be measured. It would be higher, with single stimuli, for H than for C, since the former had no errors. Results of double stimulation cannot be predicted; twelve and five-tenths per cent, of the areas gave two responses C-H, and an equal number gave two H-C. Diminution of Sensation. — Hot stimulus was always reported 'warm'. Once, patient said: "\Varm — good warm." As his English was somewhat lim- ited, this may have been merely a language habit. No analogous cool was reported. Intensification. — No evidence. Contrary to the findings reported in the history, cold sensation seems to be more interfered with than hot, in both kinds of stimulation — it is slower separately and has more tendency to extreme reac- tion — in double responses as well as in single stimuli; only cold is perverted; and in double responses, hot takes precedence. The disturbance is moderate and is general rather than local. Case V. — J. CI.; age 21; accountant; in- telligence and cooperation good. Onset. — About four years before test. Sensation. — Described in history, July, 1915: "Touch, deep sensibility, and position sense O. K. Touch uncertain {sic!) ; heat and cold not differentiated over arms and upper chest." I. Gross Disturbance. — Failure to react. Com- plete absence of sensation on upper extremities, for hot stimulus up to 52°; and for cold to 16°. Ice felt 340 Temperature Scnsiti\'ity cold when grasped in hand, and cool on right forearm after several seconds. Perversion of cold was fre- quent, but in a number of cases a dubious correction was made. Indeed, patient was generally unable to tell, when he reported first one and then the other (whether the stimulus was double or single), whether both stimuli were present or whether he had first made a mistake — the second sensation seemed to take the place of the first. II. General Character. Type of Response. — Curve shows marked predominance of H type, but greater frequency of C-H than of H-Cr C responses are infrequent except on face. Median Reaction-time. — Reaction-time is prac- tically normal. In single stimuli C seems to be a trifle slower than H, and a little delayed on abdomen. In double stimuli, second interval is slightly long in double responses; single responses are prompt except in the case of C responses on back. Extreme Reactions. — The only tendency to extreme reactions is in the case of the second interval of double responses, where half the reactions are over l.o second. Consistency. — Except for the marked inconsistency in type of separate cold stimuli (due to the frequent perversions) and a larger than normal fluctuation for this type, patient's responses are very consistent in both type and time; but in double stimuli, it is the H type, not C-H, that is consistent. Diminution of Sensation. — No diminution of sen- sation occurred in the first report; but twice on the cheeks, the double stimulus, although reported 'hot', evoked the comment: "It didn't feel either way afterward" — "it was just lukewarm afterward." A response of 'hot' was also occasionally followed by "a little cold at beginning." Intensification. — There was no evidence of this. Except for the anesthetic regions, hot seems to be intact; while cold shows some disturbance — its per- version; its slightly longer time; its inhibition in double stimuli; the fact that when the double stimu- lus was perceived, hot was more likely to follow when cold was felt first; and the few cases when cold was perceived without hot; are evidence of this. The disturbance is complete locally, and is slight generally. Case VI. — B. T. (female); age 22; stitcher; intelligence and cooperation fair. Onset. — About eight years before test. Sensation. — Described in history, Octo- ber, 1 91 7: "Over areas corresponding to sixth to eighth cervical segments of cord, on right side, there is diminished pain and temperature sensation, but touch is but slightly dimin- ished." I. Gross Disturbance. — Failure to react was the rule on left arm and was frequent on right arm. Perversion of cold was uniform on right arm. II. General Character. Type of Response. — Curve is normal except on arms and face. Median Reaction-time. — Number of cases is few, because patient came only once. Times seem to be somewhat longer than normal in all cases except the second interval of double responses. This may be due to insufficient adaptation to the task or is, per- haps, an individual characteristic. The longest times are for H responses to double stimuli. Extreme Reactions. — Occur with frequency only in single responses to double stimuli. Consistency. — Cannot be measured from one ob- servation. Diminution of Sensation. — All hot stimuli (tem- peratures 40-44) were called warm. Intensification. — There was no evidence. The perversion of cold and the fact that single responses to double stimuli were more often 'hot' than 'cold', suggest that cold is somewhat more dis- turbed than hot. The disturbance is local and quite marked. TABES DORSALIS The five cases studied show some distur- bance of a general character — marked in two; slight in two; and moderate in one. The disturbance is shown by some tendency toward perversion; by a marked deviation from the normal type of response; by a ten- dency toward somewhat lengthened reaction- times — especially with the double stimulus; by the number of extreme reactions; the in- consistency of response; and by a tendency toward intensification of the sensation. Anal- yses of the cases follow: Case VII. — G. C; age 39; barber; intel- ligence rather inferior. Onset. — About two years before test. Sensation. — Described in history, June, 191 7: "Hypesthetic to pin-prick and to cotton wool." I. Gross Disturbance. — No failure to react. Perversion of cold occurred consistently on lateral surface of both legs. Temperature Sensitivity 341 II. General Character. Type of Response. — Curve shows more single reactions than normal on every region except, perhaps, the face; being par- ticularly high on legs, abdomen and back. H-C re- sponses are slightly more frequent than C-H. There is a considerable number of doubtful responses. Median Reaction-time. — In single stimuli all regions give lengthened times except neck, abdomen and back for H; and C is somewhat slower than H. In double stimuli, C-H responses show much lengthened first interval and long second interval on left leg and back. H-C responses are long in both intervals. Single responses are also long, H some- what less than C. Extreme Reactions. — Extreme reactions are fre- quent — more so in double than in single stimuli, and more for the first interval and for single responses than for the second interval. The tendency for both interv'als is more evident in H-C than in C-H type. Consistency. — Reaction to all kinds of stimulus is ver>' inconsistent in type and fluctuating in time, except for the first interval of C-H responses. Diminution of Sensation. — Diminution of hot occurred to some extent in both single and double stimuli; cold was diminished to a much greater degree than in other subjects, chiefly in double stimuli. Intensification. — Some refle.xes were present but were not ver>' marked. They accompanied cold stimulation and double stimulation, and were mostly on abdomen and back. Cold is evidently somewhat more disturbed than hot — it is more lengthened; shows some perversion; occurs somewhat less frequently than H as the first response; and is considerably diminished in inten- sity. The disturbance is quite marked and is general rather than local. Case VIII. — P.O.; age 36; railroadman; intelligence and cooperation good. Onset. — About three years before test. Sensation. — Described in history, May, 191 7: "Touch and pain showed hypesthetic region on back, with bordering hyperesthetic area. Vibratory sensation diminished on lower extremities." (Temperature not men- tioned.) I. Gross Disturb.\nce. — No failure to react. Per\^ersion of hot on soles of feet; of cold on back. II. General Ch.^racter. Type of Response. — All are rather frequent. On the legs, C-H and C predominate; on neck and back, H. Median Reaction-time. — In single stimuli medians are about normal, but H is somewhat quicker than C. In double stimuli the reaction is slower and more variable; first inter\'al of C-H and second interval of H-C are about normal. H responses are generally slower than C. Extreme Reactions. — In -single stimuli extreme reactions are rare and occur only for C. Double stimuli increase the number of extreme reactions, particularly for single responses and for second in- terval of double. Hot, either as first or second re- sponse, is more apt to be extreme. Consistency. — In single stimuli, consistency in type is normal for both, but fluctuation in time is marked for C. In double stimuli, no type gives even a mod- erate degree of consistency; in time, the fluctuation is greater than for single stimuli and is noticeably greater with C-H than w-ith H-C responses. Diminution of Sensation. — There is no evidence of this. Intensification. — Reflex retractions, sometimes so pronounced as to interfere with stimulation, occurred on legs, abdomen and back, chiefly accompanying double stimulus. They were most violent on back, and increased rather than diminished as experiment progressed. Patient declared the stimulus felt 'scalding' and 'icy'. With separate stimuli often, and not infrequently with double, patient reported a second hotter or colder sensation from one to four seconds after first. Although in single stimuli C is somewhat slower and more apt to be extreme than H — as well as more fluctuating — in double stimuli, the opposite is the case: hot is delayed more, whether as first or second response or whether it occurs alone, and is more apt to be extreme, although it is still less fluctuating than cold. The disturbance is moderate and general, though less marked on upper extremities. Case IX. — J. Co.; age 37; tailor; intelli- gence good, cooperation fair. Onset. — About seven years before test. Sensation. — Described in history, October, 1 91 7: "Hj'pesthesia partial to pain and vibratory sensation." I. Gross Disturbance. — Failure to react oc- curred, chiefly with hot stimulus on sole of left foot. Perv-ersions of both stimuli occurred — of hot on face; more often of cold, on foot and back. II. General Character. Type of Response. — Curse shows C-H responses in general to be most frequent; H-C very infrequent; C almost absent 342 Temperature Sensitivity except on legs where they predominate; H very frequent on right arm, abdomen and back. Median Reaction-lime. — The only tendenc\- toward lengthening of reaction-time occurs for the second interval of H-C responses. Extreme Reactions. — The number of e.xtreme reac- tions is small. It is greater for single stimuli than for double, except in the second inter\al, and occurs only with hot stimuli. Consistency. — In single stimuli, consistency in both t>'pe and time is quite normal. In double responses, while no t>pe has a verj- high consistency, the C-H type attains it in forty-live per cent, of the areas. The fluctuation in time is not great. Diminution of Sensation. — There is a considerable tendency to report 'warm' instead of 'hot' — slightly greater in the case of double than of single stimuli. Twice, an initial perception of warm was followed by hot, not more than 1.5 seconds later. No diminu- tion of cold was reported. Intensification. — Stimulation on back was some- times accompanied by a reflex, but the tendency was not marked, and there was no verbal comment. The type of response suggests that on the legs the hot sensation is more interfered with than the cold; elsewhere there is not apparent a greater disturbance of one than the other, except in the rather general tendency toward a diminution of the intensity of hot. The disturbance is verj' slight, but general. Case X. — L. F. (female); age 53; cook; intelligence rather inferior; cooperation poor, influenced by suggestion.* Onset. — About two years before test. Sensation. — Described in histor%^ Sep- tember, 191 7: "Normal." I. Gross Disturb.vnce. — Xo failure to react. Per\-ersion of cold was general — the response usually H-C. This was evidently due, at least in part, to suggestibility and inference, as H-C responses were frequent to hot stimulus also. When error was pointed out to her on third day, the number of wrong responses was reduced to four, all of which occurred on arms and legs. II. Gener.\l Char.\cter. Type of Response. — Curve shows H-C t>pe to be decidedly predominant, ' Patient was unduly impressed by instruction to react as quickly as possible, and evidently inferred that nearly all stimuli were double. She seemed to think the test a 'treatment' and to consider the stimulus more important than the response. On the third day her error in giving double responses to single stimuli was pointed out. and the error was reduced, though not eliminated, by her in- creased attention. but its significance is questionable, as single stimuli also gave this response most frequently. Single C responses occur only on arms and legs; H only on abdomen and back. Median Reaction-time. — In single stimuli, while there is much variability owing to the small number of correct responses, no significant tendency to long reaction-time is apparent. In double stimuli, the only lengthening occurs for the second inter\'al of both C-H and H-C responses on the face, and in general for C responses. Extreme Reactions. — There is no ver>' marked ten- dency to extreme reaction. It is perhaps most signifi- cant in single responses to double stimuli; but while in single stimuli all extreme reactions are to hot, in single responses to double stimuli they are all to cold. Consistency. — In spite of three trials instead of two for each area, single stimuli gave a low degree of con- sistency in type and a larger than normal fluctuation in time. Double stimuli — with three instead of four trials — were quite consistent in the H-C type. Fluc- tuation in time was not marked and was normal for second inter\-al of both t>pes. Diminution of Sensation. — There is no evidence of this. Intensification. — Reflexes are ver>- numerous and occur on left arm and face as well as legs, abdomen and back. They occur more with single than with double stimulation and more with cold than hot. On abdomen, patient, asked which was more intense, called the cold (of 19°) 'worse' than the hot (44°). One can hardly say that either sensation is more disturbed. At least a tendency for hot to take pre- cedence, in spite of apparently greater intensification for cold, seems to be present. There is evidenth- a slight general disturbance the real nature of which is, however, masked by patient's desire to 'do well'. Case XI. — H. D.; age 49; metal lather; intelligence and cooperation good. Onset. — .^bout three years before test. Sens.\tion. — Described in historj', June, 191 7: O. K. (except blurring vision). I. Gross DiSTt.TRB.\N'CE. — Neither failure nor |>er- version occurred. II. Gener.\l Character. Type of Response. — Curse shows a marked predominance of C, with in- frequent double responses. Median Reaction-time. — Reaction is gencralK- quite slow. In single stimuli C is slower than H; but in double stimuli H-C and H are slower than C and C-H. Temperature Sensitivity 343 Extreme Reactions. — The tendency to extreme reactions is very marked in double stimuli, rather more when both stimuli are perceived. In all types the tendency is greater for hot than for cold, but this is not evident in single stimuli. Consistency. — Single stimuli give perfectly con- sistent responses in type, and the fluctuation, though greater than normal, is not extreme. In double stimuli, C type is quite consistent; fluctuation is high, especially for double responses. Diminution of Sensation. — There was some general diminution of hot, more in single than in double stimuli, and of cold on the left hand. Intensification. — The abdomen and back gave, not infrequently, slight reflexes; but they were evidently to considerable extent inhibited. The sensation was extremely disagreeable, according to patient's state- ment, especially cold. Although in single stimuli cold is somewhat slower than hot; in double stimuli the hot sensation is evidently more disturbed, as is shown by the greater frequency of the cold response, and the greater delay and greater tendency to extreme reactions in hot responses, whether in double responses or alone. The disturbance is marked and general. MULTIPLE SCLEROSIS The six cases studied show sHght distur- bances which are general in four, and local in two. Disturbance is shown chiefly by devia- tion from the normal type of response to double stimuli; by lengthened second inter- vals of double responses; and by some ten- dency toward diminution of sensation. Analy- ses of the cases follow: Case XII. — E. Wa. (female); age 43; housewife; intelligence good, cooperation ex- cellent. Onset. — About eighteen months before test (remission for months until a year previous). Sensation. — Described in history, January, 1918: "Intact — pain, touch, temperature." I. Gross Disturbance. — No failure to react. Perversion not present. II. Gener.\l Character. Type of Response. — The curve shows a tendency to predominance of H-C and H responses, particularly on upper part of body; the tendency is much less clear on legs, ab- domen and back, and is rather strikingly reversed on the left leg. Median Reaction-time. — In single stimuli medians are normal with slight variability, e.xcept, perhaps, on the neck and left leg for C, and abdomen for H, where there is some retardation. In double stimuli, double responses show a slight lengthening of first interval and a strikingly long second interval, the more remarkable, considering the promptness of reaction separately. C and H responses are delayed on legs and abdomen — H more than C. Extreme Reactions. — The tendency toward extreme reactions in single stimuli, and in the first interval of double responses, is negligible; while for the second interval ninety per cent, of responses are i.o second or more. Single responses to double stimuli show the tendency to a small degree. Consistency. — In single stimuli there is no incon- sistency in type. The fluctuation in time, however, is rather more than normal. In double stimuli no type has a very great consistency — H being the most consistent. In time, double responses have not more than normal fluctuation in the first interval, but fluctuate markedly in the second. Diminution of Sensation. — There is some diminu- tion of hot, considerably more frequent in single than in double stimulation. In a considerable number of these cases, however, the initial perception of warm was followed by hot, from .8 to 4.6 seconds later. In several of these cases the interval was over 2.0 seconds, which is in itself quite unusual. No diminution of cold was reported. Intensification. — There were no reflex retractions, although in this case, as in most multiple sclerosis cases, jerks of the legs were frequent — a random jerk, rather than an attempt to withdraw from the stimulus. Patient often reported, however, that sensation was a 'shock', which made the distinction difficult; but she asserted positively that it had no sting like very hot or very cold sensations. Although hot tends to take precedence, except on legs and abdomen, it also shows a tendency to diminution and to an exaggeration of what may be called the crescendo eff'ect (perception of warm fol- lowed by hot). It can hardly be said that one sensa- tion is more disturbed than the other. What is chiefly disturbed is the immediate perception of the doubleness, regardless of which is perceived first. The disturbance is slight and general, although it is more marked on legs and abdomen. Case XIII. — ^J. Q.; age 25; occupation not stated; intelligence and cooperation fair. Onset. — About three years before test. Sensation. — Described in history, June, 1916: O. K. Dorsal roots were cut, July, 344 Temperature Sensitivity 1916. Sensory examination, October, 1916, showed analgesia; temperature anesthesia on ventral surface of leg — left more than right; and hyperesthesia to cold on dorsal aspect of thighs and lateral part of legs. I. Gross Disturbance. — Failure to react oc- curred once but was insignificant. Perversion was not present. II. General Character. Type of Response. — The curve is quite fluctuating for different regions. It is to be noted that H-C responses do not occur; that the abdomen has a normal percentage of C-H responses; and that, except on face and neck, C is more frequent than H. Median Reaction-time. — In single stimuli medians are prompt, with small variability — C slightly quicker than H. In double stimuli, C responses are normal whether alone, or as first interval; while hot is delayed both in H and in C-H responses. Extreme Reactions. — The only tendency to extreme reactions was in second interval of double responses. Most marked on legs and back; least on abdomen. Consistency. — Single stimuli give very consistent responses both in type and time. In double stimuli, C-H and C responses each occur twice on nearly half the areas. In time, the only fluctuation of extent is in the second interval of C-H responses. Diminution of Sensation. — Hot stimulus was regu- larly reported 'warm' — even with temperatures up to 47°. As patient was English-speaking, and as response 'hot' did occur, this is probably not a mere language habit. No diminution of cold was reported. Intensification. — Refle.xes occurred with cold stim- uli and with double stimuli to which response was 'cold' on back. (Temperature of 47° evoked no reflex.) Jerks of legs, but not withdrawals, also occurred. While both sensations appear to be normal in single stimulation, double stimuli show some inter- ference with the hot in the long second interval, and in the small number of cases in which hot occurs alone, as well as in the delay of such responses. The disturbance is slight but general. Case XIV. — G. P. ; age 38; tailor; intelli- gence and cooperation fair. Onset. — Four years before test. Sensation. — Described in history, April, 1918: Normal. I. Gross Disturbance. — No failure to react. One instance of perversion of hot occurred. II. General Character. Type of Response. — Curve is practically normal — e.xcept on legs, where percentage of double responses falls and that of C rises. C is much more frequent than H. Median Reaction-lime. — In single stimuli medians are normal, with a normal amount of overlapping of C and H. In double stimuli, H-C typo shows some lengthening of first interval, and of second interval on legs. Lengthening of second interval on legs is also present in C-H responses, which are other- wise normal. Single responses show some delay. Extreme Reactions. — Extreme reactions are not very frequent and occur chiefly in second interval of double responses — more in H-C than in C-H. Consistency. — In single stimuli consistency is nor- mal, for both type and time. In double responses, the only abnormality in type is in the number of areas giving consistent C responses. In time the fluctuation is generally greater than normal, and is marked in the second inter\'al of H-C responses. Diminutiofi of Sensation. — 'Warm' was more fre- quent than 'hot'; and somewhat more frequent with double than with single stimuli. Occasionally, that is in three instances, the perception of warm was followed by hot — i.i, 1.5, and 2.0 seconds later. The only instance of diminution of cold occurred on the hand, where the response 'cold' followed 2.0 seconds after 'cool'. Intensification. — There were frequent jerks of legs, but not a real withdrawal. On abdomen, however, patient sucked in his breath at the cold stimulus, and expressed surprise at its intensity. The interference, in this case, is chiefly in the im- mediate perception of doubleness — whichever re- sponse came first — although when either was in- hibited, it was more likely to be hot. The disturbance is very slight, affecting chiefly the legs. Case XV. — J. T.; age 19; student; in- telligence and cooperation good. Onset. — About eleven months before test. Sensation. — Described in history, Novem- ber, 1917: O. K. I. Gross Disturb.ance. — No failure to react. Perversion not present. II. General Character. Type of Response. — Curve shows H type to be most frequent, except on face, where C-H prevails; C type is very infrequent except on face. Median Reaction-time. — In single stimuli times are normal; but are longer on legs than elsewhere. H seems to be a little quicker than C. In double Temperature Sensitivity 345 stimuli, C-H responses siiow a delay only in the second interval on right leg; H-C responses are pro- longed in both intervals on face, legs, and abdomen — and in second interval on right arm. Single re- sponses are slow on legs. Extreme Reactions. — There is some tendency toward extreme reactions, which is considerable only in the case of the second interval of double responses, especially of the H-C type. Consistency. — In single stimuli there is no incon- sistency in type, but rather more than normal fluc- tuation in time — especially for C. In double stimuli H type has a high degree of consistency, and C-H responses are fairly consistent. The fluctuation in time is not considerable e.xcept for H-C type, al- though C type shows much greater fluctuation than H. Diminution of Sensation. — There is a slight degree of diminution of both sensations, more in the case of double than of single stimuli — chiefly on legs and face for hot, and on legs and arms for cold. Intensification. — The only suggestion of this was two reflex withdrawals on back; and one verbal re- port of 'very hot'. Single stimuli show scarcely any abnormality, but suggest a slight retardation of cold and a slightly greater variability. Double stimuli confirm the view that cold is more interfered with — it is likely to be absent; if present, it is much more likely to be followed by hot than to follow it; and it shows more fluctuation and more extreme reactions than hot either as the second interval or as a single response. It is interesting to note, in this connection, the num- ber of uncertain responses, as when patient started to enunciate 'cold' but changed his mind; or said 'cold' and then thought it was a mistake. As in case of J. CI., hot, when perceived, very often seems to dis- place cold. The disturbance is moderate and general. Case XVI. — I. R. (female) ; age 29; knitter; intelligence and cooperation good. Onset. — About six years before test. Sensation. — Dorsal roots were cut Janu- ary, 1915 — L II, III, V, S I. Examination in June, 1915, showed loss of sensations of touch, heat and cold over area L I and L II segments. I. Gross Disturbance. — No failure to react. Perversions not present. II. Gener.\l Character. Type of Response. — Curve shows much apparent fluctuation between C-H and H-C type (number of cases was small as patient came only once) but is evidently normal, except for a larger projiortion of single responses than normal on legs and back. Median Reaction-time. — Single stimuli give per- fectly normal times except on left leg, where both C and H show some delay. Double stimuli appear slightly to increase the time of the first interval, and very slightly the second interval — and more for C-H than for H-C responses, except on right arm. Extreme Reactions. — Practically the only extreme reactions occurred in the second interval of double responses, more often in H-C than in C-H responses. Consistency. — No measure, as there was only one observation. Diminution of Sensation. — While there were no reports of 'warm', and only occasionally one of 'cool', patient said that sensation on legs 'felt funny . . . only a little . . . different from rest of body'. Intensification. — Reflexes were frequent on legs and abdomen, but the sensation was not described as intense. What slight disturbance there is seems to be about equal for both sensations. There is a very slight disturbance, affecting chiefly the legs. Case XVII. — J. B.; age 42; peddler; in- telligence and cooperation good, but patient had only slight knowledge of English. Onset. — About seven months before test. Sensation. — Not noted in history, April, 1918. I. Gross Disturbance. — No failure to react nor perversion. II. General Character. Type of Response. — Reaction is quite uniformly C-H, except for single C responses on legs and back. Median Reaction-time. — In single stimuli medians are normal ; H is consistently slightly slower than C. In double stimuli the only tendency to delay is in the second interval, which is greater than i.o second for all regions except abdomen and back. The prompt- ness of reaction otherwise makes this the more striking. Extreme Reactions. — In single stimuli there were no extreme reactions; in doable stimuli there were none except in the second interval where the majority were over 1.0 second. Consistency. — No measure as patient came only once. Diminution of .Sensation. — ^All responses to hot stimulation were 'warm' ; but whether this indicated 346 Temperature Sensitivity a real diminution of the sensation or was due to patient's slight ionse C-H Response Cold Med. p. E. No. of Cases 1st Interval 2d Interval No. of Cases Med. P. E. No. of Cases 3.0' + Total Med. P. E. Med. P. E. 1st Int. 3-0 • + 2d Int. i.o' + Total 3.0' + Total I ■9 .00 6 •9 •05 1-3 •15 6 8 II I.O ■05 6 I.O •15 13 •55 4 7 III IV •9 ■9 .8-1.2 3 I [..„ 1. 0-1.2 1-3 1.2-4.2 3 3 V 1.2 I.0-I.2 3 I.O .00 1-3 .10 4 4 1.2 I VI I.I .9-1.2 3 •9 .9-1.0 1.4 1. 4-1. 5 2 2 I.I .10 4 VII •9 .9-1.0 2 1.6 1. 2-1. 9 .6 .6-1.6 I 3 VIII ■9 ■9--9 2 1.2 I.I-I.3 .8 ■ 7-9 2 1.2 I Totals 26 20 29 6 Separate Hot Response H-C Response Hot I I.O .10 6 II I.I •05 6 III I.O .8-1.4 3 1.2 2.8 I I IV 1.6 I V I.I 1.0-1.6 3 VI 1.4 .9-1 .6 3 VII I.O I.O-I.O 2 2.1 3-2 I I VIII I.O I.O-I.O 2 Totals 26 2 2 * Total range given with less than four cases. t Patient came only once. Temperature Sensitivity 357 TABLE 4 MEDI.\N RE.\CTION-TIME FOR SEPARATE AND DOUBLE STIMULI IN PERIPHER.VL NEURITIS Case XVIII— L. R. . „..„. .. Double Stimuli Response C-H Response Cold Med. P. E. No. of Cases 1st I iterval 2d Inter\'al No. of Cases Med. p. E. No. of Cases 3.0' + Total Med. P. E. Med. P. E. 1st 3- Int. 0' + 2d Int. Total 3.0' + Total I 1.2 •05 11 1-4 •25 ■5 .10 2 27 II 1.2 .20 11 1.2 •05 ■4 .00 24 III I.I I 1-4 •05 ■4 •15 1 5 Totals 23 3 56 Separate Hot Response H-C Response Hot I 1.2 •05 13 1.8 .40 .6 .80 2 6 I.I I II 1.2 ■15 9 1.8 .40 .8 .00 I 4 2.6 I III 2.2 I Totals 23 3 10 2 Case XIX— M. F.f Separate Cold Stimuli Double Stimuli Response C-H Response Cold Med. p. E. No. of Cases 1st Inter\'al 2d Interval No. of Cases Med. p. E. No. of Cases 3.0' + Total Med. p. E. Med. p. E. 1st Int. 3.0 '-I- 2d Int. I.0' + Total 3-0' + Total I II 1.6 1.2 1.2-2.0* 1.2-2.0 2 3 1.4 1^5 •15 ■5 •4 •05 I 5 I 4-5 2.9-6.2 I 2 Totals 5 I 6 I 2 Separate Hot Response H-C Response Hot I II 1-3 1^3 I -3-1 -4 1. 2-1. 4 2 2 1-4 2.0 1. 3-1. 6 ■25 .6 I.O ■5--7 •.10 6 2 II 1-7 1 .6-4.0 I 3 Totals 4 1 6 13 I 3 CaseXX— H. T.t Double SUmuli Response C- H Response Cold No. of Cases 1st Interval 2d Interval No. of Cases No. of Cases Med. p. E. 3-0' + Total Med. p. E. Med. P. E. 1st Int. 3.0'-f 2d Int. I.0' + Total 3.0' + Total I I.I .00 5 1.0 •15 .6 ■05 I 4 11 I.I .00 4 1.0 ■05 .8 ■05 I 4 111 I.I .9-1.2 3 [I.I .10 .8 •05 4 IV (No stimuli) 1 V I.I I I.I •05 .8 .10 I 4 VI (No stimuli) 1.2 •05 .8 .00 I 4 Totals 13 4 20 Separate Hot Response H- : Response Hot I .8 .10 5 1.2 ■15 I.O ■30 7 12 11 ■9 .00 5 I.I •05 .8 •25 5 12 III .8 .8-1.0 3 I1.2 .20 1.0 .30 5 7 IV (No stimuli) 1 V (No stimuli) 1.4 •05 1.0 .10 I I VI 1.0 I.O-I.I 2 Totals 15 18 32 ■ Totals given with less than four cases. t Patient came only once. 358 Temperature Sensitivity TABLE 4. (Peripheral Neuritis concluded) Case XXI— B. P.f Double Stimuli Separate ^om st.mu,, Response C-H Response Cold Med. P. E. No. of Cases ist Interval 2d Interval No. of Cases Med. P. E. No. of Cases 30' + Total Med. P. E. Med. P. E. 1st Int. 3.0' + jd Int. i,o- + Total 3-0' + Total I II III 1.8 1-7 1.6 .10 •30 1.6-2.0 5 5 3 2.7 2.4 2.8 •30 •55 ■4 •5 .6 •05 •15 4 5 I 8 12 I 2.6 1.8 .10 I 4 Totals 13 9 I 21 5 Separate Hot Response H-C Response Hot I II III 1-9 1.6 2.1 .10 •15 1.5-2.6 5 5 3 2.3 I.I I I 30 .20 3 5 Totals 13 I I 3 5 t Patient came only once. TABLE 5 MEDIAN REACTION-TIME FOR SEPARATE AND DOUBLE STIMLXI IN A CASE OF THALAMIC LESION Case XXII— D. L. Double Stimu i Response C-H Response Cold Med. P. E. No. of Cases 1st Interval 2d Interval No. of Cases Med. p. E. No. of Cases 3-0' + Total Med. P. E. Med. P. E. 1st Int. 3.0' + 2d Int. 1.0' + Total 3-0' + Total I 2.2 •25 12 2.2 •35 4 30 II 2.0 .40 II 2.6 •55 I.I •35 4 14 17 III IV 2.3 2.4 •55 2.3-2.6 I 6 2 }- .40 1.2 .20 2 3 4 2-4 .00 I 4 V 2.4 •45 I 7 26 ■35 6 24 VI 2.6 .40 2 5 2.6 •65 I.I •30 5 14 15 2.8 •15 2 4 VII 2.3 .10 4 2.2 1.8-2.9 I.O .9-1.2 2 3 2.6 •25 5 VIU 2-5 •50 2 4 2.5 1.9-3.0 1.2 1. 0-2. 1 I 3 3 2.8 .70 4 II Totals 6 51 12 36 42 17 78 Separate Hot Response H-C Respon se Hot I* II 1-7 .10 I 12 2.0 •15 .8 •15 2 9 2.0 •35 5 III IV 2.6 2.0 •25 2.0-2.0 I 5 2 }- •15 I.I ■50 2 4 2.7 •15 4 V* VI 2.9 •15 2 4 2.2 ■30 .6 .10 I 2 8 VII* 2-7 I 2.0 .20 .6 •30 2 7 2.2 I VIII *i.7 I 3-2 3 -0-3 -5 .6 .6-. 7 2 2 1-5 I Totals 4 25 1 3 8 30 II * In response to hot stimulus: Region I gave 10 responses Co/J-Med. 2.8; P. E. .40. Region VII gave 2 responses Cold-Med. 2.7: Region V gave 6 responses Cold-Med. 3.6; P. E. .95. Region VIII gave i response Cold-Med. 3.8. range 2.7-3.2. Temperature Sensitivity 359 TABLE 6 PERVERSION OF RESPONSE — PERCENTAGE OF STIMULI GIVING OPPOSITE RESPONSE Syringomyelia Tabes Dorsalis P. K. E. \Vi. A.C. L. A. J. CI. B. T. G. C. P.O. J. Co. L. F. H. D. C H c H C H C H C H C H C H C H C H C H C H Hand, R L Forearm, R L Arm, R L Face, R L Neck" Leg, R L Thigh, R L Foot, R L Abdomen, R 1 L Back, R L lOO 40 40 40 35 50 50 50 100 100 100 50 67 {3, 50 1 33 50 50 1 33 67* 50* 50* 50* Us* 1 80* [ Right • arm [ 100 100 100 {'^ 50 50 33 33 17 17 20 17 17 r 50 67 100 67 33 33 {67 Multiple Sclerosis Neuritis Thai. Les. E. Wa. J.Q. G. p. J.T. I. R. J. B. L. R. M. F. H. T. B. P. D. L. C H C H C H C H C H C H C H C H C H C H C H Hand, R L Forearm, R L Arm, R L Face, R L Neck Leg, R L Thigh, R L Foot, R L Abdomen, R L Back, R L 33 33 100 100 50 100 100 100 100 50 50 '■ Includes some cases in which a very dubious correction was made, from .8 to 5.0 seconds later. 36o Hand, R L Forearm, R L Arm, R L 1^ Face, ^ M [l Xeck Leg, R L Thigh, R L Foot, R L Abdomen, R M L Back, R M L Hand, R L Forearm, R L Arm, R L Face, R L Xeck Leg, R L Thigh, R L Foot, R L Abdomen Back R -M L Temperature Sensitivity TABLE 7 FAU-CRE TO REACT— PERCENTAGE OF STIMtT-I GIVING NO RESPONSE S.\Tingoinyeiia P. K. A. C. J. CI. 25 80 67 100 67 40 100 100 75 Right arm — C & H-17% Left arm — C&H— 67% 38 63 Multiple Sclerosis C indicates cold stimuli; H, hot; and D, double stimuli. ♦Areas usually stimulated gave no response; responses were obtained on shoulder blades. Temperature Sensitivity TABLE 7. (concluded) 361 ■■ Tabes Dorsalis G. C. 1 P.O. 1 J. Co. 1 L. F. 1 H.D. C H D C H D c H D C H D C H D 67 25 Neuritis Thalamic Lesion L. R. M. F. H.T. 1 B.P. 1 D. L. C H D C H D C H D C H D C H D 1 33 33 i4t 50 50 25 hese p face o :rc£ f th ntage eleft 3 w har ere wit id. 1 temp era tures of I 6-1 7» md 46-51 )»C.; ten iperatur es of i8» and 40° evoked no respo nse on t le median 362 Temperature Sensitivity , ■ TABLE 9 DIMINUTION OF SENSATION— PER CENT. OF WARM RESPONSES Syringomyelia Tabes Dorsalis p K. E. Wi. A.C. L. A. J. CI. B. T. G.C. P.O. J. Co. L. F. H. D.J Region D S D s D S D s D s D S D s D S D s D S D I 97 64 94 58 67 50 100 100 100 100 38 31 32 17 71 II 76 92 70 28 100 67 100 100 100 100 12 23 50 47 18 III 67 -I 91 50 40 33 100 100 ICO 100 20 25 IV 100 100 50 50 100 100 100 100 75 50 V 83 67 81 83 8 100 100 100 100 18 44 75 86 60 VI 94 83 100 50 33 100 100 100 100 38 33 57 71 60 VII 38 50 88 100 60 25 TOO 100 100 100 33 VIII 100 75 53 50 25 100 100 100 100 27 29 DIMINUTION OF SENS.\TION — PER CENT. OF COOL RESPONSES Syringomyelia Tabes Dorsalis 1 p. K. E. Wi. A.C. L.A. J. CI. B. T. G.C. P.O. J. Co. L. F. H. D.; D s D S D S D S D S D S D S D S D S D S D I II III IV V VI VII VIII 7 10 ^■^ 50 25 14 16 50 50 27 14 14 13 TABLE 10 INTENSIFICATION OF SENSATION Sj-ringomyelia Tabes Dorsalis P. K. E. Wi. A.C. L.A. J. CI. B.T. G.C. P.O. J. Co. L. F. H. D. D S D S D S D S D S D S D S D s D S D s D S I II iC* III iHC IV V IH iC iCH 2C iC iC» VI 2M iC iHC iC iHC iC VII iHC iH 2CH iHC 3C 4HC 4H 6Ct 5Ht iCH '2< iC 2I 2H VIII 2H iC iC* 3H 2H iC iC iH 4HC 5H 6C11 5Ht iH Under D is shown the number of cases of reflex retraction from a double stimulus, with initials indicating the type C response; under S reflexes occurring with single stimuli — initials indicating which. [table 8 WILL BE FOUND ON PAGE 364J Temperature Sensitivity 363 Multiple Sclerosis Thalamic Lesion 1 30 in 1 Multiple Sclerosis Ne uritis Thalamic Lesion E. \Va. J.Q. G. P. J.T. I. R. J. B. L. R. M. F. H. T. B. P. D. L. D S D S D S D s D s D s D S D S D S D S D s 7 6 7 8 7 II 33 14 60 Under D is to be found the per cent, of responses to double stimuli including any perception of the hot stimulus, which were warm; under S the per cent, of warm responses to single hot stimuli. Multiple Sclerosis Neuritis Thalamic Lesion E. \Va. J.Q. G. P. J.T. I. R. J. B. L. R. M. F. H. T. B. P. D. L. D S D S D S D s D s D s D S D S D S D S D S 4C iC iH iC ICH 2C iC 4CH iC iC IC 6C IC 2C iCc^lH IC iC&iH *l 1l3 lespoi resp ise W8 jnses sH-( vvere ;orP ■i-C, H. t3 t2 respoi respoi ises w ises w ereH ereH -c. -C. 364 Temperature Sensitivity TABLE 8 CONSISTENCY OF RESPONSE Neuritis Syringomyelia Tabes Dorsalis Multiple Sclerosis Thalamic Single Stimuli Lesion P. K. E.Wi. J. Cl. G. C. P.O. J. Co. L. F. H. D. E.Wa J. Q. G. P. J. T. L. R. D. L. Per cent, of areas having Cold — 2 C responses 56 92 44 73 96 96 46 100 100 100 89 100 90 92 Hot — 2 H responses S4 73 100 50 92 96 69 100 100 100 96 IOC 100 44 Cold — range i.o' 14 17 16 24 17 23 23 8 23 20 Hot — range I.o' 10 II 6 10 4 8 28 20 12 4 19 27 Cold — range <.3" 20 47 19 32 60 33 35 65 77 50 12 56 24 Hot — range <.3" 33 50 69 35 64 52 28 24 42 73 52 69 75 36 Double Stimuli Per cent, of areas having 2 or more responses — C-H 40 48 J7 28 38 45 8 3 20 48 75 43 89 35 2 or more responses — H-C 8 18 33 20 5 70 25 ID 13 16 20 2 or more responses — C 18 13 8 18 33 28 15 65 15 48 23 6 48 2 or more responses — H 25 3 96 25 23 35 8 20 45 23 75 5 C-H, 1st — range 1.0' 19 35 27 54 6 33 100 23 29 25 50 2nd — range 1.0' 25 21 25 27 33 28 100 63 42 40 24 35 C-H, 1st — range < .3" 25 24 100 91 7 67 33 50 58 23 47 25 2nd — range < .3" 38 35 50 27 27 39 67 13 16 30 53 70 21 H-C, 1st — range 1.0" 33 29 65 38 25 10 40 13 2nd — range 1.0" 67 29 38 50 50 18 90 75 60 67 38 H-C, 1st — range < .3" 43 8 38 50 22 40 40 33 38 2nd — range < .3" 33 43 23 50 50 68 33 C — range 1.0" 28 60 57 46 54 33 44 50 63 C — range < .3" 29 14 15 46 17 8 17 68 II H — range 1.0" 40 40 44 21 50 28 13 7 100 H — range < .3" 10 100 22 50 22 29 50 13 39 50 43 The percentage of areas having two responses of a kind is calculated on the basis of all areas stimulated — forty for double stimuli, twenty-si,\ for single stimuli (except in cases where for some reason an area failed to receive two stimulations).* The percentage of areas having a given range is calculated on the basis of the number of areas giving two responses of the sort indicated. Figures in italics indicate percentages outside of normal limits — i. e., deviating from the normal average by more than two and a half times the average deviation. But C or H responses among normals were so infrequent that the time-fluctuation was not calculated. * For single stimuli, two responses of a kind indicates 100 per cent, for that area, in general; for double stimuli. 50 per cent. [owing to exigencies of M.\KE-UP, T.\BLES 9 .\ND 10 WILL BE FOUND ON P.\GES 362 .\ND 363] Temperature Sensitivity 365 TABLE II COMPARISON OF THE REACTION-TIME OF COLD AND HOT IN SINGLE STIMULI Multiple Sclerosis P. K. E. Wi. A. C. L. A. J. CI. B. T. Average ■ of C > H No. of Regions Average Amount Sec. ■50 •13 1.05 .48 .24 1.90 .48 Time of H > C No. of Regions Average Amount Sec. .23 •44 1. 00 .40 .ID Time of C = H No. of Regions E. Wa. J.Q. G. P. J.T. I. R. J. B. Average Time of C > H No. of Regions Average Amount .28 ■05 .10 •17 Time of H > C No. of Regions Average Amount No. of Regions Tabes Dorsalis G. C. P.O. J. Co. L. F. H. D. Average 6 4 3 5 3-6 2 •45 I .10 6 •32 5 •30 2 .80 3-2 •39 L. R. M. F. I H. T. 4 B. P. I Average 1-5 .30 •37 Thalamic Lesion TABLE 12 COMPARISON OF THE REACTION-TIME OF COLD AND OF HOT RESPECTIVELY IN SINGLE AND DOUBLE STIMULI C & C-H H & H-C C & C-H H & H-C C & C-H H & H-C C & C-H H & H-C P. K. + .1 +2.1 G. C. +6.3 +7^4 E. Wa. + 1^7 + 1.0 L. R. + •S + 1.2 E. Wi. + .8 + 1^4 P.O. +2.6 +4^8 .T.Q. - ^4 M. F. + .1 + .8 A.C. !• Co. + 1.6 + .6 G. P. + ■•3 +3^1 H.T. — .2 + 1.0 L.A. -1.8 + •! L. F. + -1 -1.8 J^T. +4^3 B. P. + 2.8 .!• CI. — .2 H.D. + ^4* +2.7 LR. + r9 + ^8 B.T. - •S* + -9* J.B. + -1 D.L. + .8 + .4 Av. - •S + 1.1 +2.3 +2.7 + -9 +2.3 (Neur.) + .8 + i^3 These figures represent the sum of the gains in time of medians (in seconds) for each individual in double stimuli over medians in single for same region. Starred figures, however, are gains of averages, so calculated because of too few cases to calculate by regions. 366 Temperature Sensitivity CHART A SLMMARY OF DISTURB.\XCE IN ALL CASES Syringomyelia Tabes Dorsalis P. K. E. Wi. A. C. L. A. J. CI. B. T. G. C. P.O. J. Co. L. F. H. D. Failure to react + + + + + ++++ + + + Perversion + + + + + + + + +++ + + + + + + + + + + + + + Type of response + + + + + + + + + + ++++ + + + + + + + + + + + + + + + + + + Median reaction-time + + + + + + + + + + + + + + + + + + + + + + Extreme reactions + + + + + + + + + + + + + ++ + + + + + + ++ + + + + + + + Consistency + + + + + + + No meas. No meas. ++ No meas. + + + + + + + + + + + + + + + Diminution of sensation + + + + + + + + + + + + + + + + + + + + + + + + + Intensification + + + + + + + + + Relatively greater dis- turbance CsHd H C C C C C C,Hd H c C.Hd Degree of disturbance + + + + + + + + + + ++++ + + + +++ + + + + + + + Disturbance general + + + + + + + + Disturbance local + + + Multiple Sclerosis Neuritis Thalamic Lesion E. Wa. J.Q. G. P. J.T. I. R. J. B. L. R. M. F. H. T. B. P. D.L. Failure to react + + + Perversion + + + + + Type of response + + + + + + + + + + + + + + + + + + + Median reaction-time + + + + + + + + + + + + + Extreme reactions + + + + + + + + + + -I-+ + + + + + + + + + + Consistency + + + + + + + + No meas. No meas. + No meas. No meas. No meas. + + + + Diminution of sensation + + + + + + + + + + + + + + + + + + + + + + + + + Intensification + + + + + + + Relatively greater dis- turbance C = H H H C C = H H — C C? ? H Degree of disturbance + + + ++ + + - + + + ? ? + + + + Disturbance general + + + + ? ? Disturbance local + + + + In each aspect of the experiment, as designated on the left, the relative degree of disturbance is indicated for each subject by the number of plus signs— except in the last two, where the sign signifies whether the disturbance was general or local in character. Temperature Scnsiti\ity 367 That there is in most of these cases a de- parture from the normal reaction is evident; and that the double stimulation shows a dis- turbance not always evident with separate stimuli is also clear. Perhaps the nature of the disturbance will be more quickly grasped by the summary in Chart A, in which the estimated degree of disturbance shown by each subject in the various aspects of the test is indicated by plus signs. The chart is also a convenient reference table for comparing the various syndromes. There are a few salient points to be noted : 1. Failures are by far the most frequent in syringomyelia. 2. Perversions are commonest in syringomyelia; but are frequent also in tabes,, and in the case of thalamic lesion. 3. In type of response neuritis deviates least from normal — but these cases for the most part showed no clear-cut disturbance. Other syndromes do not differ significantly in this respect one from another. 4. In reaction- time multiple sclerosis is nearest to normal. If reference is made to Tables 1-5 it will be seen that these cases are generally prompt in reaction to separate stimuli; and that they generally show a retardation, chiefly in the second interval of double responses. 5. Extreme reactions in the second interval of double responses only are characteristic of multiple sclerosis. Except in multiple sclerosis extreme reac- tions are also frequent in siitgle responses to double stimuli. 6. Diminution of intensity of sensation is least and intensification is most in tabes dorsalis. 7. Cold is more likely to show a greater distur- bance than hot in syringomyelia; hot more than cold in multiple sclerosis. 8. All factors considered, multiple sclerosis shows the least degree of disturbance. 9. The disturbance is unlikely to be local in tabes, most likely to be local in syringomyelia. Summarizing, one may say that under the conditions of the experiment — for these sub- jects at least — frequency of failures and of perversions, generally lengthened reaction- times, diminution of sensation, and tenden- cies toward marked local disturbance are indicative of syringomyelia. Perversions without failures; somewhat lengthened re- action-time — especially in double stimuli; tendencies toward intensification; and gen- eral, rather than local, character of distur- bance suggest tabes. A generally normal or nearly normal reaction — except for a lengthened interval between the perception of the two stimuli — is characteristic of multi- ple sclerosis. The lesion of the brain stem is distinguished from those of the cord by the unilateral character of the disturbance, and by the clear-cut selection of one sort of sensa- tion as evidenced by the responses. Peri- pheral disturbances, such as traumatic neu- ritis, are not adequately shown by this method of testing. There are a few points which may be of general interest to the theory of temperature sensations. One is the relation of the order of response in double stimuli to the length of reaction of cold and hot separately. It will be remembered that with normals, while the time for cold and hot was about the same, the majority of responses to the double stimulus were in the order cold-hot and that these responses were slightly more prompt and less variable, in both intervals, than those in the order hot-cold; ' that the responses of hot followed quite regularly upon those of cold except on legs and feet, where it was somewhat delayed. In testing abnormals we have an opportunity to see what happens when the times of hot and cold are more variable. Table 11 gives a summary of the relation of cold and hot on each region, for the various subjects. We find, that in spite of greater variability there is, generally considered, the same overlapping in the times of cold and hot that was found among normals. There are four subjects, however, in whom is shown a fairly distinct tendency of hot to be slower — E. Wi., J. Co., J. Q. and J. B.; and there are six who show a somewhat slower time for cold— P. K., J. CI., G. C, P. O., H. D. and J. T. Of the cases in whom cold is quicker, we find, by comparing the frequency of dif- ferent types of response to double stimuli (Tables 1-5), that C-H responses exceed H-C in all cases, and C exceed H in all but one case; that C-H is the most frequent type in all but one case, in whom C has the highest frequency. On the other hand, among those 'Neurological BuUetinrloc. cit.. Table 12, p. 156 Temperature Sensitivity 100 95 90 85 90 A •rs / \ 70 / \ 65 / \ 60 A / \ SS A / \ / p50 /\ / \ / t4.5 / \ / \ / f40 / V Y 30 ^J A /\ 25 \ '' I \ /.- y- / \ 20 IS 10 05 // • ••■• '-^ s I D m W Y VI VD P.K. E.W.. 100 35 90 85 80 15 10 65 bO SS /, / / \ i \ 50 ■■■■.' 1 K / ■. i S5 / ■• X ■■/ •. / 30 25 20 15 / '. . 'tM / 10 •^\ / 1 ,'\ 1 5 \ / ■ --.J !• ^ - ' * - \ /' n m K Y VI m Temperature Sensitivity inmiTvviwM LA. inrawYvivnviii iiiinivvviwivni 1 n in w u B m H.D. Fig. I. (Top) Distribution of Types of Response IN Syringomyelia C-H -.-H-C -- . Duobtful Fig. 2. (Centre) Distribution of Types of Response in Tabes Dorsalis C-H H-C . Doubtful Fig. 3. (Lower) Distribution of Types of Response in Multiple Sclerosis ■C-H ■H-C -- . Doubtful inmffvnvnair J.T. I D IDK7Y1TOVID I.R. 369 I n ffl H V H Vff BB Temperature Sensitivity fe ■ • • > IT HI 1 n m cr Y Yi H.T. 4. Distribution of Types of Response in Peripheral Neuritis Fig. 5. Distribution of Types — C-H -.- H-C C .... H Doubtful of Response in a Capsulothalamic Lesion fu. r^u. JsH •";; Fig. 6. Extreme Reactions A graph showing manner in which extreme reactions were distributed among the different types of response who show a slightly slower reaction to cold, H-C responses exceed C-H in only one case, though H are more frequent than C in four cases; and the most frequent type in three cases is C-H or C, although in three others, it is H or H-C. Thus, the frequency of the type of response does not invariably depend on the length of the reaction to separate stimuli. Indeed, as results show, the entire character of the disturbance may be greater for hot in double stimuli, in spite of a slower cold reaction. The opposite, however, does not occur. Another point of interest is the effect of double stimulation upon the length of reac- tion-time. If the gains in the medians for the double stimulus over the single, for each region, are added algebraically, for each sub- ject, we get a rough measure of the effect of the double stimulus. This is shown in Table 12. The same sort of measure for normals can be easily computed from Table 12 of the former article. It will show that while the re- action to cold remained practically the same in C-H responses as in single stimuli (sum of the gains, total, equals — .1 second), the reaction to H in H-C responses was somewhat length- ened (sum of gains, +2.9 seconds). The tendency of the clinical subjects is in general to increase the time for either response in double stimuli. The increase is greater for hot than for cold in tsvelve cases, and less in 370 Temperature Sensitivity five — the remaining five patients having too few H-C responses to admit of comparison. The five cases showing less increase for hot than for cold are not, however, of those who react more slowly to cold. There are four cases that show a decrease in time of cold in double stimuli; in only one, however, is the difference very considerable. There is but one case showing a decrease in time of hot; this is the subject who reacted to almost all stimuli — single and double — by H-C. The tendency is, then — as in normals — for double stimuli to increase the reaction-time of hot more than of cold. This leads us to the third point of general interest: that the C-H type of response in- dicates the least abnormality of response. It was noted even among normals that H-C responses were more irregular when they occurred than C-H, but it was assumed that this was because of the fewer number of cases of this type. Among the subjects of the present inquiry the same phenomenon ap- pears. A slighter degree of disturbance for C-H than for H-C responses is shown by every subject — whether cold or hot is more disturbed — in the time of either first or second interval, or the proportion of extreme reac- tions, or in all three ways — except in the case of D. L., in whom H-C responses, although less frequent, are shorter in both intervals and less apt to be extreme. Single responses to double stimuli, are in general, longer and more apt to be extreme than double responses; that is, they indicate the greatest divergence from the normal, except in multiple sclerosis, where the only significant divergence is the lengthening of the second interval. Here it may be observed that this seems to imply that the process of dissociation in multiple sclerosis is of a different character from that in syringomyelia or tabes in that, instead of retarding the initial perception when the double stimulus is given, the ten- dency is to lengthen, if not to inhibit, the per- ception of the second. Whereas in tabes, the prolongation of the initial perception is quite marked (c/. Table 12). In syringo- myelia, however, the reaction is slow even for separate stimuli, and the curious result is that in three subjects, C-H responses are quicker than responses to cold separately — another proof of the slighter disturbance indicated by the C-H type of response. It may be that this retardation of the perception of the second stimulus is dependent on the degree of disturbance rather than charac- teristic of a particular syndrome: H. T., a neuritis case with scarcely any disturbance, and J. CI., a syringomyelia case in whom the disturbance is quite limited, show the same general tendency. Nevertheless, since other subjects in whom the disturbance is slight do not show it, and since all the cases of multiple sclerosis tested — even those in whom the dorsal nerve roots had been severed — do show it, it seems fair to regard it as characteristic of multiple sclerosis. To return to the points of theoretical inter- est. The three tendencies mentioned — that, even when cold is separately more delayed, hot is less apt to predominate in double stim- uli; that in general, the perception of hot in double stimuli is more increased in time than cold, and that the exceptions to this are not the cases in which cold is separately slower; that C-H responses are nearest to normal whichever sensation is more disturbed — sug- gest that the double stimulation acts in a way which is not merely the additive effect of the two stimuli, but which indicates some relation between the two tending to delay the perception of hot. The theory that best explains these facts, is, in my opinion, that which regards the per- ception of hot as a fusion of impulses from cold and warm spots. On this hypothesis, the double stimulation requires the cold spots to do double duty, so to speak, and naturally the fulfillment of their secondary function is delayed. This hypothesis also explains most easily the cases of perversion — at least those which do not occur with complete consistency. If cold takes part in the perception of hot, then it is not difficult to imagine the fusion dis- turbed and cold perceived as hot or vice versa. The tendencies are, however, not sufficiently unequivocal to make this interpretation of more than doubtful correctness — and there are two cases whose reaction is difficult, though not impossible, to explain on this assumption. These are J. CI. and J. T., 37^ Temperature Sensitivity who show a normal reaction to cold and hot separately, somewhat quicker, how- ever, for hot, but a considerable interference with the reaction to cold in double stimuli. It is difficult to see how the reaction to hot can remain intact, on our hypothesis, when cold is disturbed. It is necessary to assume that the cold, in exercising its second function, loses its identity in its first function because the second reaction comes so quickly. The number of instances for which both of these subjects started to say "cold" but said "hot" instead, makes this explanation not improb- able. CONCLUSIONS 1. Patients sufTering from syringomyelia, tabes dorsalis, multiple sclerosis, and the t\'pe of thalamic lesion presented in this study, generally show some disturbance in reactions to double thermal stimuli, whether or not evinced by ordinary clinical tests. The tem- perature test, as herein presented, is not, however, well adapted to the investigation of peripheral disturbances. 2. There is a larger proportion than normal of single responses to double stimuli. 3. There is more than a normal tendency, in many cases, to call temperatures of 41° to 44°C. or higher, warm instead of the usual hot. 4. The disturbance is generally shown by an increased reaction-time. 5. The variability in time is greater than normal: the range of medians for different regions being greater and the consistency of single areas less. 6. Reactions to double stimuli are in general more disturbed than reactions to single; and H-C responses more than C-H. 7. C-H responses indicate the nearest ap- proach to the normal reaction; they are the most frequent type in the least disturbed cases (with one exception) ; and they are nearest to normal in time for each individual. 8. The type of response to double stimuli, and the character of disturbance, cannot be predicted from the response to single stimuli. 9. The differences between the different syndromes are sufficiently distinct to be sug- gestive, although there is some overlapping. Multiple sclerosis has a particularly clear-cut characteristic — a dissociation of the percep- tion of doubkness. 10. The double stimulation acts in such a way as to delay the perception of hot more often than that of cold. EXPL.\N.\TION OF T.\BLES Tables 1-5 show the median reaction-times of each patient in the five syndromes studied, both for separate and double stimuli, and for each type of response; the P. E., the number of extreme reactions, and the total number of responses for each region. Tables 6 and 7 present the evidences of gross disturbance — the percentages of stimuli that gave failures and perverted responses. Table 8 gives the percentages of consistency in type of response and in time, as is explained above. A glance at the italicized figures shows that the inconsistency was considerable among these subjects. Table 9 shows the percentages of responses that were warm and cool, instead of Iiot and cold, in both separate and double stimuli. This is quite striking in view of the fact that seven per cent, of the total was the highest percentage reached by normals. Table 10 gives the number of cases for each region of each subject in which an intensifica- tion of the sensation might be assumed by the reflex accompaniment. This is of interest in comparing the different subjects and syn- dromes, although no standard can be offered for reference. Table 1 1 presents a comparison of the time of cold and hot responses in single stimuli. It gives for each subject the number of regions on which cold gave a longer time than Jwt with the average difference in time; the number of regions and the average time differ- ence where hot was longer than cold; and the number of regions on which the time for cold and hot was the same. Table 12 is a comparison of reaction-times of cold in single and double stimuli and similarly of hot. Figures 1-5 present graphically the manner in which responses to double stimuli were dis- tributed among the different types of response. Figure 6. Figure 6 is a graph showing the distribution of extreme reactions among the different tjpes of response in the various syndromes. VITA The author was born in New York, November 28, 1892. She graduated from Girls' High School, Brooklyn, received the A.B. degree from Cornell University where she was elected to Phi Beta Kappa, and pursued graduate study leading to the degree of Ph.D. at Columbia University. Date Due APR 1 t ?? m - c '^' ■■ 1 1 1 (|) .-^ ^ /