itaacens Sm : parte oo AeA TN te eee ee er ne eee Ae Rn art at Le ND ae eli Be es, ys ha ; OR) a a i "Us hig etl iRiefit Ay : rY>4 ie Ls Na THE AMERICAN JOURNAL OF PRVSIOLOGY. EDITED FOR Che American Physiological Society BY H. P. BOWDITCH, M.D., BOSTON FREDERIC S. LEE, PH.D., NEW YORK R. H. CHITTENDEN, PH.D., NEW HAVEN W. P. LOMBARD, M.D., ANN ARBOR W. H. HOWELL, M.D., BALTIMORE G. N. STEWART, M.D., CHICAGO W. T. PORTER, M.D., BOSTON AMERICAN JOURNAL OF Piso LOG y VoLuME XVIII. BOSTON, Us. GINN AND COMPANY 1907 YU AN a Coprrighi, 1997 Joun W ‘sie ILSON AND Son, CAMBRIDGE, U.S. AL COE E NTS. No. I, FEBRUARY I, 1907. PAGE PROCEEDINGS OF THE AMERICAN PHYSIOLOGICAL SOCIETY . . .°. ix—Xxi OBSERVATIONS ON THE EXCRETION OF CARBON DIOXIDE GAS AND THE RECTAL TEMPERATURE OF RATS KEPT IN A WARM ATMOSPHERE WHICH WAS EITHER VERY MOIST OR VERY DRY. By F. F. R. Macleod I THE RELATION OF THE ACTIVITY OF THE EXCISED MAMMALIAN HEART TO PRESSURE IN THE CORONARY VESSELS AND TO ITS NUTRITION. mgs, Canine did Ff. TH: Bike nc Se ee ee ee AN APPARENT PHARMACOLOGICAL “ ACTION AT A DISTANCE” BY METALS AND METALLOIDS. By A. P. Mathews. . . Meee. ah ee THE REACTIONS OF CyYCLops TO LIGHT AND TO GRAVITY. By C. O. Ey ee we ee 8 ee a VE A THE CAUSE OF THE PHARMACOLOGICAL ACTION OF AMMONIUM SALTS. Pera Es HEROS 6 es Ce we ee eh ed ee eee eee Pe AGO THE RHYTHM OF THE TURTLE’S SINUS VENOSUS IN ISOTONIC SOLU- TIONS OF NON-ELECTROLYTES. By H. E. Eggers : 64 ON THE MECHANISM OF THE REFRACTORY PERIOD IN THE HEART. By MESMRIRIORNGIE Sw) wed ty <6 spe) ot ae ne Sues oe A ee OE A CONTRIBUTION TO THE CHEMISTRY OF CELL DIVISION, MATURATION, AND FERTILIZATION. .By A. P. Mathews . «+ - «© «+ « «© «+ « 8 No. II, MARCH 1, 1907. CONCERNING THE EXCRETION OF PHOSPHORIC ACID DURING EXPERI- MENTAL ACIDOSIS IN RABBITS. By R. Fitz, C. L. Alsherg, and Pye WACKMETSOR 6s wk ts 113 A NEw DECOMPOSITION PRODUCT OF GLIADIN. By Thomas B. Osborne tT CLAUD | 8 gs a. a a wt, Om RP al em ee me le ee ee ARE THE INFLUENCE OF DIGITALIS, STROPHANTHUS, AND ADRENALIN UPON THE VELOCITY OF THE BLOOD CURRENT. By Charles Wallis EES, Se a ote See ee re er er a, |, vl ' Contents. PAGE ON THE MECHANISM OF THE STIMULATING ACTION OF TENSION ON THE Heart, By A. F. Carlson ~ |. + ss Weenie meen ON ABSORPTION FROM THE PERITONEAL CAVITY. By H. Gideon Wells und Lafayette B. Mendel. . § . 2 = ee ee 156 THE EFFECT OF CARBOHYDRATES ON RESISTANCE TO LACK OF OXYGEN. BywVales 01. Packard . oe Mr ss ey 164 THE EFFECT OF INJURIES OF THE BRAIN ON THE VASOMOTOR CENTRE. By W..T. Porter and 1.'A. Storey: 2 2 S555 3 rr No, TT Arena. 1907; OBSERVATIONS ON NITROGENOUS METABOLISM IN MAN AFTER REMOVAL OF THE SPLEEN. Sy Lafayette B. Mendel and Robert Banks Gibson 201 THE DETERMINATION OF WATER IN PROTEINS. By Francis G. Benedict and Charlotte R. Manning. . . . ee 8 he la Anes ee rs EXTRASYSTOLES IN THE MAMMALIAN HEART. By Arthur D. Hirsch- Felder and 7. A. BE. Eystereen - = 2p ee ey oe re CONCERNING THE NEUTRALITY OF ProTopLASM. By L. F. Henderson ama O. F. Black’ 2 TB ee Fa ee te 250 THE MECHANISM OF EXPERIMENTAL GLycosuRIA. By Hugh McGuigan and. Glyde Brooks: «se es see E> pe ON Re AN 256 THE CAUSE OF THE TREPPE.. By fvederic.S. 206 & ~ 28s e- oe CONCERNING, GLYCOLYSIS.; -By G. WitHaGiie =e a ue oe 1) HYDROLYSIS OF PHASEOLIN. By Thomas B. Osborne and S. H. Clapp . 295 EFFECT OF PARTIAL STARVATION FOLLOWED BY A RETURN TO NORMAL DIET, ON THE GROWTH OF THE BODY AND CENTRAL NERVOUS SyYS- TEM OF ALBINO Rats. By Shinkisht VT ALGAE ree ua ee ee 309 CHEMICAL STUDIES ON THE CELL AND ITS MEDIUM.— ParT II. SOME CHEMICO-BIOLOGICAL RELATIONS IN LIQUID CULTURE MeEpIA. By AMOS IY. Peters. 26 ws So ere 321 No. IV, Maxam 1907. GASTRIC PERISTALSIS IN RABBITS UNDER NORMAL AND SOME EXPERI- MENTAL CONDITIONS. By Fohn Auer ey. sae 2 5 eee - 347 THE ANALYSIS OF URINE IN A STARVING WoMAN. Sy Francis G. Benedict and A. R. Diefendorf SoA cies, 40 a ae 362 THE ELIMINATION OF CREATININE IN WOMEN. Sy Francis Gano Bene- dict and Vector Caryl Myers .. «. . ain a ees See Se THE DETERMINATION OF CREATINE AND CREATININE. By Francis Gano Benedict and Victor Caryl Myers. . . .. . ‘oe oa 397 THE ELIMINATION OF CREATINE. By Francis Gano Benedict and Victor Caryl Myers 406 Contents. CONCERNING THE EFFECT OF CHANGES OF BLOOD PRESSURE PRODUCED BY TEMPORARY OCCLUSION OF THE AORTA UPON RESPIRATORY ACTIVITY. By 7. A. E. Eyster, €. Kh. Austrian, and C. R. Kingsley PHARMACOLOGIC INVESTIGATIONS ON THORIUM. By Torald Sollmann and E. D. Brown PRELIMINARY OBSERVATIONS ON THE POISONOUS ACTION OF THORIUM. By Arthur F. Chace and William F. Gies INDEX idles [Jae * -- EeOckE DINGS OF “THE AMERICAN PHYSIQ— LOGICAL, SOCIETY. NINETEENTH ANNUAL MEETING. NEw YorRK City, DECEMBER 27, 28, and 29, 1906. . . * PROCEEDINGS OF THE AMERICAN PHYSIOLOGICAL SOCIETY. OBSERVATIONS ON NORMAL GASTRIC PERISTALSIS IN” THE, RABBIT. By JOHN AUER. NorMAL gastric peristalsis may be studied in the fed rabbit without any operative interference whatsoever. The animal is merely stretched upon its back and the hair of the epigastrium clipped. A large part of the stomach is now visible, and the peristaltic waves may be studied by inspection or by means of a tambour. By this method a number of observations have been made, some of which have been reported elsewhere. In the starving rabbit gastric peristalsis is either greatly reduced or entirely absent. If now the stomach be moderately distended with air or water, powerful gastric waves appear at regular intervals. Ether and chloroform do not have the same effect upon gastric motility. When fuily under ether, the peristalsis is regular and pow- erful, often more regular than before the anzsthesis. Chloroform, on the other hand, greatly reduces the frequency of stomach waves. Morphine subcutaneously or intramuscularly abolishes stomach movements, Section of the vagi in the neck stops gastric movements as a rule ; there are some exceptions, however. FUNCTIONS AND STRUCTURES IN AMQC:BA PROTEUS. By Dr. C. F. HODGE anp O. P. DELLINGER. THE material for this study consisted of living amoeba, specimens killed in various ways and stained whole by different stains, and serial sections, also hardened and stained by different methods. xi xii Proceedings of the American Physiological Soctety. Contraction is effected by a meshwork of fibrillze, united into heavy trabeculz with wide intertrabecular spaces in the interior (endosarc) and into very fine trabeculz over the exterior (ectosarc). This mech- anism is all that can be demonstrated to account for all movements, locomotion, ingestion, egestion, contraction of vacuole and internal circulation. Movements are co-ordinated, but no differentiation of conducting fibrilla: has been clearly demonstrated. This material, when supplied with necessary nutrient substances, must possess the function of growth. | The function of digestion is mediated in all animals by gland cells, characterized by zymogen granules. The only structure in amoeba which is definitely granular is the nucleus, and sections show these granules apparently passing out of the nucleus into the food vacuoles. This accounts for all functions of the animal, reproduction being un- differentiated from growth and respiration and excretion and circula- tion being effected by movements of the whole body and supplemented by a similar action of the contractile vacuole. Except as noted above, sections reveal no differences between ectosarc and endosarc. THE INFLUENCE OF MECHANICAL ENERGY IN PHLORHIZIN DIABETES. By GRAHAM LUSK. MECHANICAL effort in a fasting dog made diabetic with phlorhizin only slightly increases proteid metabolism, but leads to a small in- crease in the output of sugar. This extra sugar may be derived from the residue of glycogen present in the animal. THE SPARING ACTION OF GELATIN. By, J. R. MURLIN (by invitation). ReiLty, Nolan, and Lusk showed several years ago that in diabetes gelatin may yield as much as 60 per cent of its weight as dextrose. An attempt has been made to determine to what extent this purely carbonaceous part of the gelatin molecule may account for the great Nineteenth Annual Meeting. x1ll sparing of the body's proteid when gelatin is fed. Experiments on dogs show that a diet limited to a quantity of carbohydrates equal to 12 per cent of the body’s heat requirement does not spare proteid, but entails a loss of nitrogen ; whereas 20 per cent of the body’s cal- orific requirement in gelatin alone spares 30 per cent or more of the starvation nitrogen. The same negative effect has been obtained in a student weighing 46 kilograms. After securing nitrogen equilibrium on a mixed diet containing 41 calories per kilo, in which nitrogen corresponding to the quantity eliminated in starvation was taken two-thirds in gelatin and one-third in meat, the gelatin was replaced by 60 per cent of its weight of dextrose. The sparing effect of the gelatin was entirely lost. The conclusion is that the sparing action of gelatin must be accounted for entirely by its nitrogenous constituents. SOME PHYSICAL REACTIONS OF PHYSA. By JEAN DAWSON (by invitation). Reactions of the snail to oxygen and to carbon dioxide.—In the lake and river systems studied, the conditions making up the habitats of Physa vary greatly, and there is a correspondingly great variation in the number of snails found in them. A comparison of the most favor- able habitats, wherever found, show strikingly similar conditions. They contain water that has the highest percentage of oxygen and the lowest percentage of carbon dioxide. A series of laboratory ex- periments corroborates the observations made in the field, Physa reacting positively to oxygen and negatively to carbon dioxide. Physa’s reactions to chemical stimult.— Only a small portion of the snail’s body is sensitive to chemical stimuli,and the snail responds negatively or positively according to the nature of the stimulating substance. The physiological state of the snail, due to the amount of the food taken, affects its reactions to both chemical and mechanical stimuli. A positive reaction, induced by the application of a chem- ical, inhibits the reaction to a second chemical which would normally produce an immediate negative response if applied to a fresh snail and vice versa. Physa’s food is solid, and if one of these snails come in contact with food while in motion, the animal receives both a me- xiv Proceedings of the American Phystological Society. chanical and a chemical stimulus; the resulting reaction depends argely upon the momentum of the snail and the character of the food substance. THE FAILURE OF REGENERATION OF THE SUPERIOR CER- VICAL GANGLION TWENTY-SIX MONTHS AFTER ITS REMOVAL. A DEMONSTRATION. By S. J. MELTZER. TWENTY-SIX months after the removal of the superior cervical gan- glion a subcutaneous or intramuscular injection of adrenalin still causes a long lasting dilatation of the pupil on the corresponding side. THE EFFECT OF SECTION OF ONE VAGUS. UPON THE SECONDARY PERISTALSIS OF THE CESOPHAGUS. By S. J. MELTZER anp J. AUER. THE prevailing opinion is that section of one vagus has no effect upon the peristalsis of the cesophagus. In a series of experiments upon dogs it was found that section of one vagus reduces considerably or abolishes the secondary peristalsis in the thoracic part of the cesoph- agus. By secondary peristalsis it is intended to designate the peristaltic movement of the cesophagus which is not preceded by an act of deglutition and which depends upon a chain of reflexes, as was previously described by Meltzer. Besides the impairment of the secondary peristalsis, the cardia becomes relaxed soon after section of one vagus. PERISTALSIS OF THE RABBIT’S CCECUM. By J. AUER anv S. J. MELTZER. Tue literature contains no definite statements with regard to the movements of the rabbit’s ceecum. When the abdomen is opened, practically no movements can be observed. However, the ccecum Nineteenth Annual Meeting. XV possesses well-marked frequent movements which can easily be seen through the normal skin and of which graphic records can easily be obtained. They are also well seen through the abdominal muscles after removal of the skin. But they disappear completely immediately on opening the abdominal cavity. The movements are arrested by stimulation of the vagi. This effect is best seen after administration of ergol. ARTIFICVAIS REGULATION OF (THE HEART RATE. By YANDELL HENDERSON. At the meeting of the Society in December, 1905, the writer stated that in an animal under artificial respiration the heart rate varies with the rate and depth of the respiration (z. ¢., with the extent of the pulthonary ventilation). Excessive pulmonary ventilation induces extreme tachycardia. He further suggested that a diminution in the CO, content of the blood resulting from the excessive respiration is the cause of this tachycardia. In a series of experiments upon dogs, the blood gases have been determined (by means of a Hill’pump). The analytical results and pulse tracings show a constant inverse relation between the CO, con- tent of the blood and the heart rate in animals under artificial respira- tion. This relation holds true also under natural respiration, except when the animal is hyperpnoeic. In this case the respiratory ex- citement induces a rapid pulse even when the CO, content of the blood is high. When the thorax of a dog has been opened, the heart rate can be regulated at the will of the operator by the following method. The bellows is connected with the tracheal cannula by a piece of large rubber tubing (15 mm. interior diameter and 50 to 250 cm. long according to the size of the animal). At both ends of this tube are ad- justable escape vents. If the vent near the tracheal cannula be open and that near the bellows closed, a rather rapid and full movement of the bellows induces a rapid acceleration of the heart rate. If the tracheal vent be then closed and the vent near the bellows be opened, the heart rate is gradually slowed. In one experiment be- tween 10 A.M. and 5 P.M. the heart rate was four times raised above 200 per minute, and as often slowed below 50. The animal was under morphin and ether. xv1_ Proceedings of the American Physiological Society. In experiments in which, after the opening of the thorax, the animals were maintained under natural respiration of compressed air (Sauerbruch-Brauer method), the increase and diminution of the respiratory “dead space” by means of the large tube with two vents (used in these cases to connect the tracheal cannula with a gasometer into which air was pumped at a constant pressure of 10 to 13 cm. of water), results identical with those above described were obtained. In experiments in which the thorax was not opened vigorous artifi- cial respiration also induced tachycardia, as did also the hyperpnoea resulting from the stimulation of an afferent nerve. Under ether it was thereafter difficult to regain a slow pulse rate, owing appar- ently to the difficulty of diminishing the hyperpnoea. Under deep chloroform anzesthesia, on the contrary, it was found that increasing the respiratory ‘“‘dead space” by attaching a tube to the trachea slowed the heart rate to 50 or fewer beats per minute. In experiments on men it was found that voluntary forced respira- tion continued vigorously for a couple of minutes induces a heart rate of 130 or more beats per minute. These observations seem to indicate that the maintenance of the normal heart rate is in large degree dependent on the uniformity of the CO, content of the arterial blood. This view harmonizes with the conclusion of the researches of Haldane and Priestley, that the respiratory activity is adjusted primarily to maintain a uniform CO, content in the blood. PRODUCTION OF ARTIFICIAL PARTHENOGENESIS IN AS-— TERIAS THROUGH MOMENTARY ELEVATION OF TEM-— PERATURE. By RALPH S. LILLIE. Exposure of the eggs of Asterias forbsit during maturation to tem- peratures ranging from 34° to 38°, for periods varying from thirty seconds to two minutes, is followed by typical membrane formation and by cleavage, which in favorable cases is regular and leads to apparently normal development. Bipinnariz indistinguishable from those reared from normally fertilized eggs have been obtained by this method. The most favorable results are gained by transfer of eggs during the early maturation period (z. ¢., within half an hour after removal) to sea water at 35°, 36°, 37°, and 38° for periods varying from fifteen seconds to eighty seconds; the higher the temperature Nineteenth Annual Meeting. XVI of the warm sea water, the shorter the optimum period of immersion; the eggs are then transferred to a large volume of sea water at the normal summer temperature (20°-22°). Fertilization, membrane formation, and cleavage follow; development is slower than normal, and a smaller proportion of eggs reach the blastula and later stages. Membrane formation is especially readily induced by this treatment, and is often shown by eggs with germinal vesicle intact. If the con- ditions of temperature or time of exposure are unfavorably adjusted, amoeboid movements of the egg protoplasm and irregular cleavage or fragmentation occur (independently of nuclear division) in a large proportion of eggs; the protoplasm of such eggs soon undergoes a coagulative change, followed by disintegration. THE ACTION OF BLOOD SERUM AND TISSUE EXTRACTS ON THE COAGULATION OF THE BLOOD. By LEO LOEB (by invitation). CERTAIN analogies exist between the action of salts in the coagula- tion of lobster blood, in the precipitation of casein and paracasein, and in the digestion of proteid by pancreatic juice. In each case the optimal amount of calcium can be split in a fraction which can be substituted by Mg or Na and presumably other kations, and in another fraction which can neither be substituted by Na nor (with the exception of the precipitation of casein and paracasein) by Mg. Analogies exist likewise in the influence exerted by tissue extracts on the coagulation of blood, on the coagulation of milk, and on the pancreatic digestion of proteid (action of exterokinase). It has been suggested that in each case different substances com- bine to form the active ferment. In the case of the coagulation of invertebrate blood it can be shown that tissue extract (tissue coagulin) and serum (thrombin) act both independently of each other on the fibrinogen. In vertebrate blood complicating factors arise; substances are found in the blood serum which accelerate and inhibit the action of tissue coagulins. But no fact is known which makes it likely that the relations between thrombin, tissue coagu- lin, and salts differ essentially in the clotting of vertebrate and invertebrate blood. xvili Proceedings of the American Physiological Society. THE FUNCTIONS OF THE EAR OF THE DANCING MOUSE. By ROBERT M. YERKES. Botu the static and the acoustic functions of the ear of the dancer differ markedly from those of the common mouse. Orientation and equilibration are fairly good. There is no evi- dence of turning dizziness. The whirling movement which is char- acteristic of the race appears as soon as the young -mouse is strong enough to stand. It is somewhat more pronounced in the female than in the male, and occurs chiefly toward evening. With respect to this movement there are three well-defined groups of dancers: those which almost always whirl toward the right, those which whirl toward the left, and those which whirl now one way now the other. At present I have no satisfactory evidence of the inheritance of the tendency to whirl in a certain way. Direct and indirect methods of testing acoustic sensitiveness have given negative results in the case of the adult, but the young dancer responds to certain sounds for from two to five days during the third week of life. This period of sensitiveness to sounds is preceded by a marked change in behavior. THE CAUSE,OF THE” DREPPE: By FREDERIC SS. LEE: THE treppe is usually ascribed to increased irritability caused by activity. The cause of the increased irritability has remained ob- scure. In studying the depressing action on muscle of its fatigue substances the author often observed augmentation of activity instead of depression. A more careful investigation of this phenomenon shows that it may be produced by all of the three recognized fatigue substances,— namely, carbon dioxide, monopotassium phosphate, and paralactic acid. When a muscle is irrigated with an indifferent fluid containing one of these substances in small quantity, and com- pared with its mate, irrigated only by the indifferent fluid, a fatigue record being made from both, more intense contractions frequently occur in the poisoned muscle at the beginning of the experiment, and may last until exhaustion sets in. When a fatigue record is being made from a muscle with the circulation intact, intravenous Nineteenth Annual Meeting. XIX injection of a fatigue substance causes augmentation of contraction. The author concludes that the treppe is due to the augmenting action of fatigue substances in small quantities, —the same substances which in larger quantities cause depression or fatigue. An excellent mode of demonstrating the augmenting action of CO, in the cat is to record the contractions of the Tibialis anticus in the living animal, and while the record is being made, to clamp the trachea. A marked treppe follows. If two corresponding muscles be compared, one with the circula- tion intact, and the other with the arteries ligated, the latter muscle performs more intense contractions and exhibits a more rapidly developing treppe, owing to the accumulation of fatigue substances. The chemical theory of the treppe is able to explain several other known phenomena. The author has experimented on both frogs and cats. The augmenting action of the fatigue substances is observed even when curare is employed. THE FORMATION OF FAT IN ANIMALS FATTENED FOR SLAUGHTER, By GEORGE T. KEMP anv L. D. HALL. THIs research was undertaken to determine, microscopically, how fat was deposited in the “marbling” of beef, and was extended to include a study of the relation of fat to the muscle cell, both histologically and chemically. The histological reagents used to stain the fat were osmic acid, Sudan III, and Scharlach R. We agree with the majority of observers in giving preference to the last. The beef was frozen, and cut with a Bardeen freezing microtome. Some tissues were fresh, and others were fixed in formaldehyde (and sodium chloride) before freezing. Even in the fattest tissues no fat was found within the sarco- lemma. It appeared to be confined exclusively to the connective tissue, as far as microscopical observations were concerned. Some specimens of very lean meat yielded a much larger percent- age of fat, by extraction, than could be accounted for by the fat which showed under the microscope. The theory has been put forward, that the toughness of meat is partially produced by a thickening of the sarcolemma. We found nothing to support this theory. xx Proceedings of the American Physiological Soctety. The following communications were also presented : THe MINIMAL PROTEID REQUIREMENT OF SOME HIGH PROTEID ANIMALS. By R. H. CHITTENDEN. THE Rare oF Loss or WEIGHT OF NoRMAL MAN. By W. P. LomBarp. THE Errecr OF MuscuLar ACTIVITY ON KREATININ EXCRETION; WITH PRELIMINARY OBSERVATIONS ON THE EXCRETION OF KREATININ IN HEALTH AND Disease. By P. A. SHAFFER (by invitation). THE OCCURRENCE AND FORMATION OF ALKYLAMINES AND ALKYEUREAS. By O. FOLtn. THE FORMATION OF SUGAR FROM AMINO-AcIDs. By W. SALANT. THE EFFECTS OF COCAINE ON THE LIVER. By G. B. WALLACE AND J. S. DIAMOND. ON THE ELIMINATION OF RADIUM IN NORMAL AND NEPHRECTOMIZED ANIMALS. By W. SaALaNnT and G. M. MEYER. THE RELATION OF INORGANIC SALTS TO LECITHIN AND KeEpHALIN. By W. Kocu. CONTRIBUTIONS TO THE PHYSIOLOGY OF THE PHOSPHATES. By C. L. ALs- BERG, L. J. HENDERSON, H. B. WEBSTER, AND R. FIrz. CONCERNING GLyco.ysis. By C. L. ALsBeERG anpD G. W. HALL. SoME NEw OBSERVATIONS ON THE ACTION OF Lipase. By A. S. LOEVEN- HART, G. PEIRCE, AND C. G. SOUDER. Nucieiss or CoprisH Ror. By J. R. ManpeL anp P. A. Levene (read by title). GLucoTuionic Acip In Pus. By J. A. ManpeL AnD P. A. LEVENE. PRELIMINARY REPORT ON THE ENZYMES OF UNFERTILIZED AND FERTILIZED Eccs. By E. P. Lyon anp O. P. TERRY. EXPERIMENTS ON ResusciraTIoN. By G. N. Srewart (read by title). ON THE SO-CALLED “ LIGATURE OF STANNIUS IN THE MAMMALIAN HEART.” By J. ERLANGER AND J. R. BLACKMAN. On THE MECHANISM OF THE SO-CALLED REFRACTORY PERIOD OF THE Heart. By A. J. CaRLson. On THE RELATION OF THE NORMAL RHYTHM TO THE SODIUM CHLORIDE RHYTHM OF THE Heart. By A. J. Cartson (read by title). ° VASOMOTOR REFLEXES. By W. T. PorTER. THE INFLUENCE OF THE DiGIraLis SERIES UPON THE VELOCITY OF THE BLoop SrreamM. By C. W. Epmunps. METHODs OF SrupyING Faticur. By F, S. Lee (a demonstration). DEMONSTRATION OF THE ADIABATIC CALORIMETER OF RICHARDS, HENDER- SON, AND FREVERT. By C. L. ALSBERG. 4 Nineteenth Annual Meeting. XX1 ON THE ALLEGED ADAPTATION OF THE SALIvARY GLanps To Diet. By F. P. UNDERHILL aND L. B. MENDEL. ADAPTATION OF SALiva TO Dier. By C. H. Netson (by invitation). THE EFFECT OF PHOSPHORUS STARVATION ON ASPERGILLUS NIGER. By W. Kocu anv H. S. REED (read by title). New CuHemicaL Facrs asoutT TENDON AND COMPOUND PROTEINS. By Wes). «Gies- (read “by. title). A FURTHER StupDy OF Peprotysis. By W. J. Gries anp W.- N. Berc (read by title). SOME OBSERVATIONS ON THE CESOPHAGUS AFTER BILATERAL Vacoromy. By W. B. Cannon. CONCERNING THE PHARMACOLOGICAL ACTION OF SaLicyLic Acip. By L. B. STOOKEY AND M. Morais (read by title). NucLeIN METABOLISM EXPERIMENT ON A DoG witH Ecx’s Fistuta. By P. A. LEVENE AND J. E. SWEET. PROTEIN ANALYysIS. By P. A. Levene, W. A. Beatty, D. R. MacLaurin, AND C. H. RUuILLER. PRESERVATION OF BLOOD VESSELS IN COLD SToRAGE. By A. CaRReEL (by invitation). TEE American Journal of Physiology. VOL. XVIII. FEBRUARY 1, 1907. NO: \E: OBSERVATIONS ON THE EXCRETION OF CARBON DE OXIDE, GAS AND THE RECTAL TEMPERATURE: OF RATS KEPT IN A WARM ATMOSPHERE WHICH WAS Pith: VERY, MOIST OR VERY DRY By J. J. R. MACLEOD. (WITH THE COLLABORATION OF J. D. KNOX.) [From the Physiological Laboratory, Western Reserve University, Cleveland, Ohio.]| N a recent paper? Haldane calls attention to the small amount of work that has been done on the influence of a warm humid at- mosphere on man. That such an atmosphere, in comparison with a dry one at the same temperature, has a most hurtful influence is a well-known fact, and it is believed that this is owing to interference with heat loss from the body: the evaporation of sweat being re- duced to a minimum in such an atmosphere. The exact degree of humidity and temperature necessary to bring about such a condition and the influence of other factors, such as the movement of the air, etc., have been little investigated in this connection. Haldane made observations on the rectal and mouth temperatures, and on the pulse, respirations, and general sensations of men in moist and dry atmospheres at high temperatures. He found, briefly stated, that in still air during rest the rectal temperature of men lightly clothed or nearly naked rises whenever the wet bulb thermometer 1 The following investigation was started over three years ago at the suggestion of LEONARD HItt, F. R. S., but had to be discontinued on account of pressure of other work. During the past summer the research was resumed and extended so as to include observations made during muscular work. The expenses of the re- search were partly defrayed by a grant from the Royal Society of London. 2 HALDANE, J. S.: The journal of hygiene, 1905, v, p. 494. I 2 J» f- R. Marlcod: reaches 88° F. and it continues to rise for some time (two and three- quarters hours) after coming out of the hot atmosphere. The higher the temperature of the air (as read on wet bulb thermometer), the greater is the rise in rectal temperature. In moving air (fifty-one metres per minute) the rise in rectal temperature is not noticed till the wet bulb thermometer reaches 93° F. In still air even moderate muscular work causes a rise in rectal temperature when the wet bulb thermometer is at 78° F. In moving air muscular work can be performed at 85° F. without abnormal rise of rectal temperature. Haldane noticed that the uncomfortable symptoms produced by warm air depend only toa certain extent on the rise in body tempera- ture; they are also, in part at least, directly due to the high ex- ternal temperature (wet bulb). He further points out that it is the temperature registered by the wet bulb thermometer which is of importance in these observations, and not that of the dry bulb. This fact is easily explained: of the factors which control the body tem- perature in man, by influencing the heat loss, the evaporation of sweat is undoubtedly the most important at high temperatures. In the above observations of Haldane’s we can accordingly account for the rise in rectal temperature as being due to the inefficient evapora- tion of sweat from the surface of the body in the moist atmosphere. In warm atmospheres, so long as the reading on the wet bulb ther- mometer stands well below that of the skin, evaporation of sweat will proceed freely, and the temperature of the air, when dry, may rise even to 250° F. without immediately influencing the body tempera- ture. On the other hand, whenever the reading on the wet bulb thermometer nears that of the skin, evaporation of sweat is interfered with, the body is not cooled down as it should be, and the rectal temperature rises in consequence. 3 In small animals the relative amount of body heat lost by evapora- tion is, in comparison with man, much less than that lost by radiation, conduction, and convection. Thus, in a man (70 kg.), of the total heat loss 71 per cent is by conduction, radiation, and convection, and 22.9 per cent by evaporation; whereas in a guinea pig (0.55 kg.) the heat loss by radiation, conduction, and convection is 93.5 per cent and by evaporation 6.5 per cent.! These figures relate presumably to ordinary temperatures; at higher temperatures the percentage loss of heat by evaporation will be much greater. In the case of the 1 HILL: Recent advances in physiology (Longmans, Green & To., New York), 1906, p. 267. Lixcretion of Carbon Dioxide Gas. 3 rat the proportion of heat lost by evaporation will be still lower than that in the guinea pig. The absence of sweat glands in the skin of the rat further shows that the heat regulation by this means does not occur, and although in hot atmospheres the quickened respirations will drain away heat for the evaporation of moisture in the expired air, yet it is probable that conduction, etc. from the skin is a much more important mechanism. On account of these considerations it seemed to us of interest to investigate the rectal temperature and the excretion of carbon dioxide gas of small animals kept in dry and moist atmospheres at high tem- peratures. In the following observations rats were chosen for this purpose and were kept in as warm an atmosphere as was consistent with safety. In one series of observations they were kept at rest, in another they were made to do muscular work; and the atmosphere was either as dry as possible or almost saturated with moisture. The amount of carbon dioxide expired by the rat while in the chamber was ascertained and, after removal, the rectal temperature was taken. Since the temperature of the chamber—just below 36° C.— was very near that of the rat’s body — about 38.3° C.— there could not be much heat loss from the latter by the processes of radiation, con- duction, and convection; on the other hand, in order to saturate the expired air with moisture, an amount of body heat sufficient to pre- vent the body temperature from rising might yet be drained away. The experiments here reported were conducted partly to show whether this source of heat loss is of much importance in the case of small animals; if it is of importance, we should expect to find that the body temperature and consequently the excretion of carbon dioxide rise higher when evaporation is more or less prevented by saturating the inspired air with moisture than when evaporation is encouraged as in dry air at the same temperature. In observations on mice recorded by Leonard Hill and myself,! we found that the greater heat conductivity of wet as compared with dry air made it impossible for many of the animals to withstand a moist atmosphere below 24° C.; for a time, in such an atmosphere, the increased heat loss was made good by increased heat production, but after a time this failed, the rectal temperature fell, and, if not removed to a warm atmosphere, the mouse died. In several preliminary ex- periments at temperatures a little above 24° C. we did not find moist air to have any deleterious influence. We thought it of interest to see whether at higher temperatures the moisture would influence the 1 Hitt and MAcLeEop: Journal of physiology, 1903, xxix, p. 508. 4 TF. Fe Se NaC leoe. body temperature and excretion of carbon dioxide in the opposite direction to that observed at lower temperatures. THE METHOD OF EXPERIMENT. Tame white rats were employed. The respiration chamber con- sisted of a large-sized desiccator containing in the case of the dry- air experiments sulphuric acid and in the case of the moist-air ex- periments water. The inlet tube entered at the side of the desiccator, and was connected either with a tower of sulphuric acid and pumice stone or with one containing a sponge soaked in water. The respi- ration chamber and towers were — placed in a zinc tank containing water kept at a constant temper- ature by means of a flame regu- ae | lated by a thermo-regulator. The ingoing air, previously to passing through these towers, was freed of carbonic acid by passing it through towers containing soda lime. It was also passed through a worm tube submerged in the water bath so as to warm it. The outlet tube, leaving the desiccator at the same opening as the inlet tube, led to a large Woulfe bottle containing pumice stone and sul- phuric acid and then to a calcium chloride tube. Beyond this were attached the weighed absorption tubes filled with soda lime and cal- cium chloride. Great care was taken to see that these were absorbing properly. A metre was connected at the beginning of the inlet tube and an aspirator at the end of the outlet tube. For the work experiments (see Fig. 1) a lid of hard wood was tightly clamped on to the desiccator in place of the glass cover. This lid was one inch thick, so as to prevent warping, and through the centre of it was passed a brass tube carrying an accurately fitting brass shaft twelve inches long and three-eighths of an inch in diameter. At about half an inch from its lower end this shaft was attached to the centre of a circular platform which formed the floor of the wider portion of the desiccator, the end of the shaft being prolonged beyond the platform and ending in a point which rested on a hollow brass disc, The brass disc was held in position by means of wooden stays. FIGURE 1. LEixcretion of Carbon Dioxide Gas. 5 The upper end of the shaft carried a pulley connected by a belt with gear wheels and a motor so that the platform was made to revolve at the rate of about twenty revolutions per minute. A piece of sheet zinc was suspended from the lid so as to make a partition on one side of the chamber. On coming against this the rat was compelled to work. The rate of ventilation in the different experiments varied between 50 and 400 c.c. per minute. When a slow rate of ventilation was employed, the chamber was flushed out with a quick stream of air before weighing the absorption tubes, so as to remove all traces of carbonic acid gas from the chamber. In the case of the rest experi- ments the rat was kept in a flat porcelain dish covered with wire gauze, and this was weighed with the rat at the beginning and end of the experiment. In the work experiments such a scheme could not be employed, and moreover the rat’s body often got smeared with the vaseline used to keep the lid air tight, so that it was useless to record the body weight. CONSIDERATION OF RESULTS The rectal temperature of the rats.— We have noted, in recording the rectal temperature of the rats, that great care has to be taken that the thermometer is well inserted in the rectum, as, on the diet of bread and water on which the rats were kept, the large amount of faeces may. quite markedly affect the reading on the thermometer unless it is well inserted. The normal temperature of the rats used in the experiments varied from 99.5° to 101.4° F., the average being 100.2° F. This is practically the same as that given by Pembrey.! The rectal temperature of rats kept at rest in atmospheres between 30° and 33° C. for considerable periods of time did not appear to show any difference according to whether the atmosphere was dry or moist. Below 35° C. the rats could be kept for any length of time without being killed, and the rectal temperature did not rise above 103° or 104° F. By referring to Tables II and III evidence on this point - will be obtained. At higher temperatures, namely, at 37° C., the rec- tal temperature quickly rose, and the rats could not be kept at such a temperature for much more than half an hour without being killed by hyperpyrexia. After being in such an atmosphere for half an hour, the rise in rectal temperature was practically the same whether the 1 PEMBREY: art. The chemistry of respiration; SCHAFER’s Physiology; YOuNG, J. PENTLAND: Edinburgh, 1898, i, p. 790. 6 FAG tex Macleod: air was moist or dry, as will be seen from the following table of re- sults (Table I)! In these observations the work apparatus was not employed, although in practically all of them the rats were moving about actively while in the chamber. TABLE I. Rectal temperature Temperature (Fahr.). of chamber. Nature of atmos- phere. Remarks. Before. After. 101.0 104.4 In all these observations the rats were kept for half 105.3 an hour in the chamber, after which they were removed 103.0 and the rectal temperature taken. 105.2 106.0 104.4 106.2 106.2 105.0 105.8 106.5 The excretion of carbon dioxide. — The results relating to the excre- tion of this are given in Tables II, III, IV, and V. In the ex- periments of which Tables II and III give the results the rats were placed in a porcelain. basin covered with wire gauze, and this was then placed in the respiration chamber. In connection with these it will be noticed that the general average of the results is the same in the dry as in the moist air. In certain of the preliminary ex- periments which we performed there appeared to be a distinctly greater excretion in the moist as compared with the dry air, but when the average of a large number of the observations is taken, and espe- cially when the same rat is employed for both the wet and dry air observations, it is evident that no such differences exist. Such dif- ferences as are occasionally noticed (e. g. in the case of rats 1 and 2) 1 There seems, from these results, to be a greater rise in the moist than in the dry air, but we believe this is accidental. Excretion of Carbon Dtoxide Gas. TABLE II. EXCRETION OF CARBON DIOXIDE IN GRAMS PER KILO BoDY WEIGHT AND PER MINUTE AT 30° C. A. In Dry AIR. . Rate of | Loss in Ee pena ventila- ates | weight of | CO, per Averages : . Eg Ponies tion in : | rat per minute of CO, weight In vation in rat on | ° ‘ e . p ee per | minute and kg. excretion. grams. minutes. : removal. if minute. | and kilo. I. 204 30 SOE 0.069 0.034 30 5Ha6 100.8 0.080 0.034 32 180 101.8 0.065 0.032 60 60 104.4 0.031 0.030 0.0325 ID Gealpe 30 boidc 100.8 0.067 0.032 100 290 seme 0.038 0.031 120 150 100.6 0.039 0.031 0.0313 TIT. 174 30 aieyous Sale 0.025 ? 0.032 30 130 101.2 0.061 0.031 0.0315 VE S9 30 | 0.102 0.026 30 | 0.059 0.032 0.029 Average of all observations 0.0310 B. IN MotIstT AIR. I. 209 130 110 100.8 0.039 0.031 200 170 102.6 0.037 0.030 180 160 101.8 0.031 0.031 0.0306 ID eli 60 130 103.0 0.058 0.028 170 90 orls 0.028 0.028 200 120 101.6 0.026 0.028 220 180 100.7 0.026 0.029 182 70 FOaC 0.027 0.027 0.0280 III. 166 165 130 0.027 0.029 0.0290 VE S3 200 90 103.6 0.033 0.034 250 150 101.0 0.040 0.035 200 180 101.3 0.033 0.032 185 160 100.8 0.048 0.037 0.0345 Average of all observations 0.0305 S. F. R. Macleod. TABLE ILE OBSERVATIONS ON THE RECTAL TEMPERATURE AND CARBON DIOXIDE EXCRETION (PER KiLo Bopy WEIGHT AND PER MINUTE) OF RATS KEPT IN A CHAMBER Ar e350) (G. A. In DRY AIR. Le Rate of > Loss of | eae, | Epes | ventila- | pester weight of | CO, per | Averages EC a el tion in, | cae eee rat per | minute | of CO, weight in| | vation in c.c. per | tat | minute | and kilo. | excretion grams. | minutes. bre re | removal. | andi 3 I. 208 40 210 | 100.6 0.066 sie 60 360 103.8 0.064 0.036 120 260 102.6 0.050 0.033 165 300 103.2 0.075 0.038 0.036 II. 183 120 170 100.6 0.056 0.030 120 220 102.2 0.045 0.030 100 290 105.0 0.054 0.035 120 210 103.6 0.062 0.033 200 220 356 0.059 0.032 0.032 III. 160 30 160 100.8 0.072 nee 200 410 101.8 0066 | 0.033 0.033 Min dhs} 60 180 102.2 0 072 0.037 120 220 101.6 0.078 0.039 150 330 102.0 0.047 0.031 200 180 101.2 0.060 0.032 0.035 Average of all observations 0.034 B. In MotsT AIR. - 1 135 60 102.8 0.076 0.038 143 30 ye) loys2 0.065 0.035 90 160 |, 02%3 0.056 0.036 0.036 ne 180 140 101.8 0.033 0.033 60 210 102.4 0.024 0.035 0.034 lel 130 40 102.6 0.086 0.038 160 10 101.7 0.079 0.037 0.0375 IV. 50 200 | 101.2 0.061 0.042 210 250 103.4 0.063 0.043 80 270 102.2 0.063 0.046 0.043 Average of all observations 0.0370 L:xcretion of Carbon Dioxide Gas. 9 are probably to be accounted for by differences in the extent to which digestion was proceeding at the time of the observations, for, although all the rats were on the same diet (bread and water), yet some of them must have been recently feeding when removed from the cage, others not for some time. The influence of body weight on the excretion of carbon dioxide is not so distinct as when the observations are made at lower tempera- tures.! It has been found, namely, that as the body weight decreases the excretion of carbon dioxide per unit of time and body weight becomes greater. This is due to the greater surface in relation to body weight in the smaller animals, so that there is a greater loss of heat from the surface of the body by radiation, conduction, and convection, and consequently a more active metabolism to make good the increased heat loss. Now, at the temperatures worked at in the observations here described, the heat loss by these means is reduced to a minimum, so that to maintain the body temperature scarcely any more active a tissue metabolism will be required by the small than by the large rats. It will further be noticed that at 33° C. the carbon dioxide excre- tion is higher than at 30° C. For all animals there is an external tem- perature at which the metabolism is at a minimum and above or below which it becomes increased. In the case of the dog Page found this temperature to be 25° C. So far as we are aware, it is not known for small animals such as the rat. This increased metabolism at high temperatures, acting along with the diminished heat loss, accounts for the rapid rise in rectal temperature which we have seen to occur when the temperature of the chamber is above 35° C.? In the observations of which Tables IV and V give the results the rats were placed in the work apparatus already described. The temper- ature of the chamber was not kept so constant as in the rest experi- ments, and was as a rule somewhat higher than in these. In Table IV, A, the results of the dry-air experiments are given, and in IV, B, those for moist air. In Table V it will be noticed that in each obser- 1 Vide PEMBREY: Loc. cit. 2 It will be noticed (Tables II and III) that the loss of body weight is often less in the wet than in the dry air experiments. In the former the urine and feces passed during the experiment will not evaporate, and since the inspired air is already saturated, or nearly so, with water, there will be little loss of water through the lungs. In the latter, on the other hand, the urine and faces passed during the experiment will dry out and the loss of water through the kings will not be inter- fered with. 10 Y. F KR. Macleod. TABLE IV. OBSERVATIONS ON THE RECTAL TEMPERATURE AND CARBON DIOXIDE EXCRETION (PER K1Lo Bopy WEIGHT AND PER MINUTE) OF Rats DOING MUSCULAR WORK IN A CHAMBER AT TEMPERATURES BETWEEN 33° AND 36° C. A. Dry AIR. Rectal Rate of temper- | CO, per | ventila- | Aver- ature of | minute | tion in | ages for Remarks. rat on. | and kilo.| cc. per) COs: removal. minute. Dura- pa tion of Tem- Weight | ? perature obser- of rat. Z of vation : chamber. (min.). | | | 33 5008 0.059 260 social i OOGIY Wirt ee rf 101.4 0.067 300 0.062 | In chamber 2 hrs. | 172.9(B) | Seah 0.067 ifm | | 0.059 eee | In chamber 2} hrs. 176.4 Po.ceee 0.061 | 0.067 0.100 0.120 aS 0.082 : In chamber 53 hrs. 0.076 “ice 0.073 hey: _ In chamber 2} hrs. 0.064 0.066 0.077 0.076 Ste | In chamber 3 hrs. 0.084 0.069 sie. In chamber 2 hrs. 0.073 0.074 nisicls 0.091 : In chamber 3 hrs. 0.072 0.075 BOO 0.062 104.0 OOSO eect In chamber 3} hrs. | Grand average Excretion of Carbon Dioxide Gas. TABLE IV (continued). B. Moist Arr. II Dura- Reni | Rectal | f Rate of | | ppetehe | 02 28| erature | Semmes | COs per|nente’| Aver | ee || ser- | | { | ages for emarks. | vation | amber.) Tat on |andkilo.) c.c. per} COz. | (min.). | ‘| removal. minute. | | | | BIG.O(R)b 70, | - 31.5. | 0.063 350 | 24 0.083 sie 30 | 0.054 0.067 | Dura’nofex.2dhrs. 157.9(R)| 30 31.5 | 0.046 30 | 0.0687 20) -.-- | 0,044 eee 10 | 100.8 | 0.069 0.057 | Dura’n of ex. 2 hrs. 1808(B)| 60 30.5 0.063 | Chamber not _per- 60 32.5 | 0.090 fectly air-tight. 60 | .... | 0.109 ied “4 60 | 102.5 0.089 0.088 | Dura’n of ex. 5 hrs. 23:8 ())} 40 32, 0.088 125 ADL Wy a320 | eee || (0.082 |'9 “225 eee My 30 | 101.2 0.081 66 0.083 | Dura’n of ex. 2 hrs. 118.9(L)} 30 35 0.071 100 | 30 | 35 0.086 50 60 35 | 0.054 Bee 60 35 | 0.116 50 30 Sy see | OO71 250 Bort | 30 B5e5 103.5 | 0.094 0.082 | Dura’n of ex. 4 hrs. 153.6(N)| 30 | 36-37 | .... | Oo71 | 330 | .... |20m.after removal 30 | 36-37 | 1062 | 0087 | 300 | 0079 | ‘temp. was 102. Rat was dead next morning. 160.4(R)|} 30 36-37 0.112 fast | After about 30 min. | stream rat was seen to be exhausted. Died a few hours after removal. 123-015)" 50 35 C079 | 8000") 30 35 beets 0.068 | | Hosts 40 35 102.0 | 0.063 | | 0.070 Grand average 0.075 FS. F R. Macleod: 12 vation the air was changed“rom dry to wet while the experiment was in progress. This was accomplished by changing the ingoing air current from passing through a sulphuric acid tower to one contain- ing sponge soaked in water and running water at the same tem- perature as that of the chamber down the inlet tube into the latter. TABLE. V: OBSERVATIONS ON THE CARBON DIOXIDE EXCREYVION OF RATS DOING WORK IN A CHAMBER HEATED TO ABOUT 33° C.. DURING THE FIRST HALF OF THE OBSER- VATION THE AIR OF THE CHAMBER WAS DRY, DURING THE SECOND HALF MOIST. Tem- Dura- Weight of rat in grams. Nature of air. | tion of | obser- perature | of cham- | CO, per minute and kilo. Averages for COs. } Remarks. vation. ber. 125.5 Dry Dry In dry air 90 min. Interval of 15 min. Wet Wet Wet In wet air 120 min. Dry Dry In dry air 90 min. Interval of 15 min. Wet Wet In wet air 90 min. Dry Dry In dry air 90 min. Interval 15 min. Wet Wet In moist air 90 min. At the start in these experiments the chamber did not of course contain sulphuric acid but had been thoroughly dried before placing the rat init. It will be seen that, so far as the carbonic acid excretion is concerned, there is no difference between the wet and the dry air exper- iments. In general the highest results both in wet and in dry air are obtained when the temperature of the chamber is high, but the rela- tionship of body weight to the excretion is, as in the rest experiments, not evident. As a general average the excretion of carbonic acid during work is twice as great as during rest. It is surprising that the increased tissue combustion, which this points to, did not cause a much more rapid rise in rectal temperature. Excretion of Carbon Dioxide Gas. 13 CONCLUSION. The deleterious influence of a hot humid atmosphere as contrasted with a dry one at the same temperature is not evident in the case of small animals such as rats.!. This is because such animals do not depend, to any extent at least, on evaporation of moisture in regulat- ing the heat loss from their bodies. 1 Such animals, therefore, will not be able to withstand very hot, dry atmospheres as well as man can. THE RELATION OF THE ACTIVITY OF Tibk fe xeiee MAMMALIAN HEART TO PRESSURE IN THE CORO NARY VESSELS AND TO IVS NUT RIaTone By C..C. GUTHRIE Ann EK. E PIIGE: [From the Hull Physiological Laboratory of the University of Chicago. | CONTENTS. Introduction ee. oy fs aera 2 ae en Methods). f<) st.) het a ee) GLE oe aah Up Ree il, “Abie Penilakon i the PresSUTe 20S RE Sea ee 2. Dhewregulation\of the temperature | yes ere eeren nen ren ec 3. The)preparation ofthe heart: “2 le.. = cceseee set 4. "The preparation of the: fluids.= o. > =) 42%. Ge ates te ete The nutrition of the excised. mammalian hearts) c=) es) ee ane eee eigen I. Theiaction’ of ‘the salt.solutions=) 3) sy ee meee anne ne Z. Dheleftect of blood and SerumdilutionSis suse ee een ene 3. The effect of milk and whey. . . peel gels. on be Oe re 4. The effect of hydrogen gas and foe sil. Be iit. hPa esas)! The relation of the activity of the excised heart to pressure in the coronary vessels . 22 Perfusion of the excised heart through the coronary veins. . . .. . . +. « 2 The effect of temperature on the excised heart . . . . . . 2. = oS oes Miscellaneous results, and discussion «©... .. =: + = 4 te ee Conclusions = .6.06 2 Se ae ee epee ct keg INTRODUCTION. HE experiments reported in this paper were begun primarily to test the relative efficiency of various inorganic salt solutions on the isolated mammalian heart under similar conditions of temperature and pressure, with a view to determining the most suitable fluid and the optimum conditions for restoring the mammalian heart to effi- ciency zz sztw. The results of Porter! and his pupils, of Langendorff,” and other workers, as well as observations of our own on mammalian hearts after restoration, both isolated and zz sztu, had led to the belief that the activity of the isolated mammalian heart bore a close relation to the blood pressure in the coronary arteries. This question was re-investigated by direct experiments on the excised heart. 1 PORTER: This journal, 1898, i, p. 511. * LANGENDORFF: Archiv fiir die gesammte Physiologie, 1895, Ixi, p. 291. 14 Activity of the Excised. Mammatan Fleart. 15 Porter! and Langendorff? have recently given reviews of the litera- ture. Only a few papers will be referred to in connection with various special points as they arise in the paper. Two preliminary notes ® have appeared. METHODS. 1. The regulation of the pressure. — A standpipe two metres high furnished the pressure for injection. It was provided with an inlet tube connected with the water faucet and an outflow tube at the bot- tom, and overflow pipes at intervals of 20 cm. along the side, the first being 60 cm. above the outflow pipe. The outflow tube led from the bottom of the standpipe to a large glass bottle filled with air. Air pressure was transmitted through tubing from this large bottle to the bottle containing the fluid used for perfusion. The height of the column of water in the standpipe was regulated by opening or closing the orifice at the lower end of the overflow pipes, which were con- tinued downward from the point where they left the standpipe to terminate at a uniform level below the opening of the outflow tube. Opening all the overflow pipes leaving the standpipe above a certain level, or closing all below this level, gave a column of water in the standpipe of any desired height, and permitted rapid changes of pressure, either from low to high or conversely. By making the inflow from the faucet larger than any possible outflow through the pressure bottle, the height of the column could be made constant to within a few millimetres of water. If necessary, the height of the water column could be increased 60 or 70 cm. by lowering the pres- sure bottle below the lower end of the standpipe. The pressure actually used for perfusion was, in most cases, recorded on a drum by a mercury manometer connected with the perfusion cannula in the coronary artery. The pressure used varied from 40 mm. to 240 mm. of mercury. ; 2. The regulation of the temperature. — A tank holding about five gallons was elevated above the table sufficiently to allow water to flow from an opening in the bottom of it into the top of the jacket of a large condenser. The outflow was from the bottom of the jacket. The tank was filled with warm water from the laboratory supply. If necessary, the water in the tank was heated by burners 1 PoRTER: American text-book of physiology, 1900, i, p. 179. 2 LANGENDORFF: Ergebnisse der Physiologie, 1905, abth. 2, p. 764. 8 GUTHRIE-and PIKE: Science, 1906, N. S., xxiv, p. 52; Biophysikalisches Cen- tralblatt, 1906, ii, p. 151. 16 C. C. Guthrie and F. H. Prke. placed beneath it. Thermometers were placed in the outlet tube from the tank and'in the condenser. The relatively large volume of the water in the tank prevented any great or rapid changes in temperature. The fluid used for perfusion was led from the bottle containing it through the worm of the condenser and then through the glass tubes with rubber connections to the heart to be perfused. The bulb of a third thermometer was placed in the fluid at the mouth of the injection cannula. 3. The preparation of the heart.— The animals were etherized and bled until respiration ceased. The heart was then rapidly excised and the pericardium removed, the blood being defibrinated in the meantime. If the heart was sufficiently large, as in dogs, cats, and rabbits, a cannula was usually introduced into the anterior coronary artery (following Porter’s method), although sometimes into the aorta. In the case of very small hearts, e. g., of guinea pigs and kittens, the cannula was introduced into the aorta. In a few experiments the cannula was tied into the coronary sinus and the perfusion made through the coronary veins (after the method of Porter and Pratt?). The heart was suspended by. the base and connected with a writing lever by a thread usually attached near the apices of the ventricles. In some experiments auricular tracings also were taken by a second lever. The fluid used was applied to the exterior of the heart before perfusion was begun. During the perfusion the heart was usually kept sufficiently wet by fluid escaping from the coronary vessels, but, when necessary, its outer surface was moistened with it. The hearts of guinea pigs, rabbits, cats, and dogs were used, but most of the experiments were done on cats’ hearts. A few turtle hearts were used for comparison. 4. The preparation of the fluids. — (a) Of the various current salt solutions, four were mainly used. The formulas are given below. HOWELL AND GREENE’S SOLUTIONS. Percents Gm. per litre. I? NaCl... 3) se oaves 7.00 Cal, ... — <6 -aouege 0.26 CI. 4.4 so) Goes 0.30 reNaCl 3)... eee 7.00 CaCl . «5 5 eae 0.66 1) a 2 0.040 0.40 1 PRATT: This journal, 1898, i, p. 86. * GREENE: American journal of physiology, 1898, ii, p. 106. Activity of the Excised Mammahan Heart. 07 LLocKE’s! SOLUTION. Per cent. Gm. per litre. RpaGlr. Ur sl! 27% etesgdo 9.000 DExtrose 4 —) A Gsaco 1.000 CaCl ne. «wee oe20 0.200 KCI) os,-oi7 |; Fe pore zZe 0.200 II. The same solution, with the addition of o.1 gm. of sodium bicarbonate per litre. Several of Ringer’s solutions and 0.9 per cent sodium chloride were used for comparison. Asa rule, no special technique was. employed for oxygenating the solutions other than shaking them thoroughly and letting them stand in contact with the air. Locke’s and similar solutions (also blood dilutions) were in some cases oxygenated by oxygen at atmospheric pressure and the heart was kept in an atmos- phere of oxygen. . Only slightly better results were obtained than by the usual technique. (2) Blood and serum dilutions, and the proteid containing fluids. — The defibrinated blood of the animal, diluted with three to ten vol- umes of 0.9 per cent sodium chloride solution, was more frequently employed than any other fluid. Washed red blood corpuscles in 0.9 per cent sodium chloride, and blood serum as free from corpuscles as possible, each diluted with several volumes of 0.9 per cent sodium chloride, were also used. The relative proportion of corpuscles in the washed corpuscle suspen- sion, and the relative proportion of serum in the serum mixture, were the same as in the defibrinated blood diluted 1 to 5. A hemoglobin solution, obtained by laking the red corpuscles by drying” them at room temperature and extracting with 0.9 per cent sodium chloride solution, was employed in one experiment. A solution containing the inorganic salts of the blood serum was made by diluting 250 c.c. of ox serum to one litre with distilled water, boiling and filtering. The filtrate was evaporated to dryness, the residue made up to 250 gm. with distilled water, extracted for twenty-four hours, and again filtered. This second filtrate was used for perfusion. Egg-white was dissolved in 0.9 per cent sodium chloride solution and used in a few experiments, but with unsatisfactory results. 1 Locke: Centralblatt fiir Physiologie, 1901, xiv, p. 672. 2 GUTHRIE: American journal of physiology, 1903, vii, p. 241. 18 C. C. Guthrie and F. H.. Prke. Another fluid was made from cow’s milk. The milk was allowed to stand on ice for some hours, after which the portion below the cream was siphoned off, strongly acidified with hydrochloric acid to precipi- tate the casein, and filtered. The filtrate was then rendered slightly alkaline with sodium carbonate. Before using, the alkaline filtrate was diluted by adding to it about three volumes of 0.9 per cent sodium chloride solution in order to bring down the concentration of the in- organic constituents other than sodium and chlorine, as the former exist in milk in greater concentration than in blood serum. The dilution also increases the proportion of sodium and chlorine, which exist in a much lower concentration in milk than in serum. To de- termine the effect of heating on the action of milk prepared in this way, a part of the filtrate was heated to 85° or 100° C. on a water bath for an hour or more and again filtered. This second filtrate, diluted as above with 0.9 per cent sodium chloride solution, was then used for perfusion. The milk used was obtained from a very large dairy, and was therefore of average and nearly constant composition. It is not difficult to see, from the analyses of milk given by Konig and Soldner,! that the inorganic constituents, with the exception of magnesium oxide and phosphorus pentoxide, would be present in our fluid in about the same proportion found in the artificial salt solutions. Sugar would be present in a considerably greater amount than in Locke’s solution, and of a different kind, namely, lactose. Washed hydrogen gas, prepared from metallic zinc and sulphuric acid, and commercial paraffin oil were the only fluids employed which were free from inorganic salts. THE NUTRITION OF THE EXCISED MAMMALIAN HEART. I. The action of the salt solutions. — All the salt solutions employed produced beats more or less rhythmical in character, which terminated in a comparatively short time, apparently either through the production of rigor in the ventricles or from exhaustion of the heart. Delirium cordis is relatively easy to produce by means of these solutions. The beat of the heart is, in general, more rapid and far more irregular than with fluids containing serum proteid. The external application of these solutions to the heart almost invariably produced contractions, if the heart had not been excised too long (Fig. 1). 1 Cited by HAMMARSTEN: Text-book of physiological chemistry, fourth English edition, 1904, pp. 448 ef seq. Activity of the Lacised Mammalan Heart. 19 Locke’s solution, containing 0.01 per cent of sodium bicarbonate, has given better results than any of the others employed. The rela- tion of the hypotonicity ! of Howell and Greene’s solutions to their stimulating effect, as well as their composition, must be considered in forming an opinion of their action on the mammalian heart, and the FIGURE 1.—Cat’s heart. The injection cannula was full of Locke’s solution when con- nected with the tube from the milk reservoir. The first beats are due to the Locke’s solution. Up strokes indicate systole, down strokes diastole. Time trace in seconds. relative isotonicity of Locke’s solution may be one reason why it conserves the beat of the mammalian heart longer than the other salt solutions used. 2. The effect of blood and serum dilutions, — Defibrinated blood, diluted with three to ten volumes of 0.9 per cent sodium chloride so- lution, conserved the beat of the heart for long periods of time, usually longer, in fact, than we cared to continue the experiment. The ven- tricles did not go into rigor, and delirium cordis never resulted at ordinary injection pressures. Serum, freed from corpuscles and diluted with 0.9 per cent sodium chloride, gave results as good as were obtained by the use of dilutions of defibrinated blood. Red blood corpuscles, washed free from serum and mixed with sev- eral volumes of 0.9 per cent sodium chloride, did not give as good results as dilutions of defibrinated blood or of serum alone (Fig. 2). The suspension of laked corpuscles did not give as good results as 1 CARLSON: This journal, 1906, xv, p. 351. 20 C. C. Guthrie and Fv H. Prke. the other fluids, and caused much trouble by clotting during perfusion. The fluids most frequently used were defibrinated blood and serum dilutions. All the defibrinated blood or serum mixtures produced rhythmical and co-ordinated beats, slower, as a rule, than resulted from perfusion with the salt solutions, and they continued for a much longer time. The aqueous serum extract, although doubtless containing some proteid, did not sustain the activity of the excised heart as long as the unheated serum. The action of the heart, as regards regularity and co-ordination of the beat, resembled more closely that following perfusion with Locke’s solution than that produced by perfusion with the blood dilutions. Greene! has used a similar aqueous extract of the serum salts of the terrapin on strips of terrapin’s ventricle. The action was more like that of diluted serum than that of Ringer’s solution. Undiluted serum produced no beats. 3. The effect of milk and whey.— Von Ott? first showed that milk and whey would sustain the activity of the frog’s heart. Ringer,® using I to 3 c.c. of milk to each 100 cc. of saline solution, kept a frog’s heart beating regularly fora long time. Bufalini and Torsellini* found that milk and whey, even when dialyzed, neutralized, and made up to the normal sodium content, were toxic to the turtle’s heart. The original paper is not accessible to us, and we are unable to learn whether or not the milk was diluted. As will be pointed out below, un- diluted milk will not long sustain the beats of the mammalian heart. Howell and Cooke® used an aqueous extract of dried milk, which contained the inorganic salts and the sugar, and were thereby able to sustain contractions of the frog’s heart for considerable periods. Milk containing caseinogen is unsuitable for perfusion of the mammalian heart, for the reason that the caseinogen coagulates in the coronary vessels and seriously interferes with the perfusion.’ It ' GREENE: American journal of physiology, 1898, ii, p. 111. * KRONECKER: Archiv fiir Physiologie, 1881, p. 569; V. Ort, /ézd, 1883, p. 1. 8 RINGER: Journal of physiology, 1885, vi, p. 364. BUFALINI and TorSELLINI: Bolletino dalla Societa tra i cultori di scienze mediche, Anno IV, No. 5, Sienna, 1886; cited by FINN, Zeitschrift fiir Biologie, 1905, XX1X, p. 320. ® HOWELL and Cooke: Journal of physiology, 1893, xiv, p. 209. * EDMUNDS: Journal of physiology, 1896, xix, pp- 466-476, has found milk-cur- dling ferments in testis, liver, kidney, brain, mesenteric lymph glands, thyroid gland, small intestine, ovary, and in a blood clot. Judging from the behavior of milk containing caseinogen, the presence of such a ferment in mammalian heart-tissue seems probable. 4 Activity of the Excised Mammalian Heart. 2I has been found, also, that if the undiluted whey fluid, after precipita- tion of the casein and neutralization of the hydrochloric acid, is in- jected, the activity of the heart is not long maintained. The fluid prepared as previously described gives results more nearly approxi- mating those of blood and serum dilutions than any other fluid em- ployed. The beats were strong, rhythmical, well co-ordinated, and maintained as long as we desired to continue the experiment (Fig. 1). The left ventricle did not go into rigor, even after being for twenty- Bisa a Al MWe bg ————— FH A by hy FIGURE 2.— Two-thirds the original size. Kitten’s heart. Perfusion with suspension of washed corpuscles in 0.9 per cent NaCl. Note the secondary series of beats following complete stoppage of the heart after the pressure has fallen to the base line. Time trace in seconds. four hours and more in contact with the solution. If the fluid is heated to 85° to 100° C. and filtered, the filtrate does not main- tain the activity of the heart as long as the unheated fluid, and rigor of the left ventricle is more likely to come on. The external application of the milk preparation to the heart did not cause contractions. 4. The effect of hydrogen gas and paraffin oil. The injection of hydrogen gas! into the coronary arteries was sufficient to cause rhythmical beats, not very strong nor long maintained, but fairly well co-ordinated and regular. The injection of paraffin oil into the coro- nary arteries, in accordance with Sollmann’s? statement, produced rhythmical contractions. We did not, however, first perfuse the heart with Locke’s solution, as Sollmann did. There could be, therefore, in our experiments no introduction of inorganic salts into the coronary arteries except such as were already present in the tissues. 1 MaGnus: Archiv fiir experimentelle Pathologie und Pharmakologie, 1902, xlvii, p. 200. 2 SOLLMANN: This journal, 1906, xv, p. 121. 22 C. C. Guthrie and F. H. Ptke. THE RELATION OF THE ACTIVITY OF THE EXcISED HEART TO PRESSURE IN THE CORONARY VESSELS. Tschirjew! found that increase of intra-cardiac pressure up to a certain limit produced in most cases a more rapid rate of the iso- lated frog’s heart. Ludwig and Luchsinger? found that the pulse rate of the isolated frog’s heart increased with pressure. This was more particularly true of the heart deprived of its sinus and of the gan- glion-free apex. Sewall and Donaldson,® working with the isolated frog’s heart, failed to get any noticeable variation in rate with change of intra-cardiac pressure. Stewart‘ states that in the isolated frog’s heart z sztu the rate is not generally affected by change of pressure. Martin® found that variations in the blood pressure did not affect the rate of the dog’s heart after it was severed from all connec- tion with the body except the connection with the lungs. The cause of Martin’s failure to obtain a change in the rate with a change in pressure will later be considered in detail. Magrath and Kennedy,° working with the isolated heart of the cat an stu and using defibrinated blood for perfusion, were unable to get any considerable change in the rate with change in pressure in the coronary arteries. The pressures which they used in their experi- ments (50 to 90 mm. mercury) were, in general, lower than those which we used, and lower than the normal blood pressure of a vigor- ous cat. As the normal cat’s heart is capable of maintaining a blood pressure of 150 to 200 mm. of mercury, it is possible that their range of pressure was not great enough to give the change in rate. With Locke’s solution and other salt solutions, the heart becomes arhythmical, and, as stated above, goes into delirium cordis at com- paratively low pressures, and the beat soon becomes unco-ordinated, so that change in rate with change in pressure is not so noticeable. With albuminous fluids such as blood and its dilutions and the milk preparation, and also with paraffin oil, the change of rate with change of pressure at constant temperature is striking. 1 TSCHIRJEW: Archiv fiir (Anatomie und) Physiologie, 1877, p. 180. 2 LupwiGc and LuCHSINGER: Centralblatt fiir die medicinische Wissenschaften, 1879, p. 404. ® SEWALL and DONALDSON: Journal of physiology, 1882, iii, p. 357- * STEWART: Journal of physiology, 1892, xiii, p. 140. ® MARTIN: Collected physiological papers, 1895, p. 25; cited by HOWELL: Text-book, 1905, p. 528. ® MAGRATH and KENNEDY: Journal of experimental medicine, 1897, ii, p. 13. Activity of the Excised Mammathan Fleart. 23 Starting with a low pressure, the beat has at first a slow rhythm with a comparatively small amplitude. As the pressure is gradually increased, the rate becomes higher, and the strength of the contraction increases until a certain pressure, which we shall call the optimum pressure, is reached, at which there exists the maximum rate without a decrease in amplitude. If the pressure is increased still more, the rate becomes higher, but the amplitude is decreased. As the optimum pressure of injection is exceeded, the diastole of the heart becomes less and less complete, until there is produced a tracing bearing a striking resemblance to the genesis of tetanus in skeletal muscle. Delirium cordis may result if a very high pressure is used. The relation between pressure of injection and rate of beat may best be shown numerically. This relation is as follows : — Pressure in mm. Rate of beat of mercury. per minute. ge Jo 140 138 156 162 175 204 104 96 Variation in pressure, 94.4 percent. Variation in rate, 126.6 per cent. With Locke’s solution, delirium cordis may be induced at a pres- sure often lower than the normal blood pressure of the animal. Working with blood dilutions, we have induced a very high rate at high pressures in the hearts of adult cats and dogs, but have only occasionally succeeded in inducing delirium cordis. With kittens’ hearts we have been able to produce delirium cordis even with the blood dilutions. If the pressure, after it has passed the optimum, be reduced, the rhythm becomes slower and the amplitude greater until the optimum is reached, below which, as the pressure falls still lower, there is a concomitant decrease in amplitude along with the change in pressure. The rate and the amplitude at corresponding pressures when rising and falling agree very closely. Extreme pressures some- times rupture the vessels so that the heart does not recover very well when the pressure is lowered. At low pressure, when the heart is first beginning to beat, there is a gradual increase in amplitude until a maximum for that pressure is reached, when there appears in the tracing a second beat, each one occurring between two of the stronger beats, barely rising above the base line at first, but gradually becoming stronger until it equals in 24 C. C. Guthrie and F. 1. Ptke. amplitude the original beat. On lowering the pressure, the second beat appears in inverse order, becoming gradually weaker until it ceases, after which the remaining beat diminishes in amplitude until it also finally ceases. This double beat is. shown in the tracings of Magrath and Kennedy,' and has also been observed by Cushny ;? and Mathews,? following the administration of substances of the digitalis series and of aconitin respectively. Dr. Lingle* has ob- served it in strips of the turtle’s heart. When the pressure of the injected fluid is reduced to about zero, the beat of the heart entirely stops for some minutes, following which there sometimes occurs a series of peculiar beats (Fig. 2). Finally these beats cease, and no more appear until the pressure of the injected fluid is again raised to a suitable level. This secondary group of heart beats bears a striking analogy to the “ secondary ” group of respirations of the Cheyne-Stokes type following cerebral anzemia.? The optimum pressure for the heart varies with the temperature and nature of the fluid, with the individuality of the heart, and also at different times during the experiment with the same heart. A higher pressure is generally needed to produce the same effect two hours after perfusion is begun than at the start. With the blood or serum dilutions or milk the optimum pressure for hearts in good con- dition is not far from the normal blood pressure of the animal, and the rhythm at this optimum pressure is very nearly the normal cardiac rhythm of the animal. On the other hand, the optimum pressure for the salt solutions is less than the normal blood pressure. If the perfusion is begun before the heart has céased to beat, a very low pressure is the optimum, and the change in rate with change in pressure is not so apparent. A sudden and great fall from high to low pressure may be followed in the excised heart (2) by a sudden slowing or even stoppage of the heart followed by the same rhythm previously observed for that pres- sure, or (0) by a very rapid rhythm which soon decreases to that previously observed for that pressure. 1 MAGRATH and KENNEDY: Loc. cit. * CusHNy: Journal of experimental medicine, 1897, ii, p. 233. 8 MATHEWS: J/bid., p. 593. * Personally communicated. ® STEWART, GUTHRIE, BURNS, and PIKE: Journal of experimental medicine, 1906, viii, p, 300. Activtty of the Lucised Mammalan Heart. 25 The first result bears a striking resemblance to partial or complete vagus inhibition in the intact heart, and the second is much like the acceleration of the heart zz sztaz which follows a sudden fall in blood pressure in the intact animal. Stewart+ has also observed the stand- still in diastole of the isolated frog’s heart when the intra-cardiac pressure is suddenly lowered. The change in rate is due to change in pressure alone, and not to a concomitant change in temperature. An increase of pressure pro- duces an increase in rate even where the temperature of the perfused fluid is falling. PERFUSION OF THE EXcISED HEART THROUGH THE CORONARY VEINS. The cannula was inserted into the sinus, and the fluid escaped through the coronary arteries or from the capillaries when the ventricles were freely incised. When blood and its dilutions were perfused, rhyth- mical beats of the heart occurred. The character of the beat differs somewhat: from that produced by perfusion through the coronary arteries. The amplitude is less, and the rate greater than would be produced by the same pressure in the coronary artery. The relation between the rate and the pressure in the vein in a typical experiment is as follows: Pressure in mm. Beats per of mercury. minute. fo. 39 ae 57 92 66 102 78 116 120 128 126 204 144 Variation in pressure, 191.4 percent. Variation in rate, 270 per cent. From the above facts we conclude (a) that the presence of a suit- able fluid somewhere in the coronary circulation at a suitable pressure is sufficient to cause rhythmical beats of a heart in proper condition. (6) The fact that the pressure in the coronary vein necessary to pro- duce beats must be high enough to permit the passage of the fluid into the capillaries is strong evidence that it is the presence or pres- sure of the blood in the capillaries, and not its passage along or pressure in the coronary artery or vein, which is the essential thing to produce rhythmical beats of the heart. 1 STEWART: Loc. cit., pp. 143, 153. 26 C. C. Guthrie and F.-H. Pike. THe EFFECT OF TEMPERATURE. No attempt was made to study exhaustively the effect of tem- perature on the heart, but some of the phenomena incidentally observed in the course of the experiments are given here. The effect of temperature on the excised mammalian heart is probably more marked than on the frog’s heart. The heart of the mammal may not work at all at a given pressure if the temperature is low, but as the temperature is raised, it rather suddenly begins to beat. As the temperature is raised still more, the pressure remaining constant, the beat becomes faster until the point of heat standstill is reached. The most favorable temperatures for the excised mammalian heart appear to lie between 33° and 37.5° C. Heat standstill in diastole of the heart occurs at about 38.5° to 40° C. when the milk preparation is used. These limits are considerably narrower than those found by Langen- dorff and Nawrocki,! who obtained beats of the cat’s heart at temperatures of 46° C. and above, and as low as 20° C. when blood dilutions were used for perfusion. The effect of the general nutrition of the excised heart upon its resistance to change in temperature and to heat standstill may be one further cause of the difference in result. It is not probable that the nutrition of the heart is as good when the milk preparation, particularly after heating, is used for perfusion as when blood dilutions are used, and because of this poorer nutrition, it is possible that the heart may succumb the easier to the higher temperatures. Although our observations on Locke’s fluid are too few to warrant drawing conclusions, there are some indications that it causes heat standstill at lower temperatures than blood. Robertson,” from a study of the relation between temperature and the rate of heart beat in a crustacean, concludes that a chemical reaction is involved in the rhythmically contracting heart. Gaskell® has supposed that, in the perfectly quiet frog’s heart, stimulation of the augmentor nerves cannot elicit beats, and that when the heart is apparently roused to activity, slight beats of the sinus or auricles have been going on with blocking of conduction at the auriculo-ventricular groove. Whether this is correct or not for " LANGENDORFF: Archiv fiir die gesammte Physiologie, 1897, Ixvi, p. 355. * ROBERTSON: Biological bulletin, 1906, x, p. 242. ® GASKELL: SCHAFFER’S Text-book, 1900, ii, p. 217; see also HERING: Archiv fiir die gesammte Physiologie, 1906, exv, p. 354. Activity of the Excised Mammalian Heart. 7 the frog’s heart, we can state that it does not hold for the cat’s heart in heat standstill at the minimum temperature necessary to produce the standstill. There are no beats of the auricles of the cat’s heart in this condition, but the whole heart is completely stopped. Further- more, we have many times caused contractions of the completely quiescent cat’s or dog’s heart by electrical or mechanical stimulation of the nervi accelerantes. Recovery from heat standstill, when milk or serum is used for perfusion, is entirely possible if the temperature be lowered, and the (i ee Ss SS eae ? FiGurRE 3.— One-half the original size. Heat-standstill of cat’s heart. Milk solution used for perfusion had previously been heated to 85° C. Temperature at mouth of perfusion cannula was 39° C. when heart stopped. Temperature 36.5° C. when heart again started to beat. Pressure remained constant. Time trace in seconds. heart may beat strongly and regularly afterward for an hour or more (Fig. 3). Heat rigor of the ventricles does not occur in heat standstill if the temperature is not permitted to go above 40° C. when the proteid fluids are used for perfusion. We confirm Herlitzka! on this point. Although Sollmann? neither looked for nor noted the connection between pressure in the coronary arteries and the rate of the heart beat, an inspection of his protocols shows this effect of pressure, but the accompanying changes of temperature obscure it to some extent. MISCELLANEOUS RESULTS, AND DISCUSSION. The difference between the reaction to pressure of the intact heart au situ and of the excised heart is striking. It is well known that in the normal animal with the cardiac nerves intact a high blood pres- sure is associated with a slow cardiac rhythm.? Nawrocki found that after section of both vagi and cervical sympathetics, changes in blood 1 HERLITZKA: Zeitschrift fiir allgemeine Physiologie, 1go5, v, p. 265. 2 SOLLMANN: Loc. cit. 8 See AUBERT: HERMANN’S Handbuch der Physiologie, iv, 1 t., p. 397; and Cotson : Archives de biologie, 1890, x, p. 431, for the earlier literature. See also Herlitzka: Archiv fiir die gesammte Physiologie, 1905, cvii, p. 557. 28 C. C. Guthrie and F. H. Ptke. pressure had no effect on the rate of the heart zz stv. Tschirjew found that in most cases great and sudden changes in blood pressure affected the heart even after section of all the cardiac nerves. A fall of blood pressure caused an increase in pulse rate, and a rise of blood pressure caused a slowing of the heart. MacWilliam! states that the heart is insensitive to changes of blood pressure after com- plete section of the extrinsic cardiac nerves. We have produced great and rapid changes in pressure by occluding by means of a liga- ture and releasing the thoracic aorta of cats, and have found that after complete section of the extrinsic cardiac nerves there is either no change in the pulse rate, or an increase in the rate with a fall in pressure and a decrease in rate with a rise in pressure. Even when all the extrinsic nerves are cut, the heart zz sz¢z does not follow the law of the excised heart as regards pressure changes. The conclusion seems inevitable that there is a local controlling mechanism normally present in the heart zz sz¢z which is inactive in the excised heart under the conditions of our experiments. The nature of this control- ling mechanism is unknown to us. From the well-known lower re- sistance of nervous tissue to injurious influences, ¢. g., asphyxia, and the greater difficulty in keeping it active in excised organs, one is inclined to regard this controlling mechanism as nervous in nature. On this view, the greater stability of the rhythm of the heart zz sztu in the face of changes of pressure would be due to an intrinsic nervous mechanism which Kaiser? postulates. It is obvious, also, from a consideration of the response of the excised heart to pressure changes, that the all or none law fails to hold under the conditions of our experiments. We have much evidence, obtained from a study of the heart zz sztu, that the loss of function of this mechanism, whatever it may be, in the excised heart, is closely connected with the temporary stoppage of the circulation The rate of the heart 2 sz‘, when started by per- fusion of a suitable fluid after comparatively long periods of stoppage, varies directly as the pressure in the aorta. The slow rate of the intact heart with high pressure is not due to the inability of the heart muscle to contract as rapidly against a high pressure as against a low pressure.® + MacWILu1Am: British medical journal, 1904, ii, p. 739- * KalseEr: Zeitschrift fiir Biologie, 1893, xxix, p. 203; éid., 1894, xxx, p. 279. See also HENDERSON: This journal, 1996, xvi, pp. 359-362. ® One of us (P.) has found in a single experiment on a dog that, after section of all the extrinsic cardiac nerves and the intravenous injection of atropine in sufficient Activity of the Excised Mammalian Fleart. 29 If the conditions of the experiment are such that the activity of this controlling mechanism is conserved, increase in pressure of the perfused fluid will not cause an increase’in the rate of the excised heart. These conditions may have been fulfilled in-Martin’s ! experi- ments, and also in Magrath and Kennedy’s? work. On this view, we can also understand why the isolated frog’s heart zw sztz should fail to respond to changes of intra-cardiac pressure by a change in rate. In some instances, when the excised heart has been beating slowly, the coronary arteries, or the tissues immediately surrounding them, have been observed to pulsate, and the beat to spread from the ves- sels over the ventricles. There is a peculiar shortening of the vessels, all drawing up toward the common point of origin of the larger ves- sels. If two strips of kitten’s ventricle, the first one taken from the left ventricle so as to include longitudinal portions of the larger branches of the anterior coronary artery, and the second cut from the anterior margin of the right ventricle, but including longitudinal por- tions of the smaller vessels only, be suspended under a tension of a few grams’ weight, attached to a muscle lever and irrigated with 0.9 per cent sodium chloride solution, the deportment of the two strips is very different. The first strip soon goes into tonus, the curve rising rather abruptly from the base line to a height which soon becomes uniform. Small, more or less rhythmical contractions may appear superposed on the tonus curve, but are not constant. The second strip shows no tonus phenomena at all, but soon begins. to beat more or less irregularly (Fig. 4). Howell? has obtained rhytimical con- tractions similar to those exhibited by the second strip of ventricle from a strip of vena cava taken from the terrapin. A longitudinal strip of a cat’s thoracic or abdominal aorta, suspended under the same conditions as the ventricular strips, shows the same deportment as the first ventricular strip containing the longitudinal portions of the coronary artery. Irrigation with Howell and Greene’s or Locke’s solutions will produce the same tonus effect on the strip of aorta as irrigation with sodium chloride solution. amounts to throw out the inhibitory function of the vagus, the heart zz sz¢z follows the law of the excised heart, showing a higher rate during high than during low blood pressure. 1 MartTIN: Loc. cit. MAGRATH and KENNEDY: Loc. cit. % HOWELL: American journal of physiology, 1898, ii, p. 57. to 30 C. C. Guthrie and F. FH, Pike. The area at the anterior margin of the right ventricle manifests a peculiar deportment in certain other points. This area is apparently very sensitive to many different influences. It oftens remains con- tracted when other parts of the ventricles are relaxed, presenting the appearance of tonus. Contractions, more or less rhythmical in char- acter, may be seen here when the rest of the heart is quiet. Occa- sionally, also, stimulation of the accelerators will cause contraction in this area when the rest of the heart fails to respond. On opening the thorax after asphyxiation of the animal, we have often noticed fibrilla- tions in this area while no movement was visible in other parts of the heart. Martius ! showed that the frog’s heart, after exhaustion from saline solution, could be revived by perfusion with saline solution contain- ing 3 to 5 mg. of sodium carbonate in 100 c.c. Martius was of the opinion that after exhaustion from the second fluid the heart could be revived only by perfusion with a fluid containing albumin. Howell and Cooke® state that such a heart may be revived by perfusion with a saturated solution of calcium phosphate in physiological salt solution, to each 100 c.c. of which were added 3 c.c. of a one per cent solution of potassium chloride (Ringer’s solution). Howell states that the beats of a heart perfused with Ringer’s solution were not so normal as those of a heart perfused with serum. Hearts perfused with a solution of inorganic salts of serum, minus the extractives, showed a tendency to form Luciani’s ® groups, z. ¢., alternate periods of contraction and rest. Gothlin,! working with frog’s hearts, found that the addition of serum proteid to a solution of the inorganic salts of the blood (Serumsalz- fliissigkeit) would restore the activity of the heart or prolong it for several hours. Similar results have been obtained by other observers, among them being White. White answers Howell’s contention by saying that the hearts Howell worked with were not fully exhausted in Martius’ sense of the term, and reaffirms the statement of Martius that frogs’ hearts which have ceased to beat on perfusion with Ringer’s solution, and which do not respond to strong electrical stimulation, can be revived only by perfusion with a fluid containing proteid. In a later paper ® Howell reaffirms his former position. 1 MArtTius: Archiv fiir Physiologie, 1882, p. 543. * HOWELL and Cooxe : Journal of physiology, 1893, xiv, p- 200. ® Vide HERMANN: Handbuch der Physiologie, iv, 1 t., p- 363. * GOTHLIN: Skandinavisches Archiv fiir Physiologie, 1902, xii, p. I. 6 WHITE: Journal of physiology, 1896, xix, p. 344. 6 HOWELL: American journal of physiology, 1898, ii, p- 50. Activity of the Excised Mammalian Fleart. 31 Baglioni! has shown that the inorganic salts are incapable of main- ' taining the activity of the Selachian heart, but that the addition of two per cent of urea to the fluid used would give a solution capable of maintaining cardiac activity for a long time. The mammalian heart is more sensitive to the action of salt solu- tions than the frog’s or turtle’s heart, and the irregularities are more noticeable than in the frog’s heart. Group formation is particularly likely to occur with Locke’s solution. Ifa heart which is exhausted Ul U Curl \ tree eam y, b “ Se FiGurE 4. — One-half the original size. Two strips of kitten’s ventricle suspended and irrigated with 0.9 per cent NaCl. The lower strip was cut so as to include a consid- erable portion of the anterior coronary artery. The upper strip was cut from the sensitive area at the anterior margin of the right ventricle. The relative conditions of the two strips some six hours later are shown at 4. from perfusion with Locke’s solution, and which shows Luciani’s groups, is perfused with blood or serum fluids, or with milk prepared as above, the groups may continue until the heart finally stops, or, in case of restoration of the heart, until replaced by a regular, rhythmi- cal beat. If the perfusion with Locke’s fluid is stopped for a time by removing the pressure, a more rhythmical beat may result on resum- ing the perfusion, but the groups soon reappear. When the mam- malian heart is nearly exhausted and perfusion is begun with the milk dilution, one group of beats may follow at a given pressure, but on the cessation of the beats of this group the heart will not contract again until the pressure is raised, when another single group may follow, more or less tardily, the rise in pressure. The upper limit of pressure is soon reached, and no more groups follow a still further increase in pressure. There is, then, a certain limit following per- fusion with salt solutions beyond which the mammalian heart is not readily revived by perfusion with serum or milk or other albumin- containing fluid, and this limit is apparently much lower than in the frog’s heart. 1 BAGLIONI: Zeitschrift fiir Allgemeine Physiologie, 1906, vi, p. 71. 22 C. C. Guthrie and F. H. Pike. These groups are not, as Gaskell! contends, due to a blocking of conduction from auricle to ventricle. The auricle of the cat’s heart is sufficiently large to watch with a lens with a considerable degree of accuracy. There were no contractions of the auricles during the period intervening between any two successive groups of ventricular beats. Asynchrony of auricles and ventricles is very common in excised mammalian hearts during perfusion with artificial solutions, but Luciani’s groups may occur independently of this asynchrony. The difference in the rhythm and character of the beat produced by perfusion with Locke’s solution, as compared with the beat pro- duced by perfusion with such a fluid as.the milk preparation, may be shown by leaving the injecting cannula full of Locke’s solution and connecting with the reservoir containing the milk preparation. When the pressure is turned on, the first fluid passing through the heart is Locke’s solution, and it is immediately followed by the milk. The pressure of injection being the same for both fluids, the contrast is somewhat striking (Fig. 1). The slower, more regular beat under the influence of the milk is sharply marked off from the rapid, irregular rhythm due to the Locke’s solution. When the perfusion is started with the cannula full of milk, the first beats are as regular as any of the others. It may be questioned whether hydrogen gas, as used by Magnus? and in.our own experiments, is an indifferent substance for the heart. Hydrogen is slightly soluble in water and the other fluids of the body, and it may be objected that the hydrogen molecules in solution may have some chemical action, but it is improbable that this would account for the effect obtained. Again, cottonseed oil, which undergoes metabolism in the tissues, may conceivably be acted upon by the ferments in the heart tissue in such a manner as to yield energy. While this is improbable, Sollmann’s® argument for a physical factor in the cause of the heart beat would thus be weakened. When the heart is previously flushed out with Locke’s solution, as in Sollmann’s experiments, it may be objected that enough of the inorganic constituents remain in the coronary system to cause the heart to beat when perfused later with paraffin oil, though this is improbable. There is small reason for believing that paraffin oil itself is acted upon by the tissues, or acts 1 GASKELL: SCHAEFER’S Text-book, 1900, ii, p. 227. 2 MAGNUS: Loe. cit. 8 SOLLMANN: Loc. cit. Activity of the Excised Mammatan Ffleart. 33 upon them, except in a purely physical or mechanical way. If no inorganic salts exist therein except such as are normally present in the tissues before perfusion, the evidence seems very strong that a purely mechanical stimulus, applied within the coronary vessels, will cause rhythmical beats of the excised heart. Howell and Cooke! state that “if one examines carefully the various experiments which have been made from time to time with the purpose of showing that blood does not act as a chemical stimulus to the heart, the objec- tion may be justly made that they do not definitely exclude such stimulation.” On examining into the phenomena presented in our experiments, two general explanations seem open to us. As it is impossible to get rid of all the lymph lying in the tissue spaces of the mammalian heart during the time of an experiment, we may first consider the relation of this lymph to the cause of the beat. When paraffin oil is used for perfusion, it is conceivable, and indeed probable, that the pressure of the lymph upon the muscle cells and nervous apparatus of the heart is increased. This increase of pres- sure may, in itself, so modify the nutrition of these cells that they again become capable of functioning. It is possible, also, that the lymph surrounding any particular ceil may be driven away by the pressure, and a fresh supply come in to take its place. The source of such lymph, however, is problematical. The lymph might there- fore be regarded as the direct cause of the heart beat, acting inde- pendently of the nature of the fluid exerting the pressure in the coronary arteries. Since the inorganic salts of the blood are present in lymph also, we cannot, on the above view, absolutely exclude them as a possible cause of the heart beat. Since the proteids are likewise present in lymph, it is equally impossible to exclude them from a similar rdle. A crucial experiment to determine whether the inorganic salts of the blood are the direct cause of the heart beat is therefore still lacking, but it appears to us that mere intra-vascular pressure can no longer be excluded as an indirect cause of the heart- beat under the conditions of our experiments. As asecond possibility, we may consider the intra-vascular pressure to act as the direct cause of the heart beat. When the cannula is inserted directly into the coronary artery, the pressure at the mouth of this artery, and probably in its larger branches also, is the same as 1 HOWELL and CooKE : Journal of physiology, 1893, xiv, p. 218. 34 C. C. Guthrie and F. 1. Pike. that in the cannula. Ceradini! showed that, during the ventricular systole, the semilunar valves do not, as was previously supposed, close the openings of the coronary arteries. Rebatel? and Porter? have made further studies of the coronary circulation. Porter has shown that, in the living animal, the pressure curve in the larger branches of the coronary arteries does not differ materially from the pressure curve in the carotid. The pressure conditions at the mouth of the coronary artery are, therefore, essentially alike in the intact and in the excised heart during perfusion. We may, then, assume that the whole coronary circulation is essentially the same in the two cases. Rebatel has measured the velocity of the flow through the coronary arteries and finds that there is a sudden decrease during systole. According to Porter, we may consider the intra-mural capillaries as empty at the close of the systole. During diastole these capillaries are opened, and the pressure is, for a time, very low. The capillaries do not long remain empty, for the blood from the !arger branches of the coronaries rushes in, and the pressure rapidly rises. As above stated, we regard the circulation in the capillaries as the one neces- sary for the production of rhythmical beats, and the circulation in the arteries and veins as important only because it is a necessary condi- tion for the circulation in the capillaries. There would come then, sometime during the diastole, a mechanical stimulus to contraction in the form of an increase of pressure in, and velocity of flow through, the intra-mural capillaries. This stimulus would have a certain threshold value which would in general increase as the experiment proceeded. (It has already been stated that, in general, a higher in- jection pressure is necessary to produce the same effect as the exper- iment proceeds.) As the pressure for injection is increased, it is not difficult to see that the intra-mural capillaries would be filled more rapidly, and that the pressure therein would reach the threshold value more quickly after the beginning of diastole, thus causing a contrac- tion sooner than it would occur with alow injection pressure. The chief variation in the cardiac cycle is the variation in the length of diastole, and when the pulse rate is high the diastolic period is very short.‘ 1 CERADINI: Il Meccanismo delle valvole semilunari del cuore, Gazzetta Medica Lombarda, 1871; Opere, ii, p. 437; Milan, 1906. * REBATEL: Recherches experimentales sur la circulation dans les arteres coronaires, Paris, 1872. 8 PoRTER: This journal, 1898, i, p. 145. 4 HENDERSON: Loc. cit. Activity of the Excised Mammahan Ffleart. 35 The high injection pressure therefore, by causing the stimulus to reach its threshold value earlier in diastole, gives us a primal condi- tion for a high rate, — the shortened diastolic period. As the pressure is increased still more, the impulse to contraction would come before the completion of diastole, and the amplitude of the beat be diminished because of the consequent incomplete relaxation of the ventricular ~ walls. Finally, at extremely high pressures for injection, the stim- ulus to contraction would come almost at the instant diastole began. The relaxation of the ventricle would be very incomplete and the amplitude of the beat greatly decreased, but the rate would be very high. The tracing from a heart under these conditions bears a strong resemblance to the genesis of tetanus or even to incomplete tetanus. At the highest pressure the amplitude of the systole, as well as the length of the diastole, is apparently greatly decreased. It becomes a question, therefore, as to whether the ventricles are able to contract fully and completely empty the intra-mural vessels when extreme pressures are used for injection. And furthermore, it is but a short step from this condition to delirium cordis. Indeed, the heart may be said to have been in delirium at the time of the highest pressure. It is therefore not necessary, for purposes of explaining the origin of the beat and the regulation of the rate in the excised heart, to postulate any action of inorganic salts. Since blood, although containing the inorganic salts, produces no contraction of the heart when applied externally, but does produce a rhythmical beat when perfused, at a suitable pressure, through the coronary vessels, it appears to us that the pressure alone may be a factor in producing the beat of the normal heart. The pressure is, in all probability, not the only factor in the pro- duction of the normal heart beat, since the excised mammalian heart may continue to beat without perfusion for a time after removal from the body. Czermak and von Piotrowski! observed that the excised hearts of rabbits beat for periods varying from three and one-fourth minutes to thirty-six minutes after removal. The mean period for sixty hearts was eleven minutes and forty-six seconds. The pressure in the coronary arteries, if any existed, must have been extremely low, but that they were completely empty is not probable. It should, however, be borne in mind that in the intact mammalian 1 CZERMAK and VON PIOTROWSKI: Sitzungsberichte der Kaiserlichen Academie der Wissenschaften zu Wien, 1857, xxv, p. 431. 36 C. C. Guthrie and F. H1. Prke. heart the stimulus necessary to be applied within the intra-mural capillaries to cause contraction has, in all probability, a vastly lower threshold value than in the excised heart after stoppage. The mere entrance of the blood into the intra-mural capillaries may be a suffi- cient stimulus to cause a contraction. At the moment of excision we must suppose this threshold value of the stimulus to be un- changed. There is still some blood in the coronary arteries, and this may find its way into the capillaries during diastole and afford a suf- ficient stimulus to contraction. But the heart tissue, either from the lack of oxygen or impaired nutrition, possibly from both, gradu- ally loses its irritability until the threshold value of the stimulus ne- cessary to cause contraction rises above that which the blood in the capillaries is able to exert, and the heart stops. Certain it is that an extremely low pressure for injection will give as great activity of an excised heart which has barely ceased to beat as a much higher pres- sure will evoke in the same heart some hours later. If pressure is not the only factor concerned in the production of the heart beat, can we find a further important factor in the inorganic salts of the perfusion fluid? There is good indirect evidence to the contrary, and this is supported by observations on the effect of such solutions on the other tissues. As examples of such action, we may take the following instances. Of the many attempts to produce artificial parthenogenesis of the eggs of the sea urchin and other lower forms of animal life, not one has as yet resulted in the development of a single individual to sex- ual maturity, and the rdle of the inorganic salts is still far from clear. Attempts to maintain the activity of the reflex nervous centres of animals by perfusion with the salt solutions have resulted in failure. We have reviewed the literature on this subject in a previous paper,! to which the reader is referred for a fuller discussion. As already stated, such solutions are not capable of maintaining the activity of the Selachian heart for more than a short time. In view of the abnormalities and irregularities appearing in the tracing of the excised mammalian heart under their influence when in aque- ous solution, and of their general incompetence in maintaining other physiological functions, it is highly improbable that the inorganic salts of the blood are the direct cause of the normal heart beat. In our opinion, there exists in the literature a certain amount of confusion between the actual cause of the heart beat and the mainte- 1 GUTHRIE, PIKE, and STEWART: This journal, 1906, xvii, p. 344. Activity of the Excised Mammalian Feart. a7 nance of a suitable condition which permits of beats, z. ¢., the main- tenance of normal nutrition of the organ. A similar confusion exists between the stimuli which elicit physiological actions of many other tissues and the factors on which their nutrition depends. That a chemical reaction is involved in the rhythmical contraction of the heart muscle! is scarcely to be questioned. That this reaction is the direct cause of the beat is not so evident. It is our opinion that the actions of the inorganic salts upon the heart need extend no further than their action in other tissues, in which they doubtless assist in the nutritive and general metabolic processes. There are weighty reasons for considering that fluids act upon the heart in two ways: (1) Certain fluids, such as blood and its dilu- tions, and milk, properly prepared, are nutritive or physiological in their effects; (2) Certain other fluids are purely artificial or stimu- lating in their effects. We do not consider the action of fluids of this second class to be truly physiological. The train of events following perfusion of the excised mammalian heart with the inorganic salt solutions is, we believe, not a true picture of the events occurring normally in the intact heart. CONCLUSIONS. 1. Increase in pressure in the coronary arteries, up to a certain limit, causes a concomitant increase in rate and amplitude of beat in the excised mammalian heart. This increase in rate occurs with con- stant or even with falling temperature. 2. Increase in temperature, up to a certain limit, causes an increase in the rate of the excised mammalian heart. This increase occurs with constant pressure. The optimum temperature is from 35° to 37.5° C. 3. Heat standstill occurs at 39° to g4o°C. ‘The heart may recover from heat standstill. 4. Defibrinated blood and serum, diluted with 0.9 per cent sodium chloride solution,,are the best of the fluids used for perfusing the ex- cised mammalian heart. Serum, free from corpuscles, gives better results than corpuscles free from serum, each being diluted to the same degree. Milk whey, prepared by precipitating the casein with hydro- chloric acid, and diluting it with about three volumes of 0.9 per cent sodium chloride solution, gives results closely approximating those obtained by perfusion with blood and serum dilutions. 1 ROBERTSON: Loc. cit. 38 C. C. Guthrie and F. Hi. Prke. 5. The inorganic salt solutions do not maintain the activity of the excised mammalian heart as long as the albuminous fluids. Luciani’s groups appear in the tracing, and the heart soon stops, presumably from exhaustion. 6. Aqueous extracts of serum salts, and milk whey after heating, produce results more like those caused by the inorganic salts than by the albuminous fluids. 7. The heart, after stoppage from the inorganic salt solutions, may often be restored by perfusion with the albuminous fluids. 8. Since blood produces no contractions until perfused through the coronary vessels at a suitable pressure, it is possible that there is a purely physical or mechanical element in the cause of the normal heart beat. 9. In a slowly beating excised heart, the coronary arteries or the tissues about them may sometimes be seen to contract and the beat to spread from the coronary vessels over the heart. A strip of cat’s thoracic or abdominal aorta, when suspended, be- haves much like the first strip of ventricle. 10. A circulation in the capillaries of the coronary vessels is suffi- cient to produce beats of the excised heart, and pressure on the walls of the larger arteries or veins is not an absolutely essential condition. We wish to acknowledge our obligation to our colleagues in the department of physiology, and particularly to Prof. G. N. Stewart, for many helpful suggestions during the progress of the work. MieeiePARENIT PHARMACOLOGICAL “ACTION AT A DISTANCE” BY METALS. AND. METALLOIDS. BY A. Po MATHEWS: [From the Laboratory of Biochemistry and Pharmacology, University of Chicago.] SHORT note by Herbst,! in 1904, called attention to the fact that if a piece of silver or copper was placed in a dish with sea- urchin eggs some of the eggs put out fertilization membranes. A very minute quantity of the silver was sufficient to produce the result. Thus, if a dish which had had a coin in it was washed with only or- dinary precautions and then a new lot of eggs put in it, some of the latter put out fertilization membranes. Silver was more efficient in producing membranes than copper. Minute traces of silver salts, the chloride or nitrate, produced the same effect as the coins. Herbst left it an open question whether this action of the metal was due to the ions of the metal or not. : This observation appeared in so many ways interesting that I took the matter up for the purpose of finding out what other metals would produce the same result, and how the metal acted. I have accord- ingly tested the action of several metals and metalloids upon the ma- ture eggs of the starfish, Asterias Forbesii. The method consisted in placing a piece of the carefully cleaned metal or coin in about 50 c.c. of sea water in a glass dish and then introducing a large number of eggs which had been in sea water maturing for about one hour, so that the eggs fell thickly all about the piece of metal. The eggs were then left quite undisturbed. The metals used were iron, silver, copper, lead, zinc, tin, platinum, gold, and mercury, and the metalloids were bromine and iodine. As I had no pure silver, a silver coin, a ten-cent piece or a quarter, was used instead. 1 HERBST: Mitteilungen aus dem Zoologischen Station zu Neapel, 1904, xvi, Pp- 445-457. 39 40 A. P. Mathews. The result was as follows: with iron, zinc, lead, tin, gold, and plati- num I got no fertilization membranes. The iron dissolved a good deal and killed the eggs near it. The lead, tin, zinc, nickel, gold, and plat- inum seemed not to affect the eggs at all, since they matured and lived, for several hours at least, when lying almost in contact with the metal. Eggs matured and lived in a tin vessel when the eggs were lying against the tin. Copper, silver, and mercury and the metailoids bromine and iodine caused the eggs to put out fertilization membranes. Of the metals, copper was the most powerful, silver came a little after it, and mercury was efficient only when the eggs were extremely sen- sitive. Iodine is as powerful apparently as copper, and bromine is not so good as iodine. We may, therefore, add mercury to the list of efficient metals, as given by Herbst, and iodine and bromine. In working with the metalloids a slightly different procedure was adopted. A minute piece of iodine or a small drop of bromine was put in the midst of a lot of eggs. The eggs near the metalloid took up some of it and became strongly colored. A few of these eggs were then removed with a very fine pipette, washed in a little sea water, picked up again with the pipette, and allowed to fall among a new lot of eggs in fresh sea water. Each of the eggs which had the metalloid thus served as a minute source of the metalloid, and acted upon the other eggs in its immediate neighborhood. This method was adopted owing to the solubility of these metalloids when introduced in sub- stance. So much of the iodine or bromine went into solution as to kill the eggs near by unless this method was used. The time required after metal and eggs were brought together be- fore the action began was in many cases very short, an effect being visible within a minute after the eggs were introduced. In other cases, however, a longer interval of five or ten minutes elapsed before the membranes began to appear. The time appeared to depend upon the sensitiveness of the eggs. In no case were the whole or a majority of the eggs in the dish affected, as reported by Herbst. On the contrary, if the eggs were left quite undisturbed only those within a short distance of the metal were affected. Eggs a millimetre or more from the metal were very seldom acted upon. This difference from Herbst’s results may be due to the starfish eggs being less sensitive, or to the fact that in his experiments the eggs were not left undisturbed, so that many eggs in succession came within the sphere of action of the wire. The actual phenomena of the changes in the egg are interesting. “ Action at a Distance.” Al If the eggs are left entirely quiet, it will be seen that the change im the egg ts strictly polar. Each egg near the wire puts out a fertiliza- tion membrane on the side toward the wire or piece of metal, and on the side farthest away from the metalloids. The picture thus ob- tained is very striking and is represented in the diagram in Fig. 1. The membranes may ultimately extend clear around the egg, although this is more often not the case; but even when this happens, the membrane is always wider on the side toward the metal, or on the side away from the metalloid. When I first saw the eggs thus so strikingly oriented, I thought it a beautiful demon- E +h : eth ee FIGURE 1.—A, copper wire sur- Stration of the action of the metals in rounded by starfish eggs. Fer- throwing off their ions and thus bom- tilization membranes are out barding the eggs on the sides turned 0” the side of the egg toward the wire. Slight coagulation in toward them, but I afterwards had to = Sowa Gh Fa c Set : some eggs at a point farthest modify this interpretation. from wire, and liquefaction to- The second interesting observation is — ward wire. 4, starfish egg that the change does not stop ordinarily Containing iodine at @, sur 2 : rounded by other eggs extrud- with the extrusion of a membrane. A : eS, ing fertilization membranes, b, further change takes place in the proto- on the side farthest from a, and plasm of the egg, and this change also is — coagulating, ¢, on the side near- strictly polar. Thus the protoplasm on “™ “ ~the side of the egg turned toward the metal, or away from the metalloid, begins to liquefy and swell, so that the egg becomes balloon- shaped; and it coagulates on the side turned away from the metal and toward the metalloid. Still later one may get secondary coagu- lation in the previously expanded portion in the side turned toward the metal. What, then, is the means by which the metals produce these effects ? When [I started the investigation I felt confident that this would prove to be an action due to the ions of the metal. Herbst’s statement that silver salts produced the same result seemed to show this. To settle the matter, I tried the effects of solutions of the salts of the metals, but to my astonishment, IJ got entirely negative results, so far as the extrusion of fertilization membranes was concerned. All of these metal salts if used in concentrations great enough to produce any effect at all, produced always a coagulation. There was never the slightest indication of liquefaction or of membrane formation. As 42 A. P. Mathews. there was danger of overlooking some favorable concentration of the salts, I used a large number of different dilutions as follows: E m mM TE m mt m 1. Lead Salts, EDC er ia o007 20,000? 40,0009 80,000? 160,000? 320,000% WN 1,000,000° a . q AVE hs wile de: Mm mt m m 2. Copper Salts, Cue Sonn ane? 500? 666? 1:333? 2.0009 4,000? 6.666? Mt mM m mL 1 mn Mt Ut m 10,000? 20,000 40,0002 50,000? 66,666? 80.000? 100,000? 200,000? 400,000: Mm __ Hd m e72 mt me LL 160:000, 500,000, 666.666, 800,000, 1,000,000? 1,200,000, 2,000,000; mn M7 M1 m mM mM 3,200,000? 4,000,000, 5,000,000, 6:666;:666;5 10,000;000, 20,000,000, WL 40,000,000 3. Silver Salts, AgCl: Saturated solution in sea water and many dilutions ‘ el) teas a nt n from this. AgNog: in sea water, 9.900,000, 1:000,000, 500,000. As the above experiments gave a negative result, solutions of other salts were tried. These also resulted negatively. Thus in CaCl,, 2m, maturation proceeds normally and some eggs segment into four cells, but no membranes appear. MegCl,, #z 3, permitted maturation, but no membranes were formed. In sea water containing manganese chloride, so, zoo, and soo, there were no membranes formed. The same result was obtained with hydrochloric acid. The eggs formed membranes when transferred from the acid to the sea water, but not while they were in the acid containing sea water. Potassium iodide, sodium sulphate, sodium citrate, and sodium hydrate were ineffective. Of the hydrate I added 1, 2, 3, 4,:5, 6,.7,.8, 9, and- To e:c) mespeae tively to 100 c.c. of sea water in different glasses. The eggs dissolved in the Jast glass, but no good membranes appeared. The dissolution of the peripheral layer of protoplasm looks somewhat like a process of membrane formation, but I do not think it identical with it. From these experiments tt seems to be clear that the metals are not acting directly by their tons. Theions of the metals in all cases caused coagulation, but the metals themselves always cause liquefaction first in the egg protoplasm nearest the metal. It is only after many minutes that a secondary coagulation may ensue, and this is ap- parently due to the action of the metallic ions, as is indicated by the light blue color of the coagulum in the case of the eggs near the copper wire. Furthermore, eggs which had been in the more dilute solutions of cupric salts were transferred toa dish containing a copper wire, when those eggs near the wire immediately put out their mem- “ Action at a Distance.” 43 branes. Eggs from the stronger solutions put out no membranes either in the solution or when brought near a wire. Against the theory of the action of the ions the curious fact may also be recalled that, with the metalloids, the membrane is put out very quickly, but on the side farthest from the metalloid. For these reasons I feel compelled to relinquish the theory I started with, that the-action is due to the ions thrown off from the metal. The question arises, What, then, is the cause of the action? Two possibilities suggest themselves: first, that the metals throw off not their ions, but minute particles of the metals themselves, which hit the eggs and thus cause an action; or possibly there is an electro- static field about the metal in water and this produces the effect. In favor of the first supposition it may be said that certainly iodine and bromine do dissolve in this way. But the probability that silver, copper, and mercury are thus dissolving and that these particles come off with sufficient speed and in sufficient numbers to produce this action, appears to me exceedingly remote. The amount of silver thus dissolved in a metallic state in sea water must be almost infinites- imal, otherwise silver would give us solutions of metallic silver in water. Furthermore, it is improbable that these small particles would differ in the manner of action from the larger particles from which they are derived. This hypothesis appears hence very improbable. In favor of the second hypothesis, that the metals are acting by means of an electrostatic field about them, there is this evidence: first, such a field is known to exist. When a metal is placed in water free from its salts, it throws off into the water positively charged ions in virtue of its solution tension. By this means the metal becomes electronegative, the water electropositive. By this means the further solution of the metal is prevented. Now this electro- static field can have any great intensity only in the immediate neighborhood of the metal. This would accord with the fact that only the eggs close to the wire, though not in contact with it, are affected. Eggs more than a fraction of a millimetre away are not affected at all. In the second place, platinum and gold, which have almost no electrostatic field about them, are ineffective. These metals have nearly as great a tendency to throw off negative as positive ions and remain hence almost uncharged. However iron and zinc, lead and tin, should be, on this theory, more efficient, unless some secondary cause comes into 44 A. P. Mathews. play, since their solution tension is greater. In the case of zine, lead, and tin the quick coating of the metal by a non-conducting sheath of oxychloride may interfere with their action. The ions of these metals certainly do not get into solution, since, although these ions are very poisonous, the eggs will develop for some time in contact with the metal without harm, thus showing that no ions get into the solu- tion and consequently the electrostatic field is nil. As regards iron and the metals of higher solution tension, these metals throw their ions into the solution in sufficient numbers to poison the eggs. As these metals are able to discharge the hydrogen ion from the water, particularly in the presence of oxygen, it may be that their lack of action is due to this. The hypothesis that the action is of an electrostatic kind is supported also by the polar nature of the change and by the fact that the mem- brane comes out on the side foward the electropositive metal and away from the electronegative metalloid, and coagulation takes place in each case at the opposite poles. This fact clearly points toward an electrical action of some kind. Furthermore electrical stimula- tion of the eggs will cause a closely similar polar change. I passed a series of make and break induction shocks with the secondary coil pushed up over the primary, through a lot of eggs. They put out fertilization membranes and ultimately liquefied at the two poles of the egg towards the electrodes. The membrane comes out most readily on the cathode side with make shocks, and on the same side, but far better, with break shocks. The membranes come out hence on the break of the current on the cathode side. I then diluted the sea water one-half with isosmotic glucose solution, introduced a fresh lot of eggs, and sent through them aconstant current of about twenty- five volts. The membranes came out on passing the current on the anode side, and liquefaction took place here as reported by R. Lillie. This shows, therefore, that the passage of a constant current brings out the membrane on the anode side while the current passes, and on the cathode side on breaking the current. It may be, therefore, that in introducing the egg into the electro- static field about the metal the effect is the same as if a current of high intensity had been passed through it. The ions in the egg are separated, the positive ions being driven away from the metal, the negative ions toward it. By this means exactly the same separation of ions in the egg could be produced as is produced by electrical stimulation. The current would appear to pass through the egg from “ Action at a Distance.” 45 the side toward the wire, z. ¢., the anode side, and here liquefaction and membrane formation occur. Similarly for the metalloid the current would pass in the opposite direction, and here liquefaction and membrane formation take place on the side away from the metalloid. The extrusion of the fertilization membrane when the eggs are transferred from acid containing sea water to normal sea water might be explained in the same way. The rapidly moving hydrogen ion striving to get into solution leaves the surface membrane of the egg electronegative, with the result that the membranes are thrown out and slight liquefaction takes place at the periphery. I believe the electrostatic explanation of the action of the metals accords best with the phenomena observed. We have, if this be the true explanation, a real “action at adistance.” That is, the metal pro- duces an action on the egg, although neither the metal itself nor any particles from it come in contact with the egg. The metal in this case would be exerting its action by means of the effect it produces upon the ions in the eggs lying in its immediate neighborhood, thus. leading to a separation of these ions. It may be that these eggs would furnish a convenient means of demonstrating the separation of ions by electrostatic action. Possibly the. pharmacological action, the so-called oligodynamic action, of finely divided metals and colloidal solutions may involve this same principle and have a similar explanation, entirely inde- pendent of the action of the ions of the metal. Also, it occurs to me, that these observations may be of interest in connection with the formation of the cell asters. The phenomena in this case indicate some kind of an electrostatic action on the part of the astro-centres as pointed out by Hartog and R. S. Lillie. Possibly an investigation of these phenomena shown by metals when placed in water may lead to a better understanding of the method by which such an electro- static action might be produced within the protoplasm itself and in the astral centres, I hope to extend these observations further, and I publish them in an incomplete form in the hope that others also may examine these interesting phenomena. SUMMARY. 1. Confirming Herbst, it is found that metallic silver and copper cause mature Echinoderm eggs to put out fertilization membranes. 2. The action takes place readily in the eggs of Asterias Forbesii. Only the eggs near the metal are affected. 46 A. P. Mathews. 3. Besides silver and copper, mercury, iodine, and bromine were found to produce membranes. Iron, nickel, lead, tin, platinum, and gold were ineffective. Hydrogen was doubtful. Mercury acts only when the eggs are very sensitive. 4. The change in the egg is strictly polar. Membranes appear first and liquefaction takes place later, on the side of the egg toward the metal and on the side of the egg away from the metalloid. Slight coagulation occurs on the side of the egg away from the metal, and marked coagulation on the side of the egg foward the metalloid. s. Salts of the metals produced only coagulation when in concen- trations strong enough to cause any action. The action of the metal is not due, hence, in this case to the action of ions of the metal. 6. The polar nature of the change, the quickness with which it occurs, and the identity of the changes with the changes caused by electrical stimulation indicate that the cause of the change is electrical. 7, The most probable explanation appears to be that the metal is producing its action by an electrostatic field about it, thus causing a separation of the ions in the egg, positive ions being driven away from the wire and negative ions being attracted toward it. 8. If this interpretation is correct, the metals are producing their action without any particle of the metal coming in contact with the egg, and in this sense the metals are acting at a distance from themselves. *» CONTRIBUTIONS FROM THE ZOOLOGICAL LABORATORY OF THE MUSEUM OF COMPARATIVE ZOOLOGY AT HARVARD COLLEGE. E. L. MARK, Direcror. No. 186. ieibs KEACTIONS OF CYCLOPS TO LIGHT AND TO GRAVITY. BY COL ie Sir RE ye T is the purpose of this paper to report a series of experiments which appear to throw new light upon the diurnal movements of certain plankton crustaceans. Most investigators of this subject have held that the response of the organisms to light determines in a large degree the time of leaving or entering the upper strata of the water. As evidence that this is true in certain cases may be cited the experiments of Groom und Loeb (’go) on barnacle larve, of Loeb (937, 93”) on Temora longicornis and Polygordius larvz, and of Parker (:02) upon Labidocera ezstiva. All of these investigators found that. an important factor in causing the upward movement of the organisms is a positive response to light of low intensity. Con- versely, a negative response to strong light is important in causing the downward movement. The results of my experiments go to show that, in the species investigated, a phototropic response has compara- tively little to do in bringing about an upward or downward migration, though exposure to light has an important part. The observations were made on the movements of fresh-water copepods. Some of the animals were collected in October in ponds near Cambridge, Mass., and kept in aquaria for several months, Material was also obtained from the permanent fresh-water tanks in the aquarium room of the zodlogical laboratories in the Museum of Comparative Zodlogy. The animals used were all females, and belonged to the species Cyclops albidus Jurine. The work was carried on under the direction of Prof. G. H. Parker, and the writer is greatly indebted to him for valuable advice and criticism. 47 48 C.°O. “Esterdy; It was found that a suitable artificial light could be obtained from a glowing Nernst filament; this was used, accordingly, throughout the experiments. In experiments to determine the nature of the response to light, the animals were confined in rectangular glass vessels with parallel sides; the vessels were 3 cm. wide, 7 cm. long, and 7 cm. high, inside measurement. The aquarium was always placed with its length in the direction of the rays of light, and the wall away from the light was coated on the inside with paraffin and lampblack to check reflection as much as possible. The vessel as a whole was protected from reflections from the walls of the room by black screens. The Nernst filament was covered by a hood of black sheet iron, and the light reached the animals through an aperture in the hood a little wider than the filament was long. Care was taken that as far as practicable the rays of light should fall perpendicularly upon the wall of the aquarium, and the work was always done in a dark room. The candle power of the light was determined by means of a Lummer-Brodhun photometer, and from this the intensity was reckoned in candle metres. The experiments, with the exception: of one, were carried on in lights of five intensities: 8, 420, 825, 1700, and 2200 candle metres. These figures represent the intensity at the wall of the aquarium nearest the light. The lights used most were the first one and last two of those named, the others being employed to check results. One set of animals was generally subjected to three lights of different intensities on one day, and on the following days to the same three but in another order. After an experiment the animals were always allowed to rest for twenty-four hours, under con- ditions as nearly normal as possible before being used again. The vessel containing the animals for an experiment was placed in the desired position, covered so as to exclude all light and left undisturbed for from six to eighteen hours before beginning the experiment. As soon as possible after exposing the animals to the light the number in one half of the aquarium was recorded (those in the half toward the light being designated as positive, those in the other half as negative), and records were made thereafter at half- minute intervals for from half an hour toan hour. Then the intensity of the light was changed by moving the aquarium toward or away from the filament, and the same method of recording continued. It will be seen that in this way at the end of the third day of experi- menting records were obtained of the distribution of the animals in The Reactions of Cyclops to Light and to Gravity. 49 each of three intensities of light following darkness, and also after previous exposure to light. It was found necessary to use a small number of animals at one time in order to count accurately and quickly. The number usually taken was five, and the whole number of times that animals were found in one half of the aquarium was calculated from records made as above. The percentage this number bore to the whole number of records was taken as evidence whether a given set of animals was positive or negative. Reactions to light after protracted retention in darkness. — It may be said at once that, in lights of 420 candle metres and below, the animals experimented with were neutral when exposed to the light after they had been in darkness for some time. The results of such experiments show that 50.9 per cent of the total number of records made under the conditions named and in the 420 candle-metre light represent animals in the negative half of the aquarium. Usually the animals under experimentation move from one end of the aquarium to the other frequently, the movement being general and not confined to one or two individuals. Even if at a certain moment more animals are posi- tive than negative or vice versa, it is possible with a large number of records extending over a considerable time to determine whether as a whole the set is positive or negative. The lights of lower inten- sity than 420 candle metres that were used were 8 and 0.55 candle metres. The latter was used on only one occasion, and special care was taken to prevent any but the most direct rays reaching the animals. The number of negative and positive animals was recorded every fifteen minutes for a period of six hours, and after each reading was taken, the water was agitated to redistribute the animals. The results for the entire period show that exactly as many animals were negative as positive, though the number at different times in the negative half varied from one to five. In this case six individuals were under observation. During observation in the 8 candle-metre light an aquarium was used in which both ends were clear, so that it could be turned end for end. ‘he aquarium was left in position and in the dark for several hours before observations were commenced. When brought from darkness into the light, the animals were not counted for the first five minutes; then records were made every half minute for ten minutes, when the vessel was turned end for end as carefully as possible and after five minutes counting was resumed. Two sets 50 C. On sterly of records were made for each of the two positions, and the same animals were used on three different days. The results of a number of trials made in this way show that as a whole 47.1 per cent of the animals experimented with were negative. Here, then, there is some evidence of a positive response, but it is probably of slight signifi- cance, as indicated by comparison with the results obtained by the use of the lights of 420 and 0.55 candle-metre intensity. . TAB THE AVERAGE NEGATIVITY IN PER CENT OF CYCLOPS IN VARYING INTENSITIES OF LIGHT FOLLOWING PERIODS IN DARKNESS. Percentage of ani- Number of mals found in the animals tested. negative half of the aquarium. Light in C.M. 50.0 Ail. 50.9 66.6 83.8 84.7 The conclusion to be drawn from these results is that in lights of low intensity following darkness the animals are neutral. When lights of greater intensity than 420 candle metres are used following darkness, the reaction is invariably negative, the negativity increasing as the intensity of the light is increased. Table I shows the averages of a number of trials with the different intensities following darkness. This table shows that there is no clearly negative reaction until an intensity of 420 candle metres has been passed. In all trials in the 825, 1700, or 2200 candle-metre lights the negativity of the animals was clear and convincing. Reactions to light after exposure to light. — If the animals are tested after they have been exposed for some time to light of any intensity, their reactions are invariably negative. It does not appear that there is an increase in the negativity with an increase in the intensity of the light, as there was when the animals were brought from darkness into light. Table II shows the average per cent of animals that were found in the half of the aquarium away from a light of given intensity (indi- The Reactions of Cyclops to Light and to Gravity. 51 cated in the upper part of the table) after exposure to light of another intensity (in column at left). The figures presented in this table make plain that the intensity of the previous stimulation by light does not affect the response toa given light. The animals are negative to all lights whether previously subjected to light of high or low intensity within the limits employed. The variation in the figures representing the percentages does not TABLE ET: THE AVERAGE NEGATIVITY IN PER CENT OF CYCLOPS IN LIGHTS OF VARIOUS IN- TENSITIES AFTER PREVIOUS EXPOSURE TO LIGHT OF THE SAME QUALITY, BUT OF A DIFFERENT INTENSITY. Intensities of lights to which cyclops had been previously subjected. Intensities of lights in which records were taken. 420 C.M. 825 C.M. 1700 C.M. 2200 C.M. § C.M. seem significant; for the present purpose it is sufficient to note that without exception the females of Cyclops albidus are negative to light of any intensity if tested after exposure to light of the same quality. Reactions to gravity. — The species used in these tests was Cyclops albidus, and the animals were all females. It can be said at the out- set that in by far the majority of cases the animals were positively geotropic. The author had many times tried the experiment of liber- ating individuals one by one at the top of a tall column of water in a cylinder of large calibre, and almost without exception they passed at once to the bottom. As a rule they remained at the bottom, but certain ones made short excursions toward the top and then dropped down again. It was rare to find the animals anywhere except in a section two or three centimetres deep at the bottom of the vessel. If the cylinder with a number of animals at the bottom was inverted, most of them passed at once to the bottom. It cannot be said that every animal of any set was positively geotropic, but the great major- 52 Ce Os Esty. ity were. The most usual inhibition of the response occurred when an animal came in contact with the side of«he vessel after once start- ing down; it might cling to the glass for some time, but even such individuals have often been seen to reach the bottom after a few minutes. We may conclude that in daylight and under ordinary conditions the response of the females of Cyclops albidus to gravity is positive. But if the vessel containing the animals was covered for a short time so as to exclude all light, and the distribution noted at the end of this time, it was seen that there had been an upward migration toa greater extent than ever occurred if light and dark periods did not alternate. To test this carefully a tall glass cylinder was marked off into five sections each about 10 centimetres in height, and the sections were numbered from the top down. Thus the position of the animals could be noted rapidly when the jar was uncovered. A single laboratory record will show the result of keeping the animals in the dark for short periods and then exposing them to diffuse daylight long enough to note their vertical distribution. April 7, 1905. — 10.15 A.M. Cylinder with six positively geotropic animals was covered so as to exclude light. 10.20 A.M. Covering removed; one animal in Section 1, moved to bottom at once; five in Section 5. Covering replaced. 10.25 A.M. Covering removed; one in Section 2, and went down at once ; five in Section 5. Covering replaced. 10.30 A.M. Covering removed ; one in Section 2 ; three in Section 1, and all moved down at once in the light; two in Section 5. Covering replaced. 10.40 A.M. Covering removed ; one in Section 3; two in Section 2 ; two in Section 1, but one of these does not move down ; one in Section 5. Covering replaced. 10.50 A.M. Covering removed; one in Section 3; two in Section 2; two in Section 1; one in Section 5; all pass to bottom. Covering replaced. I1.00 A.M. Covering removed ; two in Section 3; two in Section 2, these two reach the bottom in thirty seconds ; two in Section 5. This experiment, which is typical of all, proves that in the absence of light the animals, which are otherwise positively geotropic, tend to become negatively geotropic, and that in daylight this response is reversed, The same is true when the animals are exposed from below The Reactions of Cytlops to Light and to Gravity. 53 to illumination of such intensity that if the rays were to strike hori- zontally the phototropic reaction would be negative. These experi- ments were carried out with a Nernst light and in an illumination of 1000 candle metres. The animals were placed in the graduated tube, and an apparatus was arranged so that light would enter the tube only from below and could be cut off bya tightly fitting shutter; the whole experiment was carried on in the dark room, extraneous light being carefully guarded against. A heat screen was interposed between the light and the bottom of the cylinder, and care was taken that the - water in the screen was cool. The animals were nearly always put into the cylinder at night and not tested until the following morning ; then they were exposed to the light for five-minute periods, alternat- ing with five minutes indarkness until records had been made for five periods or more in each state. Table III shows the distribution of the animals through the cyl- inder. Half of the records were made after a period in darkness and as soon as the light entered the vessel; the other half were made at the end of a period in the light immediately before the shutter was closed. The figures in the table are the sums of the numbers of animals found in the different sections of the cylinder for the entire number of trials on one day. It will be seen that the table contains records of six separate tests, each of which consisted of at least five five-minute periods in darkness and a like number in light. Inspection of the table shows that in the six experiments animals appeared in the uppermost section more than three times as often immediately after periods in complete darkness as they did at the end of like periods of exposure to illumination from below, and like- wise more than twice as often in the second section after darkness as after exposure to light. It will be seen that during both light and dark periods more of the animals which leave the lower section are found in sections 1 and 2 than in any other. Fifty animals were tested with the Nernst light, and only one set was used twice in succession. Discussion of results. — It has been seen that the phototropic re- sponse of these animals is negative; they are neutral to lights of low intensity if tested after retention in darkness. It cannot be claimed that the response to light was modified by handling or other means, because when tested the animals had been undisturbed and in the dark for hours. In the cases where the aquarium was turned end for end during an experiment there was of course some mechanical stim- 54 C. (O. Esterty. ulation, but it was certainly very slight and may be regarded as not affecting these results, especially since no record was made of the dis- tribution until after a period of quiescence. If it were possible that in nature light could affect the organisms apart from gravity, we TABLE IIT THE COMPARATIVE VERTICAL DISTRIBUTION OF THE FEMALES OF CYCLOPS ALBIDUS THROUGH A CYLINDER OF WATER AFTER ALTERNATING FIVE-MINUTE PERIODS IN DARKNESS AND IN LIGHT. IN THE DARK. The numbers of animals in each of the sections of the cylinder Number in each of six experiments. of sec- Saad ° Totals tion of Number of the experiment. . cylinder. 3 4 11 7 6 4 4 0 36 IN THE LIGHT. should be led to say that light alone has very little to do in causing vertical migration. > solution of ammo- nium hydrate is generally too weak to cause any contraction. The contraction of the muscle is not due either to the ammonium ion or to the hydrate ion, for neither ammonium chloride nor sodium hy- EXPERIMENT 1. Solution. Time. Minimum stimulus. Secondary || coil; cms. | from primary. Solution. Minimum stimulus. Secondary | coil; cms. |from primary. Time. NH,Cl# 11.01 86.4 | Nerve in solution at 11.02. 11.06 | 73.8 Nerve in solution at 11.02. 11.05 85.9 11.10 74.8 LU: 62.0 11.16 | | Non-irritable in one-half of nerve. IDES 74.0 45.0 63.0 11.18 Non-irritable | at tip. | WL drate of .%, strength produces the contraction, although the number of ammonium ions is vastly greater in the first, and the hydroxyl ions in the second, than in ammonium hydrate. Furthermore, the following experiment shows that the action runs part passu with the number of undissociated molecules of NH,OH present. To an ¥% solution of ammonium chloride ammonium hy- drate was added sufficient to make an 5% solution. The number of undissociated NH,OH molecules in such a mixture is greater than in the...“ ammonium hydrate by itself’ When such a solution is applied to the muscle, a marked shortening results, although ammo- nium hydrate of the same strength is ineffective, or nearly so. There can, therefore, be no doubt that an increase in the number of zzd7zs- sociated ammonium hydrate molecules means in this case an increased action. Similar experiments have been tried upon the motor nerve of the frog, with closely similar results, as shown in Experiments I-III. The addition of ammonium hydrate to ammonium chloride greatly in- 62 A. P. Mathews. creases the toxic action of the chloride upon the nerve, whereas sodium hydrate, in concentrations containing more hydroxyl ions than this strength of ammonia, does not increase the toxicity of sodium hydrate. EXPERIMENT 2. EFFECTS OF ADDITION OF NH,OH ‘ro NII4Cl on Toxiciry. Sciatic oF Froc. Minimum stimulus. Minimum : Solution. Time. stimulus. Solution. Time. 1.50 87 NH,C1¥ + : 87.0 Solution. In solution at 1.51. Time. 87 80 72 64 Non-irritable at tip. NH,OH 325 EXPERIMENT 3. Minimum stimulus. Solution. Nerve im- mersed. 86.0 72.8 62.0 Non-irritable Minimum stimulus. NaCl %. 82.5 Nerve im- mersed. 80.5 76.0 | Spontaneous contractions. 76.0 $3.0 NaCl Z + NaOH x55: 79 Nerve immersed. 75 78 Contractions. 78 90 The question which has not been solved is whether the toxic action is due to the undissociated ammonium hydrate molecules, or to the NH, formed from it. I think this question may be answered pro- visionally, although it is impossible to determine it directly, so far as incanvsee; Pharmacological Action of Ammonium Salts. 63 Ammonium hydrate is constantly dissociating into water and NH. When this dissociation takes place, there is an instant when the bonds on the nitrogen atom which formerly bound the water are free or open, — there is a moment, in other words, when the NH, and the water exist in a nascent form, before the two valencies set free saturate themselves. The chemical reactions of ammonia (NH,) are due either to the saturation of these two free valencies by the sub- stances added, as for example hydrochloric acid, or to the replace- ment of one of the hydrogen atoms. Ammonium hydrate, on the other hand, enters into combination, only by means of its NH, and hydroxyl ions, which are in this case ruled out. It is, therefore, probable that the action of the ammonium hydrate is due to the dissociated NH, present, and particularly to the action of this substance in the moment of its origin, when the valencies of the nitrogen atom are open or dissociated. These experiments show that the pharmacological actions of ammo- nium compounds are due, in part, to the ammonium and acid ions present, but that certain pharmacological effects and presumably the characteristic stimulating action of ammonium salts run _ parallel with the amount of undissociated ammonium hydrate present in the solution and formed by hydrolytic dissociation. The action of the ammonium hydrate is in its turn to be ascribed probably, as is the chemical action, to the NH, formed by dissociation of the hy- drate, and probably to the NH, in nascent state when the valencies of the nitrogen are open. The action of ammonium compounds is not, therefore, contrary to the principles of pharmacological action quoted at the beginning of the paper, but necessitate the recognition, in addition thereto, of the pharmacological action of dissociated par- ticles which are non-ionic, or rather twin-ionic, such as > NHg particles. The well-known greater activity of the free alkaloids as contrasted with their salts is, in my opinion, to be explained in the same way, the alkaloid splitting off water and having thus free nascent bonds in its nitrogen. THE RHYTHM OF THE TURTLE’S SINUS VENOSUS 2 ISOTONIC SOLUTIONS OF NON-ELECTROEVTES: ; BY He E. EGGERS: [From the Hull Physiological Laboratory of the University of Chicago.| CONSIDERABLE portion of the more recent investigations of the heart rhythm has been devoted to attempts to illustrate the mechanisms of the normal rhythm by the production of artificial rhythms in non-automatic tissues by various means. From the fact that non-electrolytes (sugar, urea, glycerine, etc.) in isotonic solutions do not produce rhythmic contractions in non-automatic tissues, and the further fact that such rhythms are produced by some of the inor- ganic salts of the blood, some physiologists have been led to look for the immediate stimulus to the heart rhythm in the inorganic salts of the blood (or rather of the heart tissues). Carlson! has recently pointed out that this view rests on very meagre evidence. It is to be hoped that the study of artificial rhythms will ultimately throw some light on the mechanisms of the normal heart rhythm, but direct evi- dence can be obtained only by studying the normal automatism. The fact that sodium chloride will produce a transient rhythm in non-automatic muscle is no more a proof that this salt bears this same relation to the normal heart rhythm than the fact that a muscle fibre can be made to contract by direct action of the electric current is a proof that the nervous impulse is an electrical current. While the non-electrolytes do not produce rhythms in non-automatic tissue, it is common knowledge that the automatic parts of the heart continue in activity for some time in isotonic solutions of such non-electrolytes as the sugars. It is usually assumed, however, in this case that the non-electrolyte is neutral, and that the rhythm is maintained by the electrolytes within the automatic cells or within the tissue spaces. But Carlson has shown, for the automatic as well as for the non- automatic tissues of the Limulus heart, that neither sugar, urea, nor 1 CARLSON: This journal, 1906, xvi, p. 221. 64 Lehythm of the Turtle’s Sinus Venosus. 65 glycerine can be considered as neutral substances or as acting by means of the osmotic pressure factor alone. In isotonic solutions these electrolytes have a purely depressant action on the heart muscle and a primary stimulating action on the heart ganglion, while the duration of the ganglionic rhythm in isotonic solutions of these sub- stances depends on the condition of the ganglion as well as on the nature of the non-electrolyte. Miss Denis,’ working in this labora- tory, has recently shown that the relative rapidity with which these non-electrolytes stop the ganglionic rhythm bears a direct relation to the rate of diffusion of the blood salts into solutions of the non- electrolytes and vce versa, but the fact that the duration of the rhythm depends on the condition of the ganglion goes to show that the direct action of the non-electrolyte on the automatic cells is a factor in the ultimate cessation of the rhythm. At Carlson’s suggestion his experiments on the heart of Limulus were repeated on the sinus venosus of the turtle in order to determine whether the conclusions based on the Limulus heart are applicable to the heart of vertebrates. The sinus venosus was chosen rather than the ventricle or auricles for the following reasons. The sinus is the automatic part of the heart par excellence. The walls of this part of the heart are the thinnest; hence when immersed in any solution the cells throughout the entire sinus will sooner be surrounded by the same conditions than would be the case with the auricles or the ventricles. In all experiments the rhythm of the sinus was recorded by the ordinary graphic method, the sinus being suspended within a glass cylinder. This rendered large sinuses particularly desirable. The material was accordingly taken from large specimens of the common snapping turtle (Chelydra serpentina). The sinus was removed with as little handling as possible, imme- diately after the death of the animal, and when not used at once, was kept in serum until used. The sinus was of course freed as com- pletely as possible from auricular tissue. The great size of the organ in this turtle made it advisable to separate the right and left halves, and this was accordingly done, the behavior of the two being recorded separately. In general thestwo halves behaved similarly, sometimes the one, sometimes the other, beating the longer, depending apparently on the degree of injury each had sustained in removal. The solutions used were quarter normal, this making them practi- 1 Denis: This journal, 1906, xvii, p. 35. 66 Ld LE SAGO HS. cally isotonic with the serum of the animal. The solutions were chemically pure, for the most part of Schuchardt’s preparation. The water used was redistilled in glass. Work was done with solutions of cane sugar, dextrose, levulose, urea, and glycerine. The action of the different sugars is practically the same both quantitatively and qualitatively. Urea and glycerine n ron AA AIA P RR SHINN TANS Tee amt . ROR Re henge etn ININ PTI IMI FUR IS SN | a pS Sj Se dS ee te ee ee Ficure 1.— Four-fifths the original size. Tracing from the turtle’s sinus immersed in an isotonic solution of lavulose, showing fibrillary contractions. act with much greater rapidity than do the sugars. The primary action of glycerine, moreover, appears to be different from that of the other non-electrolytes. In considering the action of these solutions on the rhythm of the sinus tissue, the effects on the rate and on the amplitude of the beats will be taken up separately. That to some extent the two are inter- dependent is of course obvious; but differences of such a character were obtained as could not, it is believed, be explained on this basis. AANA kit FicurE 2.— One-third the original size. Tracing from the turtle’s sinus immersed in an isotonic solution of lavulose, showing the typical periods of rapid rhythm alternat- ing with periods of quiescence. I. Effect on rate of beat.— Urea and the sugars augment the rate of the sinus rhythm. The rhythm is at first perfectly regular, but finally becomes irregular and apparently assumes the character of fibrillar contraction (Fig. 1). In other words, the sinus goes into a state of delirium. That this irregular rhythm actually partakes of the character of delirium cordis could in some instances be seen by direct observation. This irregular rhythm is followed by temporary or permanent cessation of the automatic activity. In some of the prep- arations exhibiting the temporary cessation of the rhythm the al- ternating periods of activity and quiescence were continued for a considerable period before the final stoppage of the rhythm (Fig. 2). In the case of glycerine no primary augmentation of rate, aside from delirium, was observed. The rapid onset of delirium in the glycerine solution is illustrated in Fig. 3. 2. Effect on the intensity of the beat.— All the sugars used pro- duced a primary augmentation of the amplitude of the contraction Rhythm of the Turtles Sinus Venosus. 67 (Fig. 4, 4). This augmentation appeared very soon after immersion of thesinus inthesolution. It is of rather brief duration, and in these respects different from the stimulating action on the rate, which appears later and is of longer duration. The two overlapped, how- ever. The primary augmentation of the strength of the beat is followed by a gradual depression, progressing till the final cessation of the rhythm. Needless to say, the appearance of the delirium was attended by marked decrease in intensity of the contraction. | —L wifi lil Hy NA Don I a IN NIN RRP RIE PIRININ IOSD Pom orn FicurE 3.— About two-thirds the original size. Tracing from the turtle’s sinus im- mersed in an isotonic solution of glycerine. x, application of the glycerine solution, showing rapid onset of delirium cordis in the glycerine solution. Urea apparently does not increase the strength of the contraction. The application of the solution resulted, when the sinus was origi- nally beating strongly, in an almost immediate tonus; any increase of amplitude was of too brief duration to be apparent after even the short interval necessary to readjust the recording lever to the drum. No augmentation of the strength of the beats was observed with the glycerine solution. The primary stimulating action of one of the sugars is strikingly shown in Fig. 5, A. In this case the sinus had been excised twenty- four hours previously, and had at that time been placed in the levu- lose solution till the rhythm ceased. It was then placed in serum on ice, being warmed to room temperature before using. It will be ob- served that the inactive (save for tonus rhythm) sinus, after immer- sion in the lzvulose solution a second time, resumed for a time a perfectly regular rhythm. 3. The duration of the sinus rhythm in the different solutions. — When comparing the rate of action of the sugar, urea, and elycerine solutions, some differences between the Limulus heart ganglion and the turtle’s sinus come to light. The sinus maintains its activity the longest in the sugar solutions, just as Carlson found to be the case with the automatic heart ganglion of Limulus. But glycerine stops the sinus rhythm about as soon as urea, while the Limulus heart 68 ,. £: Fggers: ganglion is brought to a standstill much quicker by the urea than by the glycerine solutions. Hence for the turtle’s sinus the duration of automatism in these solutions does not bear a direct relation to the rate of diffusion of the blood salts into the solutions or the rate of diffusion of the electrolytes themselves in distilled water, or in % sodium chloride. There is considerable individual variation in the length of time that the sinuses from different specimens continued in activity in the solu- Hi i ih il i | ' | H iit ii i) Aint i : ey FicureE 4.— About two-fifths the original size. A, tracing from the turtle’s sinus twenty- four hours after removal from animal. Sinus quiescent, but exhibiting tonus rhythm. x, application of an isotonic solution of dextrose, showing inauguration of the fundamental rhythm by the sugar solution. 4, tracing from the turtle’s sinus immersed in an isotonic solution of dextrose. ., application of the dextrose solution, showing primary stimula- tion by the sugar solution. tion of the same non-electrolyte.' Part of this difference is, in all proba- bility, due to unavoidable injuries to the tissue in preparation, as it was observed in the case of the two halves of the sinus from the same specimen. But, other things equal, the sinus from a vigorous speci- men continues in rhythm in any of the solutions for a longer time than the sinus from a specimen in poor condition. The rate of action of these non-electrolytes is therefore dependent on the condition of the sinus, just as Carlson found to be the case in the Limulus heart ganglion. This is further shown by the fact that a sinus brought to standstill in the sugar solution, then rendered active by immersion in serum, and again immersed in sugar solution, does not maintain its activity in the second sugar as long as in the first. In this connection an interesting phenomenon, observed with one of the sinuses in urea solution, deserves mention. The sinus im- mediately following excision was left in the urea solution till the ' Tn the sugar solutions a sinus has been observed to beat as long as two hours ; for urea the longest period was sixty minutes; for glycerine about sixty-eight minutes. This last sinus was from an exceptionally large and vigorous turtle. Rhythu of the Turtles Sinus Venosus. 69 rhythm had nearly ceased, was transferred to serum on ice for twenty- four hours, and was then again immersed in urea solution. Its period of activity for the second time was about one quarter longer before the same degree of exhaustion was obtained than for the first time. The most obvious explanation of this explanation to the general rule would be on the basis of ‘ shock.” If we regard the irregular rhythm or delirium of the sinus as evi- dence of stimulation, it is obvious that these non-electrolytes have the same action on the turtle’s sinus as in the Limulus heart ganglion, that is, a stimulating action followed by depression and paralysis. In the case of the sugars the primary stimulating action on the sinus ap- pears both in the rate and the intensity of the beats, and comes prior to the onset of the irregular rhythm. The urea solution apparently acts in the direction of stimulation only on the rate, but this is evi- dent before the fibrillary contractions set in, while the primary stimu- lating action of glycerine appears only in fibrillary contractions. It is conceivable that fibrillary contractions may be called forth by a depressor action on the co-ordinating mechanism; but they can also be produced by excessive stimulation, and this is in all likelihood their origin in these experiments, because the stimulating phase of the sugar and the urea action passes into this form of heart activity. The fibrillary contractions of the sinus in these solutions are to all appearance identical with the irregular contraction of the Limulus heart muscle following the immersion of the heart ganglion in similar solutions. — In the case of the Limulus heart these non-electrolytes primarily augment the rhythm when the whole heart as well as when only the heart ganglion is immersed in the solution. When the solution acts on the Limulus heart muscle alone, they depress the rhythm from the beginning. These facts show that the heart ganglion is so much more sensitive than the heart muscle that when the solution acts on both at the same time their primary action on the ganglion is first apparent. On the neurogenic theory of the vertebrate heart rhythm the action of these solutions on the turtle’s sinus is brought into complete agreement with their action on the Limulus heart. The absence of any primary augmentation of the amplitude of the sinus beats by urea and glycerine is probably due to this depressor action on the heart muscle, while the stimulating phase results from the primary action on the sinus ganglia. 70 Td, £,, Leer s: SUMMARY. 1. Isotonic solutions of sugar, urea, and glycerine have, on the whole, the same primary action on the turtle’s sinus as on the Limu- lus heart ganglion or on the entire Limulus heart, that is, in the direction of stimulation. This fact, in view of the further fact that these same solutions have only a depressor action on the Limulus heart muscle, tends to support the neurogenic theory for the vertebrate heart. 2. The ultimate cessation of the sinus rhythm in isotonic solutions of these non-electrolytes is due to some direct action of the non-elec- trolyte on the cells, because the duration of the rhythm in any one of the solutions depends on the condition of the sinus, and, further, because this duration does not stand in direct relation to rate of diffusion of the blood salt into the specific solution, or the rate of diffusion of the specific non-electrolyte into water or % sodium chloride. 3. It is not permissible, at least for the heart tissues, to assume that any of these non-electrolytes are without action on the tissues save through the osmotic factor. ON THE MECHANISM OF THE REFRACTORY PERIOD IN DHS TEAR. By Aj Jj. CARLSON. [from the Hull Physiological Laboratory, University of Chicago.) HE experiments here recorded were undertaken for the purpose of determining whether there is any causal connection between the property of so-called refractory period and the property of autom- atism in the heart. So far as these experiments include the inver- tebrate heart they are merely a continuation of the work on the systolic refractory state reported in this journal last year.! In that work it was shown that the invertebrate heart (molluscs, arthropods, tunicates) exhibits the typical refractory period at the beginning and during systole, that is, a refractory period in the sense of reduced excita- bility. But it is reduced excitability only, and not a condition of in- excitability. The experimental evidence in support of this thesis is conclusive. Observers who deny the presence of a refractory state in the invertebrate heart obviously understand by refractory period a state of absolute inexcitability. Such a condition does not exist in the invertebrate heart, although in some of the molluscs (octopus) and tunicates (ciona) the excitability of the heart at the beginning of the automatic beat is very slight. Three years ago I showed that a refractory period in the sense of absolute inexcitability does not exist in the heart of one of the lowest vertebrates, the California hagfish (Bdellostoma).2 Supermaximal contractions as well as submaximal contractions can be produced in the ventricle, auricle, and hepatic heart of this animal by strong induction shocks sent through the heart at the very beginning of systole; and not only that, but it is less difficult to obtain these super- maximal beats in the auricle than in the ventricle, although the latter 1 CaRLson: This journal, 1906, xvi, p. 67. 2 CARLSON: Zeitschrift fiir allgemeine Physiologie, 1904, ii, p. 259. 71 72 As fs aglsom, exhibits much less automatism than the former. As regards the refractory state, therefore, there is no difference, or at the most only a difference in degree, between the heart of this vertebrate and the invertebrate heart. These observations have now been extended to the higher vertebrates (amphibians, reptiles) with practically the same results. The frog heart and the tortoise heart are excitable at the beginning of systole, just as is the case with the hagfish heart and the invertebrate heart. The view that the systolic refractory state in the vertebrate heart is a condition of absolute inexcitability is therefore untenable for many vertebrates, and will probably prove untenable for all. One series of the experiments was directed towards determining whether the refractory state of the heart is a property of the heart muscle or the nervous tissue or both. The other series aimed to determine the degree of refractory state exhibited by the different parts of the vertebrate heart. If there is a causal connection between automatism and a refractory state in the sense of absolute inexcitabil- ity, we ought to find this condition in the przmas movens of the heart, the sinus, or in mammals, the mouth of the great veins, and we would expect a less degree of refractory state or even none at all in parts of the heart not automatic, for example, the tortoise ventricle, or the apex of the frog ventricle. I. THe TIsSuES IN THE HEART CONCERNED IN THE PROPERTY OF THE REFRACTORY STATE. 1. The automatic heart ganglion of Limulus exhibits the typical refractory period of the heart or a state of reduced excitability during systole. —For the experiments demonstrating this point the dorsal nerve cord or ganglion was isolated from the heart muscle except in the first two heart segments, which were used for recording the ganglionic rhythm. The heart muscle was transsected in the middle of the second segment, and the part posterior to the lesion removed. This leaves the ganglion on the posterior end of the heart free, so that it can be placed on the electrodes and stimulated without the stimulus reaching the heart muscle, while the connection of the ganglion with the musculature of the first two segments insures accurate records of the activity of the ganglion. This method of preparing the heart is described in more detail in my report on the action of temperature variations on the Limulus heart. 1 CARLSON: This journal, 1906, xv, p. 207. Mechanism of the Refractory Period in the Heart. 73 An intensity of the induced shock that produces shortening of the diastole and a supermaximal contraction of the heart muscle when sent through the ganglion towards the end of the systole of the muscle produces no visible effect when sent through the ganglion at the beginning of the muscle contraction. The beginning of the contraction of the heart muscle is, of course, not the beginning of the discharge of the impulse from the ganglion. Hence, by this method it is not possible to stimulate the ganglion at the very begin- ning of the nervous discharge. It is not known whether the time of ia TSR 7 Sa ae —— We U a b Figure 1.— One-half the original size. Tracing from the anterior end of the Limulus heart. The dorsal ganglion isolated from the heart muscle posteriorly and stimulated with induced shocks (make) of uniform intensity. @, ganglion stimulated near the beginning of systole. 4, ganglion stimulated towards the end of diastole, showing diminished excitability of the heart ganglion for some time after the beginning of an automatic discharge. discharge from the ganglion covers the entire time of systole of the muscle. But even if it does, it is obvious that the beginning and the end of the muscle systole lag behind the beginning and the end of the discharge from the ganglion, beczuse the region of the ganglion that is especially automatic is in the middle third oi the heart, that is, removed from the heart muscle in our preparation by a distance of three to four centimetres. And it has been shown in a previous paper that the dorsal nerve plexus in the Limulus heart conducts at the low rate of 40 cm. per second.1 Whatever be the exact time relations between the ganglionic discharge and the systole of the heart muscle, it is certain that the stimulus which is effective when sent through the ganglion at the end of the latter is ineffective when applied to the ganglion at the beginning. This fact can only be interpreted in one way, namely, that the ganglion exhibits diminished excitability for an appreciable time after or during the automatic discharge. Typical tracings illustrating this point are reproduced in Figs. I and 2. The make shock (down stroke of signal) is used for stimu- lus. The tracing in Fig. 1 is from a rapidly moving recording surface. The strength of the shock is the same at aand 6. At a, beginning 1 CARLSON: This journal, 1906, xv, p. 99. 74 A. F. Carlson. of systole, it is practically ineffective, while towards the end of dias- tole (4) the stimulation not only shortens the diastole, but produces a supermaximal beat. It is equally easy to demonstrate that this refractory state of the ganglion is not a condition of absolute inexcitability. If induced shocks of sufficient intensity are used, supermaximal beats are pro- duced by stimulating the ganglion at the beginning of the muscular systole. It may be urged that these experiments do not show that the ganglion is excitable at the very beginning of the discharge, since it is being stimulated an appreciable time after the beginning. I a b FIGURE 2.— Tracing from the anterior end of the Limulus heart. Heart ganglion pre- pared and stimulated as in Fig. 1. a, stimulation of ganglion and the beginning of systole. 4, stimulation of ganglion towards the end of diastole. Showing diminished excitability of the ganglion at the beginning of systole. have tried to demonstrate an absolute refractory state in the ganglion by stimulating it a fraction of a second before the beginning of the muscular systole so as to make the stimulation coincident with the beginning of the discharge, but if the induced shock was strong enough I never failed to get the supermaximal beat. The fact that the gan- glion can be tetanized with the strong interrupted current also tends to show that refractory state of the ganglion is not absolute. A typical tracing showing supermaximal beats following a strong induced shock sent through the ganglion at the beginning of the muscular systole is reproduced in Fig. 3. 2. As long as the heart ganglion is in physiological connection with the heart muscle, the heart muscle and nerve plexus exhibit a condition of reduced excitability at the beginning of systole. —This result is obtained from all regions of the heart. Nor is it necessary that the ganglion is intact in the region stimulated. The ganglion may be extirpated in the first two segments and the heart muscle removed for a distance of one or two centimetres in the third segment, leaving the anterior end connected with the ganglion by the lateral nerve plexus only, and this ganglion-free anterior end exhibits the same condition of reduced excitability at the beginning of systole as the middle region of the heart containing the ganglion. Mechanism of the Refractory Period tn the Heart. 75 The heart muscle and the nerve plexus in the Limulus heart respond to stimulation with induction shocks. “These tissues are therefore directly excitable to the induced current. It is probable, however, that both the muscle and the nerve plexus are less readily stimulated by the induced current than is the heart ganglion, because it has been shown that on extirpation of the ganglion the response of the heart to direct stimulation is greatly diminished.1 The maximal stimulus for the intact heart becomes submaximal on removal of the ganglion. This fact mizht suggest the following explanation of the apparent refractory state of heart muscle and nerve plexus just de- _ OS SERS Oe > Ula Ee Ficure 3.— Two-thirds the original size. Tracing from the anterior end of the Limulus heart. Ganglion isolated posteriorly, and stimulated with a strong induced shock near the beginning of systole. Supermaximal beat following strong stimulation of the ganglion at the beginning of systole. scribed, namely, that it is really the refractory state of the ganglion and not of the muscle and nerve plexus that is brought out by these experiments. The stimulus is too weak to act on the heart muscle and the motor nerve directly, but strong enough to act on the gan- glion and on the afferent reflex fibres in the plexus.2 Hence the re- sponse of the muscle on stimulation during diastole is due to a reflex through the ganglion. And the failure of the response at the begin- ning of systole is due to refractory state of the ganglion at this period. This theory was put to the experimental test in the following way. The ganglion was extirpated in the first two segments and the heart muscle completely dissected away in the third segment, leaving the lateral nerves intact. The two ends of the heart were suspended for simultaneous graphic registration in the usual way, and the electrodes arranged for sending the current through the anterior end from side to side. Now, if the response on stimulating during systole is due to direct action of the current on the heart muscle and the motor nerves, that response should be confined to the anterior end. If it is a reflex through the ganglion, however, the posterior end of the heart should also exhibit the extra contraction. The results contradicted the pro- posed explanation. The extra contraction was confined to the ante- rior end. It is therefore not a reflex through the ganglion, but a 1 CARLSON: This journal, 1905, xiii, p- 217. 2 [bid., 1905, xii, p. 471. 76 A. F. Carlson. direct effect of the current either on the heart muscle or on motor nerve plexus or on both. Hence it follows that @ stzmulus strong enough to produce an extra beat by acting on the heart muscle and nerve plexus fails to produce any visible effect when sent through the same tissues, at the beginning of the normal systole. — In other words, both ganglion, motor nerve plexus, and heart muscle exhibit the systolic refractory state or diminished excitability. In neither of the tissues is the refractory condition absolute. A strong induced shock FIGURE 4.— One-third the original size. Tracing from the anterior end of the Limulus heart. Ganglion extirpated from the anterior end, the lateral nerves being left intact. The ganglion-free anterior end stimulated with induced shocks of uniform intensity. a, stimulation at the beginning of systole. 4, stimulation towards the end of diastole. Showing diminished excitability of the heart muscle and nerve plexus at the beginning of a normal (neurogenic) automatic beat. sent through the anterior end of the heart (prepared as just described ) at the beginning of systole produces a supermaximal contraction just as when the ganglion is similarly stimulated. 3. Do the Limulus heart muscle and motor nerve plexus exhibit a systolic refractory state after being severed from the ganglion? Since extirpation of the ganglion abolishes the automatic contractions per- manently, this question can only be attacked by means of artificial rhythms. Attempts were made to decide the point by stimulating the isolated segments by a uniform series of induction shocks at the rate of every four or five seconds and by means of a second set of electrodes test the excitability of the heart tissues at the beginning of the contraction. The results were not conclusive, mainly because of the difficulty of getting a series of contractions of absolute uni- form amplitude from the Limulus heart after the nerve cord is removed, The interpretation of the tracings from these experiments is therefore difficult. Since these inconclusive experiments were made I have found that the sodium chloride rhythm of the Limulus heart deprived of the ganglion is idio-muscular and that the superficial or dorsal nerve plexus takes no part in the rhythm.!| The sodium chloride rhythm may, in rare instances, be perfectly regular in the rate and the amplitude of the contractions. It is obvious that this rhythm, when 1 CARLSON: This journal, 1907, xvii, p. 478. Mechanism of the Refractory Period in the Heart. 77 regular, gives the conditions for determining whether the heart muscle exhibits the refractory state during an artificial systole. It is true that during the period that the sodium chloride rhythm re- mains regular the heart muscle continues to respond to the stimu- » MUU a m Db mb meb aU m b ¢ UP II Cor Tia nA D SF = So a a ae FIGURE 5. — 4-C, tracings from the isolated frog’s ventricle. , tracing from the isolated toad’s ventricle. m, make induced shock. 4, break induced shock. Stimulation at the beginning of systole resulting in supermaximal beats at times accompanied or followed by tonus contraction. Showing excitability of these ventricles at the begin- ning of systole. lation of the motor nerves, so that on direct stimulation we do not know whether we are stimulating the muscle directly or indirectly through the nerve plexus. But that does not affect this question, as the nerves or the motor plexus are not active in this rhythm and hence cannot possibly exhibit a refractory state. The question will be settled by this method as soon as material is at hand. 4. Is the refractory state in the vertebrate heart a property of the heart muscle apart from the intrinsic heart ganglia and nerve plexus ?! (1) It is needless to say that this question has not so far, and perhaps never can be, attacked by direct experiments. Rohde has recently attempted to answer the question by studying the changes in the response of the frog’s ventricle to direct stimulation after subjecting 1 The experiments in this section were performed in conjunction with Mr. W. B. Walker. 78 A. 7. Carton: the ventricle to the action of chloral hydrate.! According to Rohde © the physiological properties peculiar to the heart tissues are abolished by chloral hydrate before contractility and irritability are abolished. The refractory state is shortened if not altogether abolished, the all- or-none law fails, and appropriate stimulation produces superposition and tetanus. Rohde interprets these results as due to the paralysis of the nervous tissue in the frog’s ventricle by the drug before the muscle is paralyzed. On this assumption the refractory state and the other properties peculiar to the heart are properties of the nervous tissue or of the nervous and the muscular tissues as long as they retain their functional relations, but when the heart muscle is isolated from the nervous tissue it responds to direct stimulation like skeletal and smooth muscle. Schultz,2 working in Howell’s laboratory, has repeated Rohde’s experiments, using the frog and the terrapin ven- tricle. Schultz's experiments practically confirm the results of Rohde, except on the one point touching the complete abolition of the sys- tolic refractory state. According to Schultz, the ventricular tissue retains the characteristic refractory state in chloral narcosis as long as contractility is retained, although the drug shortens the duration of the refractory period. Thus summation of the contractions is pro- duced by stimuli reaching the ventricle during the latter third of the systole. The tracings published by Rohde (Fig. 2 b) do not prove that the “ shortened absolute refractory period” in the sense that term is employed by Schultz is abolished, because in these records some of the stimuli that give rise to the supermaximal beat reach the heart in the last third of systole. I have shown ina previous paper that Rohde’s theory is completely substantiated by the action of chloral hydrate on the tissues in the Limulus heart, this drug paralyzing first the ganglion, then the dorsal or superficial nerve plexus, and lastly the heart muscle.? I am not yet in position to state whether or not a systolic refractory state can be demonstrated in the Limulus heart muscle after paralyzing the nervous tissue by chloral hydrate. The contractions of the heart muscle after such paralysis of the ganglion are those called forth by artificial stimulation, and it is difficult to obtain a uniform series of contractions, even when great care is taken to have the stimulus of uniform intensity. ' RoubeE : Archiv fiir experimentelle Pathologie und Pharmakologie, 1905, Ixiv, p- 104. * SCHULTZ: This journal, 1906, xvi, p. 483. % CARLSON: /d7d., xvii, p. I. Mechanism of the Refractory Period in the Fleart. 79 (2) Nerve tissue probably dies sooner than muscular tissue when respiration and nutrition are stopped. If the refractory state depends on the nervous tissue in the heart, the excised or dying heart ought to exhibit a condition of diminished or abolished refractory state in the later stages while it still retains some excitability and contractil- ity. This hypothesis was tested on the tortoise ventricle, which under normal conditions comes nearer than the frog ventricle to having an “absolute refractory period.” The excised ventricle was kept for a a b C c Figure 6.— One-half the original size. Tracing from the tortoise ventricle after being kept isolated in moist chamber at 20° C. for twenty-four hours. Rhythm started by strong break shock at 4. C, ventricle stimulated with very strong break shocks at the beginning of systole, resulting in supermaximal beats. varying number of days in the moist chamber at room temperature (18°-22° C.) or in the ice box (at a temperature of about 5° C.), and the degree of systolic excitability tested from time to time. The ventricles kept at the lower temperature, of course, retained their irritability and contractility the longest. The longest time a torteise ventricle retained excitability and contractility after isolation was six days (at 3° C.). The ventricle, isolated from the auricles, does not continue in rhythmic activity. It is therefore necessary to resort to contractions artificially produced. A series of induction shocks of uniform intensity was usually employed. Isotonic sodium chloride was also made use of. The systolic excitability was tested on the whole ventricle as well as on ventricular strips. Our results were as follows: (a) The single induced shock of an intensity that fails to produce a supermaximal contraction when sent through the fresh ventricle at the beginning of systole causes a supermaximal contrac- tion in a heart in the last stages of dying (Fig. 6). This is also true for the sodium chloride rhythm of the heart in the last stages of dying, and can therefore not be explained by the staircase phenomenon. The excitability of the dying heart at the beginning of systole is further shown by the fact of summation, two induction shocks follow- ing one another within a shorter interval than the latent period of the ventricle, producing a greater contraction than either stimulus sepa- rably. This form of summation cannot be accounted for by the phe- nomenon of staircase contractions, unless the first stimulus is too 80 A. F. Carlson. weak to produce any contraction at all by itself, which was not the case in our experiments. In the case of ventricles kept for two or three days at the low temperature a rhythmical series of beats could sometimes be started by sending a very strong induction shock through the ventricle. Supermaximal beats were always produced by a strong break shock at the beginning of systole of this rhythm. But while the tortoise ventricle in the last stages of dying is excitable at the beginning of systole, the excitability is lower at the beginning of systole than towards the end of diastole. It is therefore evident that even in the last stages of dying the ventricular tissue or tissues retain the property of systolic refractory state in the sense of diminished excitability. (0) In the last stages of dying the excitability of the tortoise ventricle to the induced current is greatly diminished if not entirely abolished before the excitability to mechanical and chemical stimulation is lost. Thus frequently the strongest induction shock available (two Edison- Lalande cells, type ““S” in circuit, 1100 windings of the secondary coil) failed to produce any contraction, while mechanical stimulation or immersion in an isotonic sodium chloride solution was effective. This might suggest that the heart muscle is, even under normal con- ditions, but slightly affected by the induced current, but because of the greater excitability of the nervous tissues in the heart this fact does not become evident till the latter are eliminated. It is practi- cally certain, however, that the heart muscle itself is more readily affected by all stimuli when in normal condition than when in the last stages of dying. (3) It has been shown in a previous paper that the sodium chloride rhythm of the Limulus heart with the automatic ganglion extirpated is idio-muscular. The dorsal or superficial nerve plexus takes at least no part in this rhythm, although the rhythm is affected by stimulation of this nerve plexus. The systolic excitability of the Limulus heart muscle in sodium chloride rhythm has not yet been studied. If the corresponding rhythm of the tortoise ventricle in Sodium chloride is idio-muscular, we might expect a diminution or abolition of the systolic refractory state in this rhythm, provided the refractory state is a property of the nervous tissue alone. Numerous experiments were carried out with strips from the tortoise ventricle to test this hypothesis. It is nearly as difficult to produce a super- maximal beat by a stimulus at the beginning of systole in the case of the fresh ventricular strip brought to uniform rhythm by sodium Mechanism of the Refractory Period in the Hleart. 81 chloride as in the case of the intact ventricle beating in response to the impulse from the auricles. It has already been pointed out that when the ventricular strips are in the last stages of dying they react differently to systolic stimulation when in sodium chloride rhythm. The fact that the fresh tortoise strips in the sodium chloride rhythm exhibit practically as great a degree of systolic refractory state as the ventricle in normal rhythm, does not prove, however, that the refractory state is a property of the muscle apart from the nervous tissue, as the sodium chloride rhythm of the tortoise ventricle may involve the nervous tissue, just as it does in the Limulus heart when the ganglion is intact. We also repeated the experiments of Rohde and Schultz with the effect of chloral hydrate on the refractory state of the tortoise ventricle, obtaining practically the same result as these observers. A systolic refractory period in the sense of diminished excitability is in evidence as long as the ventricle retains excitability and contractility. It is therefore evident that the question whether heart muscle when isolated from the intrinsic nervous tissues exhibits the property of refractory state to greater degree than skeletal and smooth muscle is still an open one, since the facts bearing on the question can be interpreted either way. In Limulus the heart ganglion exhibits the typical refractory period of heart tissue, and as this is a characteristic of at least many ganglion cells in the central nervous system of verte- brates, it is probable that the ganglion cells in the vertebrate heart possess a systolic refractory state similar to that of the Limulus heart ganglion. Il. THe “DEGREE oF REFRACTORY STATE IN. THE HEART OF DIFFERENT ANIMALS, AND IN THE DIFFERENT PARTS OF THE HEART OF THE SAME ANIMAL. While some observers have held that the refractory period in the vertebrate heart is a condition of diminished excitability only, and not a state of absolute inexcitability, most physiologists employ the term “refractory period” in the latter sense. Thus Howell’s student Schultz speaks of the “absolute refractory period ” as coincident with the time of systole in the frog and tortoise ventricle. I have shown that this conception of the refractory state is not tenable for any part of the heart of the hagfish. Supermaximal beats as well as diminu- tion of the beats are produced in this heart by strong induction shocks 82 A. F. Carison. at the very beginning of systole. In the hagfish heart the refractory state is therefore a condition of reduced excitability only. 1. The ventricles of higher vertebrates do not exhibit the same degree of systolic refractory state. —In the case of the suspended and empty ventricle of the snapping turtle and the mud turtle I have not suc- ceeded in obtaining supermaximal beats by even the very strongest induction shocks at the beginning of systole. These strong induced shocks produce, however, a diminution of the beat without necessarily producing any change in the subsequent rhythm (Fig. 7a). It is obvious that inhibition of a phenomenon is just as much an evidence of irritability of the tissue as the augmentation of it! Hence, when the strong induced shock sent through the tortoise ventricle at the very beginning of a normal automatic contraction diminishes the amplitude of that contraction, it is demonstrated that the tortoise ventricle responds to the strong induced shock at the beginning of systole. The tortoise ventricle is therefore excitable at the beginning of systole, and the refractory state in this ventricle is only a condition of greatly diminished excitability. The empty and suspended ventricle of the frog, toad, and salamander (necturus) differ from the tortoise ventricle in this regard. A break or make shock of sufficient intensity never fails to produce a super- maximal beat in the empty and suspended ventricle of the former animals when applied at the beginning of the beat (Fig. 5). The supermaximal beat may or may not be accompanied by a tonus con- traction and alteration of the subsequent rhythm. But the fact that the supermaximal beat is sometimes obtained without any demon- strable tonus contraction goes to show that it is not produced by addition of the tonus to the fundamental contraction. The fact that the frog’s ventricle responds with a supermaximal beat to the very strong induction shock at the beginning of systole prevented us from using the frog’s heart in the chloral hydrate experi- ments just described. We found, in fact, that the supermaximal beat by stimulation at the beginning of systole is almost as readily pro- duced in the normal ventricle as in the ventricle acted on by chloral hydrate. A certain intensity of the induced shock applied at the beginning of systole may diminish the strength of the beat of the amphibian ventricle instead of augmenting it, but the augmentation is the usual effect. Either phenomenon demonstrates the excitability of the ven- tricle to the induced current at the beginning of systole. This phe- Mechanism of the Refractory Period in the Heart. 83 nomenon of inhibition has now been demonstrated for the heart of all vertebrates and invertebrates tested in this regard, and is there- fore probably common to all hearts! The mechanism of this inhibi- tion is obscure. It has in all probability nothing in common with the normal inhibition of the heart. We have seen that the inhibition is obtained in the tortoise ventricle, which according to Gaskell has WAVAVAVAVAVAWAVAVAVAVAVAVAVAVAVAVAVAVAVAVA OTN Ne Se SS Ur Figure 7.— 4, tracing from the isolated ventricle of the tortoise (Cistudo). Diminution of the beat by strong induced shocks sent through the ventricle at the beginning of systole. &, tracing from the sinus of same specimen as 4. Supermaximal beat by strong induced shock at the beginning of systole. no inhibitory nervous mechanism. The same is true of the heart of many invertebrates in which inhibitory nervous mechanism appears to be absent. 2. The degree of refractory state in the different parts of the same heart. —The refractory state is usually considered to be a common property of all parts of the heart. If there is any causal connection between the property of an absolute refractory state and the prop- erty of automatism, it is evident that the sinus venosus in cold- blooded vertebrates and the mouth of the great veins in the mammals must exhibit an absolute refractory condition at the beginning of sys- tole, because those are primarily the automatic parts of the heart. The auricles must also exhibit this absolute refractory state, as the auricles are invariably more automatic than the ventricles. In the case of the hagfish heart the supermaximal contraction by stimulation at the beginning of systole is more readily obtained in the auricle than in the ventricle. It is therefore evident that the auricu- lar tissues have a greater excitability and contractility at the begin- ning of systole than the ventricle, although the latter exhibits less automatism than the former. In the amphibians (frog, toad, salamander) and reptiles (snapping turtle, mud turtle) examined by me there seems to be no appreciable difference between the degree of systolic refractory condition of the 1 CarLson: This journal, 1906, xvi, p. 100. 84 A. F. Carlson. ventricle and the auricles. Supermaximal beats, apart from tonus contractions, are, if anything, more readily produced by stimulation during systole in the auricles than in the ventricles. In the case of the auricles of the snapping turtle, however, I have been unable to produce supermaximal beats apart from tonus contraction by stimu- lation at the beginning of systole. This stimulation produces dimi- nution of the contraction just as in the ventricle of these animals. The sinus venosus of the amphibians at my disposal (frog, toad, salamander) is small and delicate. Isolation from the auricles and the veins for accurate graphic record of the rate and amplitude of the contractions is therefore difficult without injury to the sinus. My observations on the sinus venosus of these animals were therefore few in number and inconclusive in results. The sinus of the large snapping turtle is easily isolated and strong enough for the ordinary graphic method, and the same may be said of the sinus of the smaller mud turtle. Numerous experiments were made with the sinus ve- nosus of these animals. The sinus of the snapping turtle exhibits on the whole the same degree of systolic refractory state as the auricles and the ventricles of the same animal. That is to say, even the very strongest induced shock at the beginning of systole fails to produce a supermaximal beat unaccompanied by tonus contraction. But this stimulation may diminish the strength-of that beat just as in the other parts of the heart (Fig. 9). The sinus of Cistudo exhibits a differ- ent response to stimulation during systole. Even moderately strong induced shocks sent through this sinus at the beginning of systole produce a supermaximal beat not accompanied by tonus contraction (Figs. 7, 8). A stronger induction shock produces in addition to the supermaximal fundamental beat a tonus contraction lasting for many minutes. But even in this case the supermaximal beat is not the re- sult of the addition of the tonus to the normal fundamental contraction, because the added amplitude is greater than the tonus contraction, even at the end of the subsequent diastole, and the tonus contrac- tion reaches its maximum only after the lapsg of minutes. Bottazzi describes numerous smooth muscle cells among the striated heart cells of the tortoise auricles, and is inclined to believe that the tonus con- tractions of the auricles are due to the contraction of this smooth muscular tissue while the fundamental beat involves the striated cells only. In case the same anatomical conditions obtain in the sinus it is obvious that if the supermaximal beats produced by stimulation ' Borrazzi: Zeitschrift fiir allgemeine Physiologie, 1906, vi, p. 140. Mechanism of the Refractory Period in the Heart. 85 at the beginning of systole were always accompanied by tonus con- tractions the systolic excitability of the tissues concerned in the fun- damental beat would not be proved, for the supermaximal beat might be simply the addition of the tonus (smooth muscle) to the funda- ET SY I See a b Ficure §.—Tracing from the sinus of Cistudo, showing supermaximal beats following stimulation with strong induced shocks at the beginning of systole. The stronger shock (4) produces also a slight tonus contraction. mental contraction (striated muscle). In this tortoise we have, there- fore, the case of a vertebrate heart exhibiting a less degree of systolic refractory state in the automatic part (sinus)than tn the non-automatic part (ventricle). 3. Is there any causal connection between the property of refractory state and the property of automatism in the heart ?—The property of a A a FIGURE 9.— About one-half the original size. Tracing from the sinus of the snapping turtle. Showing diminution of the fundamental contraction (and tonus) by the strong induced shock at the beginning of systole. refractory state in the heart is believed by some physiologists cau- sally connected with the property of automatism.! If we conceive of the stimulus to the heart rhythm to be some chemical substance ever present in the active tissues, it is plain that the systolic refractory state of the automatic tissue would result in periodic response to this constant stimulus. Admitting for the sake of argument that the sys- tolic refractory state is absolute and that the stimulus to the heart rhythm is some chemical ever present in the tissues, it is obvious that the normal maximal intensity of that stimulus fails to affect the heart during a greater part of diastole, a period in which it is admitted by all that the heart tissues are excitable to direct stimulation. This fact shows that the time of action of this stimulus does not coincide with the absolute return of excitability of the heart tissues. The 1 Howe cv: The journal of the American Medical Association, 1906, Ixvi, Nos. 22) 23: 86 A. F. Carlson. stimulus is able to effect a response only when the excitability of the heart tissues has reached their maximum. Hencea refractory state in the sense of diminished excitability would just as surely result in rhythmic activity as a systolic refractory state of absolute inexcitabil- ity. Thus we see that even on the above assumptions an absolute refractory state of the automatic tissues of the heart is by no means a sine qua non for the rhythm. It is furthermore obvious that if the heart muscle is not the auto- matic tissue in the heart, we can account for the heart rhythm even if the automatic tissue exhibits no variation in excitability during the different phases of the heart beat and is being stimulated continu- ously by our hypothetical chemical substance. If, for example, the heart muscle possesses a systolic refractory state but no automa- tism, that is, the contractions are the result of impulses reaching it from the nervous tissues, we can conceive of this nervous tissue send- ing a continuous impulse to the muscle perhaps in a way analogous to the activity of the motor neurons resulting in the tonus of skeletal muscle, but these impulses would be effective only at the point of return of the heart muscle to a certain degree of excitability. This rhythm would not be distinguishable from a rhythm due to the automa- tism of the heart muscle itself. Yet the automatic tissue may not suffer any more demonstrable changes in excitability than is exhib- ited by a nerve fibre after stimulation. But aside from all theo- retical considerations, is the above view tenable on the basis of facts ? The following considerations seem to make the theory untenable : (1) Lhe property of refractory state 1s exhibited by tissues that do not have the property of automatism under normal conditions. — The ven- tricle of the turtle is not automatic, although automatism is induced in it by many artificial conditions; yetall parts of the turtle’s ventricle exhibit a systolic refractory state as absolute as that of the automatic parts of the heart or that of the automatic parts of the heart of any other vertebrate. The Bernstein experiment shows that the apex of the frog’s ventricle is not normally automatic, yet the degree of sys- tolic refractory state is just as marked, if not more so, in the ven- tricular apex as in the auricles; and if we look for instances in support of the above thesis outside of the tissues of the heart, we find such in the motor cells of the central nervous system and in the tissues of the small intestines (mammal). The observations of Horsley and Schafer and Richet go to show that the refractory period of the pyramidal cells of the cerebral cortex and of the motor cells in the cord may be Mechanism of the Refractory Period in the Heart. 87 as great as one-tenth to one-twelfth of asecond. This length of refrac- tory period is exhibited by the nerve cells in response to the normal or physiological stimulus. Whether the refractory state of the nerve cells is absolute or only relative will probably remain an open question forever, because it is not possible to stimulate the cells directly with a very strong induction shock without at the same time stimulating the motor fibres leading from the cells. No physiologist will contend, I opine, that the pyramidal cells of the motor areas or the motor ele- ments of the cord in connection with the skeletal muscles are in auto- matic activity under normal conditions of life. ° The case of the mammalian intestines argues to the same effect. Magnus has shown that the cat’s intestine exhibits a condition of refractory state at the beginning of a spontaneous contraction. This fact has been denied by Schultz, again reiterated by Magnus, and, at least in part, confirmed by Bottazzi.1 The work of Bayliss and Star- ling and especially Magnus has demonstrated that none of the forms of periodic activity of the intestines are due to muscular automatism. The peristalsis, at least, is a complex reflex. The same is probably true of the so-called “pendulum movements.’ It would seem safe to assume that the intestinal musculature as well as the intrinsic motor nerves possess direct excitability. Nevertheless the entire organ exhibits a refractory state at the beginning of an automatic contraction. The nerve plexus and heart muscle of Limulus are not automatic, yet as long as they are in physiological connection with the automatic ganglia, they exhibit systolic refractory state in the sense of reduced excitability. (2) Some tissues that are normally automatic do not exhibit an abso- lutely refractory state, but only a condition of greatly diminished excita- bility. — (a) The hearts of all invertebrates are automatic. Buta sufficiently strong stimulus is effective in the invertebrate heart at the very beginning of systole. This fact, however, does not prove that the automatic tissue of the invertebrate heart is devoid of an abso- lutely refractory state, because the supermaximal contraction produced by the stimulus at the beginning of systole may be due to direct stimulation of the non-automatic heart muscle, while the automatic heart ganglia remain unaffected. Fortunately this objection can be disproven by the Limulus heart. In this heart the ganglion is the automatic tissue, and we have seen that the refractory period of this ganglion is a state of diminished excitability only. 1 For the literature see Borrazzt: Archiv fiir die gesammte Physiologie, 1906, cxill, p. 136. 88 A. F. Carlson. (6) The same is true of the sinus venosus of the mud turtle, as already pointed out, and will in all likelihood be found to be true for the automatic end of the heart of other vertebrates. The excitability of this sinus at the beginning of systole is not due to stimulation of the smooth or supposedly non-automatic muscle cells of the sinus walls. (c) The tonus rhythm of the tortoise auricles does not exhibit an absolute refractory period in the systolic phase! The mechanism of the normal heart tonus is as yet largely a matter of conjecture. The Limulus heart seems to show that the motor nerve-centres in the heart have the same relation to the normal tonus of the heart muscle as has the central nervous system to the normal tonus of skeletal muscle. On the side of contraction the tonus in the heart may in- volve the muscle cells concerned in the fundamental beat, or only the smooth muscle cells scattered throughout the heart, as suggested by Bottazzi. In either case, Porter’s observation modifies the com- mon conception of the refractory state, at least as regards the tor- toise heart, a fact that appears to be overlooked by Howell and Schultz. (3) lu the same heart the parts possessing the greatest degree of automatism may exhibit a less degree of refractory state than the part of the heart not automatic, as shown by the heart of the mud turtle already described. 1 Porter: This journal, 1905, xvi, p. I. A CONTRIBUTION TO THE CHEMISTRY OF CELL DIVISION, MATURATION; AND FERTILIZATION. By A: PS MATGHEWS. [From the Marine Biological Laboratory, Woods Holl, Mass. | INTRODUCTION. HIS paper contains the results of experiments upon the eggs of the starfish, Asterias Forbesii,! and of the sea-urchin, Arbacia punctulata, to determine something of the nature of the chemical processes involved in mitotic cell division. The work was done at Woods Holl, in 1905, in the latter part of August and early Sep- tember, when the starfish eggs are at their best. While the morphology of mitosis has now been very carefully studied, very little is known concerning the chemical processes which furnish the energy for these complex movements. Changes in the amount and staining reaction of the chromatin during cell division early attracted attention and have been studied by many investi- gators, of whom particular mention may be made of Lillienfeld? and Heine.2 These authors drew the conclusion, now widely accepted, that during mitosis the chromatin split into an albuminous sub- stance and into a salt of nucleinic acid. The latter substance formed the chromosomes; the fate of the former was undetermined. It is also known that chemical changes occur in the cytoplasm coincident with these changes in the chromatin, since the cytoplasm undergoes at that time marked alteration in its staining reaction, physiological properties, and physical appearance. The nuclear wall appears to be dissolved or digested, and the astral figures make their appearance. The latter, in many instances at any rate, consist of a kind of cyto- plasm different from that formerly present, if we may rely upon the different affinity for stains it shows. Nothing, however, is known 1 Or Asterias vulgaris. 2 LILLIENFELD: Archiv fiir (Anatomie und) Physiologie, 1893. p. 395. 3 HEINE: Zeitschrift fiir physiologische Chemie, 1896, xxi, p. 494. 89 90 A. P. Mathews. concerning the nature of the chemical changes involved in its pro- duction. It is in fact still uncertain whether the astral figure and the spindle are more or less viscid than the surrounding protoplasm, although the observations of Foot and Strobell! to the effect that the spindle preserves its form after flowing out of punctured eggs of Allolobophora foetida would indicate that it was coherent or jelly- like. It is not possible to say with certainty whether the act of mitosis is accompanied by an act of clotting of the cytoplasm or not, although the evidence is perhaps in favor of such a conclusion. R. Lillie has tentatively suggested that at the astral centres and chromatin hydrogen ions are set free by some auto-digestive process and by their diffusion produce the astral figure, but it cannot be said that ‘there is as yet much evidence of this. The first clear physiological evidence of the chemical nature of the processes involved is found in the work of De Moor? on the dividing cells of the stamen hairs of Tradescantia. He found that in the absence of oxygen these cells could not complete their division. If a cell was dividing when placed in an atmosphere of hydrogen, it went through the division but came to rest without the formation of cell wall. This observation indicated that free oxygen was one of the essentials of mitosis, but it did not conclusively show at what stage it was necessary. Loeb,? by similar observations on animal cells, found that free oxygen was necessary for the division of certain echinoderm eggs and eggs of the fish Ctenolabrus, but that it was not necessary for many divisions of the eggs of the minnow, Fundulus heteroclitus. In Ctenolabrus not only was cell division prevented, but the blastomeres alreacy formed in some cases re-fused in the absence of oxygen. The dependence of mitosis upon oxygen was more carefully ex- amined by Lyon,? who found that oxygen was more necessary at certain definite periods of the process of fertilization. For the eggs of the sea-urchin, Arbacia punctulata, he showed that if the eggs after fertilization were placed in sea water deprived of oxygen the division came to a stop. On removing the eggs to oxygenated sea water after several hours’ immersion in the oxygen-free water, the division in some cases recommenced and proceeded normally, while ' Foor and SrropeLL: American journal of anatomy, 1905, iv, 199. De Moor: Archives de biologie, 1895, xiii, 163. to ® Logs: Archiv fiir die gesammte Physiologie, 1895, Ixii, p. 249. * Lyon: American journal of physiology, vii, 1902, p. 56. Cell Division, Maturation, and Fertilization. gI in other cases the eggs did not recover. On further investigation it developed that if the eggs were placed in hydrogen about the time of the first cleavage they never recovered. Exactly the same phe- nomenon was discovered for their susceptibility to sea water con- taining potassium cyanide. About the time of the first division they were highly susceptible. Lyon was uncertain of the exact period at which they were susceptible, but thought it immediately after cell segmentation. Spaulding,! however, working with ether and hydro- chloric acid, came to the conclusion that maximum susceptibility was just before or during segmentation, and this conclusion I have recently confirmed? As ether, acids, and cyanides all interfere pri- marily with cell respiration, these results all show that about the period of first segmentation, when the aster is growing with its greatest rapidity and the nuclear wall dissolving, any agent which interferes with cell respiration is peculiarly fatal to Arbacia eggs. The significance of this fact will appear later. Lyon® in a subse- quent paper studied the production of carbonic dioxide by these eggs during fertilization, and succeeded in showing that carbon dioxide was not produced evenly throughout the process, but that coincident with the entrance of the sperm and about the time of the first segmenta- tion there was a marked increase in its production.* These facts, which so clearly pointed to the conclusion that mitosis was accompanied, if not caused, by modifications in the respiratory activity of the egg, were the starting-point of this investigation in which I have tried to discover whether these respiratory changes had any causal relation with cell division. I hoped to get some evidence, also, of the cause of the differences of electrical potential assumed to exist in the egg and to cause cell division by R. Lillie, Spaulding, and Hartog. THe THEORY OF RESPIRATION. The work of many authors on the nature of respiration has led to certain general conclusions which I have embodied in my paper ® on the theory of cell respiration. That theory is a modification of the 1 SPAULDING: Biological bulletin, 1903, vi, p. 97. 2 MATHEWS: /did., 1906, xi, 137.- 8 Lyon: American journal of physiology, 1904, xi, p. 52. 4 Lyon: Loc. cét., p. 57, says that one would infer that the energy for cell division comes from fermentative rather than oxidative processes. *’ MATHEWs: Biological bulletin, 1905, vili, p. 331. 92 A. P. Mathews. earlier theory of Hoppe-Seyler based chiefly upon the work of Nef. The general conclusion of all the work on respiration is to the effect that respiration is probably due to at least three distinct factors: First, to a strong reducing substance, which must be constantly pro- duced by cell metabolism. This substance is inall likelihood a carbon compound, and has the power, possibly by setting free two of the valencies of a carbon atom, of attacking water, oxidizing itself and setting free nascent hydrogen. What this substance is, where it is located, and whether there is one or several such substances is unknown, except that it is probably produced by, and exists primarily in, the region of the nucleus, which, as is well known, generally lies in the most strongly reducing part of the cell. Second, to the cell oxidases. The manner of action of these substances and their com- position is still uncertain, but there is no doubt that there exist in cells substances whose presence enormously hastens the action of atmospheric oxygen on the cell contents. These oxidases are also probably formed by, or are a constituent of, the cell chromatin, since the observations of Spitzer,! Lillie,? and Croftan,? among others, indi- cate that they accompany the cell chromatin in its chemical separa- tion. Croftan’s observations, particularly, seem to show that one at least of these oxidases consists of nucleinic acid in combination with , an albumose. And ¢7rd, to atmospheric oxygen. If cell division is at its basis a respiratory process, we have to look for and locate in the cell the first two factors and demonstrate the importance of the third. The mitoses chosen for study were the maturation mitoses and subsequent phenomena of the starfish egg, and the first segmenta- tion of the starfish and sea-urchin egg. THE CHEMICAL PROCESSES INVOLVED IN THE MATURATION OF THE STARFISH Eaa.* The living, ovarian eggs of the starfish when fully developed con- sist of a clear, viscid protoplasm containing a multitude of partially transparent granules’and at one pole a very large germinal vesicle * SPITZER: Archiv fiir die gesammte Physiologie, 1895, lx, p. 303; /ééd , 1897, Ixvii, p. 615. 8 Littie: American journal of physiology, 1902, vii, p. 412. ® CROFTAN: Medical record, New York, 1903, Ixiv, p. 9. * For a description of the morphological processes see the paper by WILSON and MATHEWS: The journal of morphology, 1994, x, p. 319. Cell Division, Maturation, and Fertilization. 93 with a large nucleolus. The eggs do not, as a rule at any rate, mat- urate in the ovary, but only after shedding into the sea water. As soon as the eggs are shed they become a little more opaque, and if thoroughly ripe, the germinal vesicle wall becomes irregular and disappears in the course of ten minutes. The nuclear sap then mixes with the cytoplasm; the nucleolus dissolves and disappears; the chromatin dissolves for the most part; the maturation spindles are formed, containing a small part of the chromatin; and two polar bodies are extruded. The whole process to this point takes between two and three hours. The nucleus then re-forms at first as a little group of vesicles, but later these fuse, and the nucleus grows greatly in size and moves toward the centre of the egg. For the next three or four hours it continues to increase in size until its diameter may become more than half that of the germinal vesicle. Throughout this period the cell protoplasm remains clear and transparent, but about eight or twelve hours after the maturation has begun, the cytoplasm becomes more and more opaque, and if not disturbed it is finally con- verted into a coarsely granular, opaque, and dead mass. The essence of the process of maturation consists morphologically, as pointed out by Delage,! in the discharge into the cell cytoplasm of the whole of the nuclear sap, the greater part of the chromatic network and the dissolved nucleolus, and the formation of the small maturation spindles. At maturation, therefore, two portions of the egg, z. e., nu- cleus and cytoplasm, which have formerly been held separate, are intermingled. All that happens to the unfertilized egg thereafter — its division, its early death, its growing opacity and wholly changed physiological potentialities — may be ascribed to this admixture. The early death and opacity of the egg after maturation occurs only if the egg be not fertilized or if the cell be left undisturbed. It may be wholly prevented by the entrance of the sperm, or by the various means producing artificial parthenogenesis. The early death of the maturated, unfertilized egg is in striking contrast to the fate of the eggs of which the germinal vesicle remains intact, and no ad- mixture of nuclear and cytoplasmic matter takes place. Such eggs remain clear, transparent, and living for two or more days. The first questions to be answered are, what causes normally the dissolution of the nuclear membrane, and thus inaugurates the process of maturation by mixing cytoplasm and nuclear contents? and what change in the egg is produced by its maturation, which leads to its early death? 1 DELAGE: Archives de zoologie expérimentale et générale, Igol, ix, p. 285. 94 A. P. Mathews. As regards the answer to the first question, my results and conclu- sions confirm those of Loeb! on the same object. The cause of the beginning of the process of maturation is to be found primarily in the presence of free oxygen in the sea water. If eggs are shed into water deprived of its oxygen, they will not mature. However, hy- droxyl ions in any numbers are not necessary, as Loeb suggests, since eggs will mature in pure solutions of calcium chloride of 2 mol con- centration. J attempted, however, to cause maturation in the ovary by exposing this to oxygen directly, but without success. This negative result, however, I do not regard as conclusive, owing to the fact that it was tried early in the summer at a time when the ovaries were not in very good condition. There is, however, no doubt that free oxygen is essential to the beginning of normal maturation, and that the first visible sign of maturation consists in the disappearance of the membrane of the germinal vesicle at a point where it comes nearest to the surface of the egg. It may be recalled in this con- nection that mechanical shock is able to inaugurate maturation in eggs which normally do not maturate, and that such eggs show on section ruptured germinal vesicles.” DoES AN OXIDASE ESCAPE FROM THE NUCLEUS ON THE DISSOLUTION OF THE NUCLEAR MEMBRANE ? We come now to the second question as to the nature of the changes produced in the cytoplasm by the admixture of nuclear mate- rial, —a change which leads to the early death of the egg, if the latter is not fertilized, and which so totally changes the potentiality of the cytoplasm. That the cytoplasm after maturation does differ markedly from that existing before fertilization is shown by its behavior. It acquires the power of forming a fertilization membrane; under suit- able conditions it can generate asters; it becomes capable of cell division; and, if left alone, it rapidly dies. The early death of the egg after maturation was studied and de- scribed by Loeb,? but he undertook no experiments to determine its Cause or nature. He says, however:* ‘I can only suggest that the 1 Lors: Biological bulletin, 1g02, iii, p. 295; Archiv fiir die gesammte Physi- ologie, 1902, xciii, pp. 59-76; also The dynamics of living matter. In his last statement Lorn lays more stress on the oxidation phenomena, but suggests that certain splitting phenomena are favored by it. 2 MorGAN: Anatomische Anzeiger, 1894, lx, 150. \ 8’ LoEr: Biological bulletin, 1902, iii, p. 306. A -PEne tips 500. Cell Division, Maturation, and Fertilization. 95 processes underlying maturation are at least in some form of a de- structive nature, one might think of autolytic processes, which the egg cannot withstand for an indefinite length of time without dying.” He conceived that the spermatozoon saved the life of the egg by ac- celerating a “‘ series of chemical changes, syntheses, in the egg which do not occur sufficiently rapidly without spermatic, chemical, or os- motic fertilization.” Again he says: ‘“ The chemical processes under- lying maturation are not identical with those which bring about fertilization.” Delage also is uncertain of the nature of the processes involved, and makes only the very broad suggestion that possibly fer- ments have something to do with it. Delage did, however, clearly recognize the fundamental fact of the importance of the admixture of nuclear and cytoplasmic substance. I first tried to discover whether oxygen had anything to do with the early death of the egg. This is shown conclusively and very simply to be the case if the eggs after maturing are brought into water freed from its oxygen by hydrogen, or if they are left ina dense mass at the bottom of a dish so that the underlying eggs are deprived of their oxygen. Eggs so treated do not become opaque for twenty hours, or even longer, but remain clear and, so far as can be judged, alive. For example, a lot of eggs were shed into sea water at 9 A. M. At 9.30 nearly all had begun to mature. At 10.10 I transferred one portion to sea water through which a stream of hydregen gas was passing. At 2 and 4 p.m. some of the eggs were retransferred from the hydrogen to fresh sea water. The remainder were left in the hydrogen for twenty-four hours. When examined at the end of this time, all eggs which had been transferred to oxygenated water were found dead and opaque. The eggs which had remained in the hydrogen, on thecontrary, although mature, were perfectly normal in appearance. The cytoplasm was clear, fertilization membranes were out, the nuclei had re-formed. On transferring these eggs to fresh sea water a few became swollen in one hour; the remainder retained their living appearance for nearly three hours and then became opaque. The foregoing experiment shows that the early death of the eg¢e after maturation occurs only if free oxygen is present. It may be concluded, then, that death is brought about in this case by an oxida- tion of the cell cytoplasm, and that this takes place much more rapidly after the germinal vesicle contents have been discharged into the cytoplasm than before, since if the nuclear wall remains intact the egg does not become opaque even in the presence of oxygen. 96 A. P. Mathews. There can be no doubt, I think, from these experiments, that at maturation a substance which greatly hastens the action of atmos- pheric oxygen upon the cell cytoplasm is poured into the cytoplasm from the nucleus, or at any rate becomes active in the cytoplasm as a result of such an admixture. A substance which thus facilitates oxidation is called an oxidase. Inasmuch as no means are known of producing such a change in the cytoplasm in the absence of nuclear admixture, I believe the most probable conclusion is that at matura- tion an oxidase escapes from the nucleus and spreads throughout the egg cytoplasm, and this oxidase, tn the presence of oxygen, leads, tf tts action zs not checked, to the death of the egg. Is THE OXIDASE THUS THROWN INTO THE CYTOPLASM ONE OF THE FACTORS IN THE PRODUCTION OF THE ASTRAL FIGURE? That the production of the sperm aster after the entrance of the sperm is conditioned by the same factors as condition the early death of the egg after maturation, is shown by the following evidence. Sometimes sperm will enter non-mature starfish eggs, or they enter before maturation takes place. In all such cases I have found, by an examination of both living and sectioned material, that the sperm does not form its ordinary large aster, but instead all that is produced is a very minute series of radiations in the neighborhood of the sperm head and formed presumably about the middle piece or centriole of the sperm. If, however, the germinal vesicle wall has disappeared spontaneously or been ruptured by mechanical shock, the aster begins to form as soon as the sperm enters, and normally its radiations shortly extend far out into the cell protoplasm. This observation shows very clearly that there is a fundamental chemical difference between the protoplasm of a mature and an immature starfish egg. Although the sperm carries a centriole into the egg or forms a centriole soon after entering, this is in- capable of forming a good-sized aster unless the nuclear wall has disappeared and the nuclear sap has spread through the cell. Indeed it may be said that, from this point of view, the egg partially fertilizes itself by the contents of its germinal vesicle. The impossibility of forming an aster in the immature cytoplasm, as contrasted with the cytoplasm after maturation, has been found also in artificial fertilization. Thus starfish eggs do not undergo artificial parthenogenesis by the action of acids or other agencies, unless maturation has occurred or at least begun. Similarly Yatsu! 1.YATSU: Journal of experimental zodlogy, 1905, ii, p. 287. Cell Division, Maturation, and Fertilization. 97 found that in pieces of the cytoplasm cut from mature eggs of Cerebra- tulus aster formation was easily caused by magnesium chloride. It was impossible to form asters in pieces which had been cut from the eggs before maturation had occurred. Miss Foot! has made similar observations in Allolobophora. The cone and aster depend on the stage of maturation. No matter how far the sperm penetrates, one never sees a sperm attraction sphere before the anaphase of the first polar spindle. Similar observations have been made by many other students. It is therefore clear that the cytoplasm becomes capable of forming a good-sized aster only after the mixture with it of the nuclear contents. I have collected many other facts which show this same relationship between aster formation and the discharge of nuclear material. Thus, in Toxopneustes, as described by Wilson,” and in Arbacia after fertil- ization and conjugation of the pronuclei, there ensues a stage lasting thirty to fifty minutes in the period after fertilization during which the astral radiations are diminishing and ultimately the coarser . radiations die out. This is a period of retrogression of the asters, during which the nuclei are greatly increasing in size, but are sharply cut off from the cell cytoplasm. At the end of this period the nuclear membrane disappears with some abruptness opposite the asters, so that the nuclear sap here comes directly in contact with the astral substance. There ensues a sudden outburst of activity on the part of the asters. They grow enormously, and the radiations extend throughout the cell. I have observed closely parallel phenomena in the starfish egg. Starfish eggs were shed in water and fertilized at 8.45 A.M. At 10.30 the pronuclei were lying in contact side by side. At 10.45 the nuclei fused. The astral radiations were small, extend- ing but a short distance from the centres. About 10.50 the nuclear membrane rapidly disappeared, and immediately following this disap- pearance great rays developed through the egg, reaching a maximum at II to 11.15, when the division took place. This disappearance of the rays after the nuclear membrane has re-formed and their sudden and great development after the solution of the nuclear membrane has been described for a variety of eggs. In some cases it may go so far that the whole astral structure disappears together with the centrioles, while in other cases only a reduction in the size of the asters and 1 Foor: Journal of morphology, 1897, xii, p. 809. 2 WILSON: Journal of morphology, 1895, xv, p. 452. 8 See especially Cork’s observations on various echinoderms and Cerebratulus, Journal of morphology, 1899, xvii, p. 455. 98 A. P. Mathews. radiations is noted. Thus, Miss Foot! states that, after or about the time of the conjugation of the pronuclei, the sperm astral centres totally disappear; the asters of the first segmentation nucleus are new formations. In Thalassema, according to Griffin,” the asters become smaller and less distinct and grow again greatly after the dissolution of the mem- branes of the nuclei. F. R. Lillie® states for Unio that the sperm asters undergo retrogressive metamorphosis, and as they disappear the yolk granules flow in and gradually obliterate all traces of the clear areas where the asters were. The centrosome disappears completely. The first segmentation asters appear as new formations. In Bufo,* when the pronuclei are in contact, every trace of radial systems disappears. The asters are round masses surrounded by pigment granules which formerly made the rays. When the nuclear wall disappears, the astro- spheres increase enormously in volume and rays appear. These observations indicate clearly that one of the factors of the formation of asters, both in the normally fertilized egg and in cells not fertilized, is to be found in the nucleus, $resumably in the nu- clear sap, since the change in the protoplasm extends throughout the egg substance. The admixture of this sap enormously stimulates the activity of the aster, whatever be the origin of the latter, whether from sperm or from the egg itself. It is, therefore, the action of the nuclear sap on the cytoplasm which enables the sperm to perform its essential act of fertilization, namely, the development of asters, and which leads to the death of the egg if this act is not performed. I will now show that the essential pro- cesses involved in these two cases are also parallel in that each is dependent for its fulfilment upon atmospheric oxygen. This has already been shown for the death process; it is also necessary for the astral formation. THe ASTRAL FIGURE DEPENDS FOR ITS EXISTENCE IN ARBACIA AND ASTERIAS UPON THE PRESENCE OF FREE OXYGEN ON THE SURFACE OF THE EGG. Wilson ® showed that ether and presumably other anesthetics caused the astral rays in Toxopneustes eggs to disappear. The ' Foor: Journal of morphology, 1897, xii, p. 809. ? GRIFFIN: Journal of morphology, 1898, xv, p. 594. ® LILLIE, F. Rs 72a. 1690, SVilg px Saas *ISING ? 202d., Pp. 333- 5 WILson: Archiv fiir Entwickelungsmechanik, 1902, xiii, p. 365. Cell Division, Maturation, and Fertilization. 99 rays reappeared when the anesthetic evaporated. This observa- tion is suggestive, since ether checks cell respiration. It suggested to me that the absence of oxygen should produce the same result. I examined accordingly the action of hydrogen gas on a lot of living starfish eggs and sea-urchin eggs at the time of the first cleavage. I used a Leitz ,4 oil immersion ocular 4, the eggs being under slight compression by the cover glass. The eggs of Arbacia and Asterias were fertilized and transferred after different intervals to sea water which contained no oxygen. The sea water had been boiled for two minutes, and hydrogen gas passed through for two to three hours, precautions being taken to avoid evap- oration. After remaining in the oxygen-free water for ten to thirty minutes, the eggs were lifted out with a little of the water, placed quickly on a slide, and covered at once with a cover glass and exam- ined. The result was always the same. If the eggs were introduced into the hydrogen before the formation of a large aster, they did not develop the aster; if they had been introduced during the height of astral formation, the rays had faded out and disappeared, and only clear areas in the egg marked where the nucleus, spindle, and aster substance was. I then lifted the cover glass, mixed fresh water with the eggs, left them exposed to the air for a minute to permit the entrance of oxygen, and re-examined. In from three to five minutes the radiations began to appear, and in ten to fifteen min- utes or longer, depending somewhat on the length of time they had been exposed to the hydrogen, the egg became filled with great radiations. These experiments show clearly that the coarse, peripheral, astral radiations of the first segmentation amphiaster of these eggs depend for their existence upon the presence of oxygen. They disappear if the oxygen is cut off; they reappear if the oxygen is readmitted. We see, therefore, that even though a substance be discharged from the nucleus into the cytoplasm, this is unable to form a large aster in these eggs unless oxygen is present. TZhzs observation indt- cates that the substance escaping from the nucleus and active in the formation of an aster is an oxidase, since it ts active only im the pres- ence of oxygen. As additional evidence of this conclusion, the peculiar action of quinine toward these eggs may be cited. Quinine, as is well known, has a very remarkable power of checking certain oxidations.. Thus 100 A. P. Mathews. very small amounts of it prevent the oxidation of guaiac by ozonized turpentine and blood. The Hertwig brothers! long ago observed the peculiarly poisonous action of quinine on dividing ova. In repeating and extending their observations, I was astonished at the very rapid fading out of the astral figure produced by this poison. The radia- tions disappeared as if by magic, and the cell disintegrated very rapidly. No other poison which I examined was quite so fatal to dividing cells as quinine. The following experiment illustrates the action of various drugs on the astral radiations and the life of dividing starfish eggs. A very fine lot of starfish eggs which maturated very uniformly were fertilized at 8.30. They were placed in the various solutions at 11.14, about ten minutes before division. Large asters had appeared in some cells; in others the nuclear membranes had not disappeared, and the asters appeared as small, clear areas with small radiations at the nuclear poles. SOLUTION. RESULT. 1. Quinine sulphate .1% . Radiations disappear at once. Eggs become opaque ‘and swell in 20 minutes. “ s 01% . Astral radiations disappear at once. No segmentatiagn. Cells darker. 24 hours all dead. 5 “005% . Radiations disappear. Astral centres as clear areas. No segmentation. 24 hours all dead. . “001% . Three-fifths of eggs slowly segment. Far later than control. A few not killed in 24 hours. 2. Atropine sulphate .1% . About } segment. 24 hours dead. cs se 05% . Allsegment and form morule. Next morning dead. me ss 02% . Segmentation somewhat retarded. 24 hours nearly ail dead. a - 01% . Allségment. Majority form abnormal swimming blastulz. 3. Caffeine hydrochlorate.1% Eggs remain clear many hours. Several centres in unseg- mented eggs. + segment in 2 hours. Next morning many living. me 05% . Segmentation delayed. Radiation somewhat reduced. Many divide at once into 4 cells. 24 hours living. - 02% . Segmentation retarded, but all develop. Abnormal gas- trule in 24 hours. ¥ «01% . Segmentation at first not retarded. Gastrulae not so good as control. 4. Physostigmine (alkaloid) Many eggs partially divide, but blastomeres refuse. At Fresh solution: .05% . 1.53 no centres visible in any eggs, but periphery of egg buds off small pieces of protoplasm. as uf 025% . A few cells divide completely. Radiations strong. Remainder show incomplete division. 1n 24 hours all dead. 1 HeRTWIG, O. and R.: Jenaische Zeitschrift, 1887, xx, p. 120. Cell Division, Maturation, and Fertilization. IOI Fresh solution. .001% . . Many divide. Centres visible. Stop after one division. Blastomeres concave toward each other. Dead in 24 hours. 3 « 0005% . First division in all, but no further. Peculiar change in cytoplasm as noted. 5. Aconitine sulphate .05% Nearly all divide and develop. a of .025% Nearly all divide and develop to gastule. Die after 48 hours. P on .001% Ail develop. 48 hours alive and swimming gastule. s .005% Development nearly normal. 6. Cocaine hydrochloride .1% Division retarded. Centres divide. Rays reduced. Dead in 24 hours. “ ee .05% First segmentations retarded, but not prevented. Rays reduced. 24 hours dead. é sf 02% Like control at first. Dead in'24 hours. “s es .01% Some retardation. Dead in 24 hours. ‘7. Strychnine sulphate .05% Centres divide. Eggs not. Radiations reduced. as < .025% A few segment. Majority not. a i 01% Nearly all divide, but retard. Alive after 24 hours. ss 3 .005% Like.control, but dies in 48 hours. 8. Adrenaline chloride .007% First two divisions take place, but protoplasm altered like physostigmine. All dead in 24 hours. fs “ 0035% Some 4cell. Do not develop farther. Same protoplasmic change. 9. Veratrine chloride .03% . All divide and develop, but retarded. Dead in 24 hours. The preceding experiment shows the astonishing fact that quinine sulphate toward dividing eggs is more poisonous than atropine, pilo- carpine, caffeine, physostigmine, aconitine, cocaine, veratrin, or strych- nine. Its power of checking cell division is even greater than sodium cyanide. Ht is indeed surprising that quinine, a drug which may be taken with impunity in relatively enormous doses by mammals, should be so extraordinarily poisonous toward dividing cells of the sea-urchin and starfish. The effect of this drug on the astral figure is almost instantaneous. A dilution of one part of the sulphate to 17,000 of sea water is sufficient to wipe out the large astral radiations and to prevent segmentation. On the other hand, veratrin and aconitin, two of the most fatal and powerful poisons known, are toward the process of segmentation relatively inert. Tue EFFECT OF COLD ON THE ASTRAL RADIATIONS. It is well known that cold checks the respiratory processes. It ought, hence, if the aster is a dynamic and not a static structure, to cause a disappearance of the rays.! 1 Bovert has tried similar experiments from a different point of view (Sitzungs- berichte der physikalischen-medicinischen Gesellschaft, Wiirzburg, 1897). 102 A. P. Mathews. A lot of ripe starfish eggs were fertilized just after the germinal vesicle membrane had disappeared. Just before division, when most of the eggs had very large asters in them and radiations extending to the periphery of the egg, I put them in sea water cooled to +2° C. After twenty-five minutes’ exposure to the cold I examined them. Nearly all eggs were undivided. The great astral radiations had dis- appeared, and only clear areas without rays could be seen, generally two in each egg. I then carefully warmed the slide on my hand. The rays began to appear almost at once, and very soon enormous astral radiations appeared about each centre, and division took place. Some eggs were left for three hours at +2° to+5° C. In these eggs three or four clear areas without ravs could be seen. On warming rays appeared about each centre. THE NATURE OF THE CENTRIOLE. We have thus far been able to trace two elements of respiration in cell division: the oxidase, formed by the nucleus and thrown into the cytoplasm, and free oxygen. There remains the other element, z. e., the reducing substance, the substance producing nascent hydro- gen, to be accounted for. It occurred to me that that substance might be the centriole. If that were the case, the astral centre should show a strong reducing action. I can find very little evidence of the chemical nature of the cen- triole. From its staining reaction and digestibility in pepsin-hydro- chloric acid, it is not apparently a nuclein, but, as Zacharias ! maintains, an albuminous substance. It may, however, be like the oxidase, an . albuminous derivative of the chromatin. It is not even certain that all the centrioles in the cell have the same chemical composition. If, as is possible, the centriole has a reducing action and causes the astral figure in virtue of that fact in the way shortly to be considered, any reducing substance, whatever its chemical composition, would have a similar action. As there are certainly several reducing sub- stances in every cell, it will be seen that it will not do to assume that all centrioles have the same composition. In the literature I find but a few scattered observations on the possible reducing action of the centriole, or centroplasm. It is of course well known that the interior of the egg in which the asters normally develop has during life a strong reducing action. Ehrlich’s 1 ZACHARIAS:. Berichte der deutschen botanischen Gesellschaft, 1898, xvi pp- 185-198. Cell Division, Maturation, and Fertilization. 103 work clearly shows this fact, which was indeed known long before. Evidence for it in the starfish egg will shortly be presented. The centrioles first appear ordinarily close to the nuclear wall or even in the nucleus. It is exactly about the nucleus that the strong- est reducing action is to be found. Thus Beyer! observed. that if tellurium oxide was injected it was deposited as metallic tellurium about the nucleus. Methylene ‘blue and other stains are reduced par- ticularly in this region. The centrioles and asters therefore normally arise in the region of most intense reduction, even though they may not be the cause of that reduction.” Foot and Strobell? report that in the eggs of Allolobophora fcetida, fixed in chrome-acetic and then exposed to osmic acid vapor, the centrosomes appear as brownish yellow granules, not so black as certain other granules in the cytoplasm. These observations indicate that the centrosomes fix and reduce osmic acid. Field * reports also that if the sperm of echinoderms be exposed to osmic vapor the tip and middle piece of the sperm reduce the osmic acid far more than the nucleus. To determine this point more directly, eggs with large asters were placed in a O.O1 per cent solution of methylene blue in sea water. The methylene blue enters the egg with great ease. In two to three minutes the granules in the outer third of the protoplasm become stained a light blue. If the eggs are left in the solution beneath a cover glass, the interior of the egg does not color even after the outer third is an intense blue. This fact indicates that the interior of the cell is strongly reducing, and as rapidly as the stain penetrates it is reduced. The clear protoplasm at the centre or at one side of the egg which represents the asters, spindle, and the chromatin will not stain in methylene blue for a very long period. Inasmuch as there is no membrane about the spindle or in the egg to prevent the pen- etration of the stain, and as the cell stains rapidly and completely if boiled first, I am of the opinion that the centres and the spindle will not stain, owing probably to their strong reducing action. However it might also be urged that the stain will not attack the centres, because there are no granules in them with which it ordinarily combines. This evidence is not, therefore, conclusive. 1 Beyer: Archiv fiir Physiologie, 1895, p. 231. 2 It is possibly on account of this intense chemical action that the chromatin will not stain during “life.” 3 Foor and STROBELL: Zodlogical bulletin, 1899, ii, p. 131. 4 FreLp: Journal of morphology, 1895, xi, p. 240. 104 A. P. Mathews: Further evidence of some value was obtained in the following way. The protoplasm of these eggs consists of a homogeneous matrix in which myriads of minute granules of varying sizes are embedded. If we knew whether these granules were electronegative or electro- positive, we might infer from their behavior whether the astral centres were oxidizing or reducing, since all oxidizing regions would be electropositive. The behavior of these granules is extremely interesting. Originally they are distributed evenly all through the cytoplasm. Just as soon, however, as the centriole appears, the clear homogeneous protoplasm begins to gather around it, and the granules move away or disappear in the neighborhood of the aster. The result is that finally the asters and spindles are left in the midst of a clear homogeneous pro- toplasm which is entirely free from these granules. This clear area can only have arisen by the granules moving out- ward and the clear substance inward toward the centre or by the dis- solution of the granules in the clear substance. I was unable to satisfy myself which of these processes took place, but I am inclined to think that both occur, although I could not actually observe with entire certainty the solution or movement of any single granule. It is well known that if the centre is reducing, it will act as a nega- tive electrode; if it is oxidizing, as a positive electrode. If the granules are electropositive and the centre is a reduction centre, the granules should be attracted toward the centre, and should fuse into larger granules near the centre, following the well-known behavior of colloidal particles and suspensions toward electrodes. If, on the other hand, the centres are reducing, the granules, if electronegative, will be repelled from the centre, they will move outward, and should grow larger toward the periphery of the end, and those near the centre should dissolve if they cannot move away. The first point, therefore, was to determine accurately the character of the charges on the granules. To determine this point I used the various stains, and I quickly found, as had been discovered for other cells already by Fischer and Overton, that the granules united only with basic dyes. Indeed this is the easiest method I know of for telling an acid from a basic dye. One puts the starfish eggs in a .03 per cent solution of the dye, al- lows them to remain five minutes, and then examines in fresh sea water. No acid dye stains. I used the following dyes: Celi Division, Maturation, and Fertilization. 105 Basic. ACID. Methylene blue. Methyl] blue. Thionin. Acid fuchsin. Dahlia. Acid violet. Methy] green. Acid green. Methyl violet. Carminate of sodium. Safranin. Indigo-carmine. Toluidin blue. Erythrosin. Neutral red. Kosin. That the basic stains form a combination with the granules and are not simply dissolved in an oil drop, as suggested by Overton, is shown by the fact that in some cases, particularly with neutral red, the gran- ules form insoluble red granules. These observations show that the granules in the living cell without exception, so far as I could see, are electronegative. These same granules are electropositive in fixed specimens, owing tothe action of the acid in the fixing fluids. Their behavior toward the astral centre indicates, therefore, that this is the region of intense reduction, since here they are either dissolved or are repelled from it, or, as is more probable, are in part dissolved and in part repelled. Conversely, when the asters die out, the granules move again toward the centre or reappear there, as is seen, for example, in the eggs of Unio, and as may easily be seen in many other eggs where the sperm or maturation asters die out before the segmentation spindle is formed. — The behavior of these granules, in connection with the other obser- vations already mentioned, indicates, in my opinion, that the substance of the centrioles, or, at any rate, the substance of the astroplasm and presumably of the centriole is a reducing agent, and is accordingly, like all reducing agents, probably setting free nascent hydrogen. CONCLUSIONS. The foregoing observations strongly support the view that the chemical processes involved in mitosis are the processes of respiration, and that the phenomena shown are the result of localized changes in respiratory activity. All the facts indicate that the formation of an aster in the starfish and sea-urchin eggs is not a purely physical pro- cess, but that the physical change is underlain by a more fundamental chemical change which supplies the energy for the process. It 105 A. P. Mathews. appears that the protoplasm of these eggs cannot produce a typically large aster, unless at least three factors coexist, (1) a substance derived from the nucleus and only active in the presence of oxygen, probably therefore an oxidase; (2) free oxygen; and (3) the sub- stance of the centriole. No two of these factors alone can in these eges produce the formation of an aster. Concerning the nature of the substance of the centriole, no conclusive evidence exists, but what facts there are indicate that it is probably an albumin and that it has a reducing action. The facts suggest that the egg during its period of growth forms in the nucleus, by its metabolism, two important substances, an ox1- dase and the substance known as the centriole. As long as the egg remains in the ovary of the starfish, the germinal vesicle wall persists, and these substances cannot find access to the cell protoplasm with sufficient rapidity to produce cell division. When the egg is shed, however, oxygen gets entrance to the egg, and owing to the eccentric position of the nucleus, it gets access to the wall of the germinal vesicle. A chemical change, possibly an oxidation in the latter, is at once set up, leading to its dissolution. Upon the dissolution of the membrane a mingling of the nuclear contents with the cell cytoplasm takes place. Among the substances thus set free are the substances of the centriole! and an oxidase. The latter spreads through the cell protoplasm, and in the presence of oxygen produces in it an oxidation, which if not checked leads to the death of the cell. Meanwhile the centriole acting on the cytoplasm in the presence of the oxidase and free oxygen gives rise to the astral figure, possibly owing, as Lillie suggests, to differences in electrical potential At the close of maturation the centriole substance is itself either completely oxidized or otherwise rendered inert. It disappears. The nucleus re-forms and at once begins the same cycle over again. It grows actively; it forms more oxidase and particularly more reducing substance. There is, however, now no agent to cause the necessary destruction of the nuclear wall. While the nucleus continues to form its me- tabolic products, the cytoplasm is being oxidized by the oxidase discharged into it and is finally killed. If, however, the nuclear wall be ruptured by mechanical shock at the right instant, when the re-formation has reached the proper stage, and the protoplasmic 1 Earlier observations show that the centriole substance precedes aster forma- tion and that it escapes from the nucleus. MATHEWS: Journal of morphology, 1594, Pp. 335- Cell Division, Maturation, and Fertilization. 107 changes have not gone too far, the centriole or reducing substance gets free in the egg. It sets up cell division of a whole or a part of the egg substance, and thereafter alternations of activity carry the process on to the formation of an embryo. Confirming this conclu- sion are my observations on artificial parthenogenesis by mechanical agitation, shortly to be considered. The action of the sperm is clear on the basis of these conclusions. The sperm brings into the egg cytoplasm, which already contains the oxidase, two important substances. In the first place it brings a reducing substance —the centriole — which by its activity counter- acts the action of the oxidase of the egg cytoplasm; and in the second place it brings a very active nucleus, which grows rapidly when immersed in the egg cytoplasm, and forms more reducing sub- stance and possibly oxidase. By the entrance of the sperm there is thus set up that extraordinary series of opposite actions of oxidation and reduction which accounts for the sudden outburst of respiratory activity coincident with the entrance of the sperm, and which prob- ably underlies many of the most important syntheses and chemical transformations in protoplasm (see Drechsel, Baumann, Hoppe- Seyler). The main difference between parthenogenetic and non-partheno- genetic ova upon this theory would lie in the different activity of their respective nuclei after maturation. This is well illustrated if we contrast Arbacia with Asterias eggs. The former is a typical non- parthenogenetic ovum; the latter is almost parthenogenetic. The nucleus of an Arbacia egg after maturation remains small for days, and shows almost no growth or other indication of activity; the nucleus of an Asterias egg, on the other hand, grows with great rapidity after the extrusion of the second polar body and moves toward the centre of the egg. In fact, if the egg is partially deprived of oxygen so as to prolong its life, the diameter of the re-formed nucleus may ultimately be half that of the germinal vesicle. In fact, there is little doubt that the nucleus is almost able by itself to in- augurate cell division in this egg and to counteract the destructive action of the oxidase thrown into the cytoplasm from the germinal vesicle. My observations on the production of artificial parthenogenesis in these eggs by mechanical shock clearly support this conclusion. Loeb has suggested that this parthenogenesis is not really due to the shock, but to changes in its gaseous environment. I have, 108 A. P. Mathews. since the publication of that paper, made additional observations which throw some more light on this process. I have found that this parthenogenesis is brought about with greatest ease about six to nine hours after the eggs have begun to maturate. If one keeps the eggs in a fairly thick layer so that they are not so freely exposed to oxygen and then waits until the egg nucleus is re-formed and grown to a large size, a slight disturbance is all that is necessary to set up cell division in some of the eggs. If the eggs are transferred very carefully to a watch glass and then the watch glass struck sharply on the desk two or three times, the changes in the eggs may easily be seen, under the microscope, beginning immediately after the shock, The changes involve the throwing out of the fertilization membranes and the solution of the nuclear membrane, and often, at any rate, the development of a large monaster about the nucleus. A con- siderable portion of the cytoplasm, particularly in the periphery of the egg, often disintegrates, while that about the nucleus, which is less oxidized, remains clear, rounds itself off, divides, and forms swimming embryos, generally dwarf and abnormal in shape. The essential point in this observation is to show that the change brought about by mechanical shock involves primarily the nucleus and leads to the mingling of its contents with the cytoplasm, the indirect result being cell division. It is impossible to cause artificial partheno- genesis in Arbacia by mechanical shock, for the reason that such shock does not disrupt the nucleus, and the nucleus itself is ap- parently quite inert. There is no question of the efficacy of me- chanical shock in producing artificial parthenogenesis in the starfish egg, and I think there can be no doubt that it is not a question of CO, formation, as Loeb suggests, but the shock itself which causes this result. The conclusions of this paper are also of interest in the matter of artificial parthenogenesis by salt solutions and the formation of asters by such solutions in pieces of the mature egg, as described by Wil- son. In my opinion the action of the various agencies which set up artificial parthenogenesis cannot be explained if these fundamental chemical changes are neglected. The opinion expressed by Morgan that we are dealing with one fundamental chemical process, and that this may be set up in a variety of ways by various means or stimuli, appears to me well founded. Anything is a stimulus which effects this fundamental process, the respiratory process, in a certain way, although the direct physical effect of the agent on the protoplasm cannot, of course, be neglected. Cell Division, Maturation, and Fertilization. 109 I would suggest the following explanation of the physiological and morphological phenomena observed in cell division and artificial par- thenogenesis. The reducing substances destined at times to give rise to the centrioles are being formed constantly by the metabolism of the nucleus. These substances are normally shed into the cyto- plasm so slowly and in such small quantities that they suffice only to keep up normal respiration and to keep the cell body in its reduced state. They are consumed at the periphery of the cell, or otherwise rendered inert as rapidly as they are produced, so that no local ac- cumulation of them in the cytoplasm in an active form is possible. The amount of these substances in the nucleus is greater than that in the cytoplasm, and at times a larger mass than usual, owing to the dissolution of the nuclear wall, gets into the cytoplasm. Under fa- vorable conditions of oxygenation this mass sets up the localized, marked disturbance in respiration and forms the aster, it itself being the centriole. The centriole, in other words, is simply an abnormal amount of the active reducing substance localized in one place. It may also be assumed, from analogy with other chemical substances, that this centriole substance exists in two forms, an active dissoci- ated and an inactive undissociated form, and that it readily passes from the inactive to the active condition. It is, for example, possibly inactive when hydrated, just, for example, as NH,OH is inactive and the dehydrated radical NH, is active. It is converted into the active form by taking out water by strong solutions or in other ways, and the active form is the strong reducing substance as assumed for cell respiration. This is, of course, only a suggestion to explain the pro- duction of asters in enucleated fragments. These two states of the centriole or reducing substance may be in equilibrium with the water present in the cell and with each other. That is, the larger the active mass of the water, the more of the particles will be in the hydrated or inactive form; the smaller the active mass of the water, the more of the particles which will become active, that is, strongly reducing. By the addition of salts such as MgCl, the active mass of the water is reduced, owing to the combination of the water with the ions and molecules of the salt, and consequently more of the particles become active, with the result that they form scattered asters, or one large monaster if they are diffused. The effects of the dehydration in in- creasing the per cent of active particles would be the same as the production and discharge by the nucleus of a large quantity of the stuff composing the particles. We cannot cause asters in immature LIO A. P. Mathews. eggs, owing to the fact that not sufficient oxidase and centriole sub- stance is in the cytoplasm. According to this view, then, all agencies for artificial partheno- genesis are effective primarily because they succeed in liberating in the cell cytoplasm active reducing centriole substance. They may liberate it there either by causing a rapid discharge of it from the nucleus or by rendering the inert substance already present active.? The sperm is able to fertilize because it brings in active centriole sub- stance itself, and a nucleus capable of forming more. This hypothesis appears to me to give a tentative explanation of both natural and’ artificial fertilization and to harmonize the experi- mental results with the morphological changes observed. I may be permitted to point out, in closing, that these results are supported by the observations of F. R. Lillie? on the discharge of substances from the cell nucleus into the cytoplasm, and that they furnish a basis for the explanation R. Lillie has given for the astral rays. The rays appear to stretch from regions of intense reduction to regions of oxidation, —in other words, between an oxygen and a hydrogen electrode, — and hence possibly to be of electrostatic origin. SUMMARY OF OBSERVATIONS AND CONCLUSIONS. 1. The inauguration of the maturation of the egg of Asterias when shed into sea water consists in the dissolution of the nuclear mem- brane at the point where the germinal vesicle comes nearest to the surface of the egg. The dissolution of the wall is due to the oxygen in the sea water, since it does not occur if oxygen is absent. 2. By the dissolution of the nuclear wall all but a minute portion of the contents of the vesicle are mixed with the cytoplasm. Coin- cident with this admixture the cytoplasm undergoes chemical and physical changes. The maturated egg becomes opaque and dies in ten hours; the egg where the germinal vesicle remains intact lives for days. The early death of the maturated egg is greatly delayed if oxygen is prevented access to the egg. It is therefore clear that among the substances set free through maturation is one which greatly facilitates the action of atmospheric oxygen on the egg. Presumably, 1 See WILSON’s account of Asters in fragments of eggs, Archiv fur Entwick- elungsmechanik, 1901, xii, p. 4. 2 LILuIk, F. R.: Journal of experimental zodlogy, 1906, iii, p. 153. Cell Division, Maturation, and Fertilization. egy therefore, an oxidase escapes from the cell nucleus into the cytoplasm on rupture of the nucleus. 3. The cytoplasm of the mature egg can form asters when the sperm enters or when subjected to dehydration. The cytoplasm of the non- maturated egg will not form asters under these conditions. The cytoplasm of the maturated starfish and sea-urchin egg will only form large asters after entrance of the sperm if free oxygen is present. The astral radiations fade out and disappear if oxygen is withdrawn, if quinine is given, if cold or ether is applied. They reappear if oxygen is readmitted. I infer from these observations that the sperm requires to develop an aster oxygenated cytoplasm of a suitable kind containing an oxidase. The astral figure is hence the product in these eggs of three factors: (1) centriole substance; (2) an oxidase; (3) free oxygen, ry 4. Thecentriole substance is probably a strong reducing substance and the substance which lies at the bottom of the cell respiration. This is indicated by the fact that the centrosome lies in the part of the cell where reduction is strongest; by the behavior of the electro- negative granules in the cell which are repelled from or dissolved by it; and by other observations of its action on osmic acid. 5. The chemical basis of cell division is probably the process of respiration, the astral figure being a localized region at the centre of which is intense reduction, 6. It is suggested that the various methods employed to produce artificial parthenogenesis do not do so by their direct physical action on the cell, but indirectly by producing in one way or another active centriole substance in the cell cytoplasm, or by causing its discharge from the nucleus. 7. A basis is thus given for the electrostatic differences in poten- tial assumed to exist by Hartog, Lillie, and Spaulding to explain the astral figure. 8. These conclusions support those of Delage in most particulars, but in addition give some evidence of the nature of the substances acting on the cytoplasm and formed by the nucleus. CONCERNING THE EXCRETION OF PHOSPHORIC ACID DURING EXPERIMENTAL, ACIDOSIS IN RABBITS. Byake Pitz, C. L. ALSBERG, Anpsk: J; HENDERSON, [rom the Laboratory of Biological Chemistry of the Harvard Medical School.] a previous paper! one of us has pointed out certain reasons for seeking in the phosphates a special activity in neutralizing acids in the body. Briefly these reasons are as follows: In protoplasm phosphates are present in very great amount, undoubtedly as mixtures of mono- and di-potassium phosphates and similar salts; such mix- tures constitute a nearly neutral solution which has the remarkable property of being able to take up large quantities of acid or alkali without becoming acid or alkaline. This behavior is easily explained by the facts that acid sufficient to convert all the di-potassium phos- phate of such a mixture into mono-potassium phosphate must be added before the slight acidity of the pure mono-potassium phosphate is obtained, and that enough alkali to convert all the acid-potassium phosphate into di-potassium phosphate must be added before the faint alkaline reaction of the latter substance is obtained, while, in accord- ance with the requirements of the concentration law, all mixtures of the two substances are much more nearly neutral than either alone. Accordingly, if acid is introduced into protoplasm in a form more strongly acid than acid potassium phosphate, it must immediately react with di-potassium phosphate and similar substances forming a salt and ‘acid potassium phosphate according to the reaction. ‘ kK, HPO), -+- HX = KH,PO, + KX. Other substances, of course, such as carbonates and proteids, are not without concern in the readjustment of equilibrium, but on the whole there is good reason to attach to them a minor importance, at least in the first stages of acid intoxication when little acid has been neu- tralized. Nevertheless a proper understanding of the division of 1°L. J. HENDERSON: This journal, 1906, xv, p. 257. 113 114 K. fitz, C.-L. Alsberg, and L. 7. Fendersoun: function among such substances is an important matter, and at present it is being investigated in this laboratory. In the process of getting rid of acid from the body the next step would be to remove from the cells the excess of acid potassium phos- phate or more precisely of H,POy ions formed by the above process, thereby restoring the equilibrium between mono- and di-potassium phosphate at its former level. If such a process should occur, either by a selective activity of the cell, or, in accord with the experiments of Maly,! through the great diffusibility of acid phosphates, that is to say, of H,PO; ions, this phosphate, entering the blood stream, should rapidly pass out of the body through the kidneys. It can hardly be doubted that the cell will endeavor to restore the normal ratio between mono- and di-potassium phosphates in view of the desirability of restor- ing the full normal power to neutralize acids. Moreover, though a considerable increase in the amount of acid phosphate at the expense of the other salt must involve a very slight change in hydrogen ioni- zation at most,” even this slight change can hardly be without influence upon the delicate catalytic reactions of protoplasm. Moreover, too, slight changes of this sort probably influence the colloidal organiza- tion of protoplasm to a certain extent. Finally, it is clear that if acid is to be removed from protoplasm without preparatory chemi- cal change it must leave in the form of mono-potassium phosphate chiefly, for there can be no doubt that in this form chiefly it exists in protoplasm. Concerning the excretion of phosphates in acidosis, there is little satisfactory information, and nothing which conclusively indicates a direct influence of acid upon phosphate excretion. Interesting in this connection are the observations of Teissier,? who noted in certain cases of diabetes inverse variations in the excretion of glucose on the one hand at=+phosphoric acid on the other. It is far from improbable that in these Sat < the period of diminished glucose excretion and sycreased excretion \¢ phosphoric acid was also a period of increased ‘4 formation: above cons aa sncreerations it seemed possible that acid feeding ight be e followed * ed phosphoric acid excretion in the urine, body could 1a falling off in the excretion of phosphoric see Guonger spare that important constituent. aan Lo4ysiologische Chemie, 1877, i, p. 174. ; 2 ee Du diabéte F m. 260. ihe ue: Paris, 1876. acl The Excretion of Phosphoric Acid. 115 The following experiments have been carried out to decide this question. Four rabbits were fed on a nearly constant diet of cabbage, such variations as were incidentally introduced being quite too small mate- rially to influence the results. They received gradually increasing amounts of 0.9 per cent hydrochloric acid through a stomach tube daily, and in the urine the phosphoric acid was estimated day by day. The choice of 0.9 per cent hydrochloric acid was made in accordance with the experiments of Walter.1 Rabbits were used because it was desired so far as possible to obviate the neutralization of acid by ammonia so readily accomplished by the carnivora. Phosphoric acid was estimated by diluting the urine to 500 c.c. and titrating the amount of phosphoric acid in 50 c.c. of the solution, using uranyl nitrate and potassium ferrocyanide as an indicator, a high degree of accuracy in the determinations being unnecessary for the matter in hand. The rabbit’s weight, food, acid ingestion, and phosphoric acid excre- tion day by day are recorded in Table J. It is clear that in all these cases there occurred, for a period of varying length, not long after the beginning of the acid feeding, a material increase in the phosphoric acid excretion by the kidney. The average increase for all four rabbits during the first fifteen recorded days after the beginning of the acid treatment is 55 percent. On closer scrutiny it is easy to see that there is a considerable regularity both in increase and in a later decrease in phosphate excretion in the ~ cases of all-four rabbits. At first the excretion is relatively little increased, in the gray rabbit, No. 2, indeed, slightly diminished ; later, as the amount of hydrochloric acid ingested rises, there is a large increase in the excretion of phosphoric acid. There follows a period of diminished excretion, the diminution in some cases being so great that materially less phosphoric acid is excreted than normally. Finally, there is to be noted in two of the three rabbits, No. 1 and No. 3, just before death a marked increase in the phosphoric acid excretion, not easily to be explained by the present considerations. This phe- nomenon we hope to study in connection with the nitrogen metabolism in a later investigation. One rabbit,? No. 4, bore a relatively large 1 WALTER: Archiv fiir experimentelle Pathologie und Pharmakologie, 1877, vii, p. 148. 2 It is thought that this rabbit was a young one. Experiments to test the influ- ence of acid upon the excretion of phosphates in young animais will soon be, undertaken in this laboratory. r16, RR. fitz, C. L. Alsberg, and D.. 7. Henderson. STAUB ISS ale LARGE WHITE RABBIT. Date. Weight. clbbige. | nea P,O,; in urine. C 1906 grams grams fe cc. grams August | 1585 150 O 0.077 2 Zs eects 350 0) 0.06+ iz 3 350 25 0.064 : 4 1550 250 25 0.085 4g 5 300 25 0.068 6 300 25 0.078 ss a 300* # 30 0.124 ef 8 300 35 0.144 9 300 | 35 0.160 6) 10, 300 | 35 0.077 ete lL 1530 | 300 35 0.164 A 300 35 0.072 fe lls} 300 | 50 0.141 “14 s00rmy al 50 0.070 ka Lele 300 70 0.114 emis 300 70 0.064 ae 300 70 lost camer allicy 1445 300 70 0.285 pe’) 5 300 70 0.064 « 99 300 70 0.136 Jie VAL 1335 300 70 0.226 eee, Dead 1 In the tables the three stars indicate that on the days indicated the rabbits received from ten to twenty grams of wheat. The phosphate content of this food is much too small to produce an appreciable effect upon the results of the investigation. Lhe Excretion of Phosphoric Acid. Pay, RABEE SUL. GRAY RABBIT. ra? | Food. | Acid ingestion. ; ; 5 : / . 2O; rine. Date. Weight. Cabbage. | 0.9% HCI. P.O; in urine 1906 grams grams (5 | grams August 2 ioe 300 | 0.107 2410 | 5 | | 0.077 | 0.043 0.145 0.068 0.068 0.038 0.052 0.053 0.045 118 &. Fitz, C. L. Alsberg, and L. F. Henderson. TA BIER erie BLACK RABBIT Food, ate. | eight. Date | Weight Cabbage. 1906 | grams. August 1 150 | Acid ingestion. 0.9%, HCl. P.O; in urine. grams 0.064 0.043 0.038 0.031 0.053 0.136 0.124 0.098 0.196 0.139 0.154 0.201 0.181 0.160 0.179 0.205 0.205 0.164 0.081 0.064 0.416 The Excretion of Phosphoric Acid. 119 TABI, Ubi SMALL WHITE RABBIT. Food. Acid ingestion. id >,O; in urine. Cabbage. 0.9%, HCl. F205 tans Date. Weight. 1906 grams cc; August 2 250 2 o 4 0.068 0.226 0.115 0.064 0.186 0.171 0.201 0.273 0.188 0.115 lost 0 262 0.215 0.128 120 KR. Pie, C. Ls Alsborg, and Le. Ficnasrsor. TABLE IV (continued). 1906 grams | grams c.c. | grams August 29 1065 200 50 0.228 30 200 50 | 0.136 31 200 50 0.053 September 1] 200 0 | 0.038 « 2 200 0 | 0.060 3 200 0 0.077 fs 4 200 0 lost “ 5 1105 | 200 0 deaum «6 200 0 ] ] : | a : | 9.058 : 8 | 200 0 J ; 9 | 200 0) | | cane S10 200 0 J d oi 200 0 —... 12 1105 200 0 “a ae es 200 0 \: ae : 14 200 0 Borel 200 0 | “ 16 200 0 - 0.309 Cie Sai 200 0 | pe ag 200 0 =. « 19 1125 200 0 4 LARBO 200 0 t ogee ay Sage 1160 200 0 | at amount of acid for a relatively long time, and manifested particularly great variation in the phosphoric acid output. In this case the acid feeding was finally discontinued, whereupon the excretion of phos- phoric acid sank even lower than before, and then gradually rose - The Excretion of Phosphoric Acid. 121 toward the normal, as would be expected if the animal had lost more phosphate than it could spare. These facts are assembled in the following table in which are recorded the average acid income and phosphoric acid output per day during the various periods above indicated. As was to be expected, SUMMARY OF RESULTS. Average Average | Acid Ingestion | P.O; excretion per day. | in urine 0.9% HCl. | per day. Rabbit. Period. ir} ne oonsto ooo OANNO grams 0.070 0.074 0.134 0.092 0.178 Table I Large White Rabbit “Ton OG) DO 0.093 0.078 0.127 0.047 Table IT Gray Rabbit SIS 0.053 0.080 0.180 0.103 0.416 Table IIT Black Rabbit Wm 0.108 0.152 0.209 0.063 0.038 0.069 bo Table LV Small White Rabbit np SKS Ms\i) these periods are of different lengths in the different animals, and in a certain degree the choice of limits for the different periods is an arbitrary one; this has, however, no material effect upon the results. These results materially strengthen the theory that the phosphates are intimately concerned not only with the neutralization of acid within the body in experimental acidosis of rabbits but also with its removal from the body. ¥ Experiments are being undertaken in this laboratory to test the effect of acid feeding on the phosphate excretion of the carnivora and also to test the effect of the feeding of alkalies on phosphate excretion. It is planned also to study the phosphate excretion during diabetic acidosis in man. 122 KR. Pie CL. Abstere, and Lf. Feast SUMMARY. It is shown that the feeding of acid to rabbits produces first a marked increase, then a marked decrease, in the excretion of phosphoric acid through the urine. It is pointed out that these observations are in accord with the theory that in the body phosphates are primarily concerned with the neutralizing of acid and with its removal from the body, and that they strongly support that theory. A sudden premortal rise in the excretion of phosphoric acid in two cases of experimental acidosis is reported. One case of experimental acidosis is reported in which after a great rise and great fall in the excretion of phosphoric acid, upon discontinu- ing the feeding of acid, the excretion of phosphoric acid sank even lower than before and then gradually rose toward the normal value. This observation further supports the theory. A NEW DECOMPOSITION PRODUCT OF GLIADIN: By THOMAS B. OSBORNE anp S. H. CLAPP. [From the Laboratory of the Connecticut Agricultural Experiment Station.| bag the action of trypsin (pancreatin) on various protein sub- stances Emil Fischer and Emil Abderhalden? obtained a substance of polypeptide nature which contained all of the proline, glycocoll, and phenylalanine that could be detected in the hydrolysis solution. Levene and Wallace? have recently investigated the prolonged action of trypsin on gelatin and succeeded in isolating an exceed- ingly interesting substance which on hydrolysis yields proline and glycocoll and -in which those two substances appear to be linked with anhydride binding. Levene and Beatty*# have further investi- gated the action of 25 per cent sulphuric acid on gelatin and reached the conclusion that after boiling a solution of 400 gm. of gelatin in 3 litres of 25 per cent sulphuric acid for twelve hours the hydrolysis - is not effected to the amino-acids, but there still remain substances of a peptide character (gelatoses). In hydrolyzing our preparation of gliadin for the tyrosine® deter- mination by protractedly boiling the solution of the protein in 25 per cent sulphuric acid, we have succeeded in isolating a crystalline substance of definite character which on vigorous hydrolysis yields proline and phenylalanine. We think it highly probable that the substance is a dipeptide, as the hydrolysis indicates, and we propose to investigate it more fully as soon as we have sufficient material at our disposal. The amount of this substance that is obtained from 1 The expenses of this investigation were shared by the Connecticut Agricul- tural Experiment Station and the Carnegie Institution of Washington, D. C. 2 FISCHER and ABDERHALDEN: Zeitschrift fiir physiologische Chemie, 1903, XKKIX, P: Ole 3 LEVENE and WaLLAcE: J/dzd., 1906, xlvii, p. 143. 4 LEVENE and Beatty: /é7d., 1906, xlix, p. 247. 5 OSBORNE and CLapp: This journal, 1906, xvii, p- 245. 123 124 Thomas B. Osborne and S. H. Clapp. gliadin is by no means insignificant, as from 1 kg. we were able to isolate about 4 gm. of nearly pure substance. One thousand grams of air-dried gliadin were treated with a mixture of 2500c.c. water and 500 c.c. of concentrated sulphuric acid and heated at 100° for about six hours, until solution was effected. The solution was then boiled in an oil bath for thirteen hours. After the quantitative removal of the sulphuric acid with baryta the filtrate was concentrated and allowed to crystallize slowly. The first crys- tallization consisted almost entirely of the new substance, while the filtrate separated more but mixed with varying amounts of tyrosine and leucine. From the latter substances the new decomposition product was freed by precipitation with phosphotungstic acid. The crude fractions were dissolved in a small volume of 5 per cent sulphuric acid and phosphotungstic acid added as long as a precipitate formed.!. The process was repeated until the filtrate no longer gave the Millons reaction. The free acid was then regenerated by decom- posing with baryta in the usual manner and after recrystallizing from water was obtained pure. It is very difficultly soluble in cold water, much more soluble in water at 100, and crystallizes from this solvent in long flat prisms, sometimes perfectly rectangular, more often with modified ends. When filtered dry by suction, the crystals exhibit a beautiful mother-of-pearl lustre much resembling valine. Dried in the air, the substance contains a molecule of water, which it, for the most part, loses in vacuum over sulphuric acid, completely at 120°. I. 0.2909 gm. substance, air dry, lost o.o1g5 gm. H.O at 127°. II. 0.3791 gm. substance, air dry, lost 0.0249 gm. H,O at 127°. III. 0.2789 gm. substance, air dry, lost 0.0182 gm. H,O at 122°. Calculated for C,,HisN.O3 ‘ H,O0 ; H.O = 6.43 per cent. Found 1. 050s 2 eee O = Giompercene . LT: As 3) eee oot == nee : TIT 5. idee MS SO5ar ew: ° I. 0.2853 gm. substance, dried at 125°, gave 0.6697 gm. CO, and 0.1745 gm. H.O. II. 0.2714 gm. substance, dried at 125°, gave 0.6362 gm. CO, and 0.1723 om. HO. III. 0.2606 gm. substance, dried at 125°, gave 0.6114 gm. CO, and 0.1653 gm. H,O. 1 Cf E. FISCHER and E, ABDERHALDEN: Zeitschrift fiir physiologische Chemie, 1904, xlii, p. 540. A New Decomposition Product of Ghadin. 125 IV. 0.2432 gm. substance, dried at 125°, gave 23.1 c.c. moist Ne at 754-5 mm. and 20.5°. V. 0.1553 gm. substance required 12 c.c. #;HCl (Gunning-Arnold). @alculated for C;,H,,N.0,; C = 64.12; H = 6.87; N 10.69 per cent. oun Sk. , C—O 4.020, 8 1,—0-.7.9, per cent: pee le mee oe, Ve Oe — On So oe Pees a? 4 eG = 02-00 rie —a7OG ce Westy us ot) eric eae, eee IN) LO.7 6 (per Cents “ Votes: 5, ee eee meme se te ING POZO 2. 0" The substance, on rapid heating, decomposes at about 249° (uncorr. ) with evolution of gas toa red oil. The melting varies, however, con- siderably with the rate of heating. Whether it thereby is converted into a piperazine we have not as yet ascertained. On boiling the aqueous solution with copper hydroxide a blue color at once appears, and on concentration the copper salt separates in well-developed crystals belonging to the orthorhombic system. On exposure to the air the deep blue crystals gradually lose their lustre and on prolonged standing disintegrate to a green powder. As a consequence the water determinations were somewhat below the calculated for three and one-half molecules. The copper salt can also be recrystallized from absolute alcohol. I. 0.2778 gm. substance, air dried, lost 0.0436 gm. H,O at 115°. II. 0.3390 gm. substance, air dried, lost 0.0535 gm. H,O at 115°. III. 0.1526 gm. substance, air dried, lost 0.0238 gm. H,O at 115°. IV. 0.2731 gm. substance, air dried, lost 0.0434 gm. H,O at 115°. Calculated for Cj,HigN.O3;Cu *- 34H.O; H,O = 16.30 per cent. once We hs i, os oe Jae Hd Or eOgi oe | er ee Meee pe ac 0 eats ley ho aan % DEES os) eat ct Sakon eel Ore borat ie ECT he ak es oe Beets wo Guire, 0.2617 gm. substance, dried at 115°, gave 0.0640 gm. CuO. 0.2341 gm. substance, dried at 115°, gave 0.4453 gm. CO, and 0.1086 gm. H,O. eematediorC,,H,,N.0,Cu; C=st.92; H=4.94; Cu=109.6s per cent. Ponceewess 2 8. oy, C= 5.88 aire: Cu =i19.53° 6 . 1 The following crystallographic measurements we owe to the kind- ness of Prof. W. E. Ford of the Mineralogical Laboratory of Yale University: 126 The The Thomas B. Osborne and S. LH. Clapp. crystals of the copper salt belong to the orthorhombic system and show the simple combinations of prism, # (110), macro-dome, @ (101) and macro-pinacoid, @ (100), as shown in the figures. The prism faces are vertically striated, and the pinacoid faces are always small, often being entirely wanting. faces were not of the character to yield very good reflections, but the results obtained by taking the average of a number of measurements cannot be far from correct. The following angles were measured, those marked with an asterisk being used as the fundamental angles from which the axial ratio was calculated : m jal: ‘m FIGURE l. FIGURE 2. mi /\iem!" 62,87" di Ned? == ony 44! in /\) (d) (= 62° 52! (caleulated = 62°46) a 2b 2C==".6070 % T0078 53545: Under the microscope with crossed nicols the crystals showed parallel extinc- tion, confirming the crystallographic evidence of their orthorhombic character. Crystals made at two different times showed the different types of development i illustrated by the figures. Fig. 1 represents a slender prismatic habit, the average dimensions of the crystals being 5 mm. X .5 to .75 mm. ; the crystals of the other type as shown in Fig. 2 were usually doubly termi- nated and much shorter and thicker, their average dimensions being 2 to 3 mm. long by 2 mm. in thickness. he free substance is readily soluble in dilute alkalies and acids. Its aqueous solution possesses no pronounced taste. It gives the xanthoprotein reaction and the pyrrol-test with the spruce splinter on sublimation. The substance is lavo-rotatory in 20 per cent hydrochloric acid. A New Decomposition Product of Ghadin. 127 0.9293 gm. dried at 115° dissolved in 17.94 c.c. of 20 per cent hydrochloric acid rotated 4.24° to the left at 20°, in a 2 dcm. tube, from which is calculated («)?° pn = — 40.93°. 1.2288 gm. of another preparation of anhydrous substance dissolved in 17.94 c.c. 20 per cent hydrochloric acid rotated 5.69° to the left at 20°, in a 2 dem. tube, from which is calculated (a) 5 =— 41.55°. After standing for two hours at 20° there was no appreciable change in rotation. Hydrolysis. — 0.9300 gm. substance were heated with 35 c.c. of 20 per cent hydrochloric acid in the sealed tube for three hours at 118° and for three hours at 125.’ The solution, which was only very _ slightly colored, was concentrated on the water bath to small volume. On cooling there separated 0.42 gm. of phenylalanine hydrochloride. This was recrystallized from strong hydrochloric acid, and 0.34 gm. obtained, which was converted to the free acid by evaporation with ammonia and the phenylalanine recrystallized from water. It sepa- rated in the characteristic crystals of phenylalanine, which heated side by side with a preparation of the pure substance from phaseolin decomposed simultaneously with the latter at about 280° (uncorr.). It gave on boiling with potassium bichromate and dilute sulphuric acid the characteristic odor of phenylacetaldehyde. For identifica- tion a small portion was converted into the characteristic copper salt.t 0.0644 gm. substance, dried at 110°, gave 0.0130 gm. CuO. Calculated for CygH2.O,N,Cu, Cu = 16.23 per cent. Foundysi-s-g.40") gs) CUl—STO. Ecc The phenylcyanate derivative melted at 179~—180° (corr.) and gave the following analysis: 0.2135 gm. substance, dried at 100°, gave 0.5298 gm. CO, and o.1103 gm. H,O. Calculated for CigH,,.N.O; ; C = 67.61, H 5.63 per cent. Found . . . . . C=67.68,H5.74 “ “ The filtrate from the phenylalanine hydrochloride was freed from hydrochloric acid and the dried crystalline residue extracted with boiling alcohol. For identification the solution of the proline in alcohol was evapo- 1 Cf. ScHuULzE and WINTERSTEIN: Zeitschrift fiir physiologische Chemie, 1902, XXXV, ps 210. 128 Thomas BL. Osborne and S. H. Clapp. rated to dryness, and the proline racemized by dissolving in water and heating with baryta in the autoclave at 140° for six hours. On boiling the racemic substance with alcohol there remained undis- solved 0.11 gm. of pure phenylalanine, while the weight of alcohol soluble product was 0.44 gm., while the calculated is 0.41 gm. of proline. The total weight of pure phenylalanine obtained was 0.39 gm., or 67 per cent of the calculated quantity for the dipeptide. The racemized proline was identified as the copper salt. It crys- tallized in plates, and when dried at 110° assumed the characteristic lilac color. It gave on sublimation the pyrrol-test. 0.0893 gm. substance, air dried, lost o.cogg gm. H,O at 110. 0.0780 gm. substance, dried at 110°, gave o.o211 gm. CuO. Calculated for Cy)HigN.Oi,Cu - 2 HzO; H,O = 11.00 per cent. Found - os) }. S«2 Seo) By GEO a09 are Calculated for Cj)H,,N.O,Cu. Cu = 21.79 per cent. Found ogoo5 Aine Ee Gy GOW ae OA ee We are at present engaged in further investigation of the proper- ties of this decomposition product of gliadin. In conclusion we wish to express indebtedness to Professor Ford for the measurements of the crystals of the copper salt. fee INFLUENCE OF DIGITALIS; STROPHANTHUS, AND ADRENALIN UPON THE VELOC Y OF THE BLOOD CURRENT. By CHARLES WALLIS EDMUNDS. [from the Pharmacological Laboratory of the University of Michigan.| INTRODUCTION AND METHOD. N considering the factors which influence the velocity of the blood current, it may be said that there are two which completely over- shadow all others in importance, namely, the efficiency of the heart and the peripheral resistance. From the pharmacological point of view, no drugs have such a characteristic action on the two factors men- tioned as the members of the digitalis series, and it seemed desirable, therefore, to study them a little further in their relation to the velocity of the blood stream and to try, if possible, to explain their effects. The first to investigate this question was Kramnik, who noted that in small doses digitalis increased the rate of flow, while in larger doses the current was slowed. Hemmeter,? using various preparations of digitalis, found only great retardation in the blood flow, the mean velocity falling from a normal of 165.16 mm. per second to 98.6 mm. He explained this as being due to a lessened activity of the heart with an increased peripheral resistance. Gottlieb and Magnus? in their works on the vascular action of this group, say that all the digitalis bodies hold this in com- mon, that they produce a strong acceleration of the blood current, first through the blood pressure increase and second through the nar- rowing of the stream beds. Both Kramnik and Hemmeter employed 1 KRAMNIK: Moskauer pharmacologische Arbeiten, p. 143; cited in Virchow- Hirsch’s Jahresbericht, 1876, ii, p. 433- 2 HEMMETER: Medical Record, 1891, xl, p. 292. 8 GortLies und MaGnus: Archiv fur experimentelle Pathologie und Pharma- cologie, 1901, xlvii, p. 163. : 129 130 Charles Wallis Edmunds. Ludwig’s stromuhr, and while this instrument has proved about as satisfactory as any devised for the purpose, many objections have been urged against it, and Stewart! has introduced a very ingenious electrical method of measuring the velocity of the blood flow which offers many advantages over the stromuhr. This method rests upon the fact that the electrical conductivity of the blood depends mainly upon the salts dissolved in it. By the injection, at one point of the circulatory system, of a small amount of a strong salt solution which will be carried in the blood stream to another point where the electrical resistance of the blood is being measured by a galvanometer, the stronger salt solution, at its moment of arrival, will alter the resistance, giving a deflection of the galva- nometer- The time between the moment of injection and that of de- flection is measured by a stop watch. It is true this does not give the actual circulation tithe,? but rather the shortest time it takes the blood to pass between any two given points in the circulatory system, and would be more nearly the velocity of the axial stream which, according to Poiseuille and v. Kries, is double the mean velocity. The method is carried out as follows, fuller details being found in the excellent description in Stewart’s article referred to. After the animal is anesthetized a cannula for injecting the salt solution is tied into any vessel that it may be desired to start measuring the time from. Another vessel where it is wished to place the electrodes for the galvanometer connection, is then dissected free from its surround- ing tissue for a distance of two or three centimetres according to its location. The best form of non-polarizable electrode is probably the kaolin and zine sulphate combination in glass tubing which may be drawn out into a hook with an opening in it against which the vessel may rest in contact with the exposed kaolin. The electrodes, held by clamps, are placed about I centimetre apart; this distance being varied according to the location of the vessel used and the length isolated. The vessel is then insulated from the surrounding tissues by means of thin rubber dam, unless, as can be done in some cases, the neighboring tissues are drawn back from all contact with the electrodes. The zinc terminals which rest in the sulphate solution of 1 STEWART: Journal of physiology, 1894, xv, p. 1. 2 It might be explained that in this paper the phrases “velocity of the blood” and “circulation time” are used in a loose sense, as convenient terms, rather than in the strict sense, as defined in works on physiology. - Influence of Digitalis, Strophanthus, and Adrenalin. 131 0 the electrodes are connected in the circuit with a Wheatstone bridge and a resistance box of the post-office type and a D’Arsonval galva- nometer. The resistance in the arms of the bridge in most of my experiments was in the ratio of 100:1000, although in a few it was 1000:1000. The galvanometer was arranged to throw a beam of light ona scale which was located about two feet from the instrument, so that the deflection could be seen by an assistant, who at the moment of injection closed the key of an electric signal which wrote on the kymograph directly under the blood pressure tracing. In my first experiments I did not record the blood pressure, but very soon found that it was necessary to do so in order to get the heart rate and the height of the blood pressure, that they might help explain the changes in the circulation time found, as well as indicating the stage of action of the drug employed. The animals used were dogs and rabbits. The latter I found very satisfactory to work with when I was investigating the effects of adrenalin and digitalis. With digitalein and strophanthin I used dogs in working on the pulmonary and peripheral circulations, but rabbits in investigating the portal area, for reasons which will be mentioned later. The rabbits were anesthetized with paraldehyde, given by the stomach, 1.7 c.c. per Kg. of body weight, while the dogs were given morphine and chloretone, about 0.20 G. of the former being administered subcutaneously three or four hours before the operation and 2 to 4 G. of chloretone dissolved in a few cubic centi- metres of alcohol given by the stomach at the time of the operation. These anesthetics have proved very satisfactory during the several years that they have been used in this laboratory, giving a uniform anesthesia and having little effect on the circulation. To maintain the body temperature of the animals and to lessen shock, I used a galvanized iron tank operating board, which was filled with warm water kept at a fairly even temperature during the experi- ment. The animals were then partially covered by another similar water tank, or, in case this could not be used for lack of room, they were covered by towels. By these means the rectal temperature was kept at about 38° or 39° C., as was proved by frequent observation, Perhaps the most important matter in the manipulations was the salt solution used. In the first place it had to be injected at as nearly a constant temperature as possible. By means of a water bath it was always kept at 39° C., thus allowing for a slight cooling during the time it was drawn up into the warmed syringe preparatory to its 132 Charles Wallis Edmunds. injection. The constant temperature was very important as a dif- ference of 1° would cause a distinct alteration in the circulation time. The strength of the solution was varied with the different animals, using it only strong enough to cause a distinct deflection of the galvanometer. With medium-sized rabbits (2500 G.) 2 c.c. of a 24 per cent solution was very good provided it had to pass through only one set of capillaries before reaching the electrodes. With larger rabbits (3500 G.) 4 per cent or 5 per cent solutions had to be used, as will be noted in the protocols to be given later. If solutions stronger than these were employed, the effects were at once shown by the animal giving a slight convulsive movement accompanied by an alteration in the respiration. These disturbances had to be avoided, and the blood pressure tracings rarely showed any irregularity except that due to mechanical influences. In the dogs stronger solutions had to be utilized. In the smaller animals 4 c.c. of a 10 per cent solution was strong enough, while in larger dogs (15 to 20 K.) it was necessary to use 4 c.c. of a 20 per cent solution. The greater capacity of the heart and the larger bulk of blood allowed so much greater opportunity for dilution of the salt solution that even in these strengths it caused no circulatory disturbance, except in one or two animals, where an occasional irregularity of the heart appeared. I found it necessary to measure the velocity of the blood stream at three points. In the greater circulation it had to be measured twice, namely, in the splanchnic area and in the peripheral vessels proper. The necessity for these two measurements is due to the fact that the resistance in these two channels may be altered in opposite directions under vasomotor influence, or constriction may occur in one area while the other may not be affected, so that the velocity of the current may be affected differently in the different parts of the vascular system. The total effect of the drug would be shown in the pulmonary circulation, which would require but one measurement, as here all the blood takes the one route. The pulmonary circulation time was measured in the same way in all the experiments —from one jugular vein where the salt solution was injected to a carotid artery where the electrodes were placed. This, as has been pointed out by Stewart, will give more than the true pulmonary time, because in rabbits there is a distance of about 1.5 to 2cm. from the cannula to the right auricle and about 4 cm. from the left ventricle to the electrodes, and figuring the velocity of the current in the arteries and veins and adding the fraction of a Influence of Digitalis, Strophanthus, and Adrenalin. 133 second it would spend in the heart, he considers the actual readings to be about 0.35 second over the exact pulmonary time. However, in my work this small excess would be of no importance, as it would be the same under the drug as it was in the controls. To measure the portal (splanchnic) circulation time, the injection was made in rabbits into the left carotid, while the electrodes were placed on the superior mesenteric vein.. This vessel was reached by a longitudinal incision made about 2 cm. to the right of the median line. The intestines being packed off to the left, the vessel could be isolated very easily, as for about 2 cm. of its course it has no side branches. Several attempts were made in dogs to get the portal time in the same way, but the great difficulty was that the mesenteric vessel was so close to the liver that it was impossible, on account of the respiratory movements, to prevent slight movement of the vessel on the electrodes, thus causing alterations in the resistance, producing constant swinging of the galvanometer. Efforts were also made to use smaller branches of the mesenteric veins, but the same difficulty was experienced, although to a less degree, depending upon the length of the mesentery. The main trouble experienced here was injury to the vein walls produced when separating the vein from the artery. This traumatism was followed at once by extensive hem- orrhage into the vascular or perivascular tissues and immediate clotting in the vessel. All the portal measurements were therefore made on rabbits. The peripheral circulation time in rabbits was taken from the left carotid artery, where the salt was injected, to the inferior vena cava, which was isolated for a distance of about 2 cm., as near to its origin as convenient. The most satisfactory point found to measure the peripheral time in dogs was from the external carotid to the femoral vein just below Poupart’s ligament. Criticisms might be made to the injections being given into the carotid on the ground that the injected salt solution is propelled by the force of the syringe and not carried by the blood stream itself. This objection cannot be entirely overcome, but was minimized by using no undue force at each injec- tion and trying to make them as uniform in this respect as possible. The only way to avoid this difficulty completely with the portal and peripheral systems would have been to inject into a vein. This could not be done for two reasons: it would have required stronger salt solution for the two sets of capillaries, and then it would not have given the pure peripheral or portal circulation time, as the pulmonary 134 Charles Walhs Edmunds. vessels would be included, and any marked change here might obscure alterations in the other two areas. That this latter objection is not purely hypothetical was proved in some of the experiments in which a modification of this method was used. ’ The drugs employed were adrenalin (Takamine) dissolved in boil- ing salt solution; the U. S. P. fluid extract of digitalis; and the glucosides, digitalein (Merck), and strophanthin dissolved in water, the solutions being made up fresh each time. A number of experi- ments were made with each drug, but only one protocol of each need be given, as they were all essentially alike, except as noted. In the protocols given, all readings are omitted which there was any reason to call in question. Difficulties arose many times during an experiment, the most common cause being slight movement of the vessel on the electrode with a consequent swinging of the galva- nometer that made accurate readings impossible. All such readings were recorded at the time as untrustworthy and are therefore omitted. ADRENALIN (ACTION ON Rassits).} ’ Protocol I.-— January 30, 1906. Pulmonary circulation time. Rabbit 2900 G. 4.9 c.c. paraldehyde. Salt solution (2 c.c., 24 per cent, tempera- ture 39° C.) injected into left external jugular vein. Electrodes on right carotid 34 cm. from arch of aorta. Blood pressure taken from left caro- tid. Adrenalin injected into jugular vein. ‘Temperature of animal during experiment 39° C. Arms of resistance box 100: 1000. Number. Time. Circulation time. Heart rate. Blood pressure. ] 3.05 24 sec. 285 66 Hg mm. Control. 2 3.10 Oye 315? ike s 3 onlZ 2 “ 285 LO in i 4 S35 Dee 285 10 e4 = 5 3.20 4 « 255 eis) Adrenalin 6 3.23 44“ 195 13S 7 3.27 3 « 210 132. a Protocol II. — January 30, 1906. Peripheral circulation time. Rabbit same as Protocol I. Salt solution (2 c.c., ro per cent, temperature 39° C.) injected into left carotid. Electrodes on inferior vena cava. Adrenalin injected into branch of external jugular vein. ' For convenience the discussion of the effect of adrenalin on dogs will be given later, p. 147. L[nfluence of Digitalis, Strophanthus, and Adrenalin. 135 Number. Time. Circulation time. Heart rate. Blood pressure. 17 4.45 52sec. * "299 120 Hg mm. Adrenalin 18 4.50 jee 255 90; << ss 19 4.52 Dees 255 34S Control 20 4.55 Zo 270 Zoe 21 4.5 Gia 255 SOme << Adrenalin 22 5.00 eo ae! 270 OF s 23 5.03 23 210 30 os Control 25 5.10 62 “ 240 124) Adrenalin Protocol III. — February 5, 1906. Portal circulation time. Rabbit 2570 G. 4.4 c.c. paraldehyde. Injection of salt solution (2 c.c., 5 per cent) into left carotid, 3 cm. from arch of aorta. Electrodes on superior mesen- teric vein. Blood pressure from right carotid. Adrenalin injected into jugular vein. Number. Time. Circulation time. Heart rate. Blood pressure. 1 2.30 3 Sec. 300 24 Hg mm. Control Z 2.40 3% 285 Uap. oe - 4 2.45 Sear e 285 235 8 2 ey ane 240 23% eee t 10 3.20 (pe 195 116 ss Adrenalin ll B25 Ole 195 Hess ‘s 113 3:39 34 210 30 “ Control 16 3.45 4¢ 225 60° * Adrenalin 17 3.52 4i « 225 SZ * 18 3.56 Ze. * 240 22% ms Control There can be no doubt but that the effect of adrenalin upon the velocity of the blood current in rabbits is to slow it very greatly. This result was confirmed in many other experiments not given here. The cause of the slowing which takes place in the entire blood stream, as shown by the pulmonary observations, must primarily lie in the greater circulation. It is there that the vessels are contracted and the blood pressure increased, mainly on account of the increased peripheral resistance. This increased pressure in the aorta, with which for a short time the ventricle is able to cope, finally prevents the heart from emptying itself completely, and it dilates, as has been shown in plethysmographic tracings by Elliott,? who found great dis- tention of the chambers, the volume of the heart increasing propor- tionately with the rise of blood pressure. The blood backs up into the pulmonary vessels, and, as Elliott found, there is a rise in the venous pressure. 1 Slow reading. 2 ELLiotTT: Journal of physiology, 1905, xxxii, p. 407. 136 Charles Wallis Edmunds. There can be little doubt that this back pressure is the cause of the retardation in the pulmonary area. Bradford and Dean! found that “back pressure” is produced if the systemic blood pressure is raised in other ways, as, for instance, by clamping the aorta. Here the systemic blood pressure is enormously raised, but no change takes place in the pulmonary pressure unless the compression lasts longer than ten seconds, in which case it will be slightly increased. This result they no doubt rightfully ascribe to ‘‘ back pressure.” A factor which would tend to overcome this obstruction to the pulmo- nary outflow would be the direct action of the adrenalin on the right side of the heart increasing its efficiency. The back pressure through the left side of the heart added to the action on the right heart would doubtless raise the pulmonary pressure considerably were the pulmo- nary vessels not able to dilate and take care of the excess of blood. Also it must not be overlooked that the right heart may be rendered less efficient by vagus action, dilate and give a rise in venous pressure, as described by Elliott. The pulmonary vessels themselves probably are not acted upon by the adrenalin or only to a very slight degree. Brodie and Dixon? say that it dilates them, while Gottlieb® argues on theoretical grounds that an action by adrenalin on the lung vessels is very improbable, and quotes Gerhardt in support of his contention to the effect that while the vagi are intact adrenalin exerts on the vessels only a very slight influence, but when the vagi are cut the vessels remain com- pletely inactive. The slowing in the pulmonary time under adrenalin can scarcely therefore be primary but secondary to the systemic changes, and here the retardation might be produced either by les- sened efficiency, or slowing of the heart, or increased peripheral resistance. The latter, being such an important result of adrenalin action, was first investigated. The readings of the three experiments were arranged in the order of blood pressure heights with the results given in table on page 137. The readings arranged in this manner show with very few excep- tions that with increasing pressure there is progressive slowing of the blood current, the two apparently standing in very close relations. This does not exclude the rate of the heart from being of importance, 1 BRADFORD and DEAN: Journal of physiology, 1894, xvi, p. 34. 2 BRODIE and Dixon: Jdzd., 1904, xxx, p. 476. 8 GoTrLieB: Archiv fiir experimentelle Pathologie und Pharmacologie, 1900, xlili, p. 300. Influence of Digitalis, Strophanthus, and Adrenalin. 137 as indeed it must be. However, in the many attempts that were made to arrange the readings in accordance with this factor, in most cases no close relation between them could be made out, probably because the pressure changes were so marked as to conceal the other influence. This was not always the case, as may be seen in revised Protocol I given below, in which with increasing rate of the heart From Protocol I. | From Protocol II. From Protocol III. Height. | Rate. | Time. |! Height. Rate. Height. | Rate. | Time. sec. 135 195 44 132 210 43 255 4 2 1 Average of controls. : 2 Recorded slowly. there was a lowering of the circulation time. Protocols II and III show no such close relation. In considering further the effect of the heart rate there were found scattered through the experiments many readings in which the pres- sure was practically exactly alike in two succeeding readings, the only difference being in the rate of the heart. In such cases, if the two rates did not differ from one another more than 5 per cent or 8 per cent, the readings were usually found the same, while if there was a difference of, say, 10 per cent or over, the velocity of the current would be found to be changed. This may be readily understood when we consider that in many of the animals the heart was beating about 250 times to the minute. Contracting at this rate, it may be questioned whether it had time to fill properly during its short dias- tole, and if not, a change of fifteen or twenty beats to the minute would have little effect on its efficiency. If, on the other hand, the slowing exceeded more than 10 per cent, then it began to make its influence felt on the rate of the blood stream, although in some of these cases even with the slower heart there was an increased velocity of the stream. 138 Charles Wallis Edmunds. This question of the influence of pressure and heart rate upon velocity was studied in a more direct way by slowing the heart with pilocarpine and maintaining at the same time a uniform blood pres- sure. The latter was accomplished by applying tightly around the animal’s abdomen a rubber bag 10 cm. wide, and then inflating the bag with air until the pressure on the animal was sufficient to raise the blood pressure enough to overcome the fall due to the pilocarpine. This experiment gave the following readings: Protocol IV.— January 11, 1906. Rabbit 3100 G. 5.2 c.c. paraldehyde. Injection 2 c.c., 23 per cent NaCl into left jugular vein. Electrodes on left carotid. Uniform blood pressure maintained by means of rubber bag to abdomen. Heart rate. Circulation time. 315 24 sec. 4 mg. pilocarpine in salt solution injected. 270 22 sec. 225 .) 150 ok 1 mg. atropine injected. 300 2# SEC. These results show in general that with slowing of the heart there is an increase in the circulation time. It will be noticed that the read- ings for the various heights are not arranged in an exactly regular order, but this point will be taken up a little later. The influence on the circulation time of changes in blood pressure with a uniform heart rate was investigated in a similar way. The rubber bag was applied to the rabbit’s abdomen as before, and was inflated to varying degrees in order to bring the blood pressure to any required point which would correspond more or less closely with a pressure which might be obtained with adrenalin. The results of these observations with a heart rate of 270 are arranged in the order of the pressure heights. Blood pressure. Pulmonary circulation time. 116 Hg mm. 23 sec. (2 readings). 112 : Sa LG a = a 104 “ 24 ae 102 . oes 96 a 2e