Wl^Jtt jtf&bl *8r*>Jfr- w! *€ f* J ■nLt*< (Bnrncll BmnerBtty library iltljaca, Netn gntb THE GIFT OF ' . ■ "* i» Carnapie, 1-w.stitu.ti'o-n. Cornell University Library QP 915.A3M643 Effect of alcohol on psycho-physiologica 3 1924 003 195 090 The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924003195090 Publications of the Nutrition Laboratory. Publications will be sent post-paid at the prices indicated by application to the Carnegie Institution of Washington, Washington, D. C. Orders should be accompanied by a remittance payable to the Carnegie Institution of Washington. No. 42. Atwatek, W. O., and F. G. Benedict. A Respiration Calorimeter with Appliances for the Direct Determination of Oxygen. Octavo, 193 pages, 49 figures. Published 1905. Price $1.25. No. 77. Benedict, Francis G. The Influence of Inanition on Metabolism. Octavo, v +542 pages, 2 figures. Published 1907. Price $2.75. No. 123. Benedict, Francis G., and T. M. Carpenter. Respiration Calorimeters forStudying the Respiratory Exchange and Energy Transformations of Man. Octavo, vn+102 pages, 32 figures. Published 1910. Price $0.75. No. 126. Benedict, Francis G., and T. M. Carpenter. The Metabolism and Energy Trans- formations of Healthy Man during Rest. Octavo, vm +255 pages. Published 1910. Price $1.75. No. 136. Benedict, Francis G., and Elliott P. Joslin. Metabolism in Diabetes Mellitus. Octavo, vi +234 pages, 2 plates. Published 1910. Price $2.00. No. 155. Benedict, Frances G., and Edgar P. Slack. A Comparative Study of Temperature Fluctuations in Different Parts of the Human Body. Octavo, v +73 pages, 38 figures. Published 1911.' Price $0.50. No. 166. Benedict, Francis G. The Composition of the Atmosphere with Special Reference to its Oxygen Content. Octavo, in, 115 pages,' 1 plate, 1 figure. Published 1912. Price $2.00. * No. 167. Benedict, Francis G., and Walter G. Cady. A Bicycle Ergometer with an Electric Brake. Octavo, in, 44 pages, 16 figures. Published 1912. Price $0.50. No. 176. Benedict, Francis G., and Elliott P. Joslin. A Study of Metabolism in Severe Diabetes. Octavo, vi+135 pages, 1 figure. Published 1912. Price $1.25. No. 187. Benedict, Francis G., and E. P. Cathcart. Muscular Work: A Metabolic Study with Special Reference to the Efficiency of the" Human Body as a Machine. Octavo, 176 pages, 1 plate, 10 figures. Published 1913. Price $2.50. No. 201. Benedict, Francis G., and Fritz B. Talbot. The Gaseous Metabolism of Infants with Special Reference to its Relation to Pulse-rate and Muscular Activity. Octavo, 168 pages, 65 figures. Published 1914. Price $1.50. No. 203. Benedict, Francis G. ' A Study of Prolonged Fasting. Octavo, 416 pages, 5 plates, 47 figures. Published 1915. Price $4.00. No. 216. Carpenter, Thorne M. A Comparison of Methods for Determining the Respiratory Exchange of Man. Octavo, 265 pages, 74 figures. Published 1915. Price $2.50. No. 231. Benedict, Francis G., and H. Mtjrschhatjser. Energy Transformations during Horizontal Walking. Octavo, 100 pages, 7 figures. Published 1915. Price $1.00. No. 232. Dodge, Raymond, and Francis G. Benedict. Psychological Effects of Alcohol. An Experimental Investigation of the Effects of Moderate Doses of Ethyl Alcohol on a Related Group of N euro-muscular Processes in Man. Octavo, 281 pages, 1 plate, 32 figures. Published 1915. Price $2.50. No. 233. Benedict, Francis G., and Fritz B. Talbot. The Physiology of the Newrborn Infant. Character and Amount of the Katabolism. Octavo, 126 pages, 10 figures. Pub- lished 1915. Price $1.00. No. 261. Benedict, Francis G., and Thorne M. Carpenter. Food Ingestion and Energy Transformations with Special Reference to the Stimulating Effect of Nutrients. Octavo, 355 pages, 2 figures. Published 1918. Price $3.50. No. 266. Miles, Walter R. Effect of Alcohol on Psycho-Physiological Functions. Octavo, 144 pages, 15 figures. Price $1.50. No. 279. Harris, J. Arthur, and Francis G. Benedict. A Biometric Study of Basal Metab- olism in Men, Women and Children. Octavo. In press. No. 280. Benedict, F. G., W. R. Miles, Paul Roth, and H. Monmouth Smith. Human. Vitality and Efficiency under Prolonged Restricted Diet. Octavo. In press. Miles, Walter R. Effect of Alcohol on Psycho-physiological Functions. (Carnegie Institution of Washington, Publication No. 266, 1918.) 144 pages and 15 figures. $1.50. This alcohol study is unique in that it is a comparison of results from two series of experiments, performed on the same subject, employing identical apparatus, technique and laboratory con- ditions, but by different investigators. Such repetition series of experiments are urged as funda- mental to the establishment of trustworthy results in this field. These data supplement and amplify those reported by Dodge and Benedict on the "Psychological Effects of Alcohol" (Car- negie Institution of Washington, Publication No. 232) and form another contribution on the psychological side under the tentative plan for alcohol investigation at the Nutrition Laboratory. The dose was 30 c.c. of absolute alcohol diluted with water, orange juice, and saccharin to a total volume of 150 c.c. The normal and alcohol data were equal in amount and control doses were used on the normal days, The subject who served in the repetition experiments treated of in this report was the one (No. VI) of the normal group employed by Dodge and Benedict who showed the smallest general effect of the alcohol. The experiments were more intensive than in the previous series, covering as they did 5 hours per day, for 6 consecutive days. Of the 30 results in the second series whioh may be taken as indicators of the alcohol effect, 27, i. e., a ratio of 9 to 1, show inferior functioning of processes after the dose. There are 16 cases in which the change wa3 less than 10 per cent and 11 cases in which it ranged from 10 to 37 per cent. No one day exercised a predominating influence on the general findings. The two series of results are in practical agreement as to the direction of the alcohol effect, and together they establish tho fact that this subject was influenced by alcohol in substantially the same way as five or six other normal men, although he clearly demonstrated certain individual peculiarities. Appended material in the report supplies data concerning the use of alcoholic beverages outside of the psychological laboratory by the subjects of Dodge and Benedict, the general effects of alcohol as reported by these subjects, and also some recalculations of the previously published data. Miles, Walter R. L'effet de Valcool sur les fonctions ^psycho-physiologiques. (Carnegie Insti- tution of Washington, Publication No. 266, 1918.) 144 pages et 15 dessins. 81.50. Cette etude a 1'egard de l'aloool est unique parce qu'elle represente une comparaison des resultats de deux series d'experienees poursuivies sur le meme sujet aveo les memes appareils, la meme technique, et dans le meme laboratoire, mais poursuivies par des. experimentateurs differents. L'auteur considere que la repetition des experiences etait necessaire dans cette re- cherche pour obtenir des resultats concordants auxquels on put accorder une pleine confiance. Ces donnees forment une nouvelle contribution aux recherches psychologiques "poursuivies con- formement au programme du "Nutrition Laboratory" dans son etude sur l'influence de l'alcool; elles completent celles de M.M. Dodge et Benedict sur "Des effets psychologiques de l'alcool" (Carnegie Institution of Washington, Publication No. 232) . Le sujet soumis a l'experience buvait 30 cc. d'alcool absolu, dilues dans un melange d'eau, de jus d'orange, et de saccharine, l'ensemble formant un volume total de 150 cc. On fit absorber au sujet deux sortes.de doses egales en volume, , les unes contenant de l'alcool, les autres n'en contenant pasl Les doses alcoolisees furent administrees en nombre egal a celui des doses privees d'alcool et sans que le sujet connut rien de la composition des unes ni .des autres. Le sujet avait ete tire d'un groupe d'individus normaux deja examines par les Professeurs Dodge et Benedict, et le choix s'etait porte sur lui parce que 1'on avait reconnu en lui une plus grande force de resistance aux effets de l'alcool. Ces dernieres experiences furent conduites d'une f aeon plus intensive que celles de la serie prece- dente; elles durerent 5 heures par jour pendant 6 jours cOnsecutifs. Sur les 30 resultats obtenus, qu'on peut considerer comme indicatifs de l'influence de l'alcool, il y en a 27 (e'est-a-dire, un rapport de 9 a 1) qui demontrent un affaiblissement des fonctions psycho-physiologiques apres l'ingestion de la dose d'alcool. On trouva 16 cas dans lesquels 1' affaiblissement fut estime a moins de 10%- et 11 cas dans lesquels il fut juge avoir varie de 10 h 37%. Les deux series de resultats sont pratiquement d'accord qiiant a l'effet deprimant de l'alcool sur les fonctions psycho- physiologiques, et, rapprochees, elles etablirent le fait que l'alcool avait presque la m6me influence sur ce sujet que sur cinq ou six autres hommes normaux, bien qu'on eut observe chez ce sujet No. VI certaines singularites individuelles. L'etude se termine par une discussion concernant l'emploi des boissons alcooliques en dehors du laboratoire psychologique par les sujets des Professeurs Dodge et Benedict; par une con- sideration des effets generaux de l'alcool basee sur les recits de ces m6mes sujets; et, enfin, par de nouveaux calculs sur des donnees anterieurement publiees. , Toutes les publications de la Carnegie Institution of Washington sont de-posees dans la plupart, des grandes bibliothegues des villes et des universiUs du monde.. Miles, Wai/teK R. Uber den Einfluss des Alkohols auf die psyehophysiotogisehen Vorgange beim Menschen. (Carnegie Institution of Washington, Publication Nr. 266, 1918.) 144 Seiten, IS Figufen. $1.S0. Diese Abhandlung ist beabhtenswert, weil sie eine Vergleichung def Resultate von zwei Ver- suchsreihen gibt, die an derselben Versuchsperson mit denselben Apparaten und derselben Technik, und unter denselben Labofatoriumsbedingungen, aber von versfihiedenen Fbrschern ausgefiihrt wurdeii. Eine solche wiederhdlte Versuehsfeihe scheint dem Verfasser unbedihgt notwendig zu sein Urn zuverlassige Ergebnisse auf diesem Gebiete zu liefern. Diese Resultate bilden die zweite psychologische Kontribution in bezug auf die nach dem Program des Nutrition Laboratory ausgefiihrte Alkoholuntersuchung. Sie dienen aueh als eine Erganzting und Erwei- terung zu den von Dodge und Benedict berichteten Ergebhissen "Uber die psydhologischen Wir kungen des Alkohols" (Carnegie Institution of Washington, Publikation Nf. 232). Der Alkohol wurde mit Wasser, Apfelsinensaft und Saceharin verdiinnt genommen, und zwar wurden 30 com 100 %igen Alkohols in einer Dosis von 150 ccm gegeben. Es wurden so viel normale- wie Alkoholversuohe angestellt, und Kontrolldosen ohne Alkohol in den Normalversuchen gegeben. Die Versuchsperson der Wiederholungsreihe war die (Nr. VI) von Dodge und Benedict in der Normalreihe beniitzte und auf welche die allgemeine Alkoholwirkung die gerihgste war. Diese VersuchSreihe war intensiver als die vorhergehenden Versuche, da sie 5 Stundeh pro Tag wahrend 6 aufeinander fdlgenden Tage einschloss. Von den 30 Resultaten der zweiten Reihe, welche als die Alkoholwirkung anzeigend angesehen werdeh konnen, gibt es 27 (d. h., ein Verhaltnis von 9 zu 1) Welche eine Herabsetzung der psycho-physiologischen Vorgange nach Alkoholzufuhr beweisen. In 16 Fallen betrug diese Veranderung weniger als 10 prozent und in 11 Fallen betrug sie von 10 bis 37 prozent. Bei alien Versuchen fand man ungefahr dieselbe Alkoholwirkung. Die zwei Reihen von Resultaten stimmen gut hinsichtlich der Alkoholwirkung uberein, und zusammen bestatigen sie die Tatsache, dass der Einfluss des Alkohols auf diese Versuchsperson bienahe derselbe war wie auf fiinf oder sechs andere normale Manner, obgleich diese VerBUchs- person gewisse individuelle Eigentumlichkeiten deutlich zeigte. Die Behandlung schliesst mit einer Besprechung uber die Beniitzung von alkoholischen Ge- tranken ausserhalb des psychologischen Laboratoriums von den Versuchspersonen von Dodge und Benedict, und die allgemeinen Wirkungen des Alkohols nach Aussage dieser Personen; sie gibt auch einige neue Berechnungen der frilher veroffentlichten Ziffern. AUe Publikationen der Carnegie Institution of Washington sind den meisten grossen Stadt-und Universitdtsbibliotheken der Welt dedizieft. EFFECT OF ALCOHOL ON PSYCHO-PHYSIOLOGICAL FUNCTIONS By WALTER E. MILES «5>" Published by the Carnegie Institution of Washington Washington, 1918 1. L t \ CARNEGIE INSTITUTION OF WASHINGTON Publication No. 266 PRESS OP GIBSON BROTHERS, INC., WASHINGTON, D. C. PREFACE. The experimental results presented in this monograph supplement and amplify the published work of Dodge and Benedict on the Psycho- logical Effects of Alcohol, and f orm another contribution on the psycho- logical side under the tentative plan for alcohol investigation at the Nutrition Laboratory. It is a pleasure to acknowledge my obligation to the above investigators, both of whom generously spent much time counseling with me on various phases of the work. The report has had the further advantage of the competent editorial supervision of Miss A. N. Darling. I was assisted in the experiments by Mr. E. S. Mills and in the preparation of the alcohol and control doses by Miss E. B. Babcock. Walter R. Miles. Nutrition Laboratory op the Carnegie Institution op Washington, Boston, Massachusetts, March 28, 1918. 3 CONTENTS. Page. Introduction 7 Subject selected for the repetition of the measurements 10 Experience of Subject VI 12 General experimental conditions 15 Amount of alcohol employed 19 General effects of the alcohol 20 Measurements employed and their sequence 23 Statistical method of presentation of results 29 Data obtained in the repetition measurements 36 Patellar reflex 37 Protective lid reflex 48 Eye reaction to peripherally appearing stimuli 59 Reaction time in reading isolated words 62 Memory 67 Sensory threshold for faradic stimulation 74 Eye movements 79 Finger movements 84 Pulse and respiration 92 Conclusions 125 Appendix I. — Data concerning the use of alcoholic beverages outside of the psycho- logical laboratory by the subjects of Dodge and Benedict 135 Appendix II. — The general effects of alcohol, doses A and B, experienced by the subjects of Dodge and Benedict's experiments 138 Appendix III. — Observations on the averages of data for normal and alcohol days presented by Dodge and Benedict 140 Appendix IV. — Observations on the association pulse data of Dodge and Benedict . . . 143 DLLUSTRATIONS. Fig. 1. Patellar-refiex records from 30 and 50 gram stimuli 44 2. Effect of alcohol on latency of the patellar reflex with two intensities of stimulation 46 3. Effect of alcohol on amplitude of the patellar reflex with two intensities of stimulation 46 4. Effect of alcohol on latency of the lid reflex 53 5. Effect of alcohol on amplitude of the lid reflex 53 6. Specimen records of lid reflex of Subject VI 54 7. Typical record of memory measurement for Subject VI 68 8. Typical eye-movement records of Subject VI reproduced unretouched from both series of experiments 80 9. Average number of finger oscillations in four successive 2-second intervals for each period on normal and alcohol days 88 10. Differences for finger oscillations in 2, 4, 6, and 8 second intervals for each period after the control and alcohol doses 89 11. Pulse and finger-movement records 96 12. General averages for pulse rate during pre-tetanus, tetanus, and post- tetanus periods on three normal and three alcohol days, after the control or alcohol dose 101 13. Average duration of pulse cycles on three normal and three alcohol days during pre-tetanus, tetanus, and post-tetanus periods after the control or alcohol dose 103 14. Duration of pulse cycles in first, second, and third periods on normal and alcohol days in pre-tetanus, tetanus, and post-tetanus periods 104 15. Average duration of pulse cycles under experimental conditions 1 to 7 on normal and alcohol days, as compared with pulse duration during word reactions (5 and 6) 115 4 EFFECT OF ALCOHOL ON PSYCHO-PHYSIOLOGICAL FUNCTIONS By WALTER R. MILES INTRODUCTION. The experiments reported in detail in the following pages give infor- mation regarding the effect of moderate amounts of ethyl alcohol upon the human organism. Since numerous workers have concerned them- selves with this problem, it may seem to many readers that nothing remains to be said. However, the fact is that no solution seems to have been sufficiently substantiated to have gained general recognition. The results of the many experiments show numerous disagreements and there seems no way of harmonizing many of the contradictions. It is unsafe to discredit totally certain experiments that do not harmon- ize with others, for each experiment must be regarded as a sincere attempt to answer scientific inquiry, and it must be remembered that the problem is by no means a simple one. There are many known variables which exercise an influence on the results and there are doubt- less as many unknown and unrecognized factors. If all the conditions of performed experiments in this field were known, doubtless unex- plained contradictions would still exist. In previous alcohol experimentation, particularly with human sub- jects, there has been much variation in the experimental conditions. The dose employed has ranged in amount from 10 to 100+ c.c. of abso- lute alcohol, and in concentration from pure alcohol through all the degrees of dilution with numerous liquids. The time of taking the dose and its relation to previous ingestion of food have varied, as well as the physical and mental condition of the subject. The processes measured, and the indicators of the alcohol effect used by the several investigators, have likewise differed. Furthermore, the subjects varied as to the pre- vious taking of alcohol; some of them were habitual drinkers (a very few), others were more or less irregular in their use of alcohol, and still others were total abstainers previous to the experiments. Since con- trol doses have rarely been used, the subjects of the experiments have, in many cases, known when the alcohol has been administered; even with control doses alcohol in any considerable quantity can generally be recognized by certain characteristic subjective sensations. Finally, the interpretation of the results is complicated by the fact that the effect of a so-called "moderate" amount of alcohol is not conspicuously large as measured by the usual methods. Taking into consideration all these variations in the experimental conditions, it would seem that sufficient complexity existed to warrant one in expecting a considerable conflict in results. The tendency has been for each experimenter in this field to try new methods of attack; the variety of results has thus continually increased. More refined methods are constantly available and should by all means 7 8 Effect of Alcohol on Psycho-Physiological Functions. be employed. Many techniques must be discarded because of faults in operation, or because later work shows them to be defective. But a certain amount of well-chosen repetition of experiments would be of great benefit to this problem. The path of repeated experiment, while in many ways unattractive, offers the only way to the harmonizing of the contradictions and to the final establishment of the solution of this problem. Certainly any one contemplating experimentation on the psychological effects of alcohol should fairly consider the possibility of making his experiments in such a way as to repeat the work of some previous investigator in this field. In the work done by Kraepelin and his pupils there was a limited amount of repetition of experiments, with the result that progress has been made. Probably criticism would have been less had there been more repetition and verification of the experi- mental findings. Of course, strictly speaking, no experiment can ever be repeated. This is perhaps as literally true when working with the nervous system of human subjects as in any other domain. Thus the complexity of the experimental material is an argument for the best possible standardization of the experimental conditions if it is hoped to isolate the effect of certain consciously varied factors. If different investigators would independently employ the same dosage, apparatus, methods, and procedure, their combined results would be of unique importance. It would then be more nearly possible to discover whether the different results actually represent (1) individual differences in sus- ceptibility to the action of the alcohol (a general conclusion which is now used to explain all sorts of contradictions in experimental results) or (2) varied experimental conditions. No very well recognized psychological measurements or procedures in taking such measurements have been advocated for use in testing for the effect of a drug on human organisms. In much of the work special apparatus has been employed; the duplication of this special apparatus is expensive and is frequently regarded as unnecessary. It has seldom, if ever, happened in this field that an investigator has had the oppor- tunity which came to the writer of stepping into a laboratory equipped and used by others and attempting independently to repeat a part of the work of his predecessor. The possible value of repetition of experiments for the thorough establishment of experimental results in this field was not so apparent to the author at the time of experimentation as at the present writing. In the spring of 1914, at the close of Professor Dodge's experimental work at the Nutrition Laboratory, the least promising of the normal group employed by Dodge and Benedict (Subject VI) 1 was chosen for repetition work. It is now regretted that diligent effort was not made to secure repeat experiments with the other sub- jects. While not all of the subjects would have been available, it is possible that some of them could have been secured for further experi- 1 Dodge and Benedict, Carnegie Inst Wash. Pub. No. 232, 1915. Introduction. 9 menting. This should not be construed as signifying in any way or manner a doubt concerning the experiments and findings of Dodge and Benedict. That these deserve the confidence of the reader has been proved by a rechecking of the experiments from original protocols and in many cases a recalculation of the results with very minor corrections, as indicated in the following pages. The new set of experiments on Subject VI has shown so many agreements with the previous findings for this subject as to convince us of the worth of this method of attack- ing the problem. Furthermore, the comparison experiments have established certain individual differences as to the effect of alcohol upon this subject which could not otherwise well have been proved. There is ample precedent for the publication of experimental results obtained on only one subject; in fact, most alcohol researches have em- ployed a very limited number of subjects. Ach, 1 Frankfurther, 2 Furer, 3 Hellsten, 4 Lombard, 5 Mayer, 6 Scheffer, 7 Totterman, 8 and Vogt, 9 as well as others, report experiments performed on one individual. The studies mentioned are among those in which the author has in each instance experimented on himself. This method of experimental self- observation, while having certain advantages, can not usually be depended upon to produce the objective results which are so much to be desired in this field. Ideally, alcohol research should be conducted in such a way that adequate data would be recorded and reported on each one of a large number of subjects who are chosen in accordance with certain classifi- cations. The results, then, for the individual and for the group would be adapted to statistical treatment. The obstacles which confront the single investigator who would complete such a research on human sub- jects, while not insurmountable, are nevertheless such as to make it a labor of years. In the meantime other investigators, who are working on the same or related problems, should have some of the data made available for their use and criticism. Logically one of the first steps toward the broader solution of the problem is to discover how large a variety and how many measurements of each process must be taken on the individual in order to obtain a satisfactory sample of his perform- ance under alcohol. The data for the individual must be subjected to statistical treatment in so far as practicable and the experiments re- peated to discover agreements or disagreements between successive sets of results. Accordingly, while reports on single subjects may be con- 1 Ach, Kraepelin's Psychol. Arbeit., 1901, 3, p. 203. 2 Frankfurther, Kraepelin's Psychol. Arbeit., 1914, 6, p. 419. 3 Furer, Ber. lib. d. V. intern. Kongr. z. Bekampf. d. Missbr. geist. Getrankes zu Basel, 1896, p. 355. * Hellsten, Skand. Arch. f. Physiol., 1904, 16, p. 139. 5 Lombard, Journ. Physiol., 1893, 13, p. 1. 6 Mayer, Kraepelin's Psychol. Arbeit., 1901, 3, 535. 7 Scheffer, Arch. f. exp. Pathol, u. Pharmakol., 1900, 44, p. 24. 8 Totterman, Finska Lakaresallskapets Handlingar, 1916, 58, p. 1527. * Vogt, Norsk Magazin for Leagevidenskaber, 1910, 8, p. 605. 10 Effect of Alcohol on Psycho-Physiological Functions. sidered uneconomical or even of minor value from some standpoints, they have a specific usefulness. There are, moreover, certain advan- tages in the individual experiment; the experiments are apt to be more carefully made, the details and special experimental incidents are more faithfully reported, and the results may be more thoroughly analyzed. The single subject of the alcohol experiments treated of in this mono- graph was an adult male, 26 years old, who had been found 1 to show the smallest general effect of the alcohol. This second series of experi- ments employed the same alcohol dose as was used for the first series, i. e., 30 c.c. of absolute alcohol, the smaller dose used by Dodge and Benedict, and it was compounded in the same way. Measurements of the various processes studied by them were repeated, with only minor changes in technique and procedure, in the same laboratory with the same apparatus and thus under almost identical conditions. Longer experimental days (5 hours in contrast to 3 hours) were used in the second series; since these days were successive, the apparatus was con- tinually in adjustment for the subject, with the result that a minimum of experimental time was lost. The data are thus more extensive than any that have heretofore been gathered on a single subject over a simi- lar period of time; 2 but still there is much to be desired. There are as many normal as alcohol data. The comparison of the results from the two sets of experiments has necessitated a large expenditure of time in working over the Dodge and Benedict material. Certain minor corrections and notes concerning their publication are embodied here which seem of significance either to the former or to the later experiments. No attempt has been made to review other recent literature in connection with these findings, and this omission here must be pardoned. The unique feature of this study is the comparison of two series of alcohol experiments, one more inten- sive than the other, performed by different investigators under other- wise identical laboratory conditions and on the same subject. The comparison has been made as simple and as uninvolved as possible. SUBJECT SELECTED FOR THE REPETITION OF THE MEASUREMENTS. The experimental measurements employed by Dodge and Benedict in their study on the psychological effects of alcohol were chosen with the definite purpose of bringing together coordinated data concerning the most fundamental aspects of neuro-muscular action. It was their conviction that the attempt to secure accurate measurements of as 1 Dodge and Benedict, Carnegie Inst. Wash. Pub. No. 232, 1915, p. 55. 2 Other individual studies on the influence of alcohol have continued over a longer period and have in this sense been more extensive, as, for example, the work of Frankfurther concerning the effect of alcohol on the complicated process of learning typewriting, and that of Vogt on memory and alcohol. But none of these studies employed such a variety of measurements which at the same time concern closely related processes. Dodge and Benedict did not use all their measure- ments on the same day, except in their 12-hour experiments. They divided the experimental processes into two convenient series (Dodge and Benedict, Carnegie Inst. Wash. Pub. No. 232, 1915, p. 15) and gave these series on different days. Selection of Subject. 11 complete a group as possible of systematically related phenomena is a sound procedure which is better adapted to revealing the effects of alcohol, particularly with human subjects, than is the measurement of special, isolated processes. The individual subjects frequently showed considerable variation which, in the case of an isolated process like reaction time, may cause confusion in the interpretation of the results. When experimenting with a group of related phenomena, the various measurements supplement each other; accordingly, while the subjec- tive variations may be as large, they do not exercise proportionately as marked an effect on the general results. Thus, a group of processes selected because of their (1) inter-relation, (2) elementary nature, and (3) theoretically practiced condition, provided the techniques and con- trols are adequate, should yield results of such definiteness that a repe- tition of the experiments with the same subject or subjects would not change the findings essentially. In choosing a subject for a repetition of the Dodge and Benedict series of experiments, we were fortunate in securing the services of the one who had shown the most irregular effects of the alcohol, and, on the average, an effect which was less than that for the other normal sub- jects. 1 This seemed the crucial subject with whom to test the general proposition of a repeat experiment, in which the technique employed in the earlier investigations should be used. On several pages in their publication 2 Dodge and Benedict call special attention to this subject, who showed rather wide variation from the average for the normal group of subjects, as will be considered later in the discussion of the results obtained with the various measurements. Dodge and Bene- dict make this general statement concerning him: "This tendency of Subject VI to differ from the group was a troublesome matter to handle. It was possible that he actually presented a physiological exception to the average effects of alcohol as represented by the rest of the group. As far as our controls went, however, it was not impossible that he was taking food or drugs that masked the effect of the alcohol. It was further possible that his was one of the cases of chance variation. In any event it appeared advisable to repeat the experiments on him after his work in the medical «chool had closed in June." 3 On the suggestion of Professor Dodge, who knew these conditions but had not fully revealed them to the writer, Subject VI was chosen for the second series of observations. It was our desire that the two series be elaborated entirely independently, that they might the better serve for a critical comparison. 1 As is found, Dodge and Benedict (Carnegie Inst. Wash. Pub. No. 232, 1915, p. 263, table 50, Subjeot VI) demonstrated an alcohol effect less than the average in patellar reflex, lid reflex, eye reactions, and word reactions, i. <•-., all of the more simple processes; see statements at bottom of page 245 of the same monograph that "the lowest centers are depressed most and the highest least." The effect was larger than the average in memory, faradic threshold, and finger move- ments. Eye movements would have been included here also, but the data for movements to the left were insufficient. 2 Dodge and Benedict, Carnegie Inst. Wash. Pub. No. 232, 1915, pp. 55, 71, 72, 133, and 209. * Dodge and Benedict, ibid., p. 55. 12 Effect of Alcohol on Psycho-Physiological Functions. EXPERIENCE OF SUBJECT VI. Prior to June 29, 1914, the date on which the second series of experi- ments began, Subject VI had served parts of 21 days, a total of approx- imately 70 hours, as a subject for Professor Dodge. The dates for these days, with information concerning the time of day for the partic- ular session, the approximate number of hours' work, the number of experimental periods of repetitions of the series of tests used on the particular day, the doses of alcohol given, together with an indication of the measurements used, are presented in table 1, sections i, n, and in. 1 This table has been compiled solely with the intention of pre- senting a statement showing the experience which the subject had had with these experiments under the general laboratory conditions which obtained. It is not a tabulation of comparable data, for not all of the measurements taken on the several dates and classified under one head are comparable. As an example, normal eye movements according to the table would appear to have been taken on October 22, November 5, November 19, 1913, and January 1, 1914. However, those so classi- fied for November 5 and 19 were corrective eye movements during the reading of moving words, and can not be compared with the eye move- ments, right and left, through an arc of 40°, the common measurement used under this name. It was merely a matter of convenience to classify these data for November 5 and 19 in this way, as it represented practice in the eye-movement measurements with the Dodge technique. As outlined in the tentative psychological program, 2 Professor Dodge proposed several measurements, data for which are not presented in the published report of Dodge and Benedict. Some of these measurements were tested, as has been instanced in this case of eye movements, but the amount of data secured with the whole group of subjects was insuffi- cient to warrant its publication. In section n the data which have already been published in Dodge and Benedict's report are indicated by the letter P. The unpublished material represented in section I of the table, December 22 to 24, 1912, was obtained in connection with a series of measurements on the effect of acidosis. These observations 1 There are certain errors concerning subject and date of experiment in the Dodge and Benedict publication. Those which are most likely to confuse the careful reader should be corrected as follows: The 12-hour experiments with Subject VI came on January 1 and 2, 1914, while those for Subject IX came on December 22 and 23, 1913. At places these dates have become confused, giving the false impression that both subjects were used on the same day in these long experiments; page 65, Subject VI, December 22, 1913, should be changed to read January 1, 1914; page 65, Subject VI, December 23, 1913, should be changed to read January 2, 1914; page 66, Subject VI, Con., December 23, 1913, should be changed to read January 2, 1914; page 67, Subject IX, January 1, 1914, should be changed to read December 22. 1913; page 68, Subject IX, Con., January 1, 1914, should be changed to read December 22, 1913; page 68, Subject IX, January 2, 1914, should be changed to read December 23, 1913; page 157, Subject IV, February 18, should be changed to read February 13; page 158, Subject IV, Con., 12-hour experiment, should be changed to read Subject VI, Con., 12-hour experiment; page 218, Subject VII, the second October 8 should be changed to read October 28 (Dose A) ; page 260, fourth line from top of page. Subject VII should be changed to read Subject IX. 2 Dodge and Benedict, Carnegie Inst. Wash. Pub. No. 232, p. 272. Experience of Subject. 13 Table 1. — The experience of Subject VI in the measurements used. 1 ja u ^ °*. M Section, date, 1-3 U .3 m P. -° s ° 8 n ■o o •g 1* 8-3 u ■9 5 and time of day. 0. O S - s if 0) ft I O 1 o u CJ 03 (0 U •O u o □3 m 03 © 1 s a >> I I a I Section I. 1912. hours. Dec. 22, a. m. 1.5 1 N N.P. N.P. N.P. N.P. N.P. N.P. Dec. 22, p. m. 1.0 1 N N.P. N.P. N.P. N.P. N.P. N.P. Dec. 23, a. m. 1.5 1 N.P. N.P. N.P. N.P. N.P. N.P. .... Dec. 24, a. m . 1.0 1 N.P. N.P. N.P. N.P. N.P. N.P. .... Section II* 1913. Oct. 7, p. m. 3.0 4 N P p P P Oct. 14, p. m. 3.0 8 A P p P P Oct. 22, p. m. 3.0 6 N P P P p" N.P. P Oct. 29, p. m. 3.0 6 A P P P P P Nov. 5, p. m. 3.0 4 N P N.P. P N.P. P Nov. 12, p. m. 3.0 6 A P N.P. P N.P. P Nov. 19, p. m. 2.0 4 N P N.P. N.P. P Dec. 2, p. m. 2.5 6 A P N.P. P Dec. 8, p. m. 1.0 1 N p Dec. 9, p. m. 2.0 6 N p N.P. Dec. 16, p. m. 3.0 2 A P NP 1914. Jan. 1, day. . 3 12.0 10 N P P P P P P P Jan. 2, day. . 3 .12.0 10 C P P P P P P P Jan. 22, p. m. 2.0+ 6 B P P P P P Jan. 28, p. m. 3.0 8 B P N.P. P P Feb. 4, p. m. 2.0+ 6 B N.P. P . . . • Feb. 12, p. m. 3.0 3 A N.P. N.P. N.P. N.P. N.P. N.P. P Mar. 2, p. m. 3.0 6 N N.P. P Section III. 1914. June 29, a. m. 5.0 4 N P P P P P P P P P June 30, a. m. 5.0 5 A P P P P P P P P P July 1, a. m. 5.0 5 N P P P P P P P P P July 2, a. m. 5.0 6 A P P P P P P P P P July 3, a. m. 5.0 6 N P P P P P P P P P July 4, a. m. 5.0 6 A P P P P P P P P P 1 The use of the letter P in the table means in section II that the data were published by Dodge and Benedict; in section III that the data were taken in the second series and published in this mon- ograph. The designation N.P. means throughout the table that the data were taken but not pub- lished, apparently because the techniques were not Btrictly comparable; see p. 12. 2 The designations in this column have meanings as follows: N, normal day; A, alcohol day with dose of 30 c.c, i. e., absolute alcohol diluted to a total volume of 150 c.c. given after the first period; B, alcohol day with a dose of 45 c.c, i. e., absolute alcohol diluted to 225 c.c. given after the first period; C, alcohol day on which 9 separate doses, each containing 12 c.c. of absolute alcohol diluted to 60 c.c, were taken. 8 The 12-hour experiment began at 8 a. m. with the first period normal. Alcohol (dose C) was taken at 9 h 15 m , 10 b 15 m , and ll h 15 m a.m. (no alcohol with dinner at 12 h 20 m p.m.), then again at l h 25 m , 2 h 20 m , S^O", 4 h 25 m , 6 h 20 n , and e^O" p. m. The experiment continued until 8 p. m. The state- ments made by Dodge and Benedict (their p. 21) are misleading here concerning the number of doses taken in the 12-hour experiments. Subject VI took 9 doses (see note 2) and Subject IX took 8 (see Dodge and Benedict's report, p. 30). These doses were diluted with water, cereal coffee, or other flavoring liquid, to a total volume of 60 c.c. 14 Effect of Alcohol on Psycho-Physiological Functions. served primarily as a test of the technique to be subsequently employed in the alcohol investigation. 1 The material given in section in of the table was obtained in the repetition series of experiments and is published in the present report. These data are comparable with each other. At the commencement of this second series of experiments it may be supposed that the subject was thoroughly familiar with the conditions of the various techniques employed, having had a rather extensive experience scattered over a period of months. Any interest due to the novelty of the measure- ments must have been lost by this time. As the disposition of the apparatus had not been altered and the general laboratory conditions 2 were the same in both series, the subject understood from the beginning the adjustments to be made on his part. None of the measurements were entirely new to him, and every effort was made to have the in- structions identical in the two series. Whereas in the first series only three or four of the tests were given on any one experimental day, ac- cording to the grouping described by Dodge and Benedict on pages 14 and 15 of their report, in the second set all the measurements were used daily, thus providing a greater variety but necessitating longer experi- mental periods. While Subject VI served willingly 3 and seemed to cooperate in the measurements, one constantly had the impression that he was not try- ing hard in the tests in which voluntary functions were involved, such as finger movements, word reactions, memory, and general tetanus for producing rapid changes in pulse rate. He of course had little, if any, personal interest in the experiments as such. The only records he ever saw were those on the kymograph, and these, to mere inspection, showed no indication of the effect of alcohol. This tendency to take things easy was natural with this subject and was favored, no doubt, by the previous experience, with its many repetitions of instructions and measurements. The use of consecutive experimental days inten- sified this difficulty. While familiarity with the conditions, by pro- ducing greater relaxation, theoretically might favor such measurements as patellar reflex and pulse, frequent repetition, as, for illustration, of the fist of 24 words used in the word reactions at the rather slow rate of one word every 10 seconds, with the considerable demand made upon the subject's attention, no doubt grew tedious. With practice there devel- oped some habits of reaction to the experimental situations which are not the most favorable to the comparison of these results with the former series, in spite of our every effort to make the conditions iden- tical. It is not possible to equate the results for these factors, and 1 Dodge and Benedict, Carnegie Inat. Wash. Pub. No. 232, 1915, p. 17. * Ibid., p. 30 ft. Also consult fig. 1 for the general plan of the psychological laboratory and apparatus. * He was paid by the hour for his services, as were all t he subjects used by Dodge and Benedict, excepting No. X. General Experimental Conditions. 15 aside from some mention of them in the succeeding pages under the discussion of particular measurements, it must be assumed that they tend in general to compensate each other. GENERAL EXPERIMENTAL CONDITIONS. At the time of the second series of observations, June 29 to July 4, 1914, the subject, having completed the year's work at the medical school, was serving as intern in a neighboring hospital. His hours on duty were from late afternoon until 1 a. m., and as he had considerable freedom from work he regarded the period as vacation. His health was excellent. On no day did he mention feeling the effects of a cold or a headache. His habits were regular with respect to food and daily routine. The last food before each experimental session was a hearty meal taken 7 hours previously, from which coffee and tobacco were excluded. This meal was taken between 12 h 30 m and 1 o'clock a. m., that is, when the subject went off duty at the hospital. While the food was varied as to kind, the amount was fairly constant. No alcohol was taken during the week, other than that given at the Labo- ratory. The subject slept approximately 2 hours in the afternoon, be- fore taking up his work at the hospital, and again from 1 until 8 a. m. Immediately after dressing in the morning and without breakfast he came to the Laboratory, and our experiment began at 8 h 30 m a. m. and continued until l h 30 m p. m., a total period of 5 hours, as contrasted with the 3-hour periods employed formerly. The longer periods were used designedly, that the effect of alcohol might be followed further than in the previous investigation. According to the statements of the subject, the longer periods were not unsatisfactory. The variety and nature of the measurements were such as to minimize the development of fatigue, and the periods were not long enough to cause serious difficulty from excessive hunger. The subject preferred to serve one long period rather than two short ones. From the experimental standpoint longer sessions are more economical, in that much less time is taken up in the preliminary adjustments, such as weighing the subject and gathering and recording data relevant to food and general conditions. The longer periods have the further advantage of calling forth a more steady, even performance on the part of the subject. In the former series the subject was working hard in regular course in a medical school; it was not feasible for him or the other student subjects to serve in the experiments on successive days, and an effort was made to guard against any prominent weekly rhythm that might exist by having all the experiments come on the same day of the week. In the later experiments the need of extending the sessions over a period of weeks to avoid any influence of a weekly rhythm seemed less urgent than at the time of the previous experimenting, as the regular 16 Effect of Alcohol on Psycho-Physiological Functions. work of the subject was lighter during the second series. It was, how- ever, more from the practical standpoints of availability of the subject and laboratory convenience that the measurements were made on consecutive days. The alcohol was given on the second, fourth, and sixth days. This frequency in administering the alcohol presents the well-known possi- bility of an additive effect. 1 On the other hand, in experiments in which the additive effect has been demonstrated, the alcohol dose has always been greater than 30 c.c. On close analysis it must be clear that such an effect, provided it exists in a measurable degree with doses of 30 c.c, would, so far as these experiments are concerned, theoretically cause a smaller resulting difference to be shown between alternate alco- hol and normal days, as such an effect from an alcohol day would apply more to the succeeding normal day than to the following alcohol day, from which it is separated by 48 hours. 2 The indicated alcohol effect is possibly somewhat reduced by the use of successive days. It might be supposed 3 that in this second series of experiments every effort would be made to exercise a rigid regulation and control of the life of the subject during the period of experimentation, especially in consideration of the fact that Dodge and Benedict felt that it was not impossible that Subject VI was taking food or drugs that masked the effect of the alcohol. 4 This matter in reference to the later experiments was carefully considered in conference with Dodge and Benedict, and it was decided not to require the subject to live at the Laboratory dur- ing the experimental days, under conditions of constant observation. Moreover, it did not seem desirable 5 to rigidly predetermine the kind, amount, and time for taking food and fluids antecedent to the experi- ment, or other conditions, such as the amount and kind of mental and physical work. As a matter of fact, these conditions were as well reg- 1 A. Smith, Beitr. u. d. V. intern. Kongr. z. Bekampf. d. Missbr. geist. Getranke, Basel, 1896; Fiirer, ibid.; Kurz and Kraepelin, Kraepelin's Psychol. Arbeit, 1901, 3, p. 417. 1 Kurz and Kraepelin, Kraepelin's Psychol. Arbeit, 1901, 3, p. 454. 3 Rivers, in a review of the Dodge and Benedict report (Science, 1916, n. s., 44, p. 102) says: "In such survey work, in which a number of subjects were employed, it was perhaps impossible to regulate their lives more completely and thus bring the research nearer to the ideal of the method of difference, but this regulation should not be neglected in more intensive work." * The original protocols for February 4, 1914 (see table 1, section II) are pertinent to this dis- cussion. The subject came to the Laboratory at 4 h 10 nl p. m.; his last food was at 12 h 30 m p. m., and consisted of chicken soup, potatoes, and plum pudding. He had drunk one bottle of beer the previous evening. During the preliminary adjustments the subject remarked : " I came prepared for the test to-day." When pressed to explain what he meant, he said, "lam not as tired as last time, and I drank all the water I needed before I came." On this day use was made of dose B, that is, 45 c.c. of absolute alcohol diluted to a total volume of 225 c.c. The subject asked per- mission to urinate at 5 h 25 m p. m. The remarks made by the subject on this occasion, together with the fact that the influence of alcohol on Subject VI seemed to be different from the average of the normal group, aroused some suspicion. Nothing has been discovered, however, since that time to substantiate this suspicion in any way, and it seems probable that the subject's "pre- paredness" consisted solely in his having taken an extra drink of water. Although the alcohol was given in a 20 per cent solution, it is evident that under these conditions of the food taken only three hours before the experiment and water immediately before, the dilution was considerable after the alcohol had reached the stomach. 6 Dodge and Benedict, Carnegie Inst. Wash. Pub. No. 232, 1915, pp. 19 and 27. General Experimental Conditions. 17 ulated during this experimental period as they would normally be in the life of any business or professional man, unless he lived under mili- tary training conditions. From the time the subject entered the Lab- oratory until the experimental session for the day was complete, he was continuously under observation. Following the set of experiments a careful physical examination was made. Information obtained from this, together with information obtained from those who knew the sub- ject best, confirmed our belief that he was sincere in his attitude toward the experiments and in his desire to cooperate in them. It is believed that the previous suspicion that the subject might be tampering with the experimental results was not well founded, or at least does not apply to the present series, and that the affidavit made by the subject is to be taken at face value. This affidavit was as follows : 1. The data given by me concerning the last ingestion of food and the use of alcohol were correct according to the best of my knowledge and belief . 2. There has been no conscious effort or intention to modify the results of the experiments. 3. There has been no discussion of the experiments and of probable results with any person outside the Psychological Laboratory. 4. There has been no habitual use of drugs during the experimental week, no use of alcoholic beverages outside the Laboratory, and no occasional use of any drug that might modify the effect of the alcohol, as far as I know. In order that the conditions for the comparison of the results of the two series of experiments may be more fully before the reader, the fol- lowing section of personal history given by Dodge and Benedict 1 is here reprinted. SUBJECT VI. Date.— October 7, 1913. Family history. — Both father and mother American, Scotch descent; mar- ried 28 years. One brother, not living, 21 years. None of the family take alcohol or drugs. There is no insanity in the family, and no alcoholism in the collateral branches. Personal data. — Age, 25 years; height, 164 cm.; weight, 68 kilos. Occupa- tion, student, second year medical school. Sport, walking 2 miles a day. Education. — Oklahoma Agricultural College. Average scholarship; best in biology, worst in English grammar. Memory. — Poor, verbal. Not quick, accurate, long, or responsive. Non-abstainer. — Drinks beer, etc., at banquets; 1 or 2 glasses at a time; effect, stupefying. Largest amount ever taken, 10 or 12 glasses, mixed drinks, in the evening, one year previous; "attempted to get drunk"; stupe- fying effect; only time ever intoxicated. Last used, October 3, 1913, one glass of beer. Two glasses of beer can be taken on a full stomach without noticeable effect. First noticeable effects are drowsiness and unsteadiness. Produces no excitement, though subject is normally nervous; causes talkativeness, nor- mally moderate in speech; produces a feeling of elation, normally cheerful. No peculiar sensations. Seems to increase the flow of ideas. No effect on the 1 Dodge and Benediot, Carnegie Inst. Wash. Pub. No. 232, 1915, p. 277. 18 Effect of Alcohol on Psycho-Physiological Functions. affections, but sweetens the temper. Effect on routine work not known, as he never takes it when working. No effect on morals. One glass aids digestion; two glasses retard it; no effect upon the urine. Tea and coffee. — One cup strong coffee every morning. Life insurance. — Examined for life insurance a year previous. Northwest- ern Mutual Life Insurance Company. Accepted. This personal history must be supplemented, particularly with regard to the frequency with which alcohol was commonly taken by this sub- ject, since during the later experiments no alcohol was used outside the Laboratory or for four days previous to beginning the work. It is stated that alcohol is used "at banquets," and in the same paragraph "last used, October 3, 1913, one glass of beer." This last quotation, which really refers only to the pre-experimental condition for October 7, 1913 (see table 1, section n), might be taken as meaning a period of abstinence during the following months in which the experiments were interspersed. From the original protocols the following facts are dis- covered: 1 October 22, 1913 (normal experiment), 3 bottles of beer taken within previ- ous week, one-half bottle two days previoufe to experiment. October 29, 1913 (alcohol experiment), 3 bottles of beer and 1 bottle of wine taken within the previous week; one-half bottle of beer taken two days previous to experiment. November 12, 1913 (alcohol experiment), 1 glass of beer four days before. November 19, 1913 (normal experiment), 2 bottles of beer taken within pre- ceding week, one of them three days previous to experiment. December 2, 1913 (alcohol experiment), 3 bottles of beer taken during the previous week, two of these on the evening of November 29. December 9, 1913 (normal experiment), 2 glasses of beer on the evening of December 6. December 16, 1913 (alcohol experiment), 2 ounces of port wine taken previous evening, a total of 1 pint of port wine taken during the week previous to the experiment. January 1, 1914 (normal experiment), 2 bottles of beer in the previous week. January 22, 1914 (alcohol experiment), 2 glasses of beer taken during the week previous. January 28, 1914 (alcohol experiment), 1 glass of beer and 1 glass of wine taken January 27. February 4, 1914 (alcohol experiment), 2 glasses of beer during previous week, one of these taken the evening of February 3. February 12, 1914 (alcohol experiment), " 1 glass of ale last night." March 2, 1914 (normal experiment), "1 glass of beer last Saturday night." The notes in the protocols do not cover all the dates on which experiments occurred. The subject ordinarily drank from 1 to 6 glasses of beer and wine during Sat- urday and Sunday. The experiments were usually made on Tuesday and Wednesday. Those on December 8, 1913, and March 2, 1914, occurred on Monday. A personal communication from Subject VII adds the information that he and Subject VI roomed together during the school year in which these experiments were made. Their custom was to buy a case of beer and several bottles of wine at a time. Subject VI drank beer and wine only (see above). Subject VII drank these and stronger bev- erages which were frequently on hand. His usual consumption was, he thinks, greater than that of Subject VI. Subject IX roomed near and the three frequently spent the evenings together. 1 See p. 13, table 1, section II, for other data concerning the dates which follow. Amount of Alcohol Employed. 19 Although the data given in the preceding paragraph regarding the quantities of alcoholic beverages taken are admittedly not exact, 1 it still seems fair to assume on the basis of the usual alcoholic content of these beverages that two or three times per week during the months covered by experimentation, Subject VI had from 15 to 40 c.c. of abso- lute alcohol. The common amount taken in an evening may have been a near approximation to dose A of these experiments, that is, 30 c.c. of absolute alcohol diluted to a total volume of 150 c.c. The possibility of an additive or prolonged effect serving to reduce the contrast between normal and alcohol experimental days by alcoholic doses following within 48 hours of each other may be assumed as approximately equal in both series. In the Dodge and Benedict series the doses were close together because of the amounts taken outside the Laboratory, and also in the repetition series because the experimental days were consecutive. 2 During the week of the repetition series of experiments (June 29 to July 4, 1914) the nude weight of the subject varied slightly. The val- ues for the experimental days were in order as follows: 66.9, 67.8, 67.0, 66.6, 66.6, 66.4 kilograms. The cup of strong coffee in the morning, mentioned in the personal history, was, as previously stated, omitted from the last meal preceding the experiment. The subject took milk with this meal. The outdoor noon temperatures for the six days beginning with June 29 were 15.0, 22.8, 15.6, 14.4, 13.9, and 21.7° C. The barometer read- ings at the Nutrition Laboratory were 757.4, 760.0, 766.6, 758.2, 766.1, and 765.3 mm. The Laboratory was not heated and forced draft venti- lation was employed. AMOUNT OF ALCOHOL EMPLOYED. The alcohol for each day was given in a single portion. The dose used was that designated by Dodge and Benedict as "dose A," which contained 30 c.c. of absolute alcohol, 7 c.c. of orange infusion, 3 1 c.c. of strong infusion of quassia, a slight amount of saccharine for sweetening, and water to bring the volume to 150 c.c. A mixture of the same com- position and volume, but without the alcohol, served as the control dose for each of the three normal days. This regular use of the control mixture constitutes a difference between the two series of measure- ments. Dodge and Benedict (page 24 of their report) finally aban- doned the use of the control mixture, as it proved a failure in com- pletely masking the alcohol. From the standpoint of uniformity it is, however, desirable to give a control dose of some kind, particularly if the actual alcohol dose is not pleasant to take. 1 The quantities are probably not smaller than reported. 2 The extra-laboratory use of alcoholic beverages by other Dodge and Benedict subjects is of significance in the interpretation of their individual alcohol results. These data from the original protocols, with certain observations, are embodied in Appendix I of this monograph. (See p. 135.) 3 Rivers, The influence of alcohol and other drugs on fatigue, London, 1908. 20 Effect of Alcohol on Psycho-Physiological Functions. Every effort was made to prevent the subject from disting uishing between the alcohol and the control doses. Before giving the non- alcohol mixture, the rim of the glass was swabbed with strong alcohol that the odor might help to hide the identity during the early part of the experiment. But as the alcohol frequently produced sensations of warmth in the stomach and flatulency, the subject never failed to iden- tify the mixture by the time the experimental session was over; that is, his final impression was always correct as to its identity. He had no way of knowing before the session what he would be given, and of course he received no confirmatory information regarding his impres- sions. As may be seen from the latter part of table 1, section n, his more recent experience would not lead him to expect that normal and alcohol days would regularly alternate. Unfortunately only the 30 c.c. dose could be used in the second series. This was in accordance with the expressed desire of the subject, who insisted that the 45 c.c. dose which he had taken three times before (see table 1, section n) made h\m feel uncomfortable for the rest of the day. The larger dose was more than he was in the habit of taking, according to the data given on page 18. GENERAL EFFECTS OF THE ALCOHOL. It is of course recognized that the alcohol dose in these investigations was a 20 per cent solution by volume of raw alcohol and water mixture, strongly bitter, which was decidedly unpleasant to take. 1 Moreover, the total amount was swallowed within one minute, the subject pre- ferring to drink rapidly. The experiments of Dodge and Benedict usually began about 4 p. m., the previous food having been taken (except in the case of Subject X) nearly 3J hours before, t. c, at 12 h 30 m p. m. Under these conditions it is a fair question, and one that has been asked since the publication of the earlier report, whether or not the subjects experienced or showed any general effects which would be interpreted as evidence of incipient intoxication, other than the results given in the experimental measurements. As this matter has a bearing upon the interpretation of both sets of results, the available material for Subject VI is included here, while that for the other subjects is given in Appendix II (see p. 138). Dodge and Benedict, in their original protocols, record the following notes regarding the general effects of the alcohol upon Subject VI : October 7, 1913. No alcohol; no comments. October 14, 1913, dose A taken 5 h 12 m p. m. At 5 h 33 m p. m., subject sleepy; 6hiQm p m> during faradic threshol d measurement, subject falls asleep. 1 Rivera, The influence of alcohol and other drugs on fatigue, London, 1908, p. 81. The use of some common high-percentage alcoholic beverage would have been appreciated by the subject. This would, however, have greatly reduced the possibility of a control dose resembling it closely enough to cause any confusion to taste. Lange and Specht, Zeitschr. f . Pathopsychol., 1915, 3, p. 155, report that some of their subjects preferred to take the 40 c.c. of absolute alcohol as a drink of 99 per cent alcohol rather than to dilute it with 100 c.c. of water and fruit juice. General Effects of the Alcohol. 21 October 22, 1913. Cereal coffee control dose after fourth period. No comments. October 29, 1913, dose A taken 5 h 06 m p. m. At 5 h 30 m p. m., subject stated "feel half intoxicated." He made very vigorous taps in one of the tests and realized he was making a louder noise than usual, but did not think he was putting forth more energy. November 5, 1913. No control dose mentioned and no recorded impres- sions. November 12, 1913, dose A taken 5 p. m. No notes are given to indicate that the subject was sleepy or felt the effects of the alcohol. Protocol other- wise very complete. November 19, 1913. Control session; no dose specified or comments given. December 2, 1913, dose A taken 5 p. m. At 5 h 25 m p. m., subject felt per- fectly comfortable and sleepy, "as always after alcohol." "Relaxes between reaction tests and comes up more easily after alcohol; time goes faster between tests; gets confused and bothered less." December 9, 1913. Control session; no dose specified or comments recorded. December 16, 1913, dose A taken about 4 h 30 m p. m. No comments con- cerning effects. January 1, 1914. Control session for 12-hour experiment; no dose specified or comments recorded. January 2, 1914, dose C taken in the method previously described. (See table 1, footnotes 2 and 3, p. 13.) No comments concerning effects, other than that the appetite was not the same for lunch as the day previous, when he had had no alcohol. It is not stated whether the appetite was better or poorer, but understood to mean the latter. January 22, 1914, dose B taken 5 p. m. No comments made concerning any effect. January 28, 1914, dose B taken 5 p. m. Subject subsequently went to sleep twice during the evening, once during the finger-movement test, and movements stopped. February 4, 1914, dose B given 5 p. m. No statement concerning effect other than that subject urinated at 5 h 25 m p. m. February 12, 1914, dose A taken 6 p. m. No effects of alcohol noted. March 2, 1914, control session. No dose specified or comments recorded. In the repetition series, Subject VI was not encouraged to comment upon his feelings or impressions during the experimentation. At the close of the session he was asked if he had any remarks to make. These statements were brief and confined themselves usually to the nature of the dose given on that particular day. They are as follows : June 29, 1914, control dose taken 10 h 10 m a. m. Subject stated: "Very little alcohol." June 30, 1914, dose A taken 9 h 40 m a. m. At 10 h 37 m a. m., subject com- plained of being very thirsty; was given one-half glass of water. Session finished l h 25 m p. m. Subject stated, " dose to-day contained alcohol, smelled it, tasted it, and belched it." July 1, 1914, control dose taken at 9 h 50 m a. m. "No alcohol to-day." July 2, 1914, dose A taken 9 b 30 m a. m. 10 h 15 m a. m., urination. At close of session, subject said dose was alcohol. July 3, 1914, control dose taken at 9 h 25 m a. m. "Drink was water and flavoring." July 4, 1914, dose A taken at 9 h 30 m a. m. 10 h 10 m a. m., urination. At close of session, subject said dose contained alcohol. 22 Effect of Alcohol on Psycho-Physiological Functions. The effort in all these experiments was to present objective results, free from personal bias. It is understood that the comments con- cerning the general effects of alcohol were only incidentally recorded, with no intention of securing complete, introspective data. It is also to be considered that, for reasons which Dodge and Benedict discuss on pages 23 and 24 of their report, they abandoned the giving of a control dose on a normal day. The almost certain knowledge of having taken alco- hol may naturally enough have colored the introspection of some sub- jects, particularly in the cases of Subject VIII (total abstainer) and Subject II (very moderate user). Both of these subjects seem to have usually experienced in a more marked degree than the others the gen- eral effect with the 30 c.c. dose. Dose B (45 c.c. of absolute alcohol) produced general effects clearly evident to all subjects; dizziness, numbness, and nausea are usually mentioned as following the ingestion of this quantity. (See Appendix II, p. 138.) Subject VI, in whom we are most immediately interested, curiously enough makes his most striking comment on October 29, 1913, follow- ing dose A. Considered in connection with the comments made con- cerning the general effect of doses on subsequent days, it appears most probable that the statement "feel half intoxicated" is an exaggeration, unless by this he means simply sleepy. Otherwise some unusual condi- tion must have made the subject more susceptible to the alcohol on this day, as he makes almost no comment concerning any effect from dose B taken three times later in the Dodge and Benedict series. It will be recalled, however (see p. 20), that the subject objected to taking dose B in the second series on the plea that it incapacitated him for work during the remainder of the day. This subject may have been in cer- tain respects somewhat more resistant to the effects of alcohol than some of the others. The comments of Dodge and Benedict on his experimental results give weight to this point of view, while his seeming lack of experiencing the general effects may have been due to his giving less expression to introspection, or the effects were possibly more delayed in his case. From the data at hand concerning the amounts of alcoholic beverages used by the different subjects in the week immediately preceding the experiment (see Appendix I, p. 135), it is clear that Subjects III, VI, VII, and IX, consumed the largest amounts. These same subjects, according to their comments (see Appendix II, p. 138), experienced the milder effects as a result of the doses given in the Laboratory. It of course can not be stated whether they were actually less affected by the doses given or made fewer comments because of their familiarity with the effects. However this may be, three of the same group of sub- jects—that is, III, VI, and VII— show experimental results as to the effect of alcohol which are below the average. (See Dodge and Bene- dict's report, p. 264, fig. 32.) On the other hand, Subject IX, who Measurements Employed. 23 perhaps had been in the habit of taking more alcohol than any other one of the normal group and who recorded the most mild general effects from the Laboratory doses, demonstrates according to this same sum- mary experimental effects of the alcohol which are above the average. The case of Subject X was peculiar in that this subject ate a full meal immediately before the experiment began, and only dose A, 30 c.c, was given. The basis for judging the quantity of alcohol which constitutes a moderate dose must ultimately be the experimental evidence of an effect plus the general subjective effect which the alcohol produces. Obviously for any one amount, as, for example, dose A of these experi- ments, the general effect will vary in degree with the subject and the particular group of conditions which are present at the time of any specific experimental session. For the abstainer 45 c.c. is perhaps im- moderate, while for Subject IX (and probably for Subject VII), judged on the basis of usual habits, it certainly could not be so considered. 1 MEASUREMENTS EMPLOYED AND THEIR SEQUENCE. Since Dodge and Benedict have carefully described the apparatus employed and the technique of the various measurements which have been used, it is not necessary to duplicate these descriptions and dis- cussions here. The purpose, as stated, was to repeat in part their observations. Therefore, the same apparatus was used and only a few minor changes were permitted in the procedure. Most of these changes were kindly suggested by Dodge on the basis of his experience with the normal group of subjects. They were not proposed as adjust- ments required specifically for Subject VI. It is likely that some specific changes should have been made for this particular subject, as, for example, more insistent instructions to be very careful and look exactly at the fixation marks in the eye-movement measurements. (See p. 79.) The routine of the experimental day was as follows: Immediately upon reaching the Laboratory the subject was seated and the pulse was counted at the wrist. The assistant adjusted the electrodes to their places of contact with the body for the taking of electro-cardiograms and fastened the respiration recorder in place. Meanwhile data were secured concerning the amount of sleep in the previous 24 hours; the last food taken, time of eating, and approximate amount eaten; general condition of the subject, and any other details that seemed pertinent. 2 1 The consensus of opinion as expressed in experimental studies on human beings is that any amount of alcohol between 15 and 60 c.c. may be considered as moderate. Kurz and Kraepelin (Psychologische Arbeiten, 1901, 3, p. 418) gave 100 c.c. of absolute alcohol and still claimed that it was a quantity such as is generally supposed to be moderate, that is, equaling 2 liters of bp.er. For information concerning the amount of absolute alcohol in usual retail measures of the ordinary alcoholic beverages, see the useful book, D'Abernon and others, Alcohol; Its action on the human organism, London, 1918, Appendix 5. 2 The nude weight of the subject was secured at the end of the day's session. 24 Effect of Alcohol on Psycho-Physiological Functions. Then the measurements, to be described in more detail in succeeding paragraphs, were made. The time required to complete the group of observations was approximately 50 minutes. The dose for the day con- tained alcohol or was simply a control mixture, as the case might be, prepared as stated on page 19. The total volume of liquid contained in either case was 150 c.c. It was given the subject at this point in the experimental program for the day. He drank it rather rapidly while he was reclining in the steamer chair. Following the taking of the dose the measurements were repeated in identical order as many times as the 5-hour b'mit would permit. It will be convenient to term each repetition of the group of measurements a "period." Thus, the dose is regularly given between periods 1 and 2, and from three to five periods were completed following the dose. The measurements may be de- scribed in order as follows : (1) Pulse. — An experimental period began with the subject seated in a steamer chair. He was instructed to relax as much as possible. After three or four minutes the first pulse record, designated as No. 1, was taken. All of the pulse records, except the count made at the beginning of the session, were electro-cardiograms from body leads. Eleven of these records were taken in each experimental hour. In the subsequent paragraphs their order and location will be indicated by numbers. The photographic paper moved 1.5 cm. per second, and the records are long enough to permit the counting of 15 seconds of pulse. Respiration curves were incorporated directly with the photographic pulse records, to show the time relations between pulse and breathing. A large tambour, buttoned under the vest of the subject, was connected to a small tambour carrying a registration point so located as to cast a shadow on the cylindrical lens of the camera. Connections for the pulse and respiration records could be established or broken very quickly. The body electrodes and respiration tambour did not hamper the breathing or other body movements and produced no discomfort whatever. (2) Patellar reflex. — Two similar stimuli, separated by an interval of 0.5 second, so that the latter came within the relatively refractory period of the reflex, were delivered to the tendon of the left leg by pen- dulum hammers. Ten records with 30-gram stimuli were followed by 10 records with 50-gram stimuli. Appropriate levers with a magnifi- cation of 6 times wrote directly the reflex thickening of the quadriceps muscle on a kymograph rotating at a rate of 100 mm. per second. From this record both the latency and amplitude were measured. The subject reclined in a steamer chair. He was instructed to relax and to say "Ha" immediately following each stimulus blow, that the respira- tion phase might be controlled. After the patellar-reflex measurements the subject remained quiet and relaxed for pulse record No. 2. (3) Sensory threshold for faradic stimulation. 1 — The assistant placed a suitable arm-rest, bearing the non-polarizable electrodes, in a conven- 1 Martin, The measurement of induction shocks, New York, 1912. Measurements Employed. 25 ient position for the subject, who remained seated in his chair in the same position as for the patellar reflex. The first and second fingers of the right hand, immersed in a normal salt solution to the depth of 2 cm., were used as receptors. The key in the primary circuit of the inducto- rium was of the general type used by Martin, but operated by a small motor at the regular rate of two breaks per second. Dodge had pre- viously used a hand or foot control for breaking the primary circuit, since the subject was in the room with the apparatus and the sound of the regular Martin key gave secondary criteria when used in such proximity. When the shocks are near threshold value, and particularly if they are separated from each other by more than a second, the sub- ject is apt to confuse them with the sensation from pulse beat, slight finger movement, and changes of pressure. Furthermore, the attention may be very unfavorable for observing the sensation at the actual time of the shock. The regular, more rapid shocks produced by the me- chanically operated key were substituted that they might supply more favorable conditions and hence more consistent threshold values than were produced by the previous method. Special photographic meas- urements made with the aid of the string galvanometer across the sec- ondary coil showed that two makes and breaks per second in the pri- mary did not cause any decrement in the strength of the shock deliv- ered by the secondary of the inductorium. In fact, there was no appre- ciable decrement when the rate of make and break in the primary of this coil was changed to 10 per second. The motor which operated the key in the primary circuit was used always at the same speed, excepting for changes brought about by the slight fluctuations in fine voltage. The make-shocks were not short-circuited from the fingers. They were subthreshold, however, as the method employed (see Dodge and Bene- dict's report, p. 135) was always to increase the stimulus strength of the break-shock until the threshold was reached. The threshold value was determined by averaging the first three ingoing threshold positions of the coil. Three such average threshold values were determined, one when the resistance in the circuit was simply that of the tissue and the secondary coil, and the others with added resistances in the secondary circuit of 20,000 and 40,000 ohms, respectively. From these three values for Z, two /3 values were found according to the regular formula of Martin. 1 The average of the two /? values is the significant measure for the period. Pulse record No. 3 followed the threshold measure- ments. The subject had been sitting quietly for approximately 15 minutes previous to this pulse record. This condition favored relax- ation. (4) Reaction time in reading isolated words. — The subject changed to an ordinary chair and sat upright in front of an exposure apparatus 1 Martin, Bigelow, and Wilbur, Am. Journ. Physiol., 1914, 33, p. 416; Dodge and Benedict's report, p. 143. 26 Effect of Alcohol on Psycho-Physiological Functions. with the voice key in hand. 1 Pulse record No. 4 was then made. An offset from the shaft of the kymograph momentarily opened a circuit and exposed a stimulus word which remained in view. The instruc- tions were to keep the voice key pressed lightly against the lips, and to pronounce the words as quickly as possible after their appearance. One word appeared every 10 seconds. Both exposure and reaction were registered by the same signal magnet, the rate of the kymograph being 100 mm. per second. The kymograph drum was visible to the subject. A signal was given him before the exposure of each word, and care was exercised to make sure that the correct word was pronounced. The set of 24 four-letter words previously used was employed in this- series. The 24 small cards bearing the words were well shuffled before each set of reaction measurements, in order that the subject might have no tendency to anticipate the sequence of their exposure. As con- nections were already made with the string galvanometer and the respi- ration recorder, pulse records Nos. 5 and 6 were taken by the assistant during the latter third of the reaction measurements, and occasioned no interruption. After 30 seconds of quiet, following the reactions,, pulse record No. 7 was made, the subject sitting upright and still holding the voice key. (5) Finger movements. — The movements were not interrupted by mechanical limitation of their amplitude. The subject sat in a steamer chair, leaning forward. The right arm and hand were supported by a rest and in a comfortable position, suitable for the free oscillation of the middle finger in a horizontal plane. A light system of levers (total mass, 7 grams) attached to the finger operated in front of the cylin- drical lens of the pulse-recording camera, and made possible the photo- graphic registration of the oscillations reduced to one-fifth their actual amplitude at the same time that pulse and respiration curves were recorded. The uniform instruction was to make the movements as rapidly as possible; they were continued for about 9 seconds, a period too short for development of conscious fatigue. Two records were taken,, with an intervening rest of 1 minute. Thus pulse records Nos. 8 and 9 are incorporated directly with the finger-movement records. (6) Voluntary tetanus pulse. — During the finger movements the sub- ject sat upright in the steamer chair, leaning slightly forward. After this measurement was concluded he was instructed to recline and relax until a given signal, when he was expected to tetanize voluntarily the- muscles of arms, legs, and trunk (he sat up, both fists were clenched, and arms and legs were extended directly in front), and to retain this posi- tion until the second signal, a period of 5 seconds. A continuous pulse tracing was taken covering the latter part of the preliminary period of relaxation, the tetanus, and the 20 seconds following, a period during, 1 For a diagram indicating the positions occupied by the subject and the general distribution of the apparatus, see Dodge and Benedict's report, p. 95. The construction and sensitivity of the- voice-reaction key can be judged from their figures 16 and 17 to 20. Measurements Employed. 27 which the heart recovered approximately its normal relaxation rate. Two similar tracings, designated as pulse records Nos. 10 and 11, were taken as outlined. The time interval between the tetanus signal for No. 10 and that for No. 11 was 1 minute. The actual taking of these two records after the subject was in position and had relaxed required approximately 2 minutes. In the Dodge and Benedict observations, 1 the pulse records taken with the purpose of showing the change in heart rate with muscular activity were with the subject sitting in position, immediately after standing, 60 seconds after standing, immediately after two double genuflections, and 60 seconds thereafter. In our series this group of five records was, at the suggestion of Professor Dodge, shortened to the three-part continuous record described. (7) Memory. — A series of twelve 4-letter words, carefully printed on a strip of white paper, was made to encircle the kymograph drum. 2 A circular screen hid the words from view, except as they could be seen through a window 15 mm. wide, which permitted a view corresponding to the width of two of the letters used in the words. When the word series had been placed upon the kymograph and properly covered, the subject was disconnected from the string galvanometer and the respira- tion recorder, and changed chairs to the position previously occupied during the reactions to isolated words (measurement No. 3). The drum rotated at a rate of 10 mm. per second in a clock-wise direction, i. e., from left to right of the subject. Hence the latt letter of a word appeared first in the window, which limited the subject's view, and not until the first letter of the word had appeared could the subject pro- nounce it correctly in the voice key, which recorded the reaction. The instructions were as follows: "There are twelve words; each is composed of four letters; watch carefully and at the first exposure speak each word as soon as you are sure of its identity, also try to remember each word in its order that at the second revolution of the drum you may speak each particular word immediately before any part of it has appeared in the window. If you can not recall a word to speak it before it has partly appeared, let as little as possible of it appear before you do speak it. If you speak the wrong word correct yourself immediately by speaking the right one." The distance on the kymograph record between the first and second reactions to a particular word was taken as the measurement of the residual memory for that word. For the twelve such values an average was found. Dodge and Benedict 3 had used different sets of word series for the different experimental days, but came to the conclusion "that for purposes of determining the effect of a drug we would have done better to use only one group for all days alike." This method was employed here. The same series of twelve words was used for period 1 1 Dodge and Benedict's report, p. 223. 2 Ibid., p. 129, and fig. 26. 3 Ibid., p. 130. 28 Effect of Alcohol on Psycho-Physiological Functions. of each day, another series for period 2, and so on. In the Dodge and Benedict experiments the series was exposed three times to the subject, but in the repetition experiments only two complete exposures were used, as mentioned, in order to reduce the possibility of the words pre- sented being completely learned by the subject and remembered from day to day. (8) Eye reaction to peripherally appearing stimuli. — A slight move- ment of the subject's chair brought him into position for this measure- ment and the two which follow. The sudden fall of a shutter carrying a fixation mark exposed another mark or letter, which was 1, 2, or 3 cm. to the right or left of the aforesaid mark; the distance of the object from the eye was 48 cm. The subject, who was in ignorance of the exact place where the stimulus mark would appear, was instructed to look from the fixation mark to the stimulus as soon as it should appear to him, and to pronounce the letter aloud if it were a letter that was exposed. A beam of light, interrupted by a tuning fork and of com- fortable intensity, was made by reflection to fall upon the eye at the moment of appearance of the stimulus. From the cornea this light was reflected through an enlarging camera and registered itself on a falling photographic plate. The resulting string of dots with the sharp bend at the moment the eye moved to the exposed letter or mark is a record from which the reaction time can be easily and accurately counted. Five such reaction records were taken side by side on the same plate. (9) Eye movements. — The general apparatus in this measurement was the same as that which was used for the eye reactions. At the instant that the interrupted light was turned upon the eye a dot ap- peared, this time in an expected position. From this the subject looked 40° on his arc of vision to the left to a second dot, and then back to the first, and continued these movements as rapidly as possible throughout the period of 5 seconds during which the photographic plate was falling. In the records each dash with one interspace represents 0.01 second. Precautions were taken to guard against head movements by the use of a suitable head-rest. The left eye was covered with an opaque screen. Uniform instructions preceding each record were to make the move- ments as fast as possible, but to look carefully at the fixation marks. With this particular subject, these instructions should have been re- versed and strong emphasis placed on the careful fixation of the marks in their turns. (10) Protective lid reflex. — An artificial eyelash consisting of a strip of black paper 1 by 16 mm. was attached to the left lid so that it re- mained in a horizontal position when the eye was open and viewing a fixation mark. This could be worn without discomfort for an indefi- nite period, except possibly when the subject rubbed his face with his hand. When a beam of interrupted light was reflected across the face, Statistical Method. 29 the eyelash cast a shadow upon the slit of a camera, behind which a sensitive plate could be made to move in a horizontal direction. The shadow was so placed that the usual closure of the lid would not com- pletely withdraw it from the camera slit, causing a break in the picture. Thus the conditions for recording the amplitude of movement were standardized. The stimulus for the wink reflex was produced by the sharp snap of a spring hammer against a sounding board. This latter member has a short projection reaching to the slit of the photographic camera, and the vibration of this pointer at the moment of sound regis- tered the instant of stimulus on the photographic record. Two stimuli were used, separated by an interval of approximately 0.5 second, which was within the relatively refractory period. The instruction was to count backward verbally, not too rapidly, from a number which was given by the operator just prior to the stimulus. Some time after the subject had commenced to count, the plate was noiselessly released and its movement opened in turn two circuits, which released the spring hammers and delivered the stimuli. The photographic curves were read for latency and amplitude. Four such records were made for each period. The ten measurements just described completed the cycle of observa- tions taken during an experimental period (approximately SO minutes). 1 The periods followed each other without intermission, excepting the slight break which always came between periods 1 and 2 when the dose was given. Theoretically, for purposes of comparison the same num- ber of periods should be taken each day and the periods should be of equal length ; this ideal of uniformity is hard to attain with a complicated group of measurements. Much depends upon the facility with which the subject can adjust himself to the apparatus and perform his part. During the six days there was no accidental interruption to cause the omission of any measurement or enforce a delay of more than ten minutes for the adjustment or repair of apparatus. The number of experimental periods completed on the respective days was as follows: June 29, four periods; June 30 and July 1, five periods; July 2, 3, and 4, six periods. STATISTICAL METHOD OF PRESENTATION OF RESULTS. In comparing normal and alcohol experiments we are interested in any differences which may exist between the two groups of data, and in the "probable correctness" of such differences as evidence of an alcohol effect. Any group of measurements is most conveniently described in terms of (1) central tendency and (2) variability. In the following 1 Dodge and Benedict's periods were approximately 30 minutes in length, for, as stated in their report (pp. 14 and 15), in their experiments all of the different measurements (reflexes, reactions, coordinations, etc.) were not made on the same day, as was the case in this second group on Subject VI. 30 Effect of Alcohol on Psycho-Physiological Functions. pages, the average (arithmetical mean) is used for the statement of central tendency and the mean variation (average deviation) as a measure of variability. 1 It is conceivable that alcohol may affect not only the average, which represents speed, amount, or sensitivity accord- ing to the process measured, but the mean variation, also, since in psychological data this stands for uniformity of conditions, consistency of judgment, and controlled attention of the subject. Wherever possi- ble, differences between normal and alcohol data will be given for the mean variation as well as for the averages. The "probable correct- ness" will be computed only for differences between averages. Dodge and Benedict assumed that the values obtained in measuring any process were functionally related to the particular day on which the subject and process were tested. 2 Hence, in addition to providing for an equal number of normal and alcohol days, it will be seen that the first period of each alcohol day was also normal, since the dose was not given until the beginning of the second period. These first-period values serve to indicate the neural condition, disposition, or level of excitability on the different days for the different processes measured. In the presentation of results emphasis is placed upon the differences, plus or minus, between the first-period values of a particular measure and those obtained for the succeeding periods of the same day. A comparison of the differences obtained for normal days with those obtained for alcohol days provides true differences between the two groups of data and supplies a numerical statement of any effect of alcohol. This is also the method used in this monograph; it may be illustrated simply and concretely as follows: A series of eye-reaction time averages with their mean variations, given in thousandths of a second, is shown in table 2. The average reaction time (R. T.) for the first period of the normal day is 212 a. (See section i.) At the close of this period and just previous to the beginning of the reactions of the second period, the control mixture was taken. Following this we have the average values for succeeding 1 See Whipple, Manual of mental and physical tests; simpler processes, Baltimore, 1914, Chap. Ill, for a very readable account of the usual statistical methods. 2 The statement is made by Dodge and Benedict (p. 28 of their report) that their tables "'are commonly accompanied by a statement of average measurements, but the latter are regarded as of relatively little importance. They are only given as details that may be of interest to some future investigators who may be measuring similar processes." The actual average measurement values were not included in the case of the patellar reflex, but only the differences were given. While the average values would have considerably enlarged the table, their omission is already a source of regret, as it greatly hampers the comparison of other data with the material collected by Dodge and Benedict, which is the most extensive concerning this human reflex that is readily available. Dodge and Benedict found it necessary in some cases, notably in the memory (see Dodge and Benedict's report, p. 130) and pulse sections, to make the averages the basis for comparison. Also, their tables 27 on eye movements and 30 and 32 on finger movements are entirely devoted to them. That wide differences in nervous excitability will be found is illustrated in Dodge and Benedict's table 1, page 46 of their report, and can be seen in many places throughout their tables where the measured values are given. The differences between days are probably larger when the experi- ments occur once a week and are scattered over a month or possibly several months than when consecutive days are employed. Statistical Method. 31 periods of 194, 192, and 189 a in order. Similarly on the alcohol day a first period average of 190 a is followed after the taking of the 30 c.c. of alcohol by 223, 214, and 201 „ = 1.0 (i. e., the probable error of difference is just Jr. JBj.d as large as the difference), the value of the probability integral for this ratio is 0.50. 2 In other words, there are 50 chances out of 100 that the difference will deviate from the observed value by an amount greater than that value. In only one-half of these 50 chances, or in 25 cases, will the change be a decrease (a difference with opposite sign). There- fore the probable correctness of this difference would be 100—25=75, usually written 0.750. A probable correctness of 0.500 represents pure chance, while 1.000 is certain cause. Usually values above 0.900 are considered to indicate reliable, significant differences. The probable correctness figures in table 2, section in, are all above 0.900, notwith- standing that only 5 reactions were taken in each experimental period, 1 The mean variation, as has been noted, expresses the degree of variability of the individual reaction times from the average. Likewise every individual average also varies somewhat from the theoretically true average which would result from several series of experiments. The ex- pression="= — -. — : — ', where M. V. represents the mean variation and n the number of obser- V n vations (reaction times) is known as the probable error of the mean (P. E.m) and is a measure of this variability of the average. The probable error of difference, P. E.s =-\/P. JE.Vi+P. JSAia, i. e., the square root of the sum of the squares of the probable errors of the means. 2 Boring has recently contributed two articles which illustrate very clearly the method of obtaining and the usefulness of the "probable correctness of the difference"; see "The number of observations upon which a limen may be based," Am. Journ. Psychol., 1916, 27, p. 315, and "On the computation of the probable correctness of differences," Am. Journ. Psychol., 1917, 28, p. 454. 34 Effect of Alcohol on Psycho-Physiological Functions. which would produce a much larger P. E. M than would result, other things being equal, from series of 25 reactions each. 1 The "certain cause" indicated by a probable correctness of 1.000 is of course not synonymous with specifying that cause as a 30 c.c. dose of alcohol. Any difference between two averages may be produced by one cause, but more likely by several factors, the resultant of which changes the phenomena under observation. In these experiments it was naturally the a-im to make the alcohol dose the critical change and cause. While the probable correctness supplies an estimate of the reliability of particular differences in results as being produced by some change or other, it is not by itself a measure of success in isolating the alcohol effect. The mean variations (M. V.) are included in table 2 in parallel col- umns with the averages to which they apply. The differences are com- puted for these exactly as was done for the averages (see p. 32). In section in of the table, the effect of alcohol upon the M. V. is found to produce +11 and +7o-in periods 2 and 3, an increase in the variability of the reaction time following the dose, while period 4 shows a decrease. It should be borne in mind that the plus and minus signs are employed in their usual meaning throughout this monograph. For example, a plus (+) difference or percentage means a longer reflex time, a larger ampli- tude, a larger number of finger-movements, a larger number of elec- trical units required for threshold stimulation, a larger variability be- tween measurements, etc., whereas the minus (— ) sign in these same connections would have the reverse meaning. The effect of alcohol must, of course, always be interpreted according to the nature of the measurement; e. g., a + reaction-time percentage (slower reactions) would not ordinarily be associated with a + percentage for finger movements (faster movements). It will be seen by consulting Dodge and Benedict's report, page 29, that they computed the average difference (Av. D.) of the day's meas- urement according to the formula 2 , n (l-2) + (l-3) + (l^) + (l-tt) Av. D.— • n According to our method for determining the differences outlined in previous paragraphs, the formula for Av. D. would be: (2-l) + (3-l) + (4-l) + (n-l) Av. £>=■ n 1 This condition of a small n (5 eye reactions, 10 patellar reflexes, 4 lid reflexes, etc., taken within any one period) is operative throughout all our measurements and increases the size of the probable errors. The usual method of obtaining the P. E. M for the 3-day averages for any period has been to find the P. E.m for each of the three values which are averaged. The sum of the three P. E M is divided by three to obtain the typical P. E. M ; then the result is divided byv% since the n for the three averages together is three times as large as for one. 2 It is obvious that n in their formula refers to the number of experimental periods following and excluding the first period, or normal of the day. Statistical Method. 35 It may appear of little consequence as to whether one subtracts the preliminary first-period values from those which succeed the giving of the dose (the method employed in this paper) or subtracts succeeding values from the preliminary first-period values (the method employed by Dodge and Benedict) . However, by the latter method the resulting signs have, when taken in connection with the discussion in the text, exactly the reverse meaning from that which is usually associated with them. According to Dodge and Benedict (see their page 29) : "If the Av. D. has a minus sign it shows that the measured values are larger as the session progresses. Conversely, if the Av. D. has a positive sign, it shows that on the average the subsequent measurements are less than the normal of the day." It is a real disadvantage for clear understanding of data to have to carry in mind continually this reversal of meaning for + and — which runs through the Dodge and Benedict -publication; accordingly, at the risk of some confusion in a com- parison of results, the method represented by the second formula given and illustrated in table 2 has been adopted. Of course the results themselves are not affected by either method of computation, but that method which most facilitates clear understanding and interpretation is to be desired. Wherever data from the Dodge and Benedict report are reprinted for comparison, the liberty is taken to change the signs. Dodge and Benedict took the average of the individual differences for all periods of an experimental day as the significant result. The individual periods received no consideration. Our experimental day was 5 hours long as contrasted to the 3 hours used previously. As will be seen later in this paper, the differences produced by the ingestion of alcohol do not always with this subject retain the same sign (plus or minus) throughout the experimental day. For example, a series of plus differences during the first three periods following the dose may change to minus differences during the last two periods of the session. To average together the five differences under these conditions would mask important tendencies and give quite misleading comparisons. The 5-hour experiment was adopted in this repetition series in order that the effect of alcohol might be traced for a longer time after the ingestion of the alcohol. The results are therefore presented as illus- trated by table 2 in a more analytical form and chief attention is given to a comparison by periods following the alcohol. 1 The method used in this report has been to average the differences for homologous periods on normal days and subtract these from a series of such averages for alcohol days in order to obtain an expression of the effect of alcohol for the different periods. This method does not preclude comparison of the averages for individual days, when this is desirable. Another item of statistical nature should receive attention. A com- 1 Dodge and Benedict's report, p. 210, table 39, is the one instance in which these authors have used this form of presentation of their data. 36 Effect of Alcohol on Psycho-Physiological Functions. mon footnote accompanying all of the Dodge and Benedict tables states that "the values for the first period of the alcohol experiment were obtained before the alcohol was given, and are therefore not included in the averages," whereas on normal days the values for the first period of the day were included in the average. It is clear that the alcohol- day average and the normal-day average would have been more strictly comparable if the latter had also been computed without including the first-period values. Wherever averages for the day are given in the following tables, usually in the next to the last column, these aver- ages always include only the values which follow the giving of the alcohol dose or the control dose, that is, the above quoted footnote is applied to all of the normal days as well as to the alcohol days. The results of recomputing some of the Dodge and Benedict material on this basis may be seen in Appendix III. (See p. 140.) In succeeding pages, where such terms and expressions as "inferior functioning of processes," "less efficient performance," "depression," and the like, are employed, it is assumed that these terms and their opposites properly describe the changes and conditions found in these neuro-muscular measurements. In the case of the respiration rate and possibly of the reflexes this terminology may be open to question. DATA OBTAINED IN THE REPETITION MEASUREMENTS. In considering the effect of 30 c.c. of absolute alcohol upon some neuro-muscular processes of the subject used in this research, the vari- ous measurements used will be treated in the order in which they were presented in the earlier publication, not necessarily in the order in which they came in the experimental hour. (See p. 23.) It is not needful to discuss here the availability, apparatus, technique, and general experimental procedure for the different processes considered. This report primarily concerns results. For a more detailed treatment of the experimental methods than is given in the mere outline on pages 23 to 29, the reader must consult the full description of Dodge and Benedict. Unless definitely stated to the contrary, the experimental procedure was fundamentally the same as theirs. As this report relates to a single subject and one who was thought to vary quite widely from a supposedly normal group, certain features of the data will call for discussion in a way that was not feasible in the earlier mono- graph. So far as we know, this group of data, for the measurements involved, is as extensive as any which has heretofore been collected for one individual. Some attention should therefore be given to the nor- mal values as such, aside from their comparison with the alcohol values. Fortunately, Dodge and Benedict have usually given separate aver- ages for the different doses in their final statement of the effect of alcohol; this facilitates comparison with the data obtained in the later Patellar Reflex. 37 research, in which only dose A was used. 1 In this report an effort has been made to employ uniform tabular presentation for the different measurements, and thereby to facilitate comprehension of the data and any comparison that the reader might desire to make. PATELLAR REFLEX. The patellar-reflex data are given in tables 3 and 4, table 3 showing the values for the reflex caused by the first stimulus and table 4 those for the second reflex, which came within the relatively refractory period. Each table is divided into three sections, as follows : Section i, normal and alcohol measurement values for the latency (L) and amplitude (A) of the reflexes, i. e., the actual averages for the respective periods on the different days stated for 30-gram and 50-gram stimuli separately, and accompanied by average mean variations; section n, normal and alco- hol differences, which are found by subtracting first-period values from those which succeed; and section in, the effects of alcohol com- puted from the differences and expressed in these terms and also in per cent, together with the ratio of the differences to their probable errors and the probable correctness of the differences. The first column in table 3, headed "Experiment and stimulus," specifies the nature of the experiment (normal or alcohol), the date, and the weight in grams of the hammer falling through an arc of 90°, a vertical distance of 20 cm., which produced the stimulus for the reflex. Under the caption "Period I" are the averages for L (latency) and A (amplitude) for the first set of reflexes on the different days and with the respective intensities of stimulation. 2 The relative position of the dose, which was control or alcohol according to the kind of experiment, come between periods 1 and 2. Following the drinking of the dose, sets of reflexes were recorded for periods 2 to 6 as indicated. The results ior normal days and alcohol days are grouped and averaged separately in the different sections of the table. Table 4, showing the values for the second reflex, follows the form of the preceding table, the designations L' and A' being used for the latency and am- plitude, respectively. The patellar reflex was the first one of the group of ten different measurements which together required approximately 50 minutes for their completion. Since each measurement occupied the same position in the group, it is evident that the time elapsing from the first perform- ance of the individual test, as, for example, the patellar reflex, until the last repetition of it during the day, would be approximately 4 hours. The averages in section i for June 29, period 4 (L, 35 and 34 1 1 ++++++ ►J eoeo Of* ■*C0COUScO'*«US3'»H 1 -( eotMr-ioto^io^i^Hrt COCOCOCOCOCOeOCO COCOCOCOCOCOCOCO rot. P1HC1HOOHO ++++ ++ h n n co m « us t. iet» ■o 1 Pk H«HN i-i cm ih cm »* cn ++++ J b uses cncM e CO CO CO CO IO ei CO M in M O H co co co co co eo r1 0»)0 + + «! hoiiooook^o Hnoo^oiflceteo US i-4 00 CMC£>*4 o -* cd r~ cm >-< cm eo ih cm >o cm cm ■* o> ejs r* eo CO CO o> COCO 00 •o o ■c Ph ++++++ J b US'* 9*4 use ■ CO CO CO CO CO eO cococococococoeo 1 ++ 1 1 ++ < jt>ooio>ot>oceocco no(-n-H»8^«ieo i= 1-H 1-4 C) 1-4 i-H fH i-H ^H Ni0ON01ffl1)'O & i* co co co eci us e4 ? ■a I Ph 1 1 ++++++ i fH ho oo o> US i* »0 CO CD US US 3 *H i»* US M 00 CO 00 CONIfl oo CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO eof» g CONHHO HriO s? +++ 1 +++ s «! g'*ffl(-C10*0^<8<8 «HO]t t- US t- Ol t- 00 t> cN CO 00 iH t» US 00 —1 00 0» 1-1 CO * *H i-H i-l i-H 04 rH CM iH co ioaci i-i + 1 ++++++ "ft /? J eci cmj «n ih CO CO CO CO CO CO CO CO £0 CO CO CO CO CO CO CO I»t- OONHO i-<00 £ ++ +++ e CN CO lONNnOOOONM = i-H 1-H *H fHlHrHp* iH CO rt ^CO CN^ft- CM iHCiy ' -H CO •a 1 +11 1 + 1 1 9 9 Ph i_5 j-v »HCN OScM j-s CO eo H CO CO CO COCOCOCO CO CO CO CO CO CO CO CO e* cm o^^ih ihoo ++ 1 1 ++ ^ n co ^i i-i t- e OS ih us r-< co ■* t- la cm' eo oonoNciac-toMco R CM i-H l-H — H rHl-Hi-H^H -»! 1 i-HO coco Ph i-l b CM CM ■* ■* CO ■* ^" CO r* r* ■* ■* CO CO t~ IO US ■« rt iH CO CO CO CO CO CO CO CO CO CO CO CO CO CO 55 CO ttuaattatltttltl MbOMMMMMtOMU MMMMMMMO! oooooooooo oooooooooo oooooooo 3 3 1 D 3 I CO us « US K US CO us cc us ec us c- us « iQ CO lO CO us cc us « LO cc to 1 -4 i > ; i > 1 § a a !, ' . a IH ■O cd I" 1 M OS C 3 Section I Norma! June July July Ave Av Alcohol June July July Av Av Section 11 Normal June July July Av Patellar Reflex. 39 coooaiOHhton I ++ I ++++ 00 o OOh«« I I I 006 ceei>99co 00 >-< (N ■* IN iH rH © © I I ++ I I + I ©i-ii-ie^M I I I I I t^ 1-4 «oioaco 03 N18 OHO <4<o ■«* 00 jeo + + + + + MMMMMUMM OOOOOOOO eo»ocoiocoiQco»a a =3! "3 ■s > 01 M >, ft a a a> 0) IS fl T1 a 3 a 3 ■a .3.2 * J? 3 ° *! ■s.s 8 *£ s~ a ° ™ • « S3 feT3 ffl ■§ o"g ■§8 1 a^ a MS . Mil ■° s* s © ° w s ■3 o m * fi-S 2-a a o a a M ^ ffi u " " ■*> s 40 Effect of Alcohol on Psycho-Physiological Functions. , ^ O lAiO Hwo>cot-et> to n oo>h ^n« 60 d 1 t»aiOH es h£io hnhm i'j i-' j w'j ~& lO CO CO CO ++ +++++ T* -c ^iH«e io a o> NNOON o © CO <{ i •h NHoee in cq co TticoeoiHrM O i-c©i-( 73 o •g 1 1 oo Ph ij to i-i OHcene OHN COCBOlHPJ CO SMt> 'tfl *# CO CO CO Tjt ++++ •* i> nscowus HOSONncOSlCOH CO CO OS CO CO ■a ■o o Ph * s ff>HH«HHOO (M^rtTjiC-l-^iHTlliHCl go o o©© 1 + 1 + 1 CO ©o 1-1 b £«3HNH3|Hffl U3U)NSNNC6C0OH £o nnnce H t + + + + + £ O ■* CO CO 0«t-(6 OiOiOOOONNW oj co o> « ih^jiio 1 ■a o I Ph < g>-H cO'*-*t-coi-i'*i-iee CO IM MSI 00 C) t- 1 eo ■o o ■6 < l-HcoGmi-HiNi-Hevioi-i HHHHHCfrtNOH OHgiHo'do'e 1 + ++ 1 1 + 1. N ©» ? Ph J beoooAHH TflOOrHjJO^HOOlHlN^I MHgMOrtrtH l •S o f ++ + +++ 00 id (M COWtHIO W CO 00 W iH rH W ^ W i-H l> SON* CO i; grHCO£^^rHC go i-h i-H CO CO CO CO CO CO r* O OOSOiOfflffSOH w CO CO CO CO Tf CO CO IM «£©-! 'H©'© H ++ 1 1 ++ e»00 t-iiHNNOe Tf< •* CO MNTfOUJ o 'S Ot~l-i-lrH ^oocot-rHO0>ae»icq CO CO CO ^ *C* CO CO CO CO CO CO CO CO CO MMM6DM611MMM61) M61M606DM6U6J1MM 6 e 6 11 6 i b 6 D t> I Ml OOOOOOOOOO OOOOOOOOOO I COWCOWCOWCOIQCOW COCOCOIDCOWCOWCOW COWCOIQCOWCOIQ I 3 1 I 8 *3 1 i ; S « > 1 § •a • ►8 - « I § -s> | a ..*« - i C So A3 & ection I: Norma June July July Av Av ■o» k k. ^ t> Is -1 -1 "< « ection II Normal June July July Avi 1 4 OH, I-, ►-, 3 GO m Patellar Reflex. 41 t)i m o to to us • t»"«* £o omohoo oeB !SSS + 1 + 1 + ++ 1 +++ t- UIOOMS Qon nnneci !i e* ■* us ■* + + 1 1 + 1 1 1 1 1 HOtDOOKfi + 1.9 + 1.6 +123 +62 +93 HHMMHO + 1 ++++ e eoet t> ■*■*■* n e> d ih co e i-i '+ I I + + + + ^ddcoddo© '++ I I + I I oo ■# m H rjl OS Nte £)o onoded 1 + 1 + 1 + 1 1 eat- oo©i>t> So us t~ ^ ** fi ei HHMNN ++ 1 +++ + + + + + OHOJOTf + I + I I 0©Oi-id + ++ mi ti ti ti © o o CO lO CO IQ 2 8 II I 3 3 s s ■iS -*= to a on J ■§« IS 3 it * Q <5 a S o 9 o • 3-4 H p o § a ci> fll g fl o ^s .SSg J3 >. a Si Si o •all 3116! m'-S » -a ° S^S -*5 ■a §1^ h8t)« .2 8 £ a 5 «"9 » is a g-jj a - g 3 s 3 t o o3 03 » J3 5 © o o jd ^5 *5 "S o *o a .a DO OH SO FUSS'S £ - « - s |- ■§ 42 Effect of Alcohol on Psycho-Physiological Functions. 4.4 and 14.8 mm.), in strict accuracy should be placed under periods 5 or 6, or in an intermediate position between these two. On June 29 only four repetitions of all the measurements could be accomplished in the 5-hour session, as compared to the six periods on three of the other days. It did not seem feasible, however, to distribute the fewer meas- urements for June 29 and 30 and July 1 so that they would correspond in point of time with the averages for the periods on the other days. The values have therefore been entered under columns by periods rather than an attempt made to specify more exactly the number of minutes which elapsed between the measurements and the dose. The aver- ages of the values for L and A obtained after the dose (i. e., in periods 2 to 6) are given in the last two columns of the table. Thus, L, 34 »♦ .^ ^!0- 2. — Effect of alcohol on latency of the patellar reflex with two intensities of stimulation. 90 80 Experimental periods 'after alcohol 3 4 5 +> c V o 70 -50 gram Primary(A) Secondary (X) /_ 60 -30 " 3R — a- 123 P Fro. 3.— Effect of alcohol on amplitude of the patellar reflex with two intensities of stimulation. Patellar Reflex. 47 relaxation on the part of the subject, who previously may have stiffened the body somewhat preparatory to receiving the heavier blow on the tendon, which usually caused in his case a violent reflex. A', 30 grams and 50 grams, tend to show results which are opposed to those for the amplitudes of the first reflex. The opposition is most clearly marked in periods 2, 3, and 4, with the reflexes from 30-jram stimuli. Whether this tendency is an alcohol effect, an individual pecu- liarity, or a normal phenomenon can not be determined with certainty, as there are insufficient normal data for refractory-phase phenomena on this and other subjects with which the alcohol data may be com- pared. The original records for the second series of patellar reflexes with Subject VI show frequent instances in which an extraordinarily large A was followed by a diminished A', or vice versa, within any one set of records. The effect of alcohol as shown in percentile difference by Dodge and Benedict 1 for Subject VI is L, +0.3; A, +13; L' (illegi- ble); A', +173. Since the former investigators used experimental sessions 3 hours in length, while those here described were approximately 5 hours in length, an average for periods 2, 3, and 4 will most nearly compare with the previous results in point of time following the alcohol dose. The effect of alcohol as shown by an average of the percentile differences for these periods is therefore as follows: 30 grams, L, +2.6 A, -16.1 L', +4.5 A', +17.0 50 grams, L, +1.0 A, - 3.2 L', +1.7 A', -37.5 It might be supposed that the earlier results for L were influenced in part by the use of two intensities of stimulation, i. e., 30 grams on the normal day and 50 grams on the alcohol day. 2 However, the statement that the 50-gram hammer was used for Subject VI on October 14, 1913, is incorrect, as the original records show that 30-gram hammers were employed on this day. All of the Dodge and Benedict records for this subject are therefore entirely comparable as to stimulus intensity. The two series of experiments yield results which for the 30-gram stimulus are partially in agreement; the L values have the same sign (+) and are both small, the A' values have the same sign and are both large; those for A are of about equal size and opposed in sign, and those for 1/ in the former series were illegible because the contraction was too slight. The 30 c.c. of alcohol affects but little the patellar reflex latencies of Subject VI. The percentage effects reported by Dodge and Benedict in their summary table for all their subjects are usually very much higher, partly because they include the effects of dose B (45 c.c. of alcohol) ; but Subject VI showed a smaller effect of alcohol on the latency than any of the others in the normal group; the two series of experiments confirm this as an individual tendency. The results for the two series on A' are again peculiar, but confirm each other. 1 See Dodge and Benedict's report, p. 54, table 3. 2 See Dodge and Benedict's report, p. 50, table 2. 48 Effect of Alcohol on Psycho-Physiological Functions. The patellar-reflex data obtained in the second series are believed to be very homogeneous. There were no large fluctuations in the laten- cies, such as are mentioned by Dodge and Benedict, and the differences were usually small, as will be seen by a comparison of tables 3 and 4 with Dodge and Benedict's table 2. In only one instance (July 1, period 2, 50-gram stimulus, table 3) were the records for L and A illegible for any cause; table 4 shows relatively few cases in which it was wholly impossible to read any of the L/ and A' values in a period on account of insufficient amplitude of reflex. So far as the data taken are concerned, there is little to be desired. It is therefore significant that the effect of alcohol is so small and that the two sets of results so nearly check each other, thus indicating the difference between Subject VI and the others in the normal group. That a slightly increased latency should be shown by the average as a final effect of alcohol can not be regarded as of great importance. Approximately the same number of plus L differences occur on the three normal days as on the three alcohol days; those in the alcohol days are somewhat larger and are particu- larly influenced by the results for Jury 2. The probable correctness figures are of very uneven value, and are, in general, larger for L than for A, but not always of such size as to warrant much confidence in the differences. All that may be said, therefore, is that 30 c.c. of alcohol had but small effect on the patellar reflex of Subject VI; in the data secured within 2 hours after the dose, it tended to depress the patellar reflex, very slightly increased the latency on the average, and more noticeably decreased the amplitude (excepting A', 30 grams) . Both of the last two results are more prominent with the second reflex, in the refractory phase, than with the first reflex. The depression was more prominent on the 30-gram reflexes. PROTECTIVE LID REFLEX. The patellar reflex was the first measurement in every period, and the lid reflex was always the last measurement made. The two reflexes measured, the one representative of the lowest spinal level, the other a cephalic reflex, were therefore separated by about 40 minutes. Only one intensity of sound stimulation was employed, but the same plan was followed as in the observations of the patellar reflex in that the stimuli were arranged to follow each other with an interval of approxi- mately 0.5 second; data were thus secured concerning a group of reflexes which occurred in the refractory phase. Tables 5 and 6 give the data for the protective lid-reflex measurements, the former show- ing those for the first reflex, and the latter those for the reflex which came in the refractory period. The tables are compiled in like form and under the same captions as the patellar-reflex tables and need no special description. IAd Reflex. 49 usS | tH 00 US* © CO l> <£>' t(< 1« «N OS CO CD 03 ffiOc-N CO ■*0! 6 to < t-OON HMN^I 1 1 1 1 1 S3 CON < hi . eo t- ©en ihcjs COt-^W"* ©USOHACO i-l Tf COO O iH i-H iH • iH lO CO »H t> 1-4 ■a o •fi 1 1 : i i i 1 i- o t- CO fc J b o us uses osti ■ CO GO CO - CO ^H ^1 ffO ■ CO CO CO ■#■« ■* i-ieM HH^l • coeo 1 1 1 1 1 ++ I ■ flotoe r~ im t~ as co CO IflFl CO CO ON YH CO coco CMOS •o COUSiH HNN IN IN CO ■* MOO CO ■3 o 1 1 1 1 1 1 1 1 US eo b- pu hi b us CO N^i Hf MHNtO r~ ■* ■* N CO o»
eo od co cn © exi WO"!» CO W IN CO si eo Ph eo CO O0 asi> hi b rt< -ej Tt< i> w osoooswco IM i-l CO CM 1-1 r* iH Tfl »H 1HCV ++ 1 + 1 + + + 1 1 ^©t-»t-i-<© cocousao© |iHCOCOMUS US id CO r-l ^' iH oooo co © CO© ■«!l (H OS < 00 CO i-H iH i-i CO o© ■«» •* S ,-H ,— ; ,— | tH « T3 O •a co 1 ++ 1 1 1 1 1 1 1 fc- eo COM CO OS hi b U5^Q0CO^ NO«ft^ CO iH (NO IN CO iH US **M tH + + 1 + ++++ + + _£ CO © us CO ff\ | « os «o g" r © US I>T-H 1-1 T3 O < Oi OS i-t©U5 1-t Jrt hi b (N CO Olfl-^ CO CO ^ CO © t> us^us CO CO CO CO >> : ^! ■ + 6 f + 5 I '§ c ft 3 ) i i 1 I i CD ■ > as ^ IN hi «» a g g iH >> 3 K > 3 a. 6 0. > 4 1 1 to • 3 • 1° r CO •SI ■a t OS i . o ■ fl : as . s ■ to : 'S OS " IN eS V iH s CO > •3 co 03 £ 3 1 CD § S3 "3 o' CO CO e 3 IN 1 > ■3 co 1 hH a J. 2 a m s l ■3 ° CO ■SSI CO c g is CU o S3 S : s rs a ■° IE S3 : § 6 a shph g .2 s h,h,h! oi-iiHi .s oHh, oh,hh .2 Q Ph Sfc 3 -g % 3 -ii CD 03 DQ 50 Effect of Alcohol on Psycho-Physiological Functions. J 00 *H iH fc»C0 1 1-< co si N i-i IQ W N t>N CO IS t» CM COO on) §3 f-l + + 1 h 1 ++ 1 1 1 8 g loieiA co iNcous COM co co in eo en co t>lH CO CO OS -*t;l> CO CO CO e CO CO 1 1 us O COiH 1 1 ih c; + + gs 1> enen | o *»■*© OOCINOi o ■<*■> IN lO t-l> •>* 00 00 >o ^ co Ttiia-"* IN CO **coiH mow onoo Ncn eo 1 1 IH o ■c + + 4 1 1 + 1 coo fr ^, U! c iei> t-i> J b oset! ^ CO ^1 in c» a>t»» IN COO 00 O INN i-l CO <£ CO CO CO CO + 1 1 1 + 1 1 1 t |ridc 1> OH) 00 00 e h-1 b HNHifteo U3^ OlAtfi IN IN i-l CO Officio cot- iH ■* lO ■*^ »-l 1 1 +++ i n 1 1 00 N en s j o io oooi» e « e eot- Nil) f iJ CO 00 CO 00^1 HIOOH NHIOO oo CO CO ^1 CO IH + + 1 1 1 1 + 1 + 4 co . CO CO OO) U5 m in co coco IN CO COCO oo oo cq eo o s§ a 9 .2 <) 1 IN CO CONW s HWHHO O O HO + + 1 1 O CO INN 1 1 1 1 1 1 U3iH H Ifl e e CS ss J b CO lO COOOU! CO tO CD N^< CO W5 t-CC cc in ■^< co co n CO "dl ^1 ^1 i- + 111 1 + + + JrlOCOHOJ |Nn4itpi IN O U3 Cft ^ ■0 o I ^ IN CO CO CO i-l h5 k MOOH|. OOiNOCD -^ ^ ^ -^ ^tl CO CO CO 5 Tt ^ i c 1 5 ) 1 i 2 : S J | + ! < S 3 ! 1 ? IN 9 I = 1— > CO 1 >< > < 1 If 11 CM 1 > > a tv S t- a. > s 1 "3 a - ^ « a t r z > a 53 S3 ISO " cr en - t i a "o O a .. °3 e i-i>« I a o i Jt a =• fc a ifferences d P.E.D robable corre I* i .2 o *= h» n la m ohhh 53 DQ (-1 Pi Lid Reflex. 51 The latent time of the lid reflex is the same as that for the patellar reflex, with the technique used and for this subject. The average for L is 35 a. The stimulus interval was the same (0.5 second) for both reflex arcs, and in both instances the 1/ was longer than its respective L. The average L' for the lid reflex is 41 a, as compared with L', patellar reflex, 30-gram, 38 a. The identity in the latent time of the two reflexes for this subject may not be considered as entirely acci- dental, since Dodge 1 showed that in his own case, in which the wink reflexes were quite uninhibited, the latent time of the wink reflex and of the knee jerk were practically identical. No comparison is made on the basis of refractory phase values. That the average amplitude with the two reflexes in the data under consideration happened to be so nearly the same has, of course, no significance and is purely an acci- dental matter. There is a fairly regular decrease in the amplitude of the first reflex (A). The first record taken in a period usually showed the largest amplitude, and the other three more regularly the smaller amplitudes. Reference to table 5 shows that the average amplitude for the first period is likewise usually greater than that for succeeding periods on the same experimental day, and finally, that the average A decreases from day to day. Beginning with 22.3 mm. for the first period on the first normal day, which is the largest amplitude during the series, there is a decrease to 6.5 mm. in the first period for July 3, the fifth experimental day. The columns of average amplitudes for the normal days regu- larly show a decrease. With the alcohol days there is an exception in the case of period 1, but in general the amplitude decreased as shown by the averages 7.0, 6.5, and 4.0 mm. for periods 2 to 6 of the three days. There is a tendency for a compensating rise in the amplitude of the second reflex; for example, the normal day values for A', period 1 (see table 6, section i), show amplitudes increasing in size (2.1, 3.0, and 4.3 mm.), while those for A decrease as previously noted. The amplitude of the two lid reflexes, therefore, seem normally to approximate recip- rocal relations with one another. The reduction in the amplitude of the reflex contraction, due to the possibility that the subject became accus- tomed to the stimulus, is of course favored by the use of consecutive experimental days. In the case of Dodge and Benedict's psychopathic subjects, 2 all of whom served on consecutive days, the reduction in the amplitude of contraction is conspicuous in every case for A, while A' shows a consistent increase. These A' values are, however, very slight. The average mean variation for the latency in the case of the pro- tective lid reflex is larger than with the knee jerk. The patellar reflex gave an average mean variation of L 1.3, and 1/ 1.0 • 1 1 1 * Secondary (A')—* 1 1 1 t \ 1 1 1 i t j • % k \ 1 1 1 1 1 1 i/ i/ r^>^- 4 i \N 20 30 40« 50 *>' 70 / 1 \ \ 1 1 1 r / / \ \ 1 1 1 -•* r \ \ 1 1 1 .-" \ \ I 1 » \ 1 1 1 1 Fio. 5. — Effect of alcohol on amplitude of the lid reflex. 54 Effect of Alcohol cm Psycho-Physiological Functions. a decrease in the latency in period 4, changing again to an increase in periods 5 and 6. The curve for the secondary reflex latency follows fairly close to that of the primary reflex. The course of the alcohol effect upon the reflex amplitude, as repre- sented in figure 5, shows the greatest decrease in the amplitude in period 2, a somewhat less decrease in period 3, a rise to an absolute increase in period 4, with a return below the base line for periods 5 and 6, that is, in the case of the primary reflex. In period 5 the secondary reflex shows its greatest depression. In every period except 6 the two groups of curves, figures 4 and 5, are reciprocal, the one rising above the base line as the other falls below, and vice versa. In periods 2 and 3 taken together, which represent an interval of 80 to 90 minutes following the ingestion of alcohol, there is an average increase in latency L, 9.5 per cent, L', 7.8 per cent, and in amplitude A, —36 per cent, and A', —50 per cent. In figure 6 nine of the lid-reflex records for our subject are repro- duced full size. These records have been trimmed down so as to per- mit of a larger number being presented for comparison. They are read from left to right, the time distance between the ordinates being 0.01 second. The curves which show the large depressions are direct shadows of the artificial eyelash; the stimulus line is given below each reaction curve. The first little peak in this line at the extreme left indicates the moment of the sound stimulus. In records A to E, inclu- sive, it will immediately be recognized that the primary reflexes are all complicated by previous voluntary lid movements. The records were so selected as to show the voluntary lid movements approaching nearer and nearer the actual time of the reflex. In record F the voluntary wink and the reflex wink fall together. The sharp, characteristic drop of the reflex is, however, clearly to be seen in all of the curves. The presence of these anticipatory voluntary lid movements is an individual peculiarity for Subject VI, not exhibited in marked degree by any of the other subjects previously employed. When the fid-reflex records of the repetition series were spread out for careful inspection, it was noted that in the period following the ingestion of alcohol the voluntary lid movement preceding the primary stimulus did not occur. In other words, all of the records were such as are shown in figure 6, Nos. G, H, and /, each one of which is the first record following the dose on an alcohol day. It seemed worth while to make a careful count of the records on the following points: (A) the number of records showing voluntary lid movements in the 0.15" preceding the first stimulus, or combined with the first reflex; (B) number of records showing voluntary lid move- ments in the 0.15" preceding the second stimulus, or combined with the second reflex; (C) number of records showing voluntary lid movements following the second reflex. The results are given in table 7. Only iiniiiuiiiiiiiiiiiiiHUiiiiii"" '•"•" '■■■■Si™ .Hk iiliiiiiir . a iiiciiiiSiiiiiSiiiuiHuiuiiiiiiMiiiiiii>*iiiiiiiiii»iKiiHt(iii ■H. ".JIUiKlll WIlK^illlHm I HiffiBlBMIlS!l"jiKMMIIIlillliilIllMllllliniHIIIlllililliMlliIllll!l IHillHNHl I •- 111 1 iHUKKKKHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIllllllHinilltllllilllHIIHIIIi L..miiiii.iiiiit " iBHiiiiilJisiiiiiiiiiiiiiiiiiiiiiiiimiHiii iiihi MlillllllHllllllH illllllllllllllllllllllllHIimilllllll llllllllll 1111111111111! iBBKuHiiiiiiiiiuiiiiiiiiuiiiiiiiiiiiiiiiiiiiniiiiiiiiiiiiiiiiiiiiiii mmniimii "KKKKKKHHUne'uBKKHHuKKiiiiiiimiHiiiiiiiiiiiiiiiiieiH U..111.11111..1 IUk.« — iiiiii, ■ Mil'" ... ■ III ..IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHI in iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiimiiiiiiiiiiiiiiiiiimi -i to toes I I + I OS tO CD © rtHM CO O OO ONMI rH IN INN IN tN INt I + I I HOrtN rH rH *#iH + I ++ CO ^< «— ( CO I I I I H* CH COO rH iHH + + + + HHJMf) IN IN INN O CO CN1S cq I ++ ! «OhN NHMC4 + + + + CO CM ton I + I I icnaoo »-H + + + + CO OS CO tN ■-H CN rH I + I I 00 ^ O C4 + I + + «-H O r-i ^i O b- COX O- CO ci © 3 1 Si OS rH CO CM ^"3 nS oosm n»* • rM--)l • ++ : sort •to. Sis" ++ 33 ++ 0031 CO OB ISNt o> i-i 1 eo S -3 o cq ^i l!i^ll! 4 gl||l^illl Hsrarj U rj h, Hj O S^H^ Orjrjr, 3 .= S r-a !--> r-a ir5 V CQ Or,!-, H5 3 9 o o OS CQ o •o Ph S ta a ,§•3 g -3 8° ■S S" s § -° • en (SQCh Eye Reaction. 61 expected to produce a noticeable decrease. Particularly would this be the case when so relatively few reactions were taken, that is, only 5 every period. The general decrease in reaction time, attributed to practice for the Dodge and Benedict normal group, unfortunately rests on fragmentary normal data for Subject VI. There was but one normal day and it so happened that the records for the first period of this day were illegible. There was also only one alcohol day with dose A and the values obtained on this day were peculiar. Subject VI showed a marked variation in his average eye-reaction time between the normal day (October 22, 1913) with an average reaction time of 209 a and a mean variation of 22 a, and that recorded on the alcohol day with dose A one week later (October 29), with an average reaction time of 159 a and a mean variation of 24 a, there being a difference of 50 moo C* «0»OOt HHMH lOl- 1 +++ +111 1 1 > 00 CO rH 00 cr CO CO CM OIC t»»o • 3 coeo ■* lOTl rHF-t t- r4IOK IBrtcJrt f-t- •* t~e ■*■* r-l 00^ cntNi-i ■tf tji ■* »T + + 1 1 1 1 1 t> (N ■*« CO lO ^T ••^ OS t* iM rjl COM i-l CO FH tPH • tN • ■a Q S 1 ++ 1 +++ e "S -§" Si Ph- CO !>>-< I! CO 00 t> t"» CDiOOT t-( (N -^i-l UJi- M 1 +++ +111 1 1 1 -a e a > © CO OOlA rtiC-^C HOeOOO © CO io C tNt- : a CO CO CO ©• ^ CO ^ ^ iH i-H lH l— 1 • CO T3 + + + + ++++ h : V 2 *B S O O § g o*o Ph t» .2 OS^O*-I HIOHH co*o»-tua hooks rtftt^H i 3 CO CO I>00 b-coosr ^4 tJ< ^ ^( tJ< ^i CO T CI TjH (N hh « as a CO CO t~fc» 00 CO t^^ IH tN • ■a wcqcqes CN OS CO »c w (Nocie ■a o •a a cq cq (M w CO rH CO W lg « .2 H05 0J«C OiOOCC nl: •mal values — une 29 1 > ■3 C. 1 '■1 'o « ll tN 1 . > ■3 a t > nil: •mal difference une 29 r- > > > a b % ■< O o § S3 C ■a « 1 ■3 ! O & • -II 1*£ er cent differe: inferences divi robable correc .3 5 •"> •"» •"» obbb<; .2 3 l ~3'~='~= oh^h o ,2 Ph H Ph 3 2z 3 03 o V to 03 02 64 Effect of Alcohol on Psycho-Physiological Functions. is found to be 471, 474, and 466 a, respectively, with a total average of 470 a. For the alcohol days the comparable averages are 466, 456, and 408 a, respectively, with a total average of 443 a. On the average, therefore, the word reaction following the taking of the control mix- ture is 27 N CD tP^I CO CO CO CO CO '•J' CO (TO b- CN C0U5 CO b- 00 CO 56 : a § i-H i-H V St 03 s 3 1 1 + 1 1 + 1 1 ii : : o o to to nte 03 ««h GO 00 ■* 00fc» HWWtO OJ to -*t> i-t io -^ N O OS ^00 enosi- ei O 00 b- OS 00 o o no CN "^ r-ti-H w* (0 •a ,— I 1— 1 1— 1 *H + 111 1 1 + 1 +++ > ■*■« OHO !>*> HNOJ o5N» ; oj s NN CM COM F-l ■a a 3 co 1 1 1 1 1 + ii : : T3 -d O i-<0 _g t-t co i>o d PL| GO Oi« OCOiH COCO SOON ■* COTjt OOiOiOO in t-eo CN lO COCO csccus - V a CO CO coco i-t !-< • _g »o 1 + + I + I I mi: 03 > TO o ■* ■n t-03 CO t>en a) 03 h cd <* t~ cs b- »0 ON CD ICiM to Nac» -" so coco CMC* CM t- i-l OS CN b- *O^H »0 CM ^t t- ■* ■a 1-1 »H i-H ^H 1 1 1 + 1 + 1 + + + !> OHCOH OHOOtfi HN«ao o >h o« leioeo - Sd < o fc GO H00HM OS b- CD b- ««HO ^] iCiOb- •*t-usrt HMHC 00HHO •** •>* COlH CO CO i-H *N l-H iH ^- l-H l-H W + 111 1 1 + 1 + 1 1 8? > ©lOHUI Ui 00 b- O O CO Of >0 « rHOO le^*© • n § CN "^ CO PC CM CO CN CO i-H iH CN CO ■*« *# rH OSCC CO *0 CNCC i-H b- CO U5 .-I i-H i-H OHOJN CN iH TjlTfO ' ci 1 1 + 1 1 + 1 I +++ : •a 53 o fe . d to 1 CBODOO PL| CO b- OS cocc iOHOf- «Hfli|> (OiH IC4 ho e>5 fflnen CDiOOI> ooqc b- CO CO CMi-n-t iH iH i-( w III) MM + + + J !> OS 00 »H O* OO00CT 1 CO ■a s CO ^ COM Tfi CO -^ « OJON^ 1-t CM i-H W 1 V E/j CO CO ■"*< w NiOON fc .H fH ^ i-t r-t rH r- d o ■a 03 Q J s is l CO 03 d 1 1 1 . n to t» i^ to OJ M 0J a £ -S •1 | tS s D D 1 CJ 1=1 »- « a b I 1 as o ,3 > cr ) CM ) Tt a t a. s o in > *■ * C a b a OJ S3 Oto ■= c 1 ■4 a 1 O ■s -31 t-t "S a a t fg-3 § T3 OJ ° * 8.2 O o f & 4 .2 o >-> >-> >■> 'o © is. s. •* ■is*** d g a33 "O © »i. »a. ^ § ,8 -= >-= !-» UHj^^ %X =3 tz ^ to u 0) m CO GQ CJ CO 0J rj CO « la 3 ts a-c a j3 -CJ o a fan CI) -S c3 -*3 p. rt Jh Q .s fi =J T) u Cl 03 PQ O Tl C3 O n A CJ >, *• £> a CO CO CO CO + =J „ no XI i-H ^ a> + fi CD a o OJ sc fi a> S3 •3 =3 ■3 1 T3 a 1 CO 1 O & 5 P- O o ■*-» cu >. 0) 3 "P. ^Q A cct > ot CD a> rfi OJ u 5 # d _ < _ s tfi Tl ^j T) 1 P. -U t- a .a E3 03 o -fi CJ ea a> OJ M o •i S CN o > QJ u CO ,o OJ a; la J=t +3 H X H CO ft o? ftT3 72 Effect of Alcohol on Psycho-Physiological Functions. 1"H (-. > »-H i-t OS O »H OS t» CO 00 |> oo«h^ o> ■* io© coiccow oio e»i «e ^ t- os «h en f-H CI lO CD CO COU3-^^ + 111 MM NHH + 1 1 t> OB oj wo co CO •* -*© CO kA SO GO a CMC* cq coeo CO O "B 1 1 + 1 + + + s s OQ coco cote com HtD« co co do CD t-tj. Gonia CO OiCfi HOON »> i-ieo nhioc CS ^< co oo a "* CN« CO CO CO w l-H ■o o "S Ph 1 + 1 1 + I 1 + + od 2 2 00 co^ OHMH CO COIO i-H OS ^00 Ob-4 cN IQO0C O CO t^© CO ^U5 HO0HM CNili-l rH rH i-H w* III Mil + + + > CO i-H CO CM Q0NOI> (N-^lOO ONW^I ccet- a CM »0 COCO CO w4 HM i-i i-i ■a M + 1 1 + M + + + e 0! CO '§ CN F- O Ph 03 OJCDOff tH W 00 GC otnhc TH r-H F- wwoirt cqeqeow + M 1 MM ■^ceos cn »M f* i-l + 1 1 > HCOQf- CO CO (N es COf-OSCO COUDUD^* l^tNUS a CO tH iH O? I-l CO O •e CD MM 1 + M 1 + + is u OS CO CM l> Hb to oo cm cm CO lO IOCN lO'tf-*^ ■sets c t- CO Ofc* WOHOi HCMrt 1-4 i-H + M 1 MM + 1 1 > HNOft ioc»c»oo t-cqc»eM OOOl a CO "# CMCN (N IN il CN M + 1 1 + 1 + 00 O « o» "8 Ph 1 * 0» CO t* 00 io « ■^ io oo« CO CO CO OC CO CO ^CS 00>*rtiH OtD-flt CO IO CDM + M 1 MM U9i-He> m CNiH + 1 1 t> ^ o oofc* oo io loco CO is ^3 i— « c t I CD © Ol «* a b it © o i- c a £ a. © is CN Tt a b I ©'Oof D p g B a 1 1,3 i-S m -S O >"9 1-9 t-s 55 co a) « m CQ CO -a H cS •c > s I O S.Q to CD O C4 M fl fe sa 3T3 "°, Tt (U ^ 5 J3 5 s co §1 if •ai 3-S Memory. 73 tion of the second reaction and that point which marked the complete exposure of the word. By this method no account is taken of R ± or its position relative to the exposed words. If at the time of the second exposure the subject can speak the word before any part of it appears, he would have a score of 250; otherwise he would get credit for just that part of the word which is not exposed at the time of the second reaction. This does not of course eliminate the factor of residual memory, but on the other hand it does not oppose this factor to that of the "saving" which we desire to measure. If residual memory was present at the time of the first reaction so as to shorten or even to make it occur before the exposure of the word was complete, the second reaction should be still earlier than otherwise would be expected. The data treated in this manner are presented in table 12. Refer- ring to the column of averages at the right in section i, we see clearly that the alcohol values are larger, particularly on July 4. The average for normal days is 74, for alcohol days 94, with a difference in favor of the alcohol values of 20. The complete exposure of a word requires 250 of our units. On normal days, therefore, the second reaction occurred on the average at a time when the word had been ^ exposed, as against |=5 for alcohol, that is, approximately 70 and 60 per cent. The mem- ory score, with the exception of the first normal day, June 29, is larger for the first period; the differences are therefore nearly all preceded by the minus sign. They average —30 for the normal days and —45 for the alcohol experiments. The results for the first normal day play a particularly prominent r61e in sections n and in of the table. On the first day the score was especially low in the first period, being only 41. All the differences have the plus sign on this day, with the result that the average differences and per cent differences show a minus effect of alcohol for periods 2, 3, and 4, 1 and a total average difference of —15, or a per cent difference of —12.4. Since the value for period 1, June 29, is so decidedly out of line with the first-period values for the other days, while the scores for periods 2, 3, and 4 on June 29 are practically what would be expected judging from the other normal days, July 1 and 3, it is not thought just to trust the method of differences here, but rather to consider the simple difference between averages. The aver- age 63 for the normal period 2 is deducted from the average of 88 for the alcohol period 2, resulting in a difference of +25 (see section in). The differences between averages for the other periods are +15, +14, +20, and +48, with an average for all periods following the dose of +20. The probable correctness figures apply to these differences and are in every case of such size as to give them a good degree of reliability. The memory data, presented in whatever way we chose, shows that with this particular memory test and in the particular form used there is a better performance following the alcohol. This confirms the re- 1 It should be noted that there were only 4 periods on June 29. 74 Effect of Alcohol on Psycho-Physiological Functions. suits obtained with Subject VI and with the other normal subjects in the former series. It does not, however, explain why the memory- results after alcohol should be opposed to those found for most other processes. It is noteworthy that on the last day, July 4, the subject obtained a conspicuously higher score than on any other day. His best work in this test came in periods 4 and 5 of the day, 1 i. e., towards the last of the session. These same features were striking in the re- sults of the word-reaction test, which employed almost the identical apparatus used for the memory work and demanded somewhat the same attention and cooperation on the part of the subject. 2 SENSORY THRESHOLD FOR FARADIC STIMULATION. The Kronecker inductorium, which was calibrated for the Nutrition Laboratory by Professor E. G. Martin, and the accessory apparatus recommended by him for the determination of the sensory threshold for faradic stimulation, were used in the same general arrangement as described by Dodge and Benedict. 3 The only change consisted in the introduction of a key for the making and breaking of the primary circuit of the inductorium which was mechanically operated by a small motor at the regular rate of two breaks per second. The adoption of such a mechanical key offered two chief advantages: (1) The subject could be near the apparatus (of course it was not in his direct view) and at the same time be ignorant of the exact moment when to expect the stimulation. The key was placed in a sound-proof box in the dark room of the laboratory and operated continuously during the test. Its action could be judged from the operator's point of view by watching the needle of the ammeter. 4 The shocks were controlled in intensity by the horizontal movement of the secondary coil and were further delivered to the fingers of the subject or not, according as the short-circuiting key in connection with the secondary coil was opened or closed. The operator constantly had his hand on this key, which could be used noiselessly. (2) The increase in the rapidity and the regular rhythm of the shocks seemed to be more favorable for attention than an isolated shock, which may be easily confused with pulse-throb and muscle-twitches, particu- larly when the shocks are near to threshold value. The most unsatisfactory point in our use of this measurement was the determination of the tissue resistance, this being necessary to 1 This does not take account of period 1, which is set off from the rest of the records as preced- ing the doBe; furthermore, the factor of residual memory seems to have had a much greater effect on the results in the first period. 8 Subject VI did not employ regularly the method of fitting the words into a story during the time of their first exposure. If an average is made for the "saving" on the first word in every series on each day, the value 179 will be obtained; and in like manner the values for the second to the twelfth words will be in order as follows: 139, 100, 102, 83, 80, 73, 99, 110, 93, 83, 107. Word series No. 11 was by far the best remembered from one day to the next. 3 Dodge and Benedict's report, p. 95, fig. 14, and p. 137. 4 The current used in the primary circuit was 0.5 ampere. Sensory Faradic Threshold. 75 determine, according to Martin's formula, the number of /3 units required for stimulation. As the tissue resistance is without doubt almost entirely due to the outside horny layer of skin, it is regarded as outside resistance like the resistance of the secondary coil of the induc- torium, or the resistance of any unit put in the secondary circuit. The measurement of the tissue resistance is accomplished in practice by balancing on a Wheatstone bridge. In the very simple form used, the secondary of a Porter inductorium was employed as a source across a slide wire. A telephone was connected to the slide and also at a point between the known arm of the bridge, 20,000 ohms, and the arm which contained the tissue to be measured. The balance point was extremely indefinite. Practically it was necessary to endeavor to match inten- sities of tone on both sides of what seemed to be the indicated balance position for the slide. This method must be called in question, since the tone on one side of the balance position of the bridge has quite a different quality from that found at the other end. These quality differences play a r61e from the standpoint of intensity when the tele- phone is used as an indicator in the bridge, because of the characteris- tics of the telephone membrane. It was of course recognized that the tissue has not only resistance but capacity and that there was present a fluctuating body current. At the time of experimentation we were under the misapprehension that it was not practical to balance for capacity along with the resistance. The tissue-resistance measure was usually about 5,000 ohms. The fingers were immersed to a depth of 2 cm. and this condition was kept quite constant, as it was recognized that theoretically the amount of tissue immersed would change the resistance value. To balance about 5,000 ohms tissue resistance against a known resistance of 20,000 ohms, as was done in this case and also in the measurements of the previous series, 1 is theoretically not good practice. The balance position on the slide wire under these conditions is close to one end of the bridge. The observational error at the end of the slide wire of the bridge is propor- tionately quite large, as a change of 2 cm. in the determined balance point would here be of much more importance on the resistance figure. The employment of a known resistance of 20,000 ohms was an expedi- ent, as in practice the balance point was much sharper than when using a known resistance of 5,000 ohms in its place. Probably this condi- tion was due to a larger proportional effect of the finger capacity with a lower resistance. Another point to be criticized in the technique is the order in which the measurements were made. As the tissue resistance is due to the outside horny layer of skin, evidently there will be a decrease in resist- ance when the finger tips become somewhat soaked with salt solution. It is regrettable on this account that the order of our procedure in refer- 1 See Dodge and Benedict's report, p. 137. 76 Effect of Alcohol on Psycho-Physiological Functions. ence to included secondary resistance was not reversed, since the tissue resistance decreases during the first four or five minutes after the immer- sion of the finger-tips in the electrodes. Our first measurement should have been with the 40,000 ohms included in the secondary circuit, and the last measurement with the tissue only in the circuit followed by the Wheatstone-bridge measurements of tissue resistance, which in the test always came last. Thus, any change in tissue resistance would not have exercised so large an effect on the first threshold values taken. 1 While the measurement of tissue resistance is thus admittedly poor and the procedure is not without fault, yet the routine from period to period and from day to day in these experiments was always the same. Hence, the value of the results for a comparison of the alcohol and non- alcohol days should not be much impaired. The data and results are incorporated in table 13, which presents them by periods, the table being divided into sections according to the usual form. In order to have data which compare directly with those presented by Dodge and Benedict for Subject VI, values for Z have been included in the table. 2 Since with the technique used threshold determinations were not mul- tiplied within the period, no mean variations are given or figures for probable correctness of differences. The averages for normal days to be found in section i in the last two columns of the table, and which exclude period 1, give the impression that there was the reverse of a practice effect. The number of /? units required to stimulate increased from day to day. 3 But this is not evi- dent in the thresholds for period 1. The values for the normal days average 356 Z or 184 /? units. This average is perhaps somewhat larger than would be expected for Subject VI from the data given by Dodge and Benedict in their table 22 for normal days October 7, 1913, and March 2, 1914. It is thus to be observed that the use of the mechani- cally operating key with more rapid rythmical shocks did not serve to reduce the threshold current strength. The average of 184 for /3 is not, however, conspicuously large when compared to the values reported by Dodge and Benedict for some other of their normal subjects. The fact that Grabfield, 4 in determining the average threshold for 135 psycho- pathic cases, found the average to he at 223 /3 and came to the conclu- sion that a threshold greater than 175 ^ may be considered as definitely pathological, can only be interpreted as indicating a difference between the apparatus or technique employed by him and that used at the Nutrition Laboratory. 1 Grabfield (Boston Med. and Surg. Journ., 1914, 171, p. 883) assumed a tissue resistance of 2,100 ohms and omitted the measurement entirely. 2 See Dodge and Benedict's report, p. 140, table 22, Subject VI, October 7 and 14. It is under- stood that the Z values given here, as in our table, are the threshold measurements, with tissue resist- ance only included in the secondary circuit. 8 Martin, Withington, and Putnam, Am. Journ. Physiol., 1914, 34, p. 97. 4 Grabfield, Boston Med. and Surg. Journ., 1914, 171, p. 883. Sensory Faradic Threshold. p oioe^ 1 o»-tTt*w ^^^-^o t-iOOiO HOt> HHrtri IN rH INl-H ^ | -)- ++++ ++H 69 t- ©a ^co^ 1 ^ t~ io -^io co i-H S co co co vt tjhcococo _|__j__|_ ++++ + + 4 a c • "8 CO 03. -*>* «OH o© O 'O CD CO Tf t- • 3 pH pH OOOOS CO CO Mil CN rH o ■§ Pi IN CM • P-t W ¥H .++ :+++ i I i ■OO • OS en co^i i-irfev o ID •d •g i-i Op 4 NHHN 1 + 1 ++++ + + 4 cs e\ O a Pi M OS 00^ O 00 OSCD hhO 00««H MrtiC £ ^ OS t» pH OS ri<00 COOS Ttl 00 CM»f3 rHCN CO CO CO. Tp" CO CO CO 1 ++ + + + + + + -4 ^ § "a c a as •J 3 91 COOSt-O MS«8 £o»N ^ t^ OS W3 O CO t> 'B> -^ IO00OS00 l> 10 © 00 — rH WHISH i-H -S S 3 Si •a o pH i-H pH p» 4 -H i-H (N »-l I + + + 1 ++ +4 o C3 "fi V £ w "p P< tsi oooN nnooo -rooiio O CD CO CO CO X If .2 CSpHCOCD OOCOOOO wcOTtCN CD rH CO ■** NW« ++++ ++4 tn P o co iaigS ++++ ++4 '■is ■ ! pJ «a ^^OOTliPH COCDiOW H.O3CC05 r« t~ t- co — © ■a 1 liHP r-l rH iH »-H i 13.- N .s w jction I : Normal valu June 29 . July 1 . . Julv 3.. U a IS 3 •2 J cS O p» CO "o « °l CS 1-3 3 IN 3 . > 1 Average ction II: Normal diffe: June 29 . . pi > CO 3 1-3 as 3 »s c3-° o o 3 o CO CS c p? CM 1 > 1 a 1 $ < ction III : Effect of alco 8-3 C3 a -S i S c t- cs S cs it! K "Is .a .2 c3 U 03 as co OS to 77 78 Effect of Alcohol on Psycho-Physiological Functions. For normal days there is no prominent change from period to period. The averages for j8 as given in the table are 191, 188, 179, 180, and 187 for periods 1 to 5, respectively. Period 6 is conspicuous in showing a larger value (214), due evidently to the fact that on July 3 the threshold level was considerably higher. The alcohol days show a decided differ- ence with a marked rise in the threshold after the ingestion of the alco- hol. Beginning with period 1, the |S values for these days are 165, 211, 184, 180, 206, and 191, the average value after alcohol being 195, in contrast to the average for normal days of 184. The rise in the thresh- old is most pronounced in period 2, which represents an interval of about 10 to 12 minutes after the taking of the dose of alcohol. Values for Z show clearly that an increase in the strength of the current was necessary to stimulate after the alcohol. The average Z value of 393 may be contrasted with the average of 356 for the non-alcohol days. In section in of table 13 the effect of alcohol is given in averages, differences, and percentile differences. In periods 2, 3, 4, and 5, all signs are plus. There is one zero, but most of the values are compara- tively large if judged on the basis of the alcohol effect on other meas- urements. The depressive alcohol effect is definitely prominent in period 2, the increase in the strength of the current necessary to stimu- late amounting to +32 per cent for /3, and 17 per cent for Z. In the case of /3, which is of course the most trustworthy measurement, as was stated previously, the depression decreases progressively in periods 3 and 4, the values being +10 and +7.3 per cent. The values for Z are irregular. In connection with the results for periods 5 and 6, it is sig- nificant that the first two normal days concluded with a very low threshold, i. e., both show 153 j8, while the last normal day concluded with almost the highest threshold found, 221 and 214 /3 for the same periods. It is impossible, of course, to say whether the results for period 6 show, according to the statistical presentation, a large threshold due to extraneous factors or are an alcohol effect. In conclusion, the ingestion of 30 c.c. of alcohol is found to decrease the sensitivity of the nerve endings in the finger tips to electrical stim- ulation by the faradic method, as was the case in the first series with this subject. The decrease in sensitivity is of such an amount that on the average an additional strength of electrical stimulation of 17 per cent 03) is required in order to arouse sensation. This depression is most prominent in the periods immediately following the taking of the alcohol, in which it amounts to 32 per cent /3, and 17 per cent Z. In periods 3 and 4 the decrease is not so large as in period 2 or in period 5. In period 6 the effect is in the opposite direction, indicating greater sensitivity. Eye Movement. 79 EYE MOVEMENTS. The general apparatus employed in this measurement was the same as that for the eye reactions, and these two tests followed each other in quick succession. There was not the slightest change in the technique from the methods used in the previous research. On a black field two white dots, separated by 40° on the arc of vision, served as the fixation points. The falling of a shutter which allowed the light to fall upon the eye revealed, at the same moment, the white dot at the right. The appearance of this dot was the signal to the subject to begin the series of movements. The use of such simple fixation marks has the disad- vantage that as there is no practical need for looking accurately at them, as one would do in reading, the subject after some practice may become negligent and slight the fixation of them, thus decreasing the amplitude of his eye movements. 1 The tendency to disregard accuracy of fixation of the marks is clearly indicated in the photographic record, for, as Dodge has pointed out, the eye, in looking from one mark to another which is a considerable dis- tance away, accomplishes this by one long "saccadic" 2 movement, fol- lowed by a small readjustment, which we may term corrective move- ment. These corrective movements tend to become not only smaller in size but fewer in number and finally to disappear from successive records as the subject becomes more and more used to the measure- ment. There must be continual emphasis laid upon the necessity for the subject to look accurately at the fixation mark, and stress of this point by the operator may not always be successful in holding the sub- ject to careful work. Unfortunately, not realizing adequately the extent of this tendency, we placed speed before accuracy in our instruc- tions, whereas the emphasis should have been reversed. It is probable that speed should not have been mentioned with this subject, since he not only neglected careful fixation of the marks, but at times seemed to disregard them entirely, with the result that the eye-movements were of very irregular amplitude, mostly far short of the desired 40°. To make clear the difficulty in dealing with these records and the impossibility of presenting the data in exactly the same form as that given by previous authors, 7 eye-movement records have been placed together in figure 8. The records, which were taken on photographic plates, 2\ by 7 inches, are to be read from the bottom upward. The beginning of the first vertical line represents the moment when the light was turned on the eye by the falling shutter, which at the same 1 A technique used by Diefendorf and Dodge (see Brain, 1908, p. 458) would doubtless meet this difficulty. They exposed in succession isolated numerals in two different areas of the field about 25° apart. The reading of one numeral by the patient was the signal for the operator to cover it and to expose another on the opposite side of the field. This provided a good and suffi- cient reason for looking carefully at the marks exposed and made the eye movements between merely an incident which the subject performed without any thought of its being the important feature of the measurement. 2 Dodge, Psychol. Bull., 1916, 13, p. 422. 80 Effect of Alcohol on Psycho-Physiological Functions. time revealed the mark at the right to which the subject was to move his eye. The movement of the eye to this mark is shown in the first short horizontal line to the right connecting the first vertical line with the second vertical line. These vertical lines are of course made up of dots, the light being interrupted by the tuning fork. These dots would appear individually if the plate had been falling at a faster speed, as was the case when the record of the eye reaction was taken. Having seen the mark at the right, the subject looks to the one at the left, which was placed at a distance of 40° on his arc of vision. The line of dashes con- necting show the course of the movement and its speed, as each dash with one interspace represents 0.01 of a second. From the left the eye again moves to the right, and so on throughout the period of 5 seconds allotted for this series of movements. The record at the extreme left (A) in figure 8 may be taken as a typi- cal eye-movement record. 1 It was obtained in the first series of experi- ments for this subject on February 12, 1914, and is better than any record which the subject made in the second series and also better than 80 per cent of his earlier records. The vertical lines (periods of fixation) which separate the movements show that in almost every case there was readjustment, that is, correction for accurate fixation, following the main movement. 2 The correction indicates that usually the saccadic movement was not sufficient to bring the line of regard to the mark. Such correction Dodge has designated as "plus correction," and it is a normal phenomenon: The six other records shown in figure 8 (B, C, D, E, F, and G), which are for July 4, periods 1 to 6 in order, show clearly a general appearance which is different from record A. There are almost no corrective movements at the end of the long eye movements; the amplitude is irregular and nearly always short. There are very frequent breaks in the records caused by winking (for example, see W on record B of figure), a factor which gave considerable difficulty not only in this measurement but also in the lid-reflex records. The pho- tographic tracing immediately preceding a wink, if part of a vertical line, is usually heavy; if it is part of a horizontal fine there is a curve downward, showing a tendency for an upward movement of the eye. The records were nearly all legible, but the irregularity exemplified in those shown, which is typical for this series, made it evident that if the records were to be used it must be on a somewhat different basis from that in the previous research, as we were dealing with many differ- ent amplitudes of movement. In the six illustrated records from this series of experiments (See B to (?) it is plain that the first movement to the left is usually of fairly regular amplitude. If we use record a as a standard for amplitude, measurement shows that the amplitude in 1 The first of the record is slightly complicated, as it was re-exposed and shows part of a second series of movements. See also Dodge and Benedict, p. 154, fig. 27, for another typical record. 2 The corrections were measured in millimeters, converted into degrees, and recorded in the tables of Dodge and Benedict's report. Eye Movement. 81 the first left movement of the others is about 40° (usually a little less). The records in the figure are typical of the availability of this first left movement throughout the series. The first movement to the right which succeeded this, while often definitely less than the 40° amplitude and without correction, was also in 58 per cent of the cases accompanied by a wink, 1 which produces a break in the horizontal line of time dots. Since the irregularity beyond the first movement to the left is so prominent, it appears that the only feasible use that can be made of the records, aside from this one left movement, is to measure the total amount of horizontal movement during the 5 seconds and employ this as a general measurement of the speed of eye movements. Large un- corrected movements will of course provide the maximum horizontal distance for a given time interval, as short movements require propor- tionately more time for their performance. 2 Table 14 contains the data for the two points considered : (1) the speed of the first movement to the left (L) for each record, 3 and (2) the total horizontal distance (H. D.) covered by the eye during the period of 5 seconds. The average duration of the left movement on normal days following the control dose is 142 l-4 enee Tjl t+i CO ^ VICIOUS t-H t-I FM T-l rH tH tH HHHrt 44 + + + + 44 Q 1 o© i-t iH CO OB o oo ieuj OS OS (NtN i-t OS^tJ* 1-4 eo T3 O 'S W g OW OOOO «)ooo o o o© oo 3 © ioco^^t iQwioua W I-H F-t «H w* t-H i-HtHfHi-H HHHp" 1 + ++ + 44 1 *S o j ■*«© CO 00 t*U9 CO O CO O CI CC U5 Nt- S^SSPt* ® ® OM S^-h con U3 CO COt* Iflrt P3 •a o Hi gwNMW NWHW "ill I I i I 1 1 •g •^ ft ►4 bf)0»OM 00>ON e£lQ oco O O >OtN oeo w O w iOCO-*}< Tt< iC <© US l-H »-H i-H^4 i— 1 i— t ^H i — 1 i — I i — ( ^-t + 4 4444 44 !-s Q ■^lOOM lOOOOOO goOOCDOO «^0« IO WO |>(M i^eH CM T-l a> a ci S^CO SOl-l lOIQOt" CM Bq g o W r CJ CO CI ^ (N CX CI CN ~4 1 1 fH 1 1 1 1 1 1 1 eo •c S3 as ft h4 OOOfr* OOiOO b co *o « eo ^t* 3 91 1 as © a 8 a 1 e3 a 1 o 1 I n V o s to 6 5 "3 en rt co rf"3 © n -* 5 SgHB D ad«* § a ° s u liSOft „T3« j T3CO | ■g §3-3 ,S5>ihii o^bh s^i 1 ;! 1 ; I s5 ^ Mz 5» V OS » CQ to 03 flora Odd II § 'SI °.s P „ as cs 5 J> J § «-o ■* m 9 °J g § a^a d fi t> * -S-a 43 -*» -*3 ■Sen •o §-: •sis n?" §^ g fl O CJT3 g I II •§a B m 03 cr3 3 -^ 0+s«h tA j0 m *> M« n'C & a-2 ft • S » S JJJ3 ■ ■^■8 3+3 s > • o 5 5 2 o 3 3 o oj g m +3 to >-s ■? 3 ■^•o' fl r?S3 a ■" a 3 Sj Is S 3 IB g •S S * O -J ° °^ » s 5 a q, in ■SlB 3 « 5 ? ■ 3 5f5a^ °>a o eo so >> Ogni) *J,f3 o|gf ^ T3 m o 3.2^3 ■« S a a s* s — ^■a a ° -*3 O R 3 1-3 03 o ^1 • ■a— i u; "Bill £ M S 3 O 3 H ^> ■JS 3 s q ■a ■8 a ID :>> § [> ,►, !> -j 3 a JO s* a s S - WHS Eye Movement. 83 respectively. There is no standard with which we can conveniently compare these values, as in the previous series the duration of the test was not always exactly 5 seconds; hence it was inadvisable to measure the records for this factor. The effect of alcohol upon the eye movements is strikingly evident in table 14. The average duration of the left movement on all the alco- hol days and periods following the ingestion of the alcohol is 151 s =3 ! 3 I 6 — «i~00>COr>>«>'io^*)o* — O 0» CO K (0 lO oodff)Oia>a»c5a>c3ciaSo> -o as a^ « o' "* "" ~ 2 u \ ©V*Uj»aoui J3 6uaio jeaiunM Finger Movement. 89 second to second, and the curve for each 2-second interval tends to main- tain its relative level to the curves for the other 2-second intervals. Curves 1 and 2 cross at period 6, and curves 3 and 4 at period 3. After period 1 the general configuration of the two groups of curves are opposed in that the tendency is for one to rise where the other falls, and vice versa. The broken-line curve shows the average for each group. For the non-alcohol days the average performance in period 2 was not inferior to that of period 1. In periods 3 and 4 there was a definite increase, while period 5 is practically at the level of 1 and 2; period 6 is somewhat below the average number of oscillations in the other periods. In con- Experimental periods after dose 6 1.8 t.6 1.4 I.Z 1.0 V) O .8 it §--4 "°^ , 4) 3 I E~ °« — u. I- \.z 2.4p 6 ""* > *-^ ^^O 4" ^-^^ ^/ Ov T - — ^"">« 'ffls>>\ S \* > y// v/7 *vf Normal differences—^ ' Alcohol " P / « f 6 1.6 4" 6" 2* 8' Fig. 10. — Differences for finger oscillations in 2, 4, 6, and 8 second intervals for eaon period after the control and alcohol doses. 90 Effect of Alcohol on Psycho-Physiological Functions. trast, the average for the alcohol curves shows period 2 as clearly lower than period 1, while 3 and 4 tend to a progressive rise, but not quite to the level of period 1. The curve drops somewhat at period 5, particu- larly in the third and fourth 2-second intervals, but period 6 shows decidedly the best average performance for the alcohol days, and reaches a level which is not equaled by any period in the average for the non-alcohol days. All of the alcohol curves show a drop between periods 1 and 2, but this is much more prominent in the first 2-second interval, indicating that the alcohol had a particularly depressant effect upon the beginning of action. The differences between the periods before and after the control mix- ture or the alcohol are given in section n of table 15. Computation has been made for records of 2, 4, 6, and 8 seconds, as in the presenta- tion of data by Dodge and Benedict. An examination of these values shows that following the control dose the differences are quite regularly positive, while after the alcohol dose they are more generally negative. The course of the normal differences is shown by figure 10 1 (solid dots). As would be expected from A in figure 9, the differences are almost zero in period 2, as the performance in this period was almost identical with that in period 1 in the normal curves. In period 3 they are markedly plus, becoming continuously smaller but without change of sign in periods 4 and 5 and with a return to practically zero in period 6. The course of the alcohol differences (fig. 10, open dots) are in sharp contrast with the normal series. Period 2 is far below the base line; periods 3 and 4 show a relatively large rise over period 2, but still remain minus and are decidedly different from periods 3 and 4 on the non-alcohol days; period 5 is distinctly minus for the 6 and 8 second curves; period 6 rose far above the line. While the effect of alcohol is evident in these curves to give it a con- crete numerical statement requires an expression for the differences between the course of the two groups of curves, i. e., a statement of the true difference between the results obtained on the non-alcohol and alcohol days. This has been computed for the full 8-second performance and is presented in table 16. The differences between the values before and after the control dose or the alcohol dose are obtained in the usual manner from section n of table 15, these differences for the respective periods being —1.46, —1.50, —1.15, —1.30, and +2.65 complete oscillations. The percentile differences are obtained by dividing these differences by the average of the relevant normals, i. e., the average of the two average values for period 1 in the normal series (37.28) and in the alcohol group (37.70). Using the average (37.5), we find that the alcohol depression amounts to about 4 per cent in periods 2 and 3 and 1 Figure 10 has been drawn for 2, 4, 6, and 8 second differences, i. e., the 8-second differences in a way include those for 6, 4, and 2 seconds, and the 6 seconds include the 4 seconds and 2 seconds, etc. In A and B of figure 9, on the contrary, each" curve is for the performance of a separate 2- second interval. Finger Movement. 91 changes to approximately 3.3 per cent in periods 4 and 5. Apparently there was a facilitation in period 6, amounting to 7.1 per cent. 1 Table 16. — Effect of alcohol upon rapidity of finger movements during a performance of 8 seconds. Period 2. Period 3. Period 4. Period 5. Period 6. -1.46 -3.9 1.80 0.886 -1.50 -4.0 2.63 0.961 -1.15 -3.1 1.62 0.862 -1.30 -3.5 1.84 0.891 +2.65 +7.1 3.20 0.984 Differences divided by P.E.d Since the finger oscillations of a record were not measured indi- vidually, no mean variation was obtainable within the period. The M. V. placed in table 15 refers to the variations between the two records of the same day and period and between the records of the different days. Thus, for all the normal days taken together, there were six 8- second records which came in period 1 ; the average of these is 37.28 and their M. V. is 0.8. The mean variations are smaller on normal than on alcohol days. On the former they average 0.9, which is 2.4 per cent of the average (37.9) of all the 8-second records following the control dose. For alcohol the mean variations average 1.7 or 4.5 per cent of the average 8-second score (37.4) following the 30 c.c. alcohol dose. The smallness of the mean variations between records of homologous periods is noteworthy, indicating a very even performance from day to day. The probable correctness figures for the differences have been computed from the above-described mean variations and necessarily are quite small, since the maximum number of cases could not be more than six. Comparison of these data with those previously obtained with the same subject and on other normal subjects by Dodge and Benedict (see their report, p. 182, table 30, for 6 seconds) reveals that the aver- age number of complete finger oscillations in 6 seconds previously scored (27.8) is about the same as in these later results, to wit, 28.8 and 28.4. According to the data obtained by Dodge and Benedict, this subject was very slow, as he performed in 6 seconds on the average about 8 complete oscillations less than the average performance for their normal group. As a result of alcohol (dose A), he shows a de- pression for 6 seconds amounting to 15.2 per cent. (See Dodge and 1 From the course of the curves for alcohol days it would be interpreted offhand that the alcohol depression was decidedly most prominent in period 2. It should be remembered, however, that the effect of alcohol is a statement of the contrast between what occurred on alcohol days and what occurred on non-alcohol days, rather than simply what occurred on successive periods of alcohol days. For example, in the data under discussion, while it is apparent that the depression of alcohol was prominent in period 2, yet it is as evident that the difference in the level between the average for alcohol, period 3, and that for normal days, period 3, is as great as the difference for period 2, and hence the effect of alcohol in the final statement appears as large. 92 Effect of Alcohol on Psycho-Physiological Functions. Benedict's report, table 33, p. 185.) This was the largest depression but one (that for Subject IX) found with the normal group, which had an average depression with dose A of 8.9 per cent. Both of these figures are larger than the percentile effects given in our table 15 for 8 seconds. They contain no contradiction, however, as to the nature of the alcohol effect. Briefly, the results of this test of motor ability are as follows: (1) The dose of 30 c.c. of alcohol with this subject reduced the num- ber of free finger oscillations which the subject could perform in a period of 8 seconds by 3 or 4 per cent, roughly according to the time that had elapsed since the ingestion of the dose. (2) The alcohol effect was more pronounced in the first record of a period than in the second record; it was more pronounced in the per- formance during the first 2-second interval than in any later similar interval (except in period 5) ; both of these results tend to reduce some- what the falling off by fatigue after alcohol, and show the effect of al- cohol to be stronger on the initial performance. (3) The variability in the performance after alcohol is larger than after the control dose, this being shown by an increase in the mean variation after alcohol. PULSE AND RESPIRATION. TETANUS PULSE. It was mentioned in a previous paragraph (see p. 24) that 11 pulse records were taken in each period. Of this number, 9 were under experimental conditions which favored more or less relaxation of the subject as he reclined in a steamer chair, or under conditions of mod- erate mental and physical activity, as, for example, during the word reactions. It seemed desirable, also, to measure the effect of alcohol upon the pulse before, during, and immediately following general mus- cular tension. To this end, and with the purpose of providing condi- tions which would test the promptness of changes in heart rate, con- tinuous pulse records were taken in which the subject was (1) relaxed in the steamer chair, (2) holding the muscles of the limbs and trunk rigid for a period of 5 seconds, and (3) relaxed following the period of exertion. Dodge and Benedict (see p. 15 of their report, series 3) took pulse records of the subject immediately after standing and after 60 seconds of standing, after two double genuflections, and after 60 seconds of quiet. In comparison with this it will be seen that our pulse records are absolutely continuous. Studies on the effect of vigorous muscular exercise on the heart rate of man have usually been limited to the pulse rate before and after the activity. The most notable exception previous to the last decade is the excellent contribution of Bowen, 1 who, with a specially designed tambour 1 Bowen, Contributions to Medical Research dedicated to Victor Clarence Vaughn, June 1903, p. 462, Ann Arbor, Mich. Pulse and Respiration. 93 arrangement, was able to secure graphic records of the pulse from the carotid of subjects working vigorously on a bicycle ergometer. The curves could be read for the duration of individual pulse cycles. To minimize the disturbances from movements of the head and neck it was usually necessary for the subject to rest the forehead and shoulders against padded supports. The periods of work varied from one minute to one hour and a quarter. Among later researches which have employed sphygmographic tech- nique, those of Aulo 1 may be mentioned. Radial pulse was taken from the right arm with an arrangement similar to that described by Tiger- stedt 2 and used later by Dodge and Benedict 3 for their "association pulse." The subject reclined in an easy chair or on a bed and, when told, actively produced tension in all the bodily muscles, excepting that the right arm was kept as relaxed and quiet as possible, as this was essential for legible records. No periods of activity of less than a minute are reported. Fortunately, with much less limitation of the subject, electro-cardio- grams giving a photographic record for accurate determination of heart rate can be taken under conditions of physical activity, and there are no great complications caused by the vibration of the apparatus or the movements of the subject. Miss Buchanan 4 took continuous electro-cardiograms showing the pulse rate during moments of change from rest to muscular tension. The subject sat with one hand and one foot in vessels of salt solution connected with the terminals of a capil- lary electrometer. Upon hearing a given signal he clenched the fist that was free or gripped a hand dynamometer. The pulse-cycles were measured individually in hundredths of a second and an account was taken of the relation of the signal to the systole. Later the same author (Buchanan) 5 took electro-cardiograms from a subject who was riding on a stationary tricycle. No details of the application of the leading-off electrodes are given. Benedict and Murschhauser 6 report: "By using body-leads and the Bock-Thoma oscillograph or the Ein- thoven string galvanometer, we were able to secure graphic records of the pulse rate of the subject while he was walking upon the treadmill." The treadmill was operated by a built-in 220 v., d. c. motor. As would naturally be expected with so sensitive an instrument as the string galvanometer, much difficulty was experienced from leakage currents and static charges which found their way into the galvanometer cir- cuit. After a special metallic brush, which was earthed, was arranged 1 Aulo, Skand. Arehiv f. Physiol., 1909, 21, p. 146; 1911, 25, p. 347. 2 Tigerstedt, Hygiea, Festband, 1908. ' Dodge and Benedict's report, p. 192. * Buchanan, Trans. Oxford Univ. Junior Scientific Club, n. s., No. 34, 1909, p. 351, reprintedin The Smithsonian Report for 1910, pp. 487-505; Science Progress No. 17, July 1910, p. 60. s Reported by Krogh and Lindhard, Journ. Physiol., 1913, 47 (see p. 117). 6 Benedict and Murschhauser, Energy Transformations during Horizontal Walking, Carnegie Inst. Wash. Pub. No. 231, 1915, p. 31. 94 Effect of Alcohol on Psycho-Physiological Functions. to drag over the surface of the revolving belt, satisfactory photographic records were obtained, although but few such records are reported in their data. Professor H. Monmouth Smith, of the Nutrition Labora- tory, has made extensive use of this method and has accumulated a large amount of data on the pulse rate during walking. Dodge and Benedict 1 called attention to the unique advantage of the body-leads for leading-off electrodes in taking electro-cardiograms during muscular activity. Since these are "situated directly over relatively small trunk-muscles, even violent activity need not interfere with the rec- ords." Recently Krogh and Lindhard, 2 assisted by Fridericia and Westerlund, made some determinations of the pulse rate while the sub- ject was working on an ergometer, recording the pulse by means of a string galvanometer. For electrodes wet bandages and thin copper wire were wrapped about the wrists. Our own experimental conditions may be more clearly understood in connection with a description of the illustrated records (see fig. 11). After the finger movements the subject leaned back in the steamer chair and was told to relax as much as possible. No adjustments were necessary, as he was already connected to the string galvanometer for pulse observations and also with the apparatus for recording the respiration rate. After approximately 60 seconds, or longer if the operator saw by the deflections of the string that the subject's pulse had not become regular, the record was begun. (See left-hand end of record, A in figure ll.) 3 After 8 or 10 seconds the signal "go" was given forcefully by the operator. At this signal the subject volun- tarily contracted the muscles of arms, legs, and trunk, and retained this condition of tension until the signal "stop," at which he fell back in the chair and remained quiet. The beginning of voluntary contraction (see the vertical line drawn near the left end of the record) and the interval of its duration are clearly shown. The action currents asso- ciated with the general muscle contraction produced in the pulse tracing a prominent deflection of the galvanometer record. Further- more, the setting of abdominal muscles caused a marked disturbance in the respiration curve. The pulse curve was continuously disturbed during the interval of tetanus by the action current from contracted muscles. However, the sharp spike of the R wave 4 of the electro- cardiogram complex is clearly visible. The cessation of contraction is 1 See Dodge and Benedict's report, p. 193. ' Krogh and Lindhard, Journ. Physiol., 1917, 51, p. 182, see p. 186. 8 The camera had been in operation for some seconds previous to the beginning of the photo- graphic record in order that the actual commencement of the record should not attract the atten- tion of the subject, who was within a few feet of the apparatus. (See Dodge and Benedict's report, fig. 1, position 1, p. 31.) 4 Standard electro-cardiograms are always read from left to right. The deflections which are above the base line (positive variations) have by Einthoven been termed P, R, T, and V. The sharp R wave very slightly precedes active ventricular contraction. For fuller explanation of terminology, time relations, and interpretation of the electro-cardiogram waves, see Wiggers, Circulation in health and disease, Philadelphia, 1915. Pulse and Respiration. 95 not so clearly marked as is the beginning, partly because the contrac- tion gradually decreased in vigor from second to second and the fall due to the final relaxation was not so pronounced as was the rise at the start. The period of muscular tension (tetanus) was approximately 5 seconds in duration. A clock beating seconds was in front of the oper- ator, and by it the signals were given. Following the tetanus, the pulse tracings became regular, all waves of the electro-cardiogram being dis- cernible, and the record was continued for 20 to 25 seconds. We thus have a continuous record approximately 35 to 40 seconds in length, which makes possible the careful measurement of individual pulse cycles during the following conditions: (1) subject comfortably resting; (2) transition to severe muscular tension; (3) a section during which muscular tension is continued; (4) transition to rest; and (5) a period of recuperation. 1 It is of course recognized that with condensers in series with the sub- ject there is an alteration in the form of the waves of the electro- cardiogram. The R wave is more prominent, while P and T are rela- tively reduced. The condensers are necessary, however, to minimize fluctuations of the base line, and these curves have been used solely for determination of heart rate . With the technique as outlined, the P wave is not always visible in the record; it was therefore not practica- ble to determine the length of pulse cycles by measuring from P to P. As the peak of the R wave is always clearly marked in all the records, it seems most practical to measure from one R peak to the next. Tech- nically speaking, we are thus not measuring isolated individual pulse cycles, but the duration obtained would represent the R-P of one pulse cycle and the P-R of the next. Probably for our simple measurements of rate this method does no great violence to the data, especially in consideration of the fact that the duration between R and R can be determined much more exactly than that from P to P. Pulse records such as those under discussion and illustrated in figure 11, A and B, are naturally considered in three divisions, which we have termed for convenience (from left to right) pre-tetanus, tetanus, and post-tetanus. As previously stated, it was not difficult in practice to set the limit for the different parts, because of the disturbance in the pulse and respiration tracings occasioned by the voluntary muscle con- traction, although it was more difficult to mark the end of the tetanus than its beginning. Each pulse cycle (from R to R, as previously explained) was measured in tenths of millimeters. This distance, divided by that which represents the corresponding second or seconds of time on the time line made by a Jaquet clock below the pulse curve, gives the length of the individual pulse cycle expressed in hundredths of 1 In figure 31, p. 201 of Dodge and Benedict's report, it will be observed that frequently the pulse at the end of post-stimulation had not reached the same level as that at the beginning of pre-stimulation. For comments explaining this, see Appendix IV, p. 143, of this monograph. 96 Effect of Alcohol on Psycho-Physiological Functions. a second. For example, if the distance from jB to R is found to be 12.8 mm. and the distance on the time line for the corresponding second, that is, the second directly below the pulse cycle under measurement, is 15 mm., then the duration of the pulse cycle is 12.8 divided by 15, or 0.85 second; in terms of pulse rate this would be 70 beats per minute. 1 Obviously the longer the duration of the individual pulse cycle, the slower the rate per minute will be. A duration of 1.00 second equals a rate of 60 per minute. A duration of 1.1 seconds equals 55 per minute, and so on. For the convenience of the reader the curves which follow will be accompanied by a column at the right which will translate the pulse duration into terms of rate per minute. A pulse cycle which fell partly in two divisions of the record was counted as belonging to that one of the divisions — pre-tetanus, tetanus, or post-tetanus — which contained the greater part of its duration. A comparison of the pulse cycles in the pre-tetanus and post-tetanus divisions which are indicated in A and B, figure 11, shows clearly that 1 While it is more common to present pulse data in terms of rate per minute, when considering individual pulse cycles as in these experiments such terminology seems misleading. Fig. 11. — Pulse and Finger-movement Records. A and B. July 2, period 1, No. 11, and June 29, period 2, No. 10, tetanus pulse records to be read from left to right. In A the white vertical lines have been drawn in to make clear the divi- sions of the record and the different sections have been marked. From the left to the first line is the pre-tetanus section, during which the subject was resting; between the vertical lines occurs the tetanus section, when the muscles were voluntarily tensed; from the second line outward to the right-hand end the subject was recuperating after the activity; this section is designated post- tetanus. From the point in the record where activity began (left vertical line) six of the pre-* tetanus pulse cycles were numbered and measured. Also, from here proceeding to the right the tetanus cycles are numbered and read until reaching the point where the muscles relax. The post- tetanus cycles are numbered in order and measured from left to right. The P, R, and T waves of the electro-cardiogram have been designated for one pulse cycle in A. It can be observed that the R wave is definitely visible during the period of tetanus. The respiration-recording apparatus was so arranged that the curves rise at inspiration. The time is in seconds. C. Section of post-tetanus record July 2, period 3, No. 11. Not infrequently with Subject VI there were periods of several seconds without any appreciable respiration, as is shown in the left portion of C. When a record 15 seconds in length has no rise and fall in the respiration tracing save that the pulse is shown, it might be argued that the tambour which was placed on the sub- ject was not properly located. This criticism can not apply to figure C. D. June 30, period 2, No. 1, after taking the first alcohol dose and before the patellar-reflex measurement. Two short vertical lines have been drawn through the respiration curve to indi- cate the two points between which the pulse cycles were counted. E and F. June 29, period 3, Nos. 8 and 9. The pulse tracings in these two finger-movement records illustrate the phenomenon described in the text, p. 117. A change in the electrical axis of the heart seems the most plausible explanation for this reduction in the size and occasional disappearance of the usual electrical waves. In E the finger-movement record has been divided into four 2-seeond sections and these have been marked for the number of finger movements occurring in 2, 4, 6, and 8 seconds, respectively. G. July 4, period 6, No. 6, pulse record taken during the last word reactions on the last day. The extra oscillations of the string were caused by induction effects from the electrically driven kymograph and the action of the word-exposure apparatus. The kymograph drum made a com- plete revolution in 5 seconds (see text, p. 25), and at each revolution broke the circuit of the exposure apparatus. The moments when this event occurred are marked by 6 in the picture. The swift downward movement of the respiration curve shortly following every second break indicates the reaction (marked r in the picture) which consisted in speaking the word exposed. H. July 4, period 1, No. 7, pulse following the word reactions. Occasionally the body elec- trodes or the subject came into a position such that the respiration is seen to change the base line of the pulse curve. Observe also the right-hand portion of C in this figure. Missing Page Pulse and Respiration. 97 in the post-tetanus division the shortening of the pulse cycle is due almost entirely, so far as preliminary measurements reveal, to a short- ened duration of the diastole (distance from end of T to beginning of P), as the distance from the beginning of P to the end of T (systole) remains quite constant. It is a well-recognized fact that the diastole is the portion of the pulse cycle that shows the greatest percentage of change in length with the transition from rest to exertion and from exer- tion to rest under conditions of such short periods of activity. In our measurements, however, it was not practicable to begin or end the muscular tension with a fixed relation to diastole. Since the vagus mechanism has a latency in its action, it occasionally happened that tetanus began at such a point during systole or immediately following it that a particular pulse cycle, although counted as belonging to the tetanus division, shows no conspicuous change in the length of its diastole; i. e., the action of the vagus does not appear until the next cycle, which is the second for that section of the record. Illustrations of this are found in table 17, June 29, period 3, record 2, and June 30, period 2, both records. Usually the first cycle in the tetanus division is shorter by several hundredths of a second than that immediately pre- ceding, i. e., the last cycle in the pre-tetanus division. The fact that the P-R time is the latter part of the distance which we are measuring for each pulse cycle must be remembered when considering the data at the transition point in relation to vagus action. In tabulating these data it is not expedient to use the form employed for other measurements, for even the most condensed form of table is necessarily large and unwieldy. Table 17 follows the same general arrangement as the individual record. The left section of the table shows the duration of individual pulse cycles preceding tetanus, the central section the duration of those during tetanus, and the large sec- tion at the right the post-tetanus pulse cycles. The beginning of tetanus is made the starting-point for the measurement of the record. Six pre-tetanus cycles were measured; these from the start of tetanus are numbered in order from right to left. The cycles which came in the tetanus period are numbered from left to right and likewise for the post- tetanus period from the end of tetanus. For the sake of clearness the cycles have been numbered in A of figure 11, the same numbers being given in the headings of the columns in table 17. The pre-tetanus sec- tion of the record contained more than 6 cycles, but the cycles measured immediately preceding the beginning of tetanus were limited to this number. The tetanus occupied approximately 5 seconds; the number of pulse cycles which fell within these limits varied from 6 to 9. Of the 60 records which enter into table 17, 7 contained only 6 pulse cycles in the tetanus section, 10 records contained 9 cycles, and 16 others contained 8 cycles. On July 4, the last alcohol day, the pulse rate tended to be faster, hence during the 5 seconds of tetanus more 98 Effect of Alcohol on Psycho-Physiological Functions. Table 17. — Duration of individual pulse cycles preceding, during, and following tetanus. [Length of pulse cycles given in hundredths of a second.] Kind of experiment, date, and period. Pre-tetanus pulse cycles. Tetanus pulse cycles. 6 5 4 3 2 1 Av. M.V. 1 2 3 4 5 6 7 Av.i M.V. Normal: June 29 — 1 103 98 111 120 109 103 106 105 109 111 103 104 113 89 111 120 121 93 104 107 103 107 93 110 104 103 109 105 112 105 107 103 96 101 100 101 107 94 111 95 103 101 107 108 94 98 93 96 81 107 97 99 98 96 103 97 98 91 97 94 94 104 100 100 100 109 99 99 99 101 110 112 103 108 111 104 108 111 108 114 92 88 100 125 118 93 113 105 108 114 97 112 100 109 112 106 107 106 106 104 111 103 107 107 110 85 108 93 106 102 100 115 91 102 91 100 81 106 97 100 97 96 99 106 94 97 99 100 97 99 107 103 93 106 99 99 100 105 116 119 114 107 111 100 111 107 95 113 110 94 111 128 121 93 110 110 108 112 105 112 106 106 95 112 106 106 109 99 107 97 103 107 116 79 111 101 97 101 107 113 93 100 100 103 87 108 94 97 100 98 104 103 96 97 100 100 102 99 104 94 87 107 99 99 100 97 103 119 106 109 117 104 110 110 97 117 110 96 107 125 123 99 113 111 104 109 111 107 113 110 115 115 90 109 110 98 106 90 100 101 104 87 101 101 99 98 104 113 94 99 113 107 88 99 100 100 100 100 97 100 91 96 94 97 93 106 103 98 89 102 97 98 93 100 103 115 112 112 110 108 110 101 103 115 117 103 113 122 122 105 109 113 111 114 106 118 114 110 115 119 99 112 112 103 110 103 106 103 119 111 104 102 95 105 99 115 95 101 107 106 87 100 103 103 112 102 103 102 95 101 97 100 99 110 105 97 93 102 100 102 95. 103 107 116 111 107 112 112 111 107 100 113 113 100 111 121 127 105 110 112 103 101 96 110 110 104 110 120 117 110 111 94 107 100 100 93 103 84 107 95 100 98 100 106 96 100 99 103 80 100 97 101 101 97 105 94 94 97 97 94 96 106 100 91 91 104 97 97 98 101 108 117 109 108 111 105 110 108 101 113 109 95 109 123 122 98 110 110 106 109 101 111 108 107 109 113 105 108 109 100 106 99 103 102 110 90 107 98 100 101 103 112 94 100 100 103 84 103 98 100 101 98 102 100 95 96 98 97 97 104 103 97 92 105 98 99 3 2 4 2 3 2 2 3 3 3 4 3 6 5 3 2 2 5 2 4 3 4 6 2 4 3 5 5 7 5 4 3 3 4 3 4 5 8 3 3 3 4 3 3 1 1 6 3 3 4 2 1 3 3 2 3 2 2 2 2 2 3 2 3 3 2 2 3 94 80 87 100 85 106 105 89 95 80 97 82 111 82 90 96 109 99 87 94 90 83 80 84 84 82 86 87 92 85 91 90 81 101 102 92 93 84 84 88 83 91 86 85 81 87 87 87 73 90 86 91 88 86 86 79 94 87 94 93 80 81 94 79 81 87 87 88 77 77 79 84 81 78 70 83 79 79 81 79 87 78 85 84 87 78 81 82 80 78 77 79 82 77 79 82 81 80 80 72 86 83 82 73 81 77 77 77 77 78 79 82 76 83 77 82 73 80 79 77 78 78 77 75 81 81 75 78 75 81 82 78 75 77 78 78 76 73 77 56 75 87 89 77 77 76 80 75 80 74 78 80 81 79 75 78 74 75 73 75 74 73 79 78 78 76 77 81 80 77 79 80 77 73 75 72 72 76 75 76 76 79 74 76 73 77 76 77 75 76 75 70 75 75 74 74 74 74 78 72 75 73 74 75 73 72 74 101 75 75 74 73 79 72 73 73 79 77 78 74 77 77 75 76 71 74 73 77 75 73 73 76 74 74 76 74 77 76 79 71 76 72 72 69 74 74 74 73 76 77 73 76 79 75 74 74 74 75 74 69 75 72 72 68 73 77 74 72 69 70 72 74 73 69 73 56 74 77 75 76 72 72 77 73 73 71 75 74 75 74 73 73 69 71 69 77 73 73 71 78 74 74 73 73 73 72 79 70 71 72 74 70 70 72 71 76 74 80 71 71 81 74 73 72 71 74 71 73 69 70 69 73 71 75 75 70 67 67 71 72 73 67 73 88 72 75 71 77 76 77 73 70 73 71 75 76 73 71 72 73 70 69 67 78 73 73 72 83 74 74 74 73 75 72 79 68 78 72 78 71 71 74 70 76 76 79 72 71 73 73 73 72 70 73 71 69 69 69 69 68 72 78 71 68 66 67 70 72 71 67 73 75 75 68 85 75 77 72 85 79 79 71 78 77 71 69 73 77 70 71 55 73 70 74 69 74 70 81 67 7i 77 71 73 73 74 73 77 73 69 73 77 73 75 70 73 70 69 69 68 69 68 68 81 69 85 68 67 71 72 77 72 77 81 77 82 79 80 79 76 80 75 84 75 81 80 83 78 77 79 75 74 73 78 77 74 76 77 78 76 78 76 78 79 83 74 79 74 77 74 74 77 76 78 76 80 75 76 75 78 76 77 75 76 75 72 76 75 75 75 73 78 78 75 72 73 75 76 5 4 4 15 4 8 10 5 8 3 5 3 10 3 5 5 9 6 4 « 6 4 3 2 4 3 5 7 5 4 6 5 4 8 5 S 5 3 3 5 3 5 4 4 1 3 4 5 3 4 3 4 4 3 4 3 7 6 6 6 3 2 6 5 5 6 5 4 2 3 4 Average 2 .. July 1— 1 2 3 4 5 Average 8 .. July 3— 1 2 ,... 3» 4 S Average 2 . . Gen. av. a . Alcohol: June 30— 1 2 3 4 5 Average 2 . . July 2— 1 2 3 4 5 6« Average 2 . . July 4 2 3 4 S 6 Average 2 . . Gen. av. 2 . i Occasionally there were one or two more pulse-cycles in the tetanus period than are recorded in the table (see text). 2 These averages do not include the pulse cycles in the first period, i. e., previous to the control or alcohol dose. 1 Second record in period 3 is missing. * First record in period 6 omitted as a portion of it was illegible. The data given are for the second record. Pulse and Respiration. 99 Table 17 — Duration of individual pulse cycles preceding, during, andfolloiving tetanus. — Continued. Post-tetanus pulse cycles. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 74 83 83 76 74 75 75 75 73 73 73 74 77 80 83 87 88 100 106 105 103 87 86 86 91 85 84 81 78 77 77 78 81 82 82 89 90 81 103 97 106 106 101 86 83 87 86 85 82 77 74 74 76 80 80 79 86 104 106 107 105 108 117 76 79 86 93 89 82 84 82 79 79 77 77 73 75 73 75 73 78 89 97 100 82 77 82 81 76 75 74 75 75 73 71 71 79 78 76 81 81 78 75 75 75 72 69 72 75 79 79 86 iod 99 105 119 112 73 76 76 76 72 71 70 70 68 68 71 68 70 72 83 89 86 94 112 111 109 100 90 90 84 79 79 79 73 71 73 77 76 74 106 94 115 112 112 114 105 85 81 82 84 80 78 77 75 74 73 73 74 74 82 83 94 95 98 105 108 109 86 79 80 80 79 76 76 76 76 74 72 73 76 77 83 85 100 111 108 110 114 81 77 80 80 79 77 79 77 73 73 73 73 73 75 73 80 97 108 99 95 113 74 75 72 76 75 74 70 71 70 70 67 67 68 70 73 73 73 77 85 91 96 79 96 87 87 85 78 80 82 82 83 86 87 81 86 93 87 107 111 113 114 108 78 78 81 78 78 79 75 74 75 74 72 72 72 76 76 93 101 108 114 114 118 86 99 100 89 81 81 85 86 86 93 93 88 98 104 101 110 121 118 121 113 80 86 83 81 87 84 80 80 78 77 75 75 79 80 78 87 96 93 101 100 ioi 88 87 87 87 86 84 85 83 76 77 78 76 76 80 87 87 96 100 110 107 114 84 84 78 81 82 81 75 73 74 75 76 73 76 77 75 81 93 102 107 114 82 78 79 78 75 74 81 85 83 82 81 79 79 78 77 78 78 77 79 82 83 88 98 101 107 108 107 88 86 75 81 81 81 75 76 78 75 74 74 71 71 69 75 74 76 89 106 100 75 75 72 76 78 75 74 74 76 75 74 78 81 84 85 100 106 112 110 104 109 81 79 79 75 73 73 73 69 69 72 76 79 81 77 84 90 91 87 97 108 95 100 90 84 84 87 79 79 78 77 77 80 84 84 95 88 100 97 94 88 115 104 116 90 84 83 80 83 84 77 79 83 82 82 79 80 90 94 92 85 103 111 73 73 76 73 72 71 72 74 71 73 72 72 71 70 72 73 73 73 88 94 'gi 78 82 74 77 77 79 73 73 72 72 68 70 70 71 70 77 84 89 95 101 98 115 99 90 87 88 87 81 76 75 75 77 81 80 87 96 99 99 100 97 96 96 79 73 73 75 75 73 71 71 70 67 69 71 74 76 73 82 91 97 94 106 110 92 84 80 79 79 78 76 74 73 74 75 77 77 79 82 88 90 89 95 104 99 86 83 82 82 80 78 77 76 75 75 75 76 77 81 83 90 94 96 102 107 105 74 72 69 69 69 69 67 69 69 70 68 67 67 67 67 67 67 67 73 80 78 80 77 80 80 75 77 77 75 73 71 73 72 72 72 79 84 81 96 104 99 112 70 71 67 70 70 69 67 69 67 68 67 70 70 73 70 69 72 71 89 81 93 83 80 77 72 73 73 74 69 68 69 66 64 64 64 64 68 79 93 87 106 107 68 68 69 68 67 67 67 67 67 67 67 67 64 66 67 70 72 81 93 87 94 93 84 84 81 77 77 77 74 72 72 72 68 66 68 70 69 69 75 79 86 114 73 72 73 73 71 71 71 68 68 68 67 65 65 68 74 77 90 99 98 108 96 83 81 75 75 72 71 69 68 68 68 65 66 66 68 68 68 74 87 87 98 97 77 79 74 75 76 74 71 74 71 70 71 69 69 69 74 84 84 97 103 80 79 67 68 67 71 71 68 67 66 67 69 66 71 78 85 88 98 101 109 '96 78 77 73 73 72 72 71 70 68 68 68 67 66 68 71 74 77 88 92 96 99 78 78 78 78 74 75 74 74 71 70 70 68 66 66 67 68 71 69 83 99 97 83 83 84 79 79 78 77 72 74 72 72 77 83 95 90 99 104 103 113 110 110 70 71 70 70 68 70 68 68 65 68 66 65 68 65 66 67 70 74 75 83 87 81 86 79 86 87 79 80 79 79 74 73 77 83 83 90 103 104 110 113 104 109 74 74 72 80 73 71 71 71 72 69 71 73 72 72 74 79 93 93 97 104 97 71 74 74 74 73 71 71 71 71 71 71 71 70 70 72 74 73 82 87 78 86 75 81 79 80 79 75 72 72 72 74 73 73 75 80 84 87 93 100 97 87 100 83 80 82 81 81 80 80 79 78 76 72 73 74 84 92 104 101 112 108 110 103 81 78 75 73 71 71 69 68 67 68 70 68 68 70 71 77 73 79 82 95 90 106 97 85 79 77 76 74 72 70 70 68 68 69 69 72 84 89 87 94 92 107 84 76 79 79 77 78 76 75 72 70 70 70 71 70 73 102 102 105 103 81 80 77 78 76 75 73 73 72 71 70 71 72 74 77 84 87 93 95 95 98 74 72 70 70 70 70 71 68 70 70 70 68 67 70 72 71 74 81 94 94 107 75 74 72 72 69 69 70 69 69 66 66 63 63 64 62 64 66 69 72 87 99 72 71 67 69 69 68 68 68 68 65 68 68 67 68 68 68 74 74 74 79 81 75 78 75 74 74 74 70 70 70 67 67 67 67 67 67 73 79 81 93 99 92 69 68 69 69 69 69 68 68 64 64 62 62 62 62 61 62 62 64 68 81 79 75 78 73 73 73 73 73 70 67 68 68 67 64 64 67 73 90 90 97 103 108 69 68 69 70 69 68 69 69 69 68 69 66 65 64 65 65 67 67 73 80 80 81 76 78 75 73 72 67 67 67 64 64 65 65 67 64 64 78 88 88 96 108 68 68 67 68 68 68 68 68 67 67 67 66 64 64 63 64 67 67 73 79 91 82 82 77 78 75 71 71 70 65 65 65 67 64 61 60 60 61 71 91 85 96 72 69 71 70 68 68 68 68 67 67 67 67 67 65 68 68 74 79 79 84 100 71 73 73 73 71 67 65 66 63 67 66 62 62 65 71 76 91 101 95 98 101 7S 73 72 72 71 70 69 68 67 66 66 66 65 65 65 67 74 78 83 88 94 77 77 74 74 73 72 71 70 69 68 68 68 68 69 71 75 79 86 90 93 97 100 Effect of Alcohol on Psycho-Physiological Functions. pulse cycles occurred; usually 9 cycles were obtained on this day. In general the tetanus data beyond 7 cycles are very scattering for the other days; and averages beyond this number would be of little value. In the post-tetanus period, 21 cycles were usually counted, since at about this point the pulse seemed to have returned to near the pre- tetanus level. Two records were accidentally not completed at the time of taking, i e., those of June 29, period 3, record 1, and for July 1, period 5, record 2. The first record on July 2, period 6, was partially illegible and not included; the second record for period 3, July 3, was missing. Each record is represented by a line of values running horizontally through the table. The averages and mean variations are given for both the pre-tetanus and tetanus pulse cycles. For illustration, June 29, period 1, record 1, shows values for the pre-tetanus pulse cycles beginning just before tetanus of 95, 93, 100, 100, 99, and 103, with an average of 98 and a mean variation of 3. The values for the tetanus period for the same record are 94, 77, 76, 73, 73, 73, and 71, with an average of 77 and a mean variation of 5. There were just 7 cycles, hence all are given. Immediately after tetanus, i. e., in the post- tetanus period, we have cycle durations of 74, 83, 83, 76, 74, etc., until we reach 106, 105, and 103. The fluctuations in the pulse-length in the post-tetanus section are so large that averaging is of doubtful value. In this table, as in previous tables, the data for the alcohol and normal days are grouped separately. The line of averages under the records for the individual days include only those periods following the taking of the control or alcohol dose, i. e., all but the values for the first period. The general averages given for the normal and the alcohol days are drawn from the averages of the three normal days and the three alcohol days, respectively. Attention may first be directed to the normal change of the heart rate occasioned by the muscular effort of tetanus. The general aver- ages for homologous pulse cycles on the non-alcohol days have been plotted in the form of a curve and are shown in figure 12. The figure is so drawn that a rise in the curve denotes a change to a shorter pulse cycle (faster rate). In the column at the left durations in hundredths of a second are given, and at the right their approximate equivalent in terms of pulse rate per minute. The curve is divided by vertical lines showing the pre-tetanus, tetanus, and post-tetanus sections. In examining the normal curve, we find that apparently relaxation was not complete previous to the tetanus, as the pulse cycles were con- tinuously lengthening before the point where the signal "go" was given. The average duration for this section is 110, or 55 beats per minute. From a pulse which is relatively relaxed, the transition is rapid after the beginning of muscular work. The first cycle at the beginning of tetanus shows a change from the average level of 110 to a level of 91, or Pulse and Respiration. 101 from 55 to 66 beats per minute. The second cycle reaches 80, the third 77; the rise is then more gradual, being 76, 73, 74, and 74, or a pulse rate of 81 beats per minute. Immediately upon the relaxation of the muscles the cycle lengthens, dropping to 86. From this point during next 8 pulse cycles the duration gradually decreases (shortens) ea 72 76 66 i° -3 hi 1 Li 31 £<* x ■-- o u EI? 89 83 yS ^ u_ a. S" 79 71 92 86 -\— 65 62 60 58 56 54 "Alcoh / 104 108 112 1 \ _ No hiih L" X J 1 i PULSE CYCLE S » 65432 I 1234567 123436789 10 12 14 16 IS 20 Fig. 12. — Average duration of pulse cycles given in hundredths of a second for pre-tetanus, tetanus, and post-tetanus periods on three normal and three alcohol days, after the control or normal dose. until it reaches 75, very nearly the duration in the latter part of the tetanus section (73). It remains at 75 (80 beats per minute) for 3 pulse cycles, then in 9 cycles it decreases to the level of 107 (56 beats per minute) or very nearly the level immediately before tetanus. The curve under discussion is of course based upon the records taken on one subject only, and in the absence of similar data on other sub- jects no exact comparison can be drawn or proof given concerning the accuracy of its representation of the changes in pulse rate occasioned by the type and duration of exercise employed. However, that por- tion of the curve which shows the transition from rest to exertion is believed to be representative, since it is in perfect agreement with the data of Bowen 1 , Buchanan, 2 and Krogh and Lindhard 3 in demonstra- 1 Bowen, Contributions to Medical Research dedicated to Victor Clarence Vaughn, June 1903, p. 462, Ann Arbor, Mich. 2 Buchanan, Trans. Oxford Univ. Junior Scientific Club, n. s., No. 34, 1909, p. 351 ; reprinted in the Smithsonian Report for 1910, pp. 487-505; Science Progress No. 17, July 1910, p. 60. a Reported by Krogh and Lindhard, Journ. Physiol., 1913, 47, p. 117. 102 Effect of Alcohol on Psycho-Physiological Functions. ting that the first full pulse cycle after the beginning of "muscular work" is shortened in duration and that this decrease in pulse-length continues progressively for several cycles following. 1 The dip in the curve immediately after tetanus and the secondary rise which follows have not been observed in any other published data. In the curves of Bowen 2 and Aulo, 3 to quote the former : "Upon the cessation of work the pulse rate falls in a manner precisely similar to its primary rise (at the beginning of work) ; very rapid at first, gradually diminishing in rapid- ity as it continues, and finally returning to the normal rate." In their researches, however, longer periods of work or exertion were employed, the shortest reported being 60 seconds as compared to the 5 or 6 seconds used here. This is not the place to present the data, but some unpub- lished records on other subjects clearly show that after such short peri- ods of muscular tension the pulse rate does not fall in the manner desig- nated by Bowen. The rate remains high for several beats, sometimes exceeding that during the tetanus, and the fall is more gradual than the primary rise, as shown in figure 12. Although no sinus arrhythmia is sensibly present in this subject, doubtless a respiration phase influences to some extent the pulse duration, as there is a natural tendency to exhale immediately following the tetanus. 4 As the pulse is normally retarded during expiration, this would account perhaps in slight part for the drop in the curve immediately following tetanus, but occasionally, at this point, a particularly long cycle would occur (as illustration see A, fig. 11). Extra systoles were never found present with this subject. The curve for the average of all homologous pulse cycles measured after the ingestion of alcohol (see the broken line in fig. 12) follows, in general, the same form as that for normal days. It is, however, at every point on a higher level than the normal curve, conspicuously so in the periods for pre-tetanus and post-tetanus, the latter period showing a pulse rate which is well above that during the tetanus section, 88 beats as compared to 83 beats per minute. Although both curves are from data which do not include the first-period values and are therefore wholly comparable as alcohol and non-alcohol averages, the difference in level of the two curves must not be uncritically accepted as entirely due to the alcohol ingested. Averages, although seemingly comparable, are sometimes misleading; the data should be considered by days and by periods, the difference being noted between the normal of the day and the succeeding experimental hours. In the first place, we need to know if the difference in level between the normal and alcohol curves shown in figure 12 is due to the influence 1 Bowen records that in one case the pulse rate steadily continued to increase for a period of 4 minutes after starting vigorous work on a bicycle. 2 Bowen, Contributions to Medical Research dedicated to Victor Clarence Vaughn, June 1903, p. 462, Ann Arbor, Mich. 8 Aulo, Skand. Archiv f. Physiol., 1911, 25, p. 347. 4 See figure 11, p. 96; a rise in the respiration curve denotes expiration. See also Appendix IV, p. 143. Pulse and Respiration. 103 of any one experimental day. In figure 13 a curve is shown for each of the six days, the average for the day not including period 1. Light lines and heavy lines have been used to differentiate between the alcohol and non-alcohol days. It is apparent at once, in the pre-tetanus sec- tions of the curve, that the light-line curves for the alcohol days are grouped at a higher level than the three heavy-line or normal curves. In the tetanus section this grouping does not persist and there is great Fig. 13. — Average duration of pulse cycles given in hundredths of a second for three normal and three alcohol days during pre-tetanus, tetanus, and post-tetanus periods, after the control or alcohol dose. confusion in the crossing and recrossing of the different curves, although the alcohol curves tend to remain above the normal curves. In the post-tetanus section there is more scattering, the alcohol curves par- ticularly being less closely grouped, but always at a higher level than the normal or heavy curves throughout nearly the entire period repre- sented. In one instance the curve for the second alcohol day crosses the curve for the third normal day near the latter portion of the post- tetanus division. The curve for the last alcohol day, July 4, reaches the highest level in the post-tetanus division and is also, on the average, the highest in the tetanus section; furthermore, it shows the smallest drop immediately after the tetanus. From this diagram it is certain that the relations shown in figure 12 are not due to the fact that any one day exercises a predominating influence upon the shape of the curves. If we selected anyone of the alcohol days and anyone of the normal days 104 Effect of Alcohol on Psycho-Physiological Functions. and plotted the two curves, the form and relationship of the curves would be substantially like those of the curves in figure 12. The question which must be answered next concerns the first period for the alcohol days. As the results are obtained before the alcohol is taken, the data are normal, i. e., uninfluenced by alcohol. The natural query would be: Are the changes in pulse rate during the first period of alcohol days different from the changes during the first period of normal days? The answer to this question will be found in figure 14. Here, as 65432 I I 2345671 23456789 10 12 14 16 18 20 Fig. 14. — Duration of pulse cycles given in hundredths of » second in first, second, and third periods on normal and alcohol days in pre-tetanus, tetanus, and post-tetanus divisions. in the preceding figure, light and heavy lines have been used to repre- sent the data for alcohol and non-alcohol days, respectively. To sim- plify the figure only 6 curves are incorporated, these representing the averages for periods 1, 2, and 3 of the alcohol days and the same periods for the non-alcohol days. The two continuous-line curves, one light and the other heavy, for the first periods of the alcohol and normal days, respectively, both represent normal data, since in each case the records were taken preceding control or alcohol doses; these normal curves must be given the first attention. In the pre-tetanus section these curves lie very close together — closer in fact than any other two curves of the six. The tendency for the same two curves to fall on identical points is also prominent in the tetanus division of figure 14 Pulse and Respiration. 105 Immediately after the tetanus they separate and maintain their respec- tive levels throughout the period of recuperation. In examining these two curves for period 1 it should be remembered that the terms alcohol and normal are purely arbitrary and used simply for convenience in designating the two groups of days. In reality each curve represents values as normal as the other, as both represent data obtained during the first experimental hour before either the control or the alcohol dose had been given. No complete explanation is at hand for the fact that the two curves followed a nearly identical course during pre-tetanus and tetanus and then changed to such definitely different levels in the process of recu- peration. Before such phenomena can be properly discussed, doubt- less much more normal data must be obtained. The physical exercise which was chosen in this measurement to act as the stimulation to heart rate is not one that is susceptible of careful standardization. The chief thing which recommended it was the rapidity with which rest, muscular tension, and relaxation could follow each other. The transi- tional interval from rest to tension would naturally be much longer if the subject were required to rise, stand, and then perform genuflections or other vigorous movements; the transition to relaxation would also be lengthened. While voluntary stiffening of the muscles of limbs and trunk at a given signal is ideal from the standpoint of a quick transition from rest to activity, and then to rest, the intensity of activity or mus- cle tension is largely dependent upon the subject. The instruction to "be vigorous" avails, of course, only in a general way and does not pro- duce uniformity. The experimenter did not observe with certainty that the tetanus was more vigorous on certain days than on others, but noted that it tended to decrease somewhat from second to second during the short interval which was used. There were times when the subject seemed to make a fresh effort during the last one or two seconds of the interval as if he himself recognized that he had relaxed his muscles. Some slight indication of the vigor of tetanus is provided in the action- current curves which are shown along with the electro-cardiograms. These can be clearly seen in figure 11, A and B, and are usually fairly prominent. Later trials with this same measurement on another sub- ject, however, have shown that the pulse-rate in the post-tetanus part of a record is largely influenced by the vigor of the muscle tension dur- ing the 5-second interval. If the tension is very vigorous the rate in the recuperation period will reach a point equaling or surpassing that during the tetanus; that is, the condition will be just the same as was found for Subject VI in the pre-alcohol periods of the alcohol days. It thus appears probable that the subject, guessing that alcohol was to be given, was incited by the fact to more severe effort. That he should be more vigorous on July 4 is only to be expected from the general spurt which obtained on that day. 106 Effect of Alcohol on Psycho-Physiological Functions. Notwithstanding the fact that the considerable difference in level of the two curves in figure 12 and of the two groups in figure 13, particu- larly for the post-tetanus section, can not be considered as entirely due to alcohol, since the first period for normal and alcohol days also shows a difference in level, the alcohol did produce an effect upon the tetanus pulse. In figure 14, while the course of the curves is somewhat irreg- ular, it is clear that the curves for periods 2 and 3 of the normal days in the pre-tetanus, tetanus, and post-tetanus sections are, in general, below the curve for the first period of normal days. This means that the course of the pulse on normal days tended towards a retardation following the first period, like that found by Dodge and Benedict for the group of normal subjects, particularly on the pulse during the associa- tion experiments. The condition after alcohol, as represented in figure 14, is in direct contrast to this. In the pre-tetanus division the first- period curve, normal, for the alcohol days occupies the lowest level; that is, the curves for periods 2 and 3 do not touch any lower point. The curve for period 2 is decidedly above, showing an average duration of 98 as compared with 104 for period 1. In the tetanus division of the figure, the curve for period 2 is below 1, showing a less rapid pulse during; the activity. Following the muscle tension there is but a slight drop; the curve crosses that for period 1 and remains above it during the post-tetanus division. The curve for period 3 of the alcohol days tends; to be above that for period 1 in every section, thus showing a more rapid 1 pulse following the alcohol, a condition of definite contrast with the normal days, when the heart rate was slower after period 1, notwith- standing the fact that there was a preliminary difference in level be- tween the two groups of days in the post-tetanus section. A summary of the durations of the pulse cycles by periods for both the normal and alcohol days is given in table 18, each pulse-length rep- resenting the average of the corresponding pulse cycles shown in table 17 for the respective periods on the normal and alcohol days. The differences, calculated according to the usual method, are likewise in- cluded in the table and the effect of alcohol on all periods taken together is given in differences and per cent. It will be seen at a glance that the differences for normal days are predominatingly plus, while those for the alcohol days are, as a rule, minus. This is a restatement of what was indicated in the curves in figure 14, i. e., that the pulse-length increased after the first period on normal days and decreased after the first period on alcohol days. The average normal differences deducted from those for alcohol show the contrast or general alcohol effect. At every point (see bottom of table 18) in the pre-tetanus and tetanus divisions and with the single exception of the eighteenth pulse cycle in the post-tetanus division, the sign is minus, indicating a shorter pulse cycle (faster pulse rate) after the alcohol. The effect is most prominent in the pre-tetanus division. Here the average for all the measured Pvlse and Respiration. 107 pulse cycles preceding tetanus shows a decrease of 9 per cent in cycle length, while for the tetanus division the decrease is 2.9 per cent. The post-tetanus division, which includes 21 cycles for comparison, as the pre-tetanus contains 6 and the tetanus 7 cycles, may be averaged in blocks of 7 cycles. Beginning at the left, the decrease in length for 1 to 7 inclusive is —2.4, 8 to 14, —3.5, and 15 to 21, —1.4 per cent. To provide a convenient numerical 'statement for the effect of alcohol by periods on the tetanus pulse, table 19 has been compiled. Here the effect of alcohol is given not on the individual homologous pulse cycles, but the differences have been averaged together to obtain a statement for the pre-tetanus, tetanus, and post-tetanus divisions of the record. This has been done for each experimental hour following the ingestion of alcohol. The method of compiling table 19 may be illustrated as follows: In table 18 the differences for the first period of normal days in the pre-tetanus division are +4, —1, +8, +8, +8, and +8; the average of these differences is +6. The corresponding differences for the foist period on alcohol days in the same division are shown as —5, —7, — 7, —8, —5, and —3; their average is —6. From this latter average we subtract +6 for the normal days and obtain the difference of —12 as given in table 19. It will be seen, therefore, that the effect of alcohol in period 2 was to reduce the pulse length in the pre-tetanus division by an average of 0.12". In the tetanus division there was a slight increase in the length of 0.01", and in the post-tetanus division a decrease which was also slight, 0.02". In periods 3, 4, and 5 a clear tendency is shown for the pulse-length to decrease, this of course meaning a faster pulse rate as a result of the taking of alcohol. These values are given in terms of percentages in the right-hand columns of table 19. The values for the relevant normal were obtained by averaging the pulse- lengths in period 1 of the pre-tetanus, tetanus, and post-tetanus sec- tions of table 18. For example, it will be seen from table 18, period 1, normal, that the relevant normal for the pre-tetanus division may be obtained by adding 103.8 (the average of 104, 107, 104, 103, 104, and 101) to 103.7 (the average of the similar values for the first period on the alcohol days, that is, 102, 106, 105, 103, 105, and 101) and dividing the total by 2, giving an average relevant normal for the pre-tetanus division of 103.7. Using this average as a basal value, we may compute the percentile effect of alcohol from the values for differences given in table 19, the percentages for the pre-tetanus division being —11.6, —3.9, —14.5, and —5.8 for periods 2, 3, 4, and 5, respectively. By way of summary, therefore, it may be said that in this measure- ment, under the conditions employed, a dose of 30 c.c. of absolute alco- hol produces not only relatively but absolutely a positive rise in pulse rate per minute. The increase in rate is particularly prominent pre- vious to muscular tension. During the period of tension the pulse fails 108 I Effect of Alcohol on Psycho-Physiological Functions. I ■a § si- ll* 8 S"2 S .3 S ? s e -q ft. is f a ■« 3 s X t» HMMIOO CNt-^CNM HC0O)-*^« NNMW«f» CMN t- t» t- t- CO O t> CO l> IN CC +++ 1 + + i ++ i + ii «N CO HlOMTflO '^ t» t» t- t* t- fr» w ec CO + + + + + + 1 + 1 1 1 1 oo lO CM O ■* T(l lO ClNMCOi- COT** rH ^f (M 0C HlMHHlOf" CNN t» t- t- t» l> t^. t* f- t* l> cC s? as 3 1 + + + + + 1 + 1 1 1 1 1 *4 •* (N OS CO ■* iO r^-^cNCOM COCOM WMH COOMOflrt OJ^C in t- o t- t- t- i> t- t> b- b- i> +++++ + + 1 + II l» i> t» o b- 1 ++++ + 1 1 1 1 1 1 US IN OIhhOO wcohhn + + + + + + 1 1 1 1 CM l-t l> rH OS ■* rH TjINb-^^J T* CN rH CM b- 10 00 OS OS CO 00 X aSNMHM *H *H CO OS X OS OS +++++ ++ 1 +++ 1 1 DO ih OS oo -* « xt>cot>»e © © © rH rH rH rH rH HUX>ONO! c0«O*«0 HNONi- CO OS rH CO rH ff N>OOU}Nt< CO U5 O O O i-H o 00100)001 r-t rHrHrHrHrH 11 + 11 rH rH rH i i i i i i 1 1 i-teq O O O rH O eg b- x x o> cm U3 ** T* CO 03 U5-* rH rH rH rH rH _L 1 _|_ | _l_ rH rH 1 1 1 1 1 1 1 1 "O | a •c I st ■ 1 st u p. i V ■c J, $ X 8 O o a 13 p O 4 o • • '-J3 g > < ■S«A< a a 3rHtNC0-^»C*5tNC0^»O §0 ja gweNco-tfioce IS (N CO Tji lO CO SB H so Q & 3 Pulse and Respiration. 109 l-~ CO tJI CO CO ^t^^-^^H O O »H © © I _1_ I I i ^ rl iH iH r~t I T^ I I 1 10 ■* m ■* t-ieo tvirt I I I I + I II © Q ^H © © r-t CO lO i-l CO >-l CN IOUS I I + I + I I I I ++ I + O00NNHN Os CO GO OS OS OS (NMIMHiNO 1 I + + + 1-1 FN 1 1 H^HdmTt< CO GO CO OS 00 OS _l_ ^H i-^ + + + + + -HP] 1-1 f-t + + OOOINUS I III f— i + + + + I + + + I I + + + + + O -^1 IN iH CO rt I ++ I I CM IS CO I I + + + + + I I + I I I us US CO I I in t- « inco + + + + + «U30)QIOCO II III CO 00 I I rH ■* "* OO + + I O ** CM co coeq I I I I I I I g ++ I + I I I I I I Next I I MCCHOH + + I + IN IN IN IN IN «S| I I I I I I CO ^ I I IN lO i-H INrt + + I I + in in o e> eoeq II III OS-* I I onnne + I I cq in i-4 co ■* w I I I I I I I I © iH —l mr* + 111 I I I I I I HWHHO + + I I in in in in coeq I I I I I I I I OHOHS + + HnOHClni II III I I HOOriH + ++ OHrHHHO I + I I I I WHHHO + + I I I I + I I I I I CO CO CH iHCO +++++ MMrtHHrt I I + I I I I I 03 CO -t NH + + + + + O IN O •* CO© I + I I I + + + + NOOfflrtO -*U} I I + I II 0"° So P.T3 8 S g | tli in co •* «o i§ §Q Q a - 31 - 78 - 52 +125 - 56 + 63 + 44 - 90 - 87 + 97 - 17 + 64 - 24 - 72 -116 + 48 - 71 values. The pulse data are given and averaged separately for each of the experimental conditions 1 to 7. The to Lai averages reveal that the pulse level for the initial period is, on the average, practically identical for the two groups of days, being 1037 and 1033 a, as shown at the bottom of the table in the column headed Period 1. This average may be relied upon, as a comparison of the individual averages in this column shows that the average pulse duration for the first period of June 29, July 1, and July 3 is al- most always approximately the same as the average for the same period on the corresponding alcohol dates, June 30, July 2, and July 4. Inspection of the general averages for duration given in the next to the last column of the table 1 will immediately show that in every case the averages for the normal days are larger than the averages for dura- tion in the first period of these days. Thus, for the pulse preceding the 1 These averages are for periods 2, 3, 4, 5, and 6 and do not include period 1. 114 Effect of Alcohol on Psycho-Physiological Functions. patellar reflex (No. 1) we have the general average of 1076 for the dura- tion of the pulse cycle during periods 2, 3, 4, 5, and 6 on June 29, July 1, and July 3, and a value of 1058 for the average duration in the first period on these three days. For the other experimental conditions we have for pulse 2 the general average of 1061 as compared to 1007 for the first period; pulse 3, 1093 and 1016; pulse 4, 1104 and 1027; pulse 5, 1085 and 1060; pulse 6, 1087 and 1037; and finally pulse 7, 1096 and 1052 a, the value for period 1 being the smaller in each case. This must be interpreted as indicating that normally the pulse gradually length- ens after the first period; that is, its rate per minute becomes less, but the change is comparatively slight. The comparable values for pulse duration on the alcohol days show an exactly reverse condition; that is, in every case the average for the first period on the three days, June 30, July 2, and July 4, is larger than the average for periods 2, 3, 4, 5, and 6 of these three days. Since the pulse duration started at the same level for both groups of days, it is perfectly clear that the pulse on alcohol days following period 1 is more rapid. At the bottom of the table general averages are given for the normal and alcohol values. Each average, both normal and alcohol, is computed from the averages in the vertical column above it; they thus represent by periods the general tendency for all the pulse records taken under experimental conditions 1 to 7, inclusive. The total average duration for the pulse cycle on alcohol days is 962 as compared with the average for normal days of 1086 a. The shorter pulse length after alcohol shown by these general averages is also revealed in each section of table 20 and in almost every period of each section. This tendency is so definite as to make it unnecessary to point out the individual comparisons. The averages at the bottom of table 20 are also given in diagrammatic form in figure 15 (see solid-dot curves). The average durations of the cycles are given at the left and the corresponding pulse rate at the right. In this figure the heavy line represents the averages for the normal days. The curve starts with a duration of 1037 a, which corre- sponds to a rate of 58 beats per minute, and drops in periods 2 and 3, reaching 55 beats per minute. In the fourth period there is a slight rise to 56 beats, which is continued in period 5 ; finally there is a decided drop to 51 beats in period 6. Inasmuch as the fall in period 6 is in large part due to the more fragmentary data for this period (see table 20), the course of the normal curve shows a very slight decrease in pulse-rate throughout the experimental days. The alcohol curve starts at prac- tically the same level as the normal curve, that is, 58 beats per minute. From this point there is a rise in period 2 to 63 beats per minute, a decrease in period 3 to 60, then a return in period 4 to 63+, with con- stancy in period 5 and a drop to 61 beats in period 6. The two curves are sharply distinguished and show that the pulse of Subject VI was faster by about 6 beats per minute after the taking of alcohol. Pulse and Respiration. 115 The effect of alcohol upon the pulse rate of Subject VI is also indi- cated in table 20 by the statistical method of differences. The average differences given in the last column of the table show that on the normal days the average difference is always a plus value. These average differences for the seven experimental conditions are, for the normal days, +11, +54, +78, +77, +25, +50, and +44. The comparison values for the alcohol days are always minus, being —37, —107, —77, —83, —53, —52, and —87. The general average differences for these two series are, respectively, +48 and —71. What is thus indicated by the average differences will likewise be found to be true for the indi- vidual periods 2, 3, 4, 5, and 6 for the pulse taken under the different conditions from 1 to 7; that is, the differences for normal days have usually the plus sign, while those for alcohol days are usually minus. Experimental periods ' Dose * 3 r 4 Flo. 15. — Average duration of pulse cycle under experimental conditions 1 to 7 on normal and alcohol days, as compared with pulse duration during word reactions (5 and 6). The effect of alcohol is found by deducting the normal difference from the alcohol difference. These results are given for each of the seven experimental conditions in table 20, and are also stated in terms of per cent. They are likewise summarized in table 21, which shows by periods the average effect of alcohol for each of the conditions. There are only two values in the latter table which are preceded by the plus 116 Effect of Alcohol on Psycho-Physiological Functions. sign, i. e., +1.7 and +2.1. It is clear from table 21 that the pulse was quickened most under the experimental conditions 2, 3, and 4, that is, in rest after the patellar reflex, in rest after the faradic-threshold measurement, and with the subject holding the voice key before the word-reaction test. In the latter case the high average of —17.3 is due partly to the value of —32.6 in the sixth period, which unfortu- nately is the period where the data are fragmentary. With the subject holding the voice key after the word-reaction test the average effect of the alcohol was —11.9. Here the subject was undoubtedly not so relaxed as under experimental conditions 2, 3, and 4. During the word reactions (5 and 6) the average effect was —7.4, and —8.3. In this measurement the subject was of course mildly active, or at least he was supposed to be. The pulse durations in table 20 do not show, how- ever, that he was very active, as the word-reaction pulse was on normal days not faster during 5 and 6 than under the other conditions. Table 21. — Effect of alcohol upon the pulse, expressed in per cent differences by periods. Experimental condition. Period 2. Period 3. Period 4. Period 5. Period 6. Average periods 2 to 6. Last period. 1. Best preceding patellar + 1.7 -15.6 -17.0 -16.9 - 7.5 - 9.8 - 8.9 -10.6 - 2.8 -10.3 -15.2 -17.2 - 4.0 -10.0 -10.2 -10.0 - 7.7 -20.1 -18.8 - 7.3 - 8.6 -12.2 -15.4 -12.9 - 9.8 -18.5 - 6.3 -12.7 - 1.5 + 2.1 -13.0 - 8.5 -21.0 -19.5 -12.3 -32.6 -15.2 -11.7 -12.0 -17.8 - 7.9 -16.8 -13.9 -17.3 - 7.4 - 8.3 -11.9 -11.9 - 8.2 -14.5 -10.0 -22.3 -11.9 - 8.6 -14.5 -12.9 2. Rest after patellar re- flex 3. Rest after faradic 4. Holding voice key be- fore word reactions . . 5. During word reactions . 6. During word reactions. 7. Holding voice key after The relation of the pulse duration during word reaction (averages by periods for pulses 5 and 6 of table 20) to the averages for all pulse data 1 to 7 are given in figure 15, page 115, in which the broken-line curves represent the averages for the word-reaction pulse measurements. Apparently there was more relaxation during word reactions on normal days than existed under any of the other conditions, for we find the broken-line curve (fig. 15, normal) is slightly below the average of all pulse measurements 1 to 7 in periods 1 to 4, indicating slower pulse rate. On alcohol days the pulse rate during word reactions is faster than the average for all conditions. These facts are pertinent to the discussion of the word-reaction data on page 66, where it was found that the reactions were considerably faster on alcohol days, including the pre- liminary period 1, and where it was stated that usually on normal days the subject did not try hard to react quickly. It is unlikely that alert Pulse and Respiration. 117 attention and quick response in reaction experiments would be asso- ciated with a pulse rate slower than the average resting pulse recorded. It will be seen that the faster pulse in period 1 (pre-alcohol) is in con- formity with the results found for the same period with finger-movement pulse, and post-tetanus pulse, i. e., in all three cases the pre-alcohol pulse is faster than that before the control dose. It is probably not an accident that all these represent conditions where the subject was more active than in Nos. 1, 2, 3, 4, and 7 and the pre-tetanus period. The resting pulse preceding patellar reflex gives an average alcohol effect of —7.9, which is relatively small, and particularly so if compared with the average of —16.8 for resting after the patellar reflex. Pulse record No. 1 was of course the very first record taken in each period. Previous to it the sub j ect had been moving about more or less ; in period 2 the dose for the day immediately preceded this record. The average percentile effects of alcohol upon the pulse in the several experimental conditions for periods 2 to 6, inclusive, are given in table 21 as -10.6, -10.0, -12.9, -8.5, and -17.8, with a total average for all conditions and all periods of —11.9 per cent (decrease in duration of cycle length and increase in rate). The pulse-duration averages pre- viously discussed and the differences summarized in table 21 therefore agree in showing a faster pulse after alcohol. There still remain for consideration the pulse records 8 and 9, which were taken at the time of the finger movements. These were not usually 15 seconds long, but were regulated by the finger-movement records, which were about 10 seconds in length. A peculiarity exists in the pulse during the finger movements which may be seen in the illustrative records shown in figure 11, E and F, page 96. The electrical waves are decidedly reduced in amplitude and in some instances seem to be dropped out, particularly the P and T waves. Approximately one-third, that is 20, of the finger-movement pulse records show this phenomena in more or less marked degree. These records are well distributed throughout the various days, as there were three or four on each day except on June 30. In considering this diffi- culty it may be called to mind that the subject in performing the finger- movement measurements tended to lean forward somewhat during the exercise. Since the pulse line shows no great irregularities, as would be expected were the electrodes out of contact with the body a part of the time, it seems that the only explanation of this phenomenon which is at all satisfactory is that a change in the position of the body produced a change in the electrical axis of the heart, so that at times there was no voltage across the string galvanometer. Curves E and F in figure 11, as well as others that might be shown, indicate that the condition was also dependent in part upon the phase of respiration, as the electrical waves disappear in the pulse cycles during the moments at the height of inspiration. 118 Effect of Alcohol on Psycho-Physiological Functions. This condition with the pulse during finger movements was not found with any other subject used by Dodge and Benedict, and a careful search did not reveal that it was certain with Subject VI in the former series of experiments. It must be observed, however, that Dodge and Benedict employed their pulse-recording technique No. 2, that is, the telephone pulse-recorder of Dodge with Subject VI. (See their report, p. 189.) In fact, all of their records for him were obtained with the telephone-recorder technique, except that of the normal electro-cardio- gram taken at the beginning of the series of experiments. With the Dodge telephone recorder, the pulse during finger movements usually shows much disturbance and irregularity, and this irregularity is more marked than when using body electrodes. It was a surprise to find this to be the case, as one would think that the position of the body electrodes, because of movements of the arms and fingers, would cause other body currents to complicate the heart picture. On the other hand, one would not expect very much jarring and shaking of the head during the finger movements, but exactly this seems to have taken place and the pulse curve is complicated by many other deflections, large and small. This condition makes it impossible to compare the two groups of records taken by the different techniques. The change in the elec- trical axis of the heart which probably produced the condition under discussion seems to bear no relation to the dose of alcohol and must be regarded as without significance, aside from the fact that it made the elaboration of the records somewhat more difficult. Since the pulse records taken during the finger movements showed the peculiarity that at times even the R waves were absent, the records could not be dealt with as with pulse records 1 to 7 inclusive, but indi- vidual pulse cycles were measured as in the tetanus pulse and an aver- age was taken for the measurable pulse cycles of each record. All of the available data for the pulse during the finger movements are presented in table 22. This table is of the usual form, the results being presented by periods, with three sections showing (i) duration, (n) differences, and (in) the effect of alcohol by differences and per- centiles. The average duration of 925 a for the normal days shown in the next to the last column indicates that the pulse with the finger move- ments is considerably faster than in conditions 1 to 7 previously dis- cussed. The same is shown by the averages 906 and 861 a for period 1. The latter value is for pre-alcohol and is noticeably shorter (faster rate) than that for the normal group of days, a condition quite parallel to that which existed for period 1 in the pulse during word reactions and in the post-tetanus pulse. The change in the level of preliminary pulse for the two groups of days does not favor the comparison by averages; nevertheless the averages in the different periods show the following: (1) a slight lengthening of the pulse from period to period throughout the normal days; the durations from period 1 to period 6 are 906, 898, Pulse and Respiration. 119 909, 952, 992, and 986 a; the average for all the periods, excluding period 1, is 925 a; (2) the pulse cycle after alcohol is shorter, 770, 810, 816, 826, and 869 +88 -11 +39 -16 -11 -14 Alcohol differences — June 30 July 2 -36 + 2 P) -34 -23 -62 - 7.0 -27 -29 (') -16 -24 -10 -14 July 4 Section III: Effect of alcohol — 1 Subject broke connection to galvanometer. 2 Electrical phenomenon of heart caused no deflection of galvanometer in first period. 1 Illegible because of faulty illumination. * Second record for this date; see note 2 for information regarding first record. 120 Effect of Alcohol on Psycho-Physiological Functions. ation of the pulse-length after alcohol was considerably smaller than on normal days. It was this consistent change in the mean variation, together with other supporting evidence, which led to the belief that the effect produced by alcohol was probably a partial paralysis of the cardio-inhibitory mechanism. As a rule, the individual pulse cycles were not measured in the data for the repetition experiments on Sub- ject VI, but in the case of the tetanus-pulse records and of those rec- ords taken with the finger movements, the individual cycles were by necessity measured; a mean variation is thus available for these. The mean variations for the pulse taken during finger movements have been entered in table 22. Here it will be found that the normal values ranged about 80 a, approximately 10 per cent of the pulse dura- tion. After alcohol there was a marked decrease, the average for all periods after the dose on the alcohol days being 41 a, with an average pulse duration of 813 a. In terms of per cent differences, the effect of alcohol upon the mean variation of the pulse duration is (section in), beginning with period 2, -20, -37, —21, -17, and -14 per cent, the average for periods 2 to 6 being —34 per cent. The mean variations in the pulse-length during pre-tetanus and tetanus are given in table 23 by periods and by days. The average pre- tetanus mean variation for periods 2 to 5 of the normal days is 3.7 hun- dredths of a second and for the alcohol days, 3.0 hundredths of a sec- ond. The average normal tetanus mean variation is 5.7, exclusive of period 1, and that for the alcohol periods is 4.4 hundredths of a second. The decrease in mean variation is particularly prominent in period 2 of alcohol pre-tetanus pulse. Thus these data for the effect of alcohol on Table 23. — Mean variation c (Values / pulse duration for pre-tetanus and tetanus periods. given in hundredths of a second.] Kind of experiment and date. Pre-tetanus period. Tetanus period. Period 1 Period 2 Period 3 Period 4 Period 5 Period 1 Period 2 Period 3 Period 4 Period 5 Normal: June 29 July 1 July 3 Av. M. V Alcohol: June 30 July 2 July 4 Av. M. V 3 2 3 4 3 4 4 2 3 6 6 2 3 2 5 3 4 2 3 2 2 3 5 5 2 5 7 5 4 3 5 6 4 4 15 3 10 3 2 4 8 3 5 4 10 5 5 9 3 5 6 4 7 5 3.2 3.8 3.4 2.8 4.7 4.5 6.2 4.8 6.2 5.5 3 3 3 3 2 3 4 3 1 1 2 2 4 5 6 3 2 2 8 3 3 4 2 3 3 3 2 1 2 3 5 4 4 4 4 3 8 5 1 3 7 6 6 5 4 5 6 6 3 3 3 4 3 2 5 3 3 4 6 5 2.8 2.2 3.7 3.8 2.3 4.0 5.0 5.3 3.0 3.4 Pulse and Respiration. 121 mean variation of the pulse rate, meager as they are in comparison with those collected by Dodge and Benedict, are directly in accord with the findings 1 of those investigators in showing a decreased mean varia- tion following alcohol. Taken as a whole, the pulse data where comparable in the two series of experiments on Subject VI are in accord in every important detail. (1) There is normally, under such experimental conditions as those employed, a gradual retardation of the pulse rate from period to period throughout the experimental day, this retardation being less prominent after alcohol; the alcohol pulse is therefore relatively faster. Aside from the association pulse data of the first series, this subject varied from the normal group in that he did not show so prominent a retardation as the others. (2) The pulse rate after alcohol is absolutely faster, as the cycle duration is actually shortened after the dose in comparison to that for the pre-alcohol period of the same day. This is found in all of the records taken in this series, with the exception of those during the actual moments of tetanus. In Dodge and Benedict's table 42, page 230, where a summary of the average pulse differences is presented and where a plus value means an absolute rise in pulse rate per minute, Subject VI has an unduly large proportion of the plus signs. This table gives data for 7 subjects; of the 43 plus signs occurring, Subject VI has 16, or 37 per cent. (3) The effect of alcohol on the pulse rate is apparent in periods 2, 3, 4, 5, and 6, but in general is not conspicuously more prominent in any one of these than in the others. There is no indication that the effect had reached its limit at the end of the experimental day, which extended 4 hours after the ingestion of the alcohol. (4) The alcohol effect is greatest upon the pulse rate when the sub- ject is resting quietly, there being an average increase of about 11 per cent at such a time, as compared to an increase of about 7 per cent when the subject is mildly active, as in sitting upright and reacting by speech. In interpreting the increased pulse rate, either relative or absolute, after a dose of alcohol a question must be raised which Dodge and Bene- dict did not consider and which can not be answered here, as it involves further experimentation. With the possible exception of periods of very severe muscular activity, an individual's pulse rate is generally considered to be approximately a linear function of his metabolism, high pulse rate being associated with a high total heat production and vice versa. 2 The metabolism does not keep a dead level throughout the 1 A correction should be made in Dodge and Benedict's table 44, p. 237, as follows: The data for two genuflections, normal and alcohol, and 60 seconds after two genuflections, normal and alcohol, are recorded under Subject IV, but should be placed one column to the right, that is, under Subject VI. (See table 40, p. 217, in which the mean variation averages are first presented.) s See Benedict and Carpenter, Carnegie Inst. Wash. Pub. No. 126, 1910, p. 248; Dodge and Benedict's report, p. 253 ff., and other publications and articles from the Nutrition Laboratory; also see Henderson and Prince, Am. Journ. Physiol., 1914, 35, p. 106; Murlin and Greer, Am. Journ. Physiol., 1914, 33, p. 251, and Krogh and Lindhard, Journ. Physiol., 1917, 51, p. 182. 122 Effect of Alcohol on Psycho-Physiological Functions. 24 hours of a day. It of course varies with excitement, exercise, and relaxation and also with ingestion of food. Other things being equal, it is highest after meals and presumably the pulse follows these fluctua- tions with fair approximation. In a laboratory session, which comes between meals and lasts 3 to 5 hours with the subject under uniform conditions that demand but slight activity, the metabolism theoreti- cally should decline; pulse records taken frequently and at homologous points in the periods should therefore naturally show some progressive decrease in rate aside from any change due to psychic factors, such as a growing familiarity with the experimental environment. Both series of experiments, particularly the former, have thoroughly demonstrated this normal decline in pulse rate which probably has its basis in a change in metabolism. After the first period on alcohol days there was an energy intake of dose A, 30 c.c., approximately 260 calories, or of dose B, 45 c.c, about 390 calories, and the pulse following was found to be relatively faster than on normal days, when occasionally a control dose of cereal "coffee" was employed. The specific effect of alcohol on the total metabolism has not been conclusively established, but there appears to be slightly more evidence for a small rise than for a fall after alcohol. 1 Hence, in the absence of definite proof to the contrary, the pulse changes with it are such as to raise the question whether they are not due to normal metabolism phenomena rather than to so-called drug action. RESPIRATION DATA. As respiration records were also incorporated with the pulse records, some attention should be given these from the standpoint of a possible effect of alcohol upon the respiration rate. In the Tentative Plan 2 the effect of alcohol upon respiration-rate was but a small part of the pro- posed respiration section, and our data can be regarded as only sup- plementary. In the first place, it must be said that Subject VI was naturally very irregular in his breathing. Dr. Carpenter 3 for this cause found him quite unsatisfactory as a subject in some respiration experi- ments. In our pulse and respiration records there are frequently intervals of apnea; a record of this type is shown in C, figure 11 (see p. 96) . That the apparatus for recording the respiration was not insen- sitive is proved by the fact that the pulse shows very clearly in the curve. The average length of the respiration in seconds for the different days and the different periods of pre-tetanus and post-tetanus pulse is shown in table 24. Naturally the respiration curve could not be interpreted during the moments of tetanus. From the normal data in the upper 1 Carpenter, Am. Journ. Physiol., 1917, 42, p. 605; Higgins, Journ. Pharm. and Exp. Ther., 1917, 9, p. 441. 2 Tentative plan for a proposed investigation into the physiological action of ethyl alcohol in man, Boston, 1913; reprinted in Dodge and Benedict's report as Appendix I (see p. 269). 3 Carpenter, Carnegie Inst. Wash. Pub. No. 216, 1915. Pulse and Respiration. 123 part of the table, it can be seen that the respiration was longer in the post-tetanus division of the record, the average for period 1 being 3.7 seconds as compared with 2.8 seconds for pre-tetanus; the total average for periods 2 to 6 during post-tetanus was 3.2 seconds as compared with 2.8 seconds for the pre-tetanus division. The curves were not such as could be measured for depth of respiration. In the alcohol section of the table (periods 2 to 6 and average), it is again shown that the respiration is longer in the post-tetanus division, but the difference is not so large as with the normal values. As with the word-reaction, finger-movement, and post-tetanus pulse, there is a difference in the initial level of the respiration on the two groups of days, showing a shorter normal respiration on the alcohol days (see period 1) ; this limits our direct comparison of averages in other periods. The signs for the differences are about equally divided between plus and minus. The effect of alcohol in terms of per cent differences seems to be, in general, Table 24. — Length of respiration in pre-tetanus and post-tetanus periods. [Values given in seconds.] Kind of experiment and pulse measurement. Period 1. Period 2. Period 3. Period 4. Period 5. Period 6. Av. length. Av. diff. Av. Av. Diff. Av. Diff. Av. Diff. Av. Diff. Av. Diff. Normal : Pre-tetanus — June 29 July 1 July 3 Post-tetanus — July 1 July 3 Average Alcohol: Pre-tetanus — June 30 July 2 July 4 Post-tetanus — June 30 July 2 July 4 Average Effect of alcohol : Pre-tetanus — 2.7 2.8 2.9 2.8 4.0 3.8 3.2 3.7 2.6 2.4 2.5 2.5 2.9 2.9 2.9 2.9 2.8 2.9 2.8 2.8 3.6 2.8 3.0 3.1 2.5 2.5 2.3 2.4 3.2 2.9 2.6 2.9 + 0.1 + .1 - .1 - .4 - 1.0 - .2 - .5 - .1 + .1 - .2 - .1 + .3 - .3 - .1 - 3.8 + .5 +15.1 2.5 2.6 3.8 3.0 3.2 2.8 4.0 3.3 2.6 2.8 2.7 2.7 2.9 2.7 3.5 3.0 - 0.2 - .2 + .9 + .2 - .8 - 1.0 + .8 - .3 + .4 + .2 + .2 - .2 + .6 + .1 + .4 +12.1 2.8 2.5 2.6 2.6 3.4 2.9 3.0 3.1 2.7 3.1 2.5 2.8 2.5 3.2 2.8 2.8 + 0.1 - .3 - .3 - .2 - .6 - .9 - .2 - .6 + .1 + .7 + .3 - .4 + .3 - .1 - .1 + .5 +19.2 + .5 +15.1 2.7 2.7 3.1 2.8 3.4 2.8 3.3 3.2 2.6 3.0 2.6 2.7 2.8 2.9 3.1 2.9 - .2 + .2 - .6 -1.0 + -1 - .5 + .6 + .1 + .2 - .1 + .2 2.6 - 0.2 2.6 - .2 2.8 3.2 3.0 2.7 3.6 2.5 2.9 2.5 3.1 3.0 2.9 - 1.0 - .5 + .1 + 1.2 + .4 - .4 + .2 + .1 + .6 +23.1 + .5 +15.1 2.9 3.0 2.9 +0.5 + .5 + .5 2.7 3.4 3.0 - .2 + .5 + .1 Per cent differ- Post-tetanus — Per cent differ- 124 Effect of Alcohol on Psycho-Physiological Functions. a lengthened respiration. This is not true in the second and third periods for the pre-tetanus condition, in which we have —3.8 and per cent, but applies to the third and fourth periods of pre-tetanus (+19.2 and +23.1 per cent) and for all periods of post-tetanus ( + 15.1, +12.1, +15.1, and +15.1 per cent). The respiration data taken with pulse records 1 to 7, inclusive, of the experimental cycle are recorded in table 25. Here the averages for the first period are the same for alcohol and non-alcohol days, i.e., 3.0 seconds. The average for all of the normal periods, exclusive of the first, is 2.9 seconds, while the comparison average for alcohol days is 2.8 seconds. The average respiration on the alcohol days in periods 2 and 3 is a little longer than for the same periods on the normal days, but the difference is insignificant. The effect of alcohol as shown by per cent differences is +3.3, 0, —13.3, —16.7, and per cent for periods 2 to 6, respectively. On the whole our supplementary data for respiration- rate, like those collected by Higgins, 1 show no definite alcohol effect. Table 25. — Length of respiration during pulse [Values given measurements Nos. lio7 of experimental cycle. in seconds] Kind of experiment and date. Period 1. Period 2. Period 3. Period 4. Period 5. Period 6. Av. length. Av. diff. Av. Av. Diff. Av. Diff. Av. Diff. Av. Diff. Av. Diff. Normal: June 29 July 1 July 3 Average Alcohol: June 30 July 2 July 4 Average Effect of alcohol : 3.2 2.6 3.1 3.0 3.5 2.8 2.8 3.0 3.0 2.7 2.8 2.8 2.8 3.2 3.0 3.0 -0.2 + -1 - .3 - .1 - .7 + .4 + .2 + .1 +3.3 2.7 2.8 2.9 2.8 3.3 2.6 2.7 2.9 -0.5 + .2 - .2 - .2 - .2 - .2 - .1 - .2 3.1 3.2 2.8 3.0 2.8 2.8 2.6 2.7 - 0.1 + .6 - .3 + .1 - .7 - .2 - .3 - .4 -13.3 2.9 2.8 2.9 2.9 2.9 2.8 2.7 2.8 -0.3 + .2 - .1 - .1 - .6 - .2 2.6 3.3 2.9 2.7 2.8 2.5 2.7 + .2 + .1 - .8 - .3 - .4 - .5 -16.7 2.9 2.9 -0.2 - .2 2.6 2.6 2.6 - .2 - .2 - .2 Per cent differ- 1 Higgins, Joum. Pharm. and Exp. Ther., 1917, 9, p. 441. Conclusions. 125 CONCLUSIONS. This independent repetition series of experiments, employing equiva- lent dosage, the same apparatus and general conditions, and essentially the same technique as those used by Dodge and Benedict, was under- taken on one of their normal subjects (No. VI) for the purpose, among other things, of verifying the somewhat irregular results obtained on him in the first series. A summary of the findings in this repetition series for the different experimental periods following the ingestion of alcohol, together with the comparable averages found by Dodge and Benedict for their normal group, 1 and also for Subject VI individually, is presented in table 26. The values for the repetition series given in this table appear in section in of the foregoing tables and are the effects of alcohol stated in percentile differences for the various pro- cesses in question. 2 A plus sign denotes an increase and a minus sign a decrease (see p. 35), and these signs must be interpreted according to the nature of the measurement. For simplicity the summary table has been divided into two parts. In part i are grouped those measurements for which a plus sign sup- posedly means inferior or less efficient performance, e. g., longer reflex latency, slower movements or reactions, larger mean variation in reactions, and the like. To state this still more concretely, it will be observed that the first value in period 2 is +4.1. This is a numerical expression of the contrast between the results of two groups of days, taking into account the preliminary periods of the same. It shows that the patellar-reflex latency with 30-gram stimuli was longer by 4.1 per cent in the first measurements made after the taking of 30 c.c. of alcohol as compared with the results found in the period immediately after the control dose was taken. This lengthening of the reflex latency was also present in the third and fourth periods, but the percentages +0.9 and +2.9 show that the effect was not so great. In the fifth and sixth periods the sign changes to minus (—7.3 and —1.5), indicating a shortened reflex time. Part ii of table 26 includes those processes for which a plus sign sup- posedly signifies superior or more efficient performance. In the cases of pulse and respiration, minus denotes a decrease in length, that is to say, a faster pulse rate or respiration rate per minute. It will be seen that under period 2 and opposite "Number of finger movements in 8 sec- onds" —3.9 will be fouaeV"which means that the total number of free finger movements performed in 8 seconds was 3.9 per cent less after alcohol in contrast with the record after control doses. Similar decre- ments were present, also, in periods 3, 4, and 5, in which the percentage 1 For some measurements the normal group was composed of 7 and for others of 6 men. In every case it included Subject VI. 2 The method of computation has been fully discussed in connection with illustrative table 2, p. 31. 126 Effect of Alcohol on Psycho-Physiological Functions. reductions were 4.0, 3.1, and 3.5, respectively, while in period 6 there was an increase of 7.1 per cent. The foregoing illustrations should make it perfectly clear that a plus sign in part i and a minus sign in part ii have fundamentally the same meaning — that is, an inferior or less satisfactory functioning of a process. The general picture for the effect of alcohol on Subject "VT, as shown by the repetition series, is definitely one of neuro-muscular depression, demonstrated by the fact that in part I of table 26 the signs for the percentile differences are predominantly plus, while in part n they are as prominently minus. At the bottom of each column in the table the signs of the column have been classified under the three heads — inferior, superior, and neutral — and the entries are the percentages found for the several categories. Tabu: 26. — Percentile effects of ingestion of SO c.c. of absolute alcohol on a related group of processes. 1 Processes measured. Part I. Patellar reflex (latency of primary and secondary re- flexes) : L, 30-gram blow L, 50-gram blow I/, 30-gram blow I/, 50-gram blow lid reflex (latency) : L, primary wink I/, secondary wink .... Eye reaction: Time Mean variation Word reaction: Time Mean variation Faradic sensory threshold: Z units /S units Eye-movement time, left movements Memory saving, mean varia- tion 6 Repetition series. Experimental periods following the taking of alcohol. Period 2. + 4.1 + 2.1 + 1.3 +13.3 +15.0 +12.0 +34 + 3.1 - 2.0 +17 +32 +14 +10 Period 3. + 0.9 + 0.9 + 2.6 + 2.7 + 5.7 + 0.7 +13 +41 - 9.0 + 7.0 + 0.9 +10 + 6.4 -10 Period 4. + 2.9 +11 + 1.1 - 2.9 - 7.5 + 9.5 + 6.3 - 7.4 + 5.3 + 7.3 + 7.2 +13 Period 5. - 7.3 - 8.9 - 9.2 +21.0 + 6.4 - 3.7 + 6.2 -25 -11 +21 + 6.2 +25 + 5.0 -15 Period 6. Last period. Av. periods 2, 3, and 4. + + 1.5 4.5 5.3 16.0 4.3 3.7 1.9 47 21 25 11 7.9 2.1 5.0 - 9.8 - 12.0 - 5.3 - 15.0 - 5.8 - 6.7 + 8.1 + 22 - 9.6 + 35.0 + 0.9 + 24 + 3.0 - 10.0 + 2.6 + 1.0 + 4.5 + 1-7 + 5.4 + 2.7 +11.5 +27.1 - 4.4 + 1.7 + 7.7 +16.4 + 9.2 + 4.3 First series. 2 Results found by Dodge and Benedict with equivalent dose. Av. for normal group. + 11.1 + 4.0 + 5.9 + 6.8 - 5.4 + 21 + 3.4 Av. for Subject VI. + 0.3 (*) - 4.5 - 10.7 - 9.3 - 5 + 48 *+ 25.8 1 The values given in this table differ in certain instances from those published in the preliminary abstract (Miles, Proc. Nat. Acad. Sci., 1916, 2, p. 703), e. g., the patellar reflexes from 30 and 50 gram stimuli have been treated separately; there were also one or two errors discovered in the original computations. J The signs used in presenting Dodge and Benedict's data have been changed to make them conform with the usage in this monograph, where a plus means increase and a minus decrease in the quantities measured. 3 The records were illegible. 4 No left eye movements are given for Subject VI in Dodge and Benedict's table 28, p. 166; therefore the percen- tile difference for right movements is entered here. 6 From computations based on the "saving" between the first and second reactions; see table 11, p. 71. Conclusions. 127 Table 26 — Percentile effects of ingestion of SO c.c. of absolute alcohol on a related group of processes. 1 — Cont'd. Processes measured. Repetition series. Experimental periods following the taking of alcohol. Period 2. Period 3. Period 4. Period 5. Period 6. Last period. Av. periods2, 3, and 4. First series. Results found by Dodge and Benedict with equivalent dose. Av. for normal group. Av. for Subject VI. Part II. Patellar reflex (amplitude of primary and secondary re- flexes): A, 30-gram blow A, 50-gram blow A', 30-gram blow A', 50-gram blow Lid reflex (amplitude) : A, primary wink A', secondary wink Progress in memorizing 5 .... Eye-movement distance .... Number of finger movements in 8 seconds Pulse-cycle length: Just before tension .... During muscle tension. . After muscle tension . . . Pulse during finger movements Mean variation of fin- ger-movement pulse. Av. of other pulse data, Nos. 1 to 7 Length of respiration during pulse records Nos. 1 to 7 . Paet III. Percentage of signs denoting: Inferior or less efficient functioning* Superior or more efficient functioning Neutral -11.9 + 1.9 +19 -46 -57 +10.9 -21 - 3.9 -11.6 + 1.3 - 2.5 - 9.4 -20 -10.6 + 3.3 77 20 3 - 6.6 + 7.4 +13 -62 -33 -43 + 2.5 -17 - 4.0 - 3.9 - 4.0 - 8.7 - 5.6 -37 -10 80 17 3 -30 -19 +19 - 4.6 +13 +26 - 5.9 -22 - 3.1 -14. - 4. + 2. - 9 -21 -12 -13 70 23 7 -65 -30 + 6.5 + 8.9 -18 -69 +28 - 1.3 - 3.5 - 5.8 - 2.7 - 1.2 -12.9 -17 - 8.5 -16.7 67 33 + 4.0 + 5.6 +123 + 62 - 17 + 66 + 57 - 5.0 + 7.1 C> C) C) - 7.0 - 14 - 17.8 33 63 4 + 47 + 44 +148 + 54 + 23 - 5.7 + 18 - 10.6 + 2.0 '- 6.6 '- 6.2 '- 0.8 - 5.3 - 18.6 - 12.9 - 10.0 50 50 -16.2 - 3.2 +17.0 -37.5 -19.7 -24.7 + 2.5 -20.0 - 3.7 -10.0 - 2.2 - 2.9 - 8.1 -26.0 -11.2 - 3.3 90 10 - 60.3 - 38.0 - 10.7 -104 + 1-4 - 8.9 - 3.1 - 4.1 - 3.0 82 + 13 +173 - 18.0 - 27.0 «+ 14 - 15.2 - 4.0 + 6.0 - 3.5 12 6 50 50 6 Corrected value according to the method of not including the first-period score on normal days; see Appendix III, p. 140. 7 There was no period 6 in the normal data of tetanus pulse. 8 The computation was not carried out for each pulse cycle, but averages were found for cycles 1 to 7, 8 to 14, and 15 to 21, from the averages in the pre-tetanus and tetanus divisions, and in the post-tetanus division, and computations were made finally, taking the average percentile difference for the three groups of 7 pulse cycles of the post-tetanus divison. 9 The assumption that a faster respiration rate may be rightly classified as less efficient functioning is admittedly somewhat arbitrary. In but one instance, period 6, are there more signs indicating a supe- rior functioning of processes than were found for inferior performances. The usual effect of alcohol would seem to be reversed in the sixth period, so that the subject actually did better in the majority of pro- cesses measured at the end of the alcohol-day experiments than he did 128 Effect of Alcohol on Psycho-Physiological Functions. on non-alcohol days. Some indication of this reversal appeared to be present in the former experiments 1 and one of the reasons for using 5-hour sessions in the repetition series in place of 3 hours as formerly was to obtain data for a longer time following the ingestion of alcohol which might illuminate this problem of possible facilitation succeeding depression. For reasons which have been previously explained (see p. 29), it was practically impossible in the later experiments to have an equal number of periods on each day. Hence it transpires that for the sixth period the data are confined to one normal and two alcohol days. The normal day was July 3, on which the subject tried less to carry out the instructions in performing his part of the experiments. The gen- eral relaxation and drowsiness of the subject are confirmed by the fact that the pulse rate was decidedly the slowest on this day. This can be seen in the "daily averages " at the bottom of table 20, p. 1 13, where the general average pulse-cycle length for the day in question is 1.14 sec- onds, equaling a rate of 52.6 per minute. On the other hand, the results obtained on July 4, the last alcohol day, were considerably influenced by a spurt of effort made by the subject. These conditions seriously complicate the interpretation of the results for period 6 which might otherwise appear as a post-alcohol facilitation. Obviously, however, it would not be unfair, as the days were all 5 hours in length, to com- pute percentile differences, using as a basis the last period of each day, regardless of whether the periods were numbered 4, 5, or 6. The differ- ences are already in the tables for the various measurements. It is necessary to find the "last-period" average difference for normal and alcohol days separately, to subtract the former from the latter, and to divide the result by the average of all first-period scores. For illus- tration, the percentile difference effect of alcohol in the last period of eye reactions would be found as follows (see p. 60, table 9, section n) : The normal differences for the last period are —23, +26, and +12, average +5; the alcohol differences are +51, —16, and +32, average +22; the alcohol average minus the normal average equals +17, which, divided by the average of relevant first-period normals (211+208^-2 = 209.5), equals 8.1 per cent; this is entered in the column headed " Last period" in table 26. This percentile difference, +8.1, for the last period of eye reactions shows continued depression to the end of the experimental day. It would seem that the value — 1.9 in period 6 was largely due to the fragmentary data for this period and not to a real facilitation of the process. When the signs for all the 30 percentage values of the last period are classified, it is found that half of them repre- sent inferior and half superior performance, while in period 6 the pro- 1 Dodge and Benedict's report, p. 258 ff. 2 This comparison is not strictly accurate, as there was no tetanus pulse for period 6. Since there is a strong probability, judging by the other tetanus pulse data, that these three signs would have been minus if present, the proportion might have stood 40, 57, and 3, which would be a little less favorable to facilitation in period 6. Conclusions. 129 portion was 1 to 2, i. e., 33, 63, and 4. 2 (See table 26, columns headed "Period 6" and "Last period" and note signs for different measure- ments.) Specifically, the changes in sign (superior to inferior) from period 6 to the last period have been with eye-reaction time, faradic- sensory threshold for both Z units and ft units, eye-movement time, and lid reflex A'. There are also some changes from inferior in period 6 to superior in the last period, the reverse of those just given; these are lid reflex L and L' and lid reflex A. From the percentile differences in the last period (in contrast with period 6) certain quite definite conclusions can be drawn regarding the effect of alcohol on this subject 4 hours after the dose of 30 c.c. had been taken: (1) The reflexes are more active, their latency is shorter, and the amplitude is increased. There are 12 percentile differences for the effect of alcohol on the reflexes. It will be observed that in the last period every difference except that for lid reflex A' shows relative facili- tation, and the percentages are so large as to establish the observation. (2) The faster pulse and respiration rates continue, i. e., the direction of the alcohol effect is here unchanged. The 7 values are in agreement and there is no certain decrease in the alcohol effect found by comparing the last period with the average of periods 2, 3, and 4. (3) The measured processes, word-reaction time, and progress in memorizing, which were found to be superior in periods 2 to 4, are the same in the last period. Eye-reaction time and mean variation, word- reaction mean variation, faradic threshold Z and /?, eye-movement time and distance, and number of finger movements in 8 seconds, all of which showed depression in the earlier periods, are found to do likewise in the last period. Facilitation in these experiments therefore seems limited to the reflexes, and pending the accumulation of data with other subjects it is best to forego further discussion of the matter here. Considering again periods 2, 3, and 4, we find that the results for all of these are from 3 normal and 3 alcohol days, and that together the three periods make a time interval following the dose which corre- sponds fairly well to the sessions of the first series. An average of the results of these periods is given in table 26. Of these 30 percentile differences for the effect of alcohol, 27 (a ratio of 9 to 1) show inferior functioning of processes after the dose. The exceptions are those which the preceding consideration of the results would lead us to expect, i. e., word-reaction time, progress in memorizing, and patellar reflex, A', 30 grams. 1 Before accepting this predominance at its face value, it is desirable to know if the 9 to 1 ratio can in any way be due to the pecu- 1 It was pointed out on page 43 that as the primary reflex decreases in amplitude, the secondary one, which falls in the relative refractory phase, tends to increase. This inverse ratio between the amplitude of such successive reflexes appears to be a normal phenomenon as well as an alcohol result. 130 Effect of Alcohol on Psycho-Physiological Functions. liar results of one or two days which have exercised undue weight on the average differences. When from section n of the preceding tables all the signs for the differences in periods 2, 3, and 4 for the 30 measure- ments were counted separately for each day and classified under the three heads inferior, superior, and neutral functioning, the results in terms of per cent were as shown in table 27. Table 27. — Comparison of results for the six experimental days. June 29, normal. June 30, alcohol. July 1, normal. July 2, alcohol. July 3, normal. July 4, alcohol. Inferior functioning . . Superior functioning. . Neutral functioning . . 43 52 5 59 35 6 49 48 3 78 17 5 48 44 8 64 29 7 For normal days the condition is about 50 to 50, implying that obser- vations made following the preliminary period and the control dose stand even chances of showing superior or inferior functioning. On the other hand, it can be definitely stated that the differences on each alco- hol day show that inferiority is predominant in these periods. This condition or change to inferiority denoted by the difference between preliminary measurements and those made after the ingestion of 30 c.c. of alcohol should be designated as absolute, since it is not relative to the normal or control results. There is some indication that the alcohol effect increased from day to day. July 2 shows a higher per- centage of inferior signs than is evident for June 30; this is also true for July 4, notwithstanding a decided spurt on the part of the subject in certain voluntary processes on that day. In general it appears that although there is variation from day to day with specific measurements, no one day can be said to have exercised a predominating influence in producing the ratio of 9 to 1 for inferior functioning of processes in the repetition series. For comparison the results found in the Dodge and Benedict series have been entered at the right in table 26. In the column which gives the average for the normal group the proportion of signs indicating the alcohol effect as being inferior, superior, and neutral is 82, 12, and 6, these percentages comparing very favorably with the 90, 10, and for the average of periods 2, 3, and 4 of the repetition series. The signs for the two series disagree at three points. Dodge and Benedict's subjects were nearly all new to the eye-reaction test, and as there was a practice effect and as the first day was normal, this condition largely accounts for the percentile difference being minus. 1 Subject VI did not have a practice change in the repetition series, hence the alcohol effect had a better opportunity to show itself. The disagreement on eye reactions can therefore be discounted. On word reactions the average for the 1 Dodge and Benedict's report, pp. 89 and 250. Conclusions. 131 normal group is zero, but there were two, one of them Subject VI, who showed minus differences and 4 men plus differences. This is a real disagreement between the series, as is also that with the patellar reflex A', 30 grams, where Subject VI was the one man in 6 who had a plus per- centile difference. These disagreements in the direction of the alcohol Table 28. — Percentile effects of alcohol arranged in order of magnitude. Measurements. Second series. Av. of periods 2, 3, and 4. Dodge and Benedict series (normal group). Inferior functioning: Patellar reflex, amplitude of secondary (A') with 50- gram blow Mean variation of eye-reaction time Mean variation of pulse cycles with finger movements Lid reflex, amplitude of secondary (A') Eye movement, horizontal distance in 5" Lid reflex, amplitude of primary (A) Faradic threshold in /} units required to stimulate the fingers Patellar reflex, amplitude of primary (A) with 30- gram blow Eye-reaction time to peripheral stimuli Pulse-cycle duration, " average of other pulse data, Nos. lto7" Pulse-cycle duration immediately before muscle ten- sion Eye-movement time required for turning from left to right through 40° Pulse-cycle duration at time of finger movements . . . Faradic threshold in Z units required to stimulate the fingers Lid reflex, latency of primary (L) Patellar reflex, latency of secondary (L') with 30- gram blow Mean variation of memory saving Number of free finger movements performed in 8 seconds Length of respiration Patellar reflex, amplitude of primary (A) with 50- gram blow Pulse-cycle duration after muscle tension Lid reflex, latency of secondary (L') Patellar reflex, latency of primary (L) with 30-gram blow Pulse-cycle duration at the time of muscle tension .... Patellar reflex, latency of secondary (L') with 50- gram blow Mean variation of word-reaction time Patellar reflex, latency of primary (L) with 50-gram blow Superior functioning: Progress in memorizing four-letter words Word-reaction time for four-letter words Patellar reflex, amplitude of secondary (A') with 30- gram blow 37.5 27.1 26.0 24.7 20.0 19.7 38 104 10.7 16.4 16.2 11.5 60.3 S. 1 5.4 11.2 3.0 10.0 3.1 9.2 8.1 3.4 7.7 5.4 21.0 5.9 4.5 4.3 4.0 3.7 3.3 8.9 3.2 2.9 2.7 H60.3) 6.8 2.6 2.2 11.1 4.1 1.7 1.7 2 (4.0) 1.0 Hll.l) 2.5 4.4 1.4 17.0 2 (38) 1 Superior with Dodge and Benedict. 1 Dodge and Benedict made no distinction between intensities of stimulation for the patellar reflex in their final summaries. 132 Effect of Alcohol on Psycho-Physiological Functions. effect are not more striking than some that could be pointed out between others of the so-called normal subjects and the average for the normal group. The amount of the alcohol effect is variable. The different pro- cesses arranged in order of magnitude of the average percentile differ- ence for periods 2, 3, and 4 are shown in table 28. There are 7 instances in which the inferiority is less than 3 per cent and 16 cases in which it is less than 10 per cent. With any one of these relatively small effects, since the number of observations is also statis- tically small, the probable error would be large and a single value by itself could not be regarded as of much significance. The probability that the results indicate the true nature of the alcohol effect is greatly multiplied when many such results fall in the same direction. If the experiment had been limited to the 16 end results which showedaninferi- ority of less than 10 per cent and to the 3 instances of superiority, the case would still be quite definite for the alcohol effect. Of the 27 results showing inferior functioning, 11 range from 10.0 to 37.5 per cent. Prominent among these are 4 for reflex amplitude, with an average of 24.5 per cent, and 3 for pulse, which average 15.7 per cent. Other than this the repetition series does not indicate clearly that the more simple processes are affected most and the higher processes least by alcohol. 1 Grouping the 30 end results under convenient heads for averaging in comparison with Dodge and Benedict's average results for their normal group with dose A, the summary in table 29 is obtained. Table 29. — Comparison of alcohol effects found in two series of measurements. Percentage effect on measurements. Second series. Av. of periods 2, 3, and 4. Dodge and Benedict's normal group. 3 14 »7 «8.6 7 53 3.4 4.4 1 Average of pulse before, during, and after muscle tension, pulse during finger movements, and "Av. of other pulse data." * This includes memory, word reactions, and mean variations. The most conspicuous effect found by Dodge and Benedict was the decrease in reflex amplitude, which is confirmed by the second series, but is not relatively so large a decrement. 2 Two indications exist for the time relation of the maximum effect to the alcohol dose, viz, the predominance of sign and the average amount of the effect in the different periods. By the former it would seem, ac- 1 Dodge and Benedict's report, p. 245. ' For comments concerning patellar-reflex amplitude, see pp. 44 and 45. Conclusions. 133 cording to table 26, part in, that the maximum effect came in period 3, as here the proportion of signs stands 80, 17, and 3 for inferior, supe- rior, and neutral functioning, respectively. In no other period is there so large a percentage for inferior functioning. The average effect for each individual period, expressed in percentages and including all the 30 measurements, is: period 2, 11.7 per cent; period 3, 9.5 per cent; period 4, 4.8 per cent; period 5, 7.6 per cent; period 6 (s), 8.3 per cent. All are inferior except 8.3 per cent for period 6, which is superior, as previously noted. The maximum average amount of change is at period 2, but period 3 is not far short. The two indications for maxi- mum effect do not absolutely agree as to period, but in general it evi- dently occurs within 90 to 120 minutes after the dose. As the experi- mental periods in the repetition series were 50 to 60 minutes long and a larger variety of measurements (see p. 23) was given within the period, the data do not lend themselves readily to the problem of temporal incidence of effect. There are 16 percentile differences from the first series of observa- tions on Subject VI to compare with the results of the later experiments. Classified under the heads "inferior" and "superior," these stand exactly 50 to 50. In 11 of the 16 cases the differences have signs which the repetition series of observations verify regarding the nature of the alcohol effect on this individual. For example, the supe- rior progress in memorizing after alcohol is verified for this subject under the specified conditions employed. In the former series, how- ever, he did better than here, and was, in fact, the best of the three subjects who showed superior performance. Shorter word-reaction time after alcohol is another of these agreements. This one may more properly be looked upon as an individual peculiarity, as it was found with only one other subject (IX). The other agreements call for no special comment. There are 5 cases in which the comparable differences of the two series show conflicts as follows: patellar reflex A, 30-gram, lid reflex L and L', eye-reaction time, and pulse during muscle tension. The more or less detailed discussion of these conflicts given in connection with the presentation of data for the individual measurements may be sum- marized in part as follows: (1) The different results for patellar reflex A, 30-gram, are likely referable to the condition that the former data were quite limited, as there was only one normal and one alcohol day (dose A). The former result of slightly lengthened latency with increased amplitude in the primary reflex is abnormal according to the present basis of judgment, but this subject is unique in demonstrating a relatively small and in certain respects peculiar alcohol effect with the reflexes. It is further- more noteworthy that the reflexes alone show facilitation at the end of the day. 134 Effect of Alcohol on Psycho-Physiological Functions. (2) It was shown (p. 54 ff.) that anticipatory voluntary lid movement — another characteristic peculiar to Subject VI and one which was also very troublesome in the eye-movement measurements — changed the latency of the lid reflex, and that this factor happened to be so dis- tributed as to markedly influence the earlier results and make this con- flict probably accidental, although it must be noted that the average differences for inferiority in the second series are small, +5.4 and +2.7 per cent, and an occasional change of sign under such circumstances may be expected. (3) This subject, like the majority of others in the normal group, showed a practice change in eye reactions. He had but one normal day, which was the first for this test. The alcohol effect was therefore hidden in the former reactions and quite overbalanced by the practice change. (4) The differences opposite the heading "Pulse during muscular tension" are not strictly comparable, as is explained on page 92. The conflict is, however, not to be ignored. It is important to observe that in the repetition experiments, in which the records were continuous through rest, tension, and post-relaxation, the alcohol effect is on the average very small during muscle tension, and, in fact, is +1.3 (a slower pulse rate, the same as reported by Dodge and Benedict) in period 2, where for most other measurements the maximum alcohol effect occurs. A shorter pulse cycle during rest and a longer pulse cycle (relative to preliminary periods or to normal days) during brief periods of muscle tension are two phases of the same condition, viz, a depressed vagus control of the heart. It is not possible at present to state the time relations or the conditions which govern the appearance and predominance of one or the other of these results. The resulting data from the repetition experiments with Subject VI are the most complete thus far secured for any of the alcohol subjects used at the Nutrition Laboratory. These data show that concerning the nature of the alcohol effect this subject is not a physiological ex- ception according to present normal standards, since in 27 out of 30 measurements inferior functioning of processes was discovered after the ingestion of doses containing 30 c.c. of absolute alcohol. He does show some individual peculiarities. Conspicuous among these is the relatively 1 small effect on the reflexes. It is believed to have been primarily this, with the added conditions that insufficient data were obtained and that the differences from the reflexes made up a large pro- portion (6) of the experimental results, which raised the question about Subject VI in the minds of the former authors. 2 The two series of observations on this subject are in agreement with 11 out of 16 meas- urements which the former series contained. Analysis has proved that 1 Relative to the normal group of subjects used by Dodge and Benedict. 2 It is in connection with the patellar-reflex data that they give their long footnote concerning VI (see Dodge and Benedict's report, p. 55). Appendix I. 135 the disagreements are not without apparently reasonable explanations. It may naturally be expected that variation will exist in the amounts of the effects found in the two series. From the trend of the investiga- tion it is clear that had the first series been as extensive as the later series, a greater degree of correlation would have been found. A repetition series by an independent experimenter has proved, therefore, that this subject was affected by alcohol in substantially the same way as five or six other men previously studied. Furthermore, it has verified, in large part at least, the direction of the former results. These contributions are considered of far more significance than alcohol data equivalent in amount on some new and untried individual. The measurement of the alcohol effect from many standpoints on this one man is believed to have accomplished more in establishing results than would have been possible by following the characteristic statistical method and distributing the equivalent data in observations on a group of subjects. APPENDIX I. DATA CONCERNING THE USE OF ALCOHOLIC BEVERAGES OUTSIDE OF THE PSYCHO- LOGICAL LABORATORY BY THE SUBJECTS OF DODGE AND BENEDICT. In the personal history 1 of the Dodge and Benedict subjects, statements were made concerning the last use of alcohol. These statements refer only to the days immediately preceding the first experiment and not to the period of months during which the experiments were interspersed. They are therefore supplemented by excerpts from the original protocols. These data regarding the extra-laboratory use of alcoholic beverages are of significance in the inter- pretation of the alcohol results of the individual subjects. Subject II. — October 8, 1913 (normal), no alcohol in preceding week. No- vember 14, 1913 (normal), no alcohol during preceding week. November 20, 1913 (alcohol, dose A), no alcohol since the experiment on November 14 (a normal day). December 5, 1913 (normal), "3 glasses of cider during previous week." December 19, 1913 (alcohol, dose A), no alcohol during the week and no coffee. January 6, 1914 (normal) and January 13, 1914 (alcohol, dose A), no data given. February 3, 1914 (alcohol, dose B), no alcohol during pre- ceding week. February 17, 1914 (normal), no alcohol during preceding week. March 10, 1914 (alcohol, dose B), no alcohol during preceding week. March 17, 1914 (normal), no alcohol since March 10 (when a 45 c.c. dose was taken). These notes confirm the statement in the personal history that Subject II was a "very moderate user." They are of importance when it is remembered that of all the subjects compared for susceptibility to the influence of alcohol, 2 Subject II shows the most marked effects of the alcohol, his average percent- age being +5.1. Subject III. — October 1, 1913 (normal), no alcohol during previous week. January 5, 1914 (normal), seidel of beer and 2 highballs taken during previous week; "nothing yesterday." January 12, 1914 (alcohol, dose A), no data given. January 19, 1914, Monday (normal), "3 or 4 bottles of beer taken in previous week; last Saturday night, also 1 bottle of Burgundy." January 26, 'Dodge and Benedict's report. Appendix II, p. 276. s See Dodge and Benedict's report, p. 263, table 50, in which Subjects II, III, IV, VI, VII, IX, and X are discussed and compared. 136 Effect of Alcohol on Psycho-Physiological Functions. 1914 (alcohol, dose A), no data given. February 2, 1914, Monday (alcohol, dose B), " 1 cocktail and 1 bottle of beer taken last Saturday night." Febru- ary 9, 1914, Monday (alcohol, dose B), Wednesday, 1 cocktail, 1 pint sauterne; Monday, 1 glass beer." It is not clear which Monday is meant. February 16, 1914 (normal), no alcohol during preceding week. March 9, 1914, Monday (normal), "1 bottle of beer Wednesday night." According to table 50, pre- viously referred to, Subject III showed an alcohol effect less than the average; in fact he was below the average in 5 of 6 recorded measurements in that table. Subject IV. — September 27, 1913 (alcohol, dose A), last alcohol taken on September 24, 1 liter of beer at dinner. October 2, 1913 (normal), no alcohol taken during succeeding week. January 8, 1914 (normal), no data given. January 15, 1914 (alcohol, dose A), no data given. January 30, 1914 (normal) , "one-half pint whisky and 2 glasses of sherry taken for grippe on January 21-23." February 6, 1914, Friday (alcohol, dose B), "2 glasses of sherry; last evening 1 small glass of wine and 1 glass of beer." February 13, 1914, Friday (alcohol, dose B), "last Sunday night, 1 cocktail and beef cordial, 3 glasses of sherry; no alcohol yesterday." February 19, 1914 (normal), "2 glasses of sherry on the 17th, 1 glass yesterday." March 19, 1914 (normal), 1 glass of sherry and 2 cocktails taken during previous week; no statement made con- cerning their relation to experimental day. Subject IV ranked third in regard to the average effects of the alcohol. Subject VI. — See statements made on page 18 of this monograph. Subject VII. — October 6, 1913 (experimental data not published), 400 c.c. of beer on the afternoon of October 4. October 8, 1913 (normal), no data given. October 15, 1913 (alcohol, dose A), " 1 glass of ale last night; during the week had "ptomaine poisoning," and used 5 ounces of brandy, Jamaica ginger, and 4 ounces of wine; 3 pints of beer also taken since last experiment." The subject said on this day that he had a cold in the head and felt "dopey." October 17, 1913 (alcohol, dose A), no alcohol since last experiment. October 21, 1913 (normal), 60 ounces of beer during previous week; no statement of its relation to the experimental day. October 28, 1913 (alcohol, dose A), 60 ounces of beer during the week, 1 bottle taken preceding night. November 4, 1913 (normal), no statements recorded. November 11, 1913, Tuesday (alcohol, dose A), "1 glass of ale and one-half bottle of wine taken last Satur- day p. m." November 18, 1913 (normal), " during the week 3 glasses of beer, 3 glasses of port wine; "ptomaine poisoning" from steak Friday night, some cramps to-day." December 3, 1913, Wednesday (alcohol, dose A), " during the week 1 pint of claret, Saturday 24 ounces of beer, Monday 1 glass of ale, and Tuesday 2 glasses of ale." December 10, 1913 (normal), "2 glasses of beer," but no statement as to date. December 17, 1913 (alcohol, dose A), "1 quart of light wine and 3 ounces of claret, the latter taken last night." February 27, 1914 (alcohol, dose B), 7 glasses of wine, one of them taken last night. March 6, 1914 (normal), 2 glasses of wine and 5 glasses of beer taken during the week; 1 glass of beer taken previous night. March 13, 1914 (alcohol, dose B), 7 glasses of beer during the week, one taken each night. March 20, 1914, Friday (normal), "last Saturday, small bottle of wine; Monday, 1 glass of beer; Tuesday, 3 gin-rickeys." Subject VII ranked next to the last in regard to the average effect of alcohol. Subject VIII. — (Total abstainer.) He served on only four days, October 9, 1913 (normal), October 16 (alcohol, dose A), October 23 (normal), and October 30 (alcohol, dose A) . The series of experiments was thus incomplete and data were published for patellar reflex, faradic threshold, and finger movements only. No mention is made of him in Dodge and Benedict's summary chapter, and thus we have no detailed information as to the relative effect of alcohol Appendix I. 137 upon him. Concerning the patellar-reflex data, the authors state (p. 54 of their report, table 3, note 4), "his results are in the same direction (lengthened latency and decreased amplitude of the reflex) as the average of the group, but greater in degree." The effect of alcohol on the faradic threshold was in the same direction and practically equal to the average for the group. Subject IX. — October 11, 1913, the statement is made in the history by Dodge and Benedict as follows: Subject "drinks one-half to 1 bottle of wine or beer a day now, but previously 3 bottles a day, in the evening; no general effects. Largest amount taken, 4 bottles beer in the evening; did not feel intoxicated, but vomited. Last use, previous evening 1 bottle of beer; no effects. Never intoxicated. 2 or 3 liters of beer could be taken in the evening without noticeable effects. Sometimes produced vomiting next day. In excess of 2 or 3 liters it acted as a diuretic." October 20, 1913 (alcohol, dose A), no statements given. October 27, 1913 (normal), 3 bottles of beer 1 day before. November 3, 1913, Monday (alcohol, dose A), " during last week used 4 bottles of beer; Friday last, 3 bottles"; not discoverable whether the 3 was in addition to the 4, thus meaning a total of 7. November 10, 1913, Monday (normal), 2 bottles of beer previous Friday. November 17, 1913, Monday (alcohol, dose A) , 2 bottles of beer on the 15th, 1 bottle of beer on the 16th. No- vember 24, 1913, Monday (normal), "4 bottles of beer last Saturday night." According to the personal history in Dodge and Benedict's report, this amount was as much as the largest amount ever taken previous to the beginning of the observation. December 1, 1913, Monday (alcohol, dose A), 2 bottles of beer on the preceding Friday, and 2 on the preceding Wednesday. December 15, 1913 (alcohol, dose A), 1 bottle of beer taken 3 days before. December 22, 1913, Monday (normal, 12-hour experiment), "4 bottles of beer taken Satur- day." December 23, 1913 (alcohol, dose C, 12-hour experiment), no state- ments made. January 21, 1914 (alcohol, dose B), 6 bottles of beer during the preceding week. January 29, 1914, Thursday (alcohol, dose B), 2 bottles of beer on Sunday and 1 on previous Friday. February 20, 1914 (alcohol, dose B), 1 glass of beer two days before. March 3, 1914, Tuesday (normal), 1 glass of beer preceding Saturday. Subject IX, according to Dodge and Benedict's table 50 (page 263 of their report) ranked second in prominent alcohol effect, i. e., +2.0. Subject X. — February 11, 1913, largest amount and last time taken was on December 15, 1912, 2 glasses of champagne at dinner. No statements are made concerning the use of alcohol in the succeeding experiments, except for that on March 11, 1913, at which time it was stated that no alcohol had been used preceding the experiment. It is assumed that the same condition existed for the other days. This subject was fourth in the order of alcohol effect. It should be stated in this connection that he alone, of all the subjects, ate a full meal at the laboratory immediately before the experiment (see Appendix II, p. 139) and that only dose A was used on the 3 alcohol days of the 6 days on which he served as subject. In estimating the susceptibility of an individual to the effects of a dose of alcohol, account must be taken of the individual's habits. It is uncertain to what extent the regular use of alcohol develops a tolerance for it and so reduces the effects incident to the ingestion of a dose, or whether prolonged effects have such an influence on the normal or control experiments as to reduce the contrast between these and the alcohol experiments. The Dodge and Bene- dict series indicates that those subjects who are practically abstainers show more than an average alcohol effect. It may be said that in a short series of experiments this will be due to psychical factors. These experiments were, however, carefully designed to avoid this difficulty. Subject IV, and more 138 Effect of Alcohol on Psycho-Physiological Functions. particularly Subject IX, although regular users, were still affected more than the average, according to data in Dodge and Benedict's table 50. Judged in the light of their regular habits, these subjects are probably the most suscep- tible to the alcohol effect. Subject III is the least susceptible. He used less alcohol than Subjects VI, VII, or IX and yet shows the smallest alcohol effect. APPENDIX II. THE GENERAL EFFECTS OF ALCOHOL, DOSES A AND B, EXPERIENCED BY THE SUB- JECTS OF DODGE AND BENEDICT'S EXPERIMENTS. Dodge and Benedict made no attempt to collect complete introspective data from their subjects. Their measurements were planned to present objective results, free from the personal bias of either subject or experimenter. The protocols, however, contain the following notes, which are relevant to the im- portant question of what amount of alcohol may be considered a moderate dose. The general effects are furthermore of significance as sidelights on the susceptibility of the individual to the influence of alcohol as shown by experi- mental methods. Subject II — September 23, 1913, dose A 8 h 32 m p. m. 8 h 50 m p. m. dizzy when he moves; 9 h 10 m p. m., still feels dizzy, warm feeling in the stomach since he took alcohol; 9 h 30 m p. m., not dizzy when he moves. November 20, 1913, dose A, 5 h 15 m p. m. At 5 h 55 m p. m., dizzy when he moves; feels unnat- ural sleepiness. December 19, 1913, dose A about 5 h 30 m p. m. 6 h 12 m p. m., slightly dizzy, warmth in stomach. January 13, 1914, dose A, 5 h 30 m p. m. 5 h 55 m p. m., dizzy and sleepy; no doubt of alcohol effect; feels himself whirling around. February 3, 1914, dose B, 5 h 27 m p. m. 5 h 47 m p. m., subject tingling and numb all over, " on edge of a jag" ; later, feels dizzy and slightly nauseated. March 10, 1914, dose B, 5 h 55 m p. m. 6 h 05 m p. m., dizzy and numb, especially in arms and legs; 7 p. m., a slight nausea, sleepy, but "chipper" (cheerful); 7 h 30 m p. m., very sleepy and still somewhat nauseated. Subject III.— September 24, 1913, dose A, 7 h 50 m p. m. 8 h 02 m p. m., head feels full, face is flushed, feeling of well-being intensified; 8 h 17 m p. m., face still flushed, slightly drowsy when relaxed, not so elated as earlier; 8 h 35 m p. m., flush gone. January 26, 1914, dose A, 4 h 50 m p. m. To the experimenter the subject seemed mildly elated directly after the dose. 5 h 45 m p. m., sleepy. February 2, 1914, dose B, 4 h 57 m p. m. 5 h 50 m p. m., "feels like counting" to be sure of himself. February 9, 1914, dose B, about 4 h 30 m p. m. No marked dizziness experienced. Subject IV. — September 27, 1913, dose A, 4 h 04 m p. m. No comments made. February 6, 1914, dose B, about 4 h 30 m p. m. 4 h 57 m p. m., dizzy, head throb- bing, vision hazy; 6 p. m., urination; 6 h 18 m p. m., head somewhat clearer; doubts if he could have walked before. February 13, 1914, dose B, 5 h 05 m p. m. 5 h 25 m p. m., complained of headache; somewhat relieved by removal of telephone pulse-recorder from the head; dizzy; no elation, but depression and dullness. Subject VI. — See this monograph, page 20 ff. Subject VII. — October 15, 1913, dose A, 5 h 05 m p. m. 5 h 30 m p. m., sleepy; 6 h 20 m p. m., not so sleepy. October 17, 1913, dose A, 3 h 47 m p. m. 4 h 15 m p. m., during faradic threshold falls asleep. More extensive introspective notes are said to have been recorded on cards, but these are not available. October 28, 1913, dose A, 5 p. m. No comments made November 11, 1913, dose A, 5 p. m. 5 h 40 m p. m., notices narcotic, irresponsible feeling; 6 h 02 m p. m., less of the narcotic feeling. December 3, 1913, dose A, 4 h 40 m p. m., relatively little effect; sleepy at first; slight fermentation; hot feeling; gas after one-half hour. Appendix II. 139 5 h 37 m p. m., still some gas in stomach. December 17, 1913, dose A, 4 h 45 m p. m. No effect noted. February 27, 1914, dose B, 4 h 48 m p. m. 5 h 30 m p. m., feels hypnotic; 6 p. m., urinated March 6, 1914, normal experiment, began 4 h 25 m p. m. 6 h 20 m p. m., subject comments: "alcohol makes it easier to keep still." March 13, 1914, dose B, 5 h ll m p. m. 5 h 25 m p. m., says dose was sooth- ing; "I would know that I had taken alcohol." Subject VIII.— October 16, 1913, dose A, 5 h 20 m p. m. 5 h 50 m p. m., "sleepy since I took the alcohol"; 6 h 15 m p. m., not so sleepy. October 30, 1913, dose A, 5 h 43 m p. m. Said"the dose tasted frightful"; 6 h 04 m p. m., body numb, no control of feet, words just come out, touch blunted, stomach sensations hot or cold; 6 h 20 m p. m., marked incoordination, "extremities don't seem to belong to me"; 6 h 30 m p. m., "numb all over"; doesn't care what he does. It may seem strange that the subject makes all of these comments on this day when he made such slight comment upon taking dose A the first time. In this con- nection it is noted that on October 30, in preliminary statements, he said: "After taking the alcohol on October 16, 1 couldn't study." Subject VIII was not able to complete the experimental series. (See Dodge and Benedict's report, p. 25.) Subject IX. — October 20, 1913, dose A, 5 h 07 m p. m. No comments made on alcohol effect. November 3, 1913, dose A, about 6 p. m. No comments. November 17, 1913, dose A, about 5 h 30 m p. m. No comments. December 1, 1913, dose A, about 5 p. m. No comments on alcohol effect. December 15, 1913, dose A, about 5 h 50 m p. m. No comments. December 23, 1913, dose C, as explained in note 2, table 1, page 13, of this monograph. No mention of alcohol effect. January 21, 1914, dose B, 5 b 05 m p. m. Only comment, "subject slept after alcohol." January 29, 1914, dose B, 5 p. m. Began to perspire at 5 h 05 m p. m.; later talkative, felt fine and not so tired; spoke pre- viously of being tired because of examinations taken; 6 h 05 m p. m., urinated. February 20, 1914, dose B, 5 p. m. No mention of alcohol effect. Subject X. — February 18, 1914, dose A, 2 h 40 m p. m. Felt hot in stomach for first few minutes; unable to notice any particular change during the whole afternoon, except that in perhaps the very last period of word reactions atten- tion apt to waver; knew he was to take alcohol on this day. At end of first period after taking alcohol, distinct discomfort and passed a large quantity of urine. At end of experiment, no particular desire to urinate. March 4, 1914, dose A, 2 p. m. 2 h 30 m p. m., some gas, but not so much as on previous days and not so uncomfortable; slight headache; wouldn't have noticed it at usual work; no desire to urinate until 4 p. m. March 18, 1914, dose A, 2 p. m; 2 h 20 m p. m., gas, feeling of warmth in the stomach; none at 2 h 30 m p. m.; 2 h 45 m p. m., eyes feel heavy; tendency not to look up; in the eye movements, feels that the eyes move down instead of straight across ; 3 h 10 m p. m., slightly sleepy; 3 h 20 m p. m., wink stimulus felt to be 50 per cent louder than on previous days; 3 h 22 m p. m., very sleepy during word reactions; feels that the pause between the reactions was quite long; 4 h 21 m p. m., notices how automatic the word- reaction device is; surprised he can wake up and say the word so quickly; more sleepy than in previous experiment; poor attention. On each of the 6 experimental days on which Subject X served, he ate dinner at the laboratory immediately before the experimental session, 1 p. m. to l h 25 m p. m. The dinner was on all days as follows: cooked beefsteak, 170 grams; boiled pota- toes, 224 grams; butter, 12 grams; bread, 37 grams; water, 300 c.c. The experiments began at l h 30 m p. m. on each day and continued until 4 h 30 m p. m. In the case of all the other subjects used by Dodge and Benedict the previous food was taken from 12 h 30 m p. m. to 1p.m., and the experiments usually began from 4 to 6 p. m. Only dose A was taken by Subject X. 140 Effect of Alcohol on Psycho-Physiological Functions. APPENDIX III. OBSERVATIONS ON THE AVERAGES OF DATA FOR NORMAL AND ALCOHOL DAYS PRESENTED BY DODGE AND BENEDICT. The following discussion has absolutely no bearing or modifying effects upon any of the comparisons which Dodge and Benedict have made by what they term " the method of differences." It applies only to their discussion of aver- age values, which they have used as supplementary values in expressing the effect of alcohol on the various processes and occasionally have limited their comparisons to them. A common footnote which accompanied their tables states that "the values for the first period of the alcohol experiment were obtained before the alcohol was given, and are therefore not included in the averages." No specific statement is made as to whether the values for the first period on normal days were included or omitted in the averages for these days, which were to be compared with the average values for the alcohol days, but as their footnote implies and a recomputation of any of the data in their tables will show, the first-period values were included in the averages for the normal days. The alcohol and normal days would have been more directly comparable had the averages in both cases been computed without including the first period, which is termed "normal of the day," and is really preliminary. The point is a minor one, and produces no qualitative change in the conclusions offered by Dodge and Benedict, but as it is one the careful critic may notice, the following recomputations of the Dodge and Benedict data are presented : The pulse data obtained during mental and physical activity are given in Dodge and Benedict's table 40, pages 212 to 222 of their report. The mean variations between the lengths of individual pulse cycles for the respective periods are regarded as significant measures and are further presented in their tables 44 and 45. They form the basis of discussion of these data, with the conclusion that the pulse after alcohol shows a smaller mean variation and hence a decreased responsiveness of the cardio-inhibitory mechanism (see pages 238 and 239 of their report). A recalculation of the averages and the mean variations in their table 40 will show that the mean variation for the first period of an experimental day is usually smaller than the average mean varia- tion for the other periods, i. e., the heart rate was usually faster. If, therefore, we omit the first period in the averages for the normal day, as was done in the averages for the alcohol days, the final computed decrease of the mean varia- tions, as shown in their table 45, will be 22 per cent in place of 19 per cent. The method of presentation has to an appreciable extent masked the effect of the alcohol in this instance. The summary of the association pulse data in table 38, page 209 of their report, is an apparent exception to the usual method of computing the aver- ages; although no statement is made in the text, the averages for both normal and alcohol days apparently include the first-period values. This is the only observed instance in Dodge and Benedict's computations where the average for the alcohol experiment has included the normal of the day. These aver- ages, as presented in their table 38, are therefore directly comparable, but since the first period has been included, the alcohol section of the experiment is not so well isolated. If we accept the data in table 36, pages 202 to 205, as correct, then it is a simple matter to reconstruct table 38, as the preliminary data for it are given in the extreme right-hand column of table 36. In part i of table 30, Dodge and Benedict's table 38 is reprinted as published. Unfor- tunately this contained several errors. One of these errors is large, i. e., the average for Subject VI, Normal i and n, should be 806 instead of 876. In part ii the errors have been corrected, the method of computation employed being that used in the original table. In part in of table 30 the averages do Appendix III. 141 not include the values for the first period. It will be seen that in part in the values given in the column of normal averages are in every instance larger than their comparison values in part n of the table, which is another way of saying that the pulse was most rapid in the first period in normal days. The average values for dose A and dose B, part in, are in the majority of instances larger than the comparable ones in part n, but in general the leaving off of the first period on alcohol days has not increased the final averages, 879 and 886, as much as it did for the average of the normal days. Hence as a measure of the effect of alcohol in terms of the average we have —53 and —28 as compared to the previous —46 and —18. With the memory data the result of considering the averages in the way sug- gested is opposed to that just described. As the series of words used in these experiments varied, it seemed to the authors that the method of averaging the differences would be misleading, and that if they were to use the data at all, recourse must be made to the averages for a comparison. Reference to their table 20, page 131, will show that the "saving" in the first period is frequently much larger than in any other period during the day. It is clear that if the average for the normal day includes the "saving" of the first period and that for the alcohol day omits it, the comparison will give the normal day a Table 30. — Association pulse data of Dodge and Benedict as affected by changes in method of computation. [Values given in thousandths of a second.] Subject. Normal I and II. Alcohol. Effect of alcohol (alcohol-normal) . Dose A. Dose B. Dose A. Dose B. Av. Av diff Av. Av diff Av. Av diff Av. Av. diff. Av. Av. diff. Part I. 1 II... 1050 797 931 876 816 968 913 907 1050 797 931 806 876 968 913 897 1062 813 958 834 832 986 940 918 - 6 -10 -16 -17 - 9 -10 -16 -la - 6< -101 -14' -16' - 9 -io: -16 -12( ) 1018 5 769 i 890 2 1 799 2 862 L 862 1 867 ) 1018 5 769 L 890 T .... 1 799 ! 862 I 862 ) 867 1017 782 921 808 869 876 879 + ' - 7< -15. - 5( - 4- - 8! - 6) + i - 7! -15. -5! - 4( - 8! - 6! i 1040 ) 872 2 927 849 5 728 I 850 I .... i 878 5 1047 ) 872 > 927 848 ! 718 i 850 i » 877 1045 882 950 878 716 847 886 + 2f - 5! -11' -17: + 1 + 1! -a + k - 5! -13. -17( + i: + 1! -« 5 - 32 ) - 28 I - 41 ! 4-17 ) -106 - 51 ) - 46 ) - 32 ) - 28 ) - 41 : -17 ) -106 - 51 ; - 46 - 45 - 31 - 37 - 24 -117 - 64 - 53 +77 +27 +16 +35 +58 +73 +48 +77 +27 -11 +39 +56 +73 +43 - 10 + 75 - 4 - 27 - 88 -118 -29 - 3 + 75 - 4 + 42 - 98 -118 -18 - 17 + 69 - 8 + 44 -116 -139 -28 + 95 + 47 + 54 - 1 + 92 +121 + 68 + 79 + 47 + 9 - 9 +105 +121 + 59 Ill IV VI VII IX X Pabt II. II Ill IV VI VII IX X Pabt III. II Ill IV VI VII IX X Average 1 Part I is a reprint of the data published in table 38, page 209, of Dodge and Benedict's report, with signs as used by them. 142 Effect of Alcohol on Psycho-Physiological Functions. decided advantage. In the accompanying table, No. 31, in which the aver- ages computed by Dodge and Benedict are given in part i, it will be observed that the computation of the averages for the normal days without including the first-period values (see part n) has decreased the average saving for the effect of alcohol with all of the subjects except Subject VIII, who shows a saving of 1,027 as compared with 1,005. The only case in which the signs for the percentile effects have been changed from minus to plus is in the values for Subject IX, but all of the plus values have been increased in size and, except with Subject VIII, the minus values diminished. For purposes of comparison, average savings have been computed for the alcohol days, by including the first- period values for the respective sessions, the results being given in part in. The values in the normal column of part in correspond exactly to those in the normal column of part i. With this method of computation, the alcohol val- ues for memory are larger in every instance but one than those shown in the comparison column of part n. The final effect of alcohol, when thus more nearly isolated and given in per cent, is +1-4 as compared to +2.5 previ- ously reported. Computing the average normal in the way suggested has sometimes pro- duced only minor changes, as in the case of eye reactions and finger move- ments, showing the effect of alcohol as a little larger than stated; at other times, as in the case of word reactions and eye movements, the effect of such recomputations has been to show a smaller effect for the alcohol. With the association experiments there is no apparent difference; the averages are not discussed for the faradic threshold. These points have been raised solely with a view to j ustif ying the method of elaboration used in the second series of experi- ments and of emphasizing that when differences are available they should be given chief consideration. It must again be urged that changes in the averages of the normal days change in no way the comparisons which are based on average differences between the normal of the day and succeeding periods. Table 31. — Memory data of Dodge and Benedict of computation. as affected by changes in method Subject II. Subject VI. Subject VII. Subject VIII. Subject IX. Subject X. Average. Pabt I. Average saving: 932 939 +7 +0.6 858 939 +81 + 6.9 932 963 +31 + 2.6 761 856 +95 + 8.7 711 856 +145 + 13.3 761 914 +153 + 14.0 1,382 1,189 -193 - 13.1 1,298 1,189 -109 - 7.4 1,382 1,247 -135 - 9.2 1,005 974 -31 - 4.1 1,027 974 -53 - 6.9 1,005 909 -96 -12.6 921 881 -40 - 3.9 862 881 +19 + 1.8 921 903 -18 - 1.7 980 1,071 +91 + 7.0 978 1,071 +93 + 7.2 980 1,176 +1% + 15.2 -12 - 0.8 +29 + 2.5 +22 + 1.4 Effect of alcohol (alcohol- Part II. Average saving: Effect of alcohol (alcohol- Part III. Average saving: Effect of alcohol (alcohol- Appendix IV. 143 APPENDIX IV. OBSERVATIONS ON THE ASSOCIATION PULSE DATA OF DODGE AND BENEDICT. The measurements of the association pulse by Dodge and Benedict were divided into three divisions — pre-stimulus, stimulus to reaction, and post- reaction. The latter is sometimes mentioned as post-stimulation. (See their page 198.) In their figure 31 they have embodied diagrammatically the records of Subject VII. In these curves it may be clearly seen that an increase in pulse rate follows the stimulation of the association words. One point, however, deserved consideration in reference to these curves. It appears that the right end of the curve almost invariably fails to reach the level of the left end. The large fall which frequently occurred between the pre-stimulus and post-reaction ends of the curves in figure 31, taken in connection with the statement on page 198 of Dodge and Benedict's report that all the pulse waves were read and the data are capable of any other arrangement that statistical interest might demand, led us to reread the record which forms the basis for Dodge and Benedict's table 34, page 197. It will be observed from their figure 30, page 195, that there are unbroken horizontal lines which separate the reactions into groups of five. This was produced by breaking the circuit of the pulse-recorder so that the marker point gave a straight line. Sometimes this break in the circuit came quite soon after the fifth reaction, hence there are fewer post-stimulus cycles after the fifth reaction, and furthermore, the last figure in a fifth reaction was not joined directly to the first measured value of the next reaction. In Dodge and Benedict's table 34, page 197, it will be seen that the fifth reaction rarely has more than four pulse cycles in the post-reaction phase. In three cases, the fifth or last reaction in a set contained only three pulse cycles in its post- reaction phase. The two readings of the record checked remarkably well. In the final tables 35 and 36, from which figure 31 was drawn, many measure- ments are dropped out by limiting the number of columns. For example, the number of columns from stimulus to reaction is given in the average tables as 2, while in table 34 there were many times when three or more pulse cycles came within that period. The same may be said for the pre-stimulation and the post-reaction divisions of the tables. The averages at the foot of table 34 show that in 26 cases pulse cycles were omitted from the final averages, and that in 21 cases pulse cycles were omitted from the pre-stimulus section. Four of these cycles were of longer duration than the last which was included in their table; 12 were shorter and 5 of the same length. It is therefore cer- tain that if we plotted table 34 with the additional data, curve 2 for the first day in figure 31 (page 201) would connect at its ends on the same level. The discrepancies between the level of the ends of these curves and also in the data in table 36 are largely to be explained in this way. One point in the technique of Dodge and Benedict intensified this tendency to drop out data at the extreme ends of the table. This was the fact that there was no experimental break between the post-reaction and the pre-stimulation pulse cycles, since the records were taken on a kymograph, although an appa- rent break was produced when the record was cut to be removed from the drum. Since this apparent break was not always the same in its time rela- tion to the three phases of the record, the post-reaction phase was occasionally shortened and, in consequence, the pre-stimulation phase was lengthened, and trice versa. When the final tables were made up these extra long ends were not included in the averages, thus omitting the essential transitions between the two ends of the curves under discussion. 144 Effect of Alcohol on Psycho-Physiological Functions. In connection with the post-stimulation pulse acceleration reported by Dodge and Benedict in their association experiments, some account should be taken of the respiration phase. As stated by them (page 196 of their report) respi- ration records were incorporated with the association pulse curves. These are indicated in their illustrative figure 30 (page 195) by the long broken lines in the record. These must not be confused with the very regular short dotted lines which are the control time. In the discussion of the association pulse no account seems to be taken of the respiration records. Sample association pulse records were selected for each of the normal subjects, and these were prepared by scratching with a stylus a dot at the beginning of each inspiration. (The pointer of the respiration recorder was in contact with the drum during inspiration.) These dots were clearly marked and could easily be counted. The records, 8 in number, were then placed one above the other, so that their stimulus columns formed practically a straight line. A cord was stretched along the line so as to represent as nearly as possible the stimulus point of all the records. This cord was marked S. Other cords were placed parallel to S on both sides at distances of 5 centimeters, which represents the duration of 1 second. The dots representing the beginnings of inspiration were then counted and recorded. The results are given in table 32. Table 32. — Relation of respiration phase to association puke. Before stimulus. 1 to After stimulus. 4th sec- ond. 3d sec- ond. 2d sec- ond. 1st sec- ond. 1st sec- ond. 2d sec- ond. 3d sec- ond. 4th sec- ond. 5th sec- ond. 6th sec- ond. No. of inspirations begin- 62 5.6 139 12.5 108 9.8 81 7.3 140 12.6 207 18.6 56 5.0 95 8.6 155 14.0 68 ; 6.1 1 Total number of inspirations counted 1,111. It required 10 seconds for the kymograph drum to complete one revolution. According to the statement of Dodge and Benedict in table 13 (p. 114) the average reaction time in these experiments was 2.16 seconds. There was therefore a marked tendency for inspiration to occur or at least to begin after the stimulus word and previous to the reaction, since the percentages 12.6 and 18.6 are largest at this section of the table. The smallest percentage of inspira- tion occurs during the 1-second interval following the reaction. This same relationship of respiration to the association experiment is logical. Prepara- tory to reacting by vocalization the subject will naturally inspire and will tend to hold the breath until he reacts. The vocalization constitutes the first phase of the expiration. If the subject in this experiment tended to have in any degree sinus arrhythmia, there would be a more or less rapid pulse from this cause alone during and just after the reaction which would tend to be counted as pulse acceleration and due to the stimulation word of the associa- tion experiments. It therefore seems probable that the post-stimulation acceleration is due in part to the stimulation of the vagus by inspiration as well as the word stimulation of the experiment; this respiratory influence ap- plies also to the post-tetanus phase of the tetanus pulse records of Subject Vl.