exert their action on the tissue, and through this of learning more to ...

ON THE RELATIONS TO ELECTROLYTES OF THEHEARTS OF DIFFERENT SPECIES OF ANIMALS1.I. ELASMOBRANCHS AND PECTEN. BY GEORGERALPH MINES, Fellow of Sidney Sussex College, Cambridge.

(From the Physiological Laboratory, Cambridge, and the Laboratoryof the Marine Biological Association, Plymouth.)

CONTENTS.PAGE

Introduction and methods. 467Experiments on elasmobranch hearts .470Interpretation of the action of polyvalent kations on the

elasmobranch heart .487Hydrogen ion concentration of the plasmata of elasmobranchs 492Experiments on the heart of Pecten .494General remarks on the action of ions on the heart 503Summary ..505References ..506

Introduction. In previous comimunications I have insisted on thevalue of comparison of the action of related chemical substances on atissue as a means of getting information as to how these substancesexert their action on the tissue, and through this of learning moreabout the mechanism of the tissue itself(,2,3).

The object of the present paper is to describe some experiments.which illustrate a complementary method of attacking the sameproblem: the comparison of the action of the same substances onsimilar tissues taken from different species of animals. The interest ofthe results obtained by this second comparative method is not confinedto the light thrown by them on the problem which led to the ex-periments undertaken. They indicate a line of work which will be

1 A preliminary account of part of this work was given to the Physiological Sooietyin October 1911 (This Journal, ExLi.) and in the Journal of the Marine BiologicalAs8oCiatiOn, ix. p. 171. 1911.

31-2

G. R. MINES.

followed up for its general biological importance; for it reveals func-tional differences between corresponding tissues of different animalswhich would otherwise escape notice and yet which may in certaincases serve as clues to genetic affinities. They further suggest a verydefinite way in which differences in the chemical constitution oftissues may lead to differences in behaviour. This paper is concernedwith experiments on hearts, chiefly on those of elasmobranchs and ofthe mollusc Pecten.

It may be recalled that numerous experiments on the action ofelectrolytes on the heart of the frog led to the conclusion that onecondition essential for its continued functional activity is the main-tenance of a certain electric charge on surfaces forming part of themechanism of the muscle, this being necessary in order that they maypossess a particular differential ionic permeability. A full account ofthe work on which this conclusion was based will be found in otherpapers(l, 3).

The electric charge of a suirface depends on certain features of thesolution in contact with it and on the nature of the membrane itselfIt was shown that with respect to the hydrogen ion and to simple andcomplex trivalent kations, the heart of the frog behaved so as toindicate that certain essential surfaces in its mechanism possessed thecharacters of a particular class of emulsoid colloids, to which there isindependent evidence(4) for thinking many of the body proteins belong.In the present paper it will be shown that this conception may beextended to the hearts of different animals and that certain of thedifferences in the behaviour of different hearts towards electrolytes findin it a simple interpretation.

Methods. A number of experiments made by the commonlyemployed device of suspending strips cut from the auricle and recordingtheir behaviour in different solutions show that this method cannot berelied upon to give consistent results in cases where one is concernedwith relatively small changes in the composition of the fluids used.Even though a large builk of solution is used to surround the strip, andit is kept stirred by a stream of air, it is difficult to ensure the constantrenewal of the layer of fluid in immediate contact with the muscle;and it is impossible to avoid a certain amount of creasing or foldingof the strip, reducing the exposed surface to a different extent indifferent preparations. Moreover, where as is often the case thesurrounding fluid is intended to remove some substance from themuscle tissue, it is obvious that after a short time the fluid surrounding

468

ELdUECTROL YTES ON THE HEA RT.

the muscle will no longer be free from the substance in question andthus the concentration gradient will be altered. In work of this kind,which is aiming at the quantitative expression of results, it is clearly offirst importance to know as precisely as may be the composition of thesoluition actually in contact with the muscle. By the perfusion methodthis object is much more closely attained. The heart being gentlydistended presents a maximal surface and this is laved by a constantlyrenewed supply of solution uncontaminated by previous contact withthe muscle, of the solution that is to say whose composition is known.

Experiments show repeatedly that changes in the composition ofthe solution, which produce characteristic modifications of the heart-beat in a few seconds when tested by the perfusion method, cause nochange in the behaviour of suspendedstrips for several minutes-and eventhen the change, if it appears, is lessdefinite. By perfusing a suitablesolution the hearts of elasnmobranchsand of Pecten may be kept beatingstrongly for many hours. In perfusingthe cavities of the heart it is of courseimportant to avoid changes of pres-sure in changing from one solution toanother. I have found the special Fig. 1. Diagram of perfusion cannula.form of cannula shown in Fig. 1 ofconsiderable help in this work. It consists of a small glass bulbinto which open five tubes. Of these, three are for connection withthe Mariotte's bottles which hold the perfusion fluids, olne is drawnout with a neck, for insertion into the vein, while the fifth whichis at right angles to the other four tubes forms a short chimney,about 1- cms. in height, rising from the summit of the bulb. In usingthis cannula it is easy to secure, bv adjustment of the Mariotte's flasks,that the fluid shall rise just to the top of the vertical tube. It is thusarranged that the pressure at which each solution is supplied shall bethe same. Any accidental rise in pressure, caused for example bymanipulation of the clips on the connecting tubes, causes over-flow atthe vertical tube and is not transmitted to the heart. Further, airbubbles, which despite precautions are occasionally left in the con-necting tubes in setting up the apparatus, arriving at the cannula atonce rise to the top of the bulb and escape by the chimney, instead ofentering the heart and spoiling or at least interrupting the experiment.

469

G. R. MINES.

The same form of cannula was used for experiments on elasmobranchfishes and on Pecten.

The chemicals used were, in all cases except where speciallymentioned, Kahlbaum's purest. Solutions were prepared at first withwater redistilled in glass, but it was found that the use of the ordinarydistilled water of the laboratory (tin condenser) did not affect theresults, and since large quantities of solutions were needed this waterwas in general used.

Solutions were prepared by weighing, except in the case of deli-quescent salts (lithium, magnesium, calcium and cerium chlorides). Ofthese the solutions were standardised by titration with silver nitrate.As a rule solutions were prepared of concenitration M/2. They werekept in vessels of Jena glass with rubber stoppers, with the exceptionof the NaCl which was kept in a large glass aspirator. This solutionwas frequently renewed, as it formed the basis of all the perfusionfluids. Solutions were oxvgenated by bubbling air through them forat least five minutes before use. The use of oxygen instead of air didnot alter the beating of the perfused heart and was tried only occasion-ally.

In work on the heart the hydrogen ion concentration of perfusionfluids used is a miiatter of profouind importance. In order to control thisfactor, standard solutionis of di-sodium and rDono-potassium phosphateswere prepared, solutions of a number of indicators including p-nitro-phenol, rosolic acid, neutral red, a-naphthol-phthalein, phenol-phthaleinand thymol-phthalein were made and sets of test-tubes of even borewith special racks for colorimetric comparison were procured, and wherepracticable these were employed according to the method so admirablydescribed by Sorensen(s).

ExPERIMENTS ON ELASMOBRANCH HEARTS.

Preparation. In experiments on elasmobranchs, the operativeprocedure I have found most convenient is as follows. The centralnervous systenm of the fish is destroyedl. The pithing instruments are

1 In pithing the spinal cord of an adult male ray the behaviour of the copulatoryorgans or " claspers " affords a striking spectacle. When the region of the cord at aboutthe level of the pelvic girdle is reached, these smooth cylindrical bodies open up like fansand display their sharp cutting edges. By moving the wire up and down, so as to stimu-late mechanically the nerve roots, the phenomenon may be repeated perhaps a dozentimes.

470

ELECTROLYTES ON THE HEART.

introduced in the mid line of the dorsal surface just behind the level ofthe spiracles. Turning the fish over, the front of the pericardial cavityis removed. The upper part of the abdominal cavity is then opened bya semi-circular incision convex to the anterior extremity of the animaland the resulting flap turned back. With strong scissors the cartilaginousbar lying between the pericardial and the abdominal cavities is cut away,taking care to avoid injury to the veins which lie beneath it.

Moving the liver aside, a ligature is drawn by a curved needle aroundthe cardinal sinus on one side, and this is tied. On the opposite side avery little dissection enables a ligature to be put round the cardinal veinclose to the point where it enters the heart. If both cardinal veins areligatured close to the heart the beat is usually stopped, since it is herethat the contractions originate. For this reason the vein on one sideshouild be ligatured in the abdominal cavitv. The cartilaginous andmuscular tissue at the side is cut away to allow the cannula to resthorizontally, and the nozzle of the cannula is then introduced into thecardinal vein. Traction on the vein by the cannula is prevented bytying the latter down to the body of the fish.

If it is desired to collect the fluid pumped out of the heart, a cannulais tied into the median branch of the ventral aorta, and the lateralbranches are ligatured. In other cases these vessels are cut soas to allow free egress to the fluid. The miiovements of auricles,ventricle or both are recorded in the usual way by connecting theseparts of the heart to levers by fine silk threads and small platinum hooks.

Graphic records were taken in most cases on a slowly moving paper.During the progress of the work the great importance of frequent andcareful examination of the heart became clear: without it there is riskof misinterpretation of the graphic records. Mere inspection of a graphicrecord on a slowly moving surface does not reveal for examnple so im-portant a condition as reversal of beats, a condition which was notinfrequently encountered.

General observations on elasmobranch hearts. The elasmobranchsexperimented on included Baia clavata and Raia blanda, the dog-fishScyllium canicula and the angel-fish or monk-fish Rhina squatina.More experiments were made on rays than on either of the other species.In some experiments in which large specimens were used, the veryobvious coronary system gave rise to a doubt as to whether the perfusionfluid ought not rather to be circulated through the coronary vessels thanthrough the cavities of the heart. A number of experiments directedespecially to test this point showed that the heart-beat is much more

471

G. R. MINES.

readily affected by change in the fluid perfusing the cavities than bysimilar changes in the fluid circulating in the coronary system.

The arrangement of the coronary vessels shows considerable variationin different specimens; the coronary artery or arteries arising sometimeson one side, sometimes on both sides from the gills and running downon the front of the aorta to the ventricle. In large specimens wherethe vessel happens to be single it is possible to introduce a fine cannula.Fig. 2 is a tracing from one such experiment where the double perfusionwas carried out. At the beginning of the tracing both cannulae con-veyed a solution (to be described in a monment) which allowed the activity

- ~~~~~~~~~~~~~~~Fig. 2. Raia clavata. Separate perfusion of cardinal vein and coronary artery. Cardinal

vein perfused with Knowlton's solution at A, B,.C, D and F; with the same solutionbut containing *025 M Mg at E. Coronary artery perfused with Knowlton's solutionat A, C, D, E and F; with the same solution but containing *025M Mg at B. Topline auricle, 2nd line ventricle, 3rd line time marking, 4th line signal.

In this and in all tracings in this paper the time is marked in intervals of 65 secs.

of the heart to continue regularly for hours. At the signal the solutionperfusing the coronary artery was changed to one containing an excessof magnesium. It is apparent that the effect produced was trivial. Thesecond part of the tracing shows the effect of substituting the fluid withexcess of magnesium in the venous cannula, the coronary artery beingstill perfused with the " normal" solution.

Injection with coloured gelatine shows that the coronary vesselsare confined to the ventricle. In some cases a circulation through thecoronary vessels was secured by leaving untied those branches of theventral aorta which supplied the gills from which the coronary vesselsarose and putting another large branch of the aorta in connection witha vertical glass tube, so that the fluid was pumped out against consider-able pressure.

472


As it was found, however, that the ventricle would continue to beatregularly for a long while even though there was no circulation in thecoronary vessels, provided the fluid perfused by the veins was suitablein character, it is probable that the ventricle is not entirely or evenchiefly dependent on the coronary system for its chemical milieu.And as the substances whose influence on the heart we have studiedexert their most striking effects on the region of the heart where therhythmic contractions originate-in the roets of the great veins and thesinus-it is clear that the method of venous perfusion is best suited toour purpose.

Perfusion of the heart with sea-water stopped the beat at once; andvarious mixtures of sodium and calcium chlorides which I tried wereinadequate to maintain the beat for more than a few minutes. It wasnoted that the heart, excised and placed in such solutions or in sea-water,would often continue to beat for a considerable time. In the light ofwhat follows it will be seen that this indicates, not that the perfusionmethod is in itself hurtful, but that the immersion method is a veryinadequate way of controlling the immediate chemical environment ofthe heart. For on perfusion of a suitable solution, the heart could bekept beating vigorously for at least four hours.

I am much indebted to Dr F. P. Knowlton for the formula of asolution which he found suitable for the heart of the dog-fish in thecourse of investigations to be published shortly in the American Journalof Physiology. I used a slightly modified Knowlton's solution as thestarting point of my experiments; it was prepared as follows:

NaCl M/2 440 c.c. = *22 MKCI M/2 14 c.c. = 007 MCaCl2 M/2 8 c.c. = 004 M

MgCL2 M/2 10 c.c. = 005 MUrea 10 0/0 200 c.c. = *333 M (2 °/0)Water to 1000 c.c.

The solution was aerated by a fine glass jet connected with thecompressed air supply of the laboratory. This solution was found tomaintain the heart-beat of the various elasmobranchs very efficiently.Before discussing the significance of its constituents I shall referbriefly to some phenomena observed from time to time in hearts treatedas described; not with the suggestion that these phenomena form partof the normal behaviour of the heart in the living fish, but becausetheir occurrence under any defined conditions is of interest and because

473

G. R. MINES.

their recognition is necessary for the interpretation of facts to bepresented in the main part of this paper.

Grouped contractions (Luciani) were observe(d on several occasions.(Fig. 3.) Here the contractions are proceeding from the sinus toventricle in normal sequence, but from time to time the excitationslapse.

Fig. 3. Raia blanda. Auricle and ventricle separately recorded. Perfused with Knowlton'ssolution. The concentration of urea had been reduced to 1 0/0 for some minutes andthen restored.

Fig. 4. Raia clavata. Record of ventricle. Heart perfused with Knowlton's solution.At signal concentration of Mg increased. To show outbreak of ventricular contrac-tions initiated from bulbus aortte.

Several anomalous rhythms which were recorded could be traced tothe tendency on the part of the aortic end of the heart to originateimpulses. This was much more marked in these elasmobranch heartsunder the experitnental conditions than in other hearts which I haveseen. It was revealed when the normal pace-maker of the heart hadbeen arrested by a liminal concentration of a salt such as magnesiumchloride. After a pause a group of contractions would break out, thebulbus aortle contracting first, then the ventricle and then the auricle.Fig. 4 shows an instance of this. Another case is shown in Fig. 11.

It is noticeable that when the bulbus aortas originates beats, theirfrequency is often greater than the usual frequency of the sinus beats.Sometimes such bulbus contractions would start while the rest of theheart was engaged in a series of contractionw in normal sequence. In

474

E,LrECTROLYTES, ON THE HEART.45

such cases, if the contractions succeeded in getting through to theventricle, there would be an abrupt change in the frequency of theventricle and auricle. Fig. 5 gives an interesting case in the heart ofScyllium. The record is from the ventricle and it will be noted thatthe rhythm changes abruptly several times. These changes mark thechange in point of origin of the beat. The slower beats originated inthe sinus; the faster beats originated in the bulbus. During thisperiod of alternation the heart wvas in a condition physiologically likethe heart of the tunicate, where opposite ends of the heart becomepre-potent alternately, each initiating a group of twenty beats or so.The fastest active part sets the time of the whole. In Figr. 5 it willbe noticed that after a while the bulbus became more continuous in itsactivity and set the pace of the heart for a considerable time.

.~~~~~~~~~~~~~~~~~~~~~~~~~~~12.

Fig. 5. Scyllium canicula. Record of ventricle. Heart perfused. Showing alternationof beats originating in sinus and in bulbus.

Fig. 6 shows a case similar, but complicated by the fact that thesinus beats were themselves grouped. The sinus and the bulbus havebeen spoken of as giving rise in various cases to groups of contractions.In some instances it was clearly seen that the effect was in reality dueto a periodic failure of conduction from bulbus to ventricle or fromsinus to auricle as the case might be. Thus for instance in a case ofgrouped ventricular contractions originating in the bulbus, the bulbusgave quite regular contractions, but the ventricle responded to anumber of these beats and then failed to respond to a number and soon. It is possible that this is the usual state of affairs in groupedcontractions of the type we have been considering, but that the sinusor bulbus beats are so slight as to escape detection. The failure ofconduction is comparable with the condition of heart-block betweenauricles and ventricles or with the artificial block produced in Gaskell's

475

G. R. MINES.

m

a1)

C)*_e

.;

C;

P.

0

0

0

PA

476

00

0

Cs

P~ 0

0.-

0

0Ca

*_

Cs

8tl

*-4

X C;


beautiful experiment in which the auricle of the tortoise heart is slitup so that only a narrow bridge of tissue connects the two portions.

The very curious rhythm of Fig. 7 occurred in a case of partialinco-ordination of auricle and ventricle; there appear also someoscillations of tone which further complicate the tracing.

Importance of urea in the perfusion fluid. It has been known forsome time' that the blood of selachians contains a remarkably highconcentration of urea. Baglioni showed in 1905 that it was necessaryto add urea to a neutral perfusion fluid in order to maintain the heart-beat of these fishes. This conclusion has been confirmed by Knowlton2who found also that it was possible to prepare an alkaline perfusion fluidcontaining dextrose which would keep the heart of the dog-fish beatingthough it contained no urea. For mny purpose it was generally im-portant to avoid an alkaline solution. I made about a dozen experimentson the effects of removal of urea and on its replacement by othersubstances, which confirm the results ofBaglioni and of Knowlton. Inone experiment complete removal of urea from the perfusion fluidcaused arrest of the heart in about six minutes: the urea restored,recovery started in five minutes and was comaplete in 12 minutes. Intwo experiments reduction of the urea from 2 0/0 to 1 /o was followed byslowing of the beats, which soon became irregular: there was completerecovery on restoring the original concentration of urea. In one of theseexperiments the output of the heart against a pressure of 12 cms. salinesolution was measured. The mean of several observations gave theseresults: outflow in 1O secs. with 2 0/0 urea = 13 c.c., with 1 0/0 urea = 5 5 c.c.

That the effect of the rernoval of urea is in part due to the changein osmotic pressure of the fluid is shown by substituting certain othernon-electrolytes such as cane sugar3 or glucose. When these are sub-stituted in equi-molecular proportions for the urea there is no immediatefalling off of the beat, indeed with cane sugar there was some increase,but after a few minutes irregularity sets in and the beats becomeweakened, to be restored by the return to urea. Irregularity of beatsresulted also after replacement of half the urea by cane sugar.Replacement of the. urea by glycerol or by ethyl alcohol caused promptstoppage of the beats.

Using the ordinary hypobromite method of estimation I could not

1 For literature see Macallum, Proc. Roy. Soc. B. LXXXI. p. 616. 1910.2 Private communication.3 The eane sugar used was coffee-sugar " in large crystals, well washed with distilled

water.

477

G. R. MINES.

detect any disappearance of urea from the fluid which had been cir-culated several times through the heart or in which hearts had beenleft beating for over, twenty hours. The relation of these hearts to ureaoffers a fascinating problem which might well be attacked by seeinghow far substances chemically akin to urea can take its place. Thepresence of pure urea does not disturb the hydrogen ion concentrationf' the perfusion fluid, and so I have been able to use it in the experi-ments to be described.

The general relations of the elasmobranch heart to electrolytes are-so similar to those of the frog's heart (except of course in respect to theactual concentration with which it is in equilibrium), that it seemsprobable that the existence of this special chemical condition-theneed for the presence of urea-denotes no. very far-reaching differencein the cardiac mechanism.

Replacement of sodium by lithium. A few experiments on this pointshowed that, as in the frog's heart so in the heart of the ray, lithium isunable completely to replace sodium. The substitution of a solutioncontaining Li in place of Na led to weakening of the beats, increasedtone and arrest in systole. It may be re-called in this connection, thatlithium will replace sodium very perfectly in the skeletal muscle of thefrog(6, 2).

Importance of calcium and its replacement by strontium. Removal'of the 004 M calcium from the perfusion fluid is followed immediatelyby mnarked weakening of the heart-beat, and complete arrest in a stateof relaxation ensues within one to two minutes. For reasons statedelsewhere (2) calcium is believed to form some loosely bound compoundwith some constituent of the tissue, this being in equilibrium with Ca inthe fluid in contact with the cells. The fact that removal of the Cafrom the perfusion fluid is followed so promptly by a marked functionalchange in the heart indicates that this perfusion fluid comes into promptequilibrium with the layer of fluid immediately in contact with thecells.

As Ringer (7) showed in 1883, in the frog's heart a very perfectsubstitution of calcium by strontium is possible. The perfused rayheart, stopped by the removal of calcium, will start beating again andcontinue beating if a concentration of strontium equivalent to thecalcium removed is added to the fluid. In experiments where the Caand Sr mixtures were perfused alternately the beats while the Srwas perfused were weaker than with the Ca. A typical experimentmay here be quoted. /

478

ELECTROLYTES ON THE HEART. 479

Raia clavata. (Small specimen, about 120 grams.)

Usual operation.12.27 Ca mixture perfused.12.34Chane to r miture Output in c.c. per minute12.34 Change to Sr mixture. pumped to a height of 17 cms.

12.47 ... . .. . .. 5

12.55 ... ... ... 3-512.56 ... ... ... ... ...3

Change to Ca mixture.12.57 ... ... ... ... ... 571.3 ... ... ... ... 5,7

Change to Sr mixture.1.5 .. .. ....... 5-21.7 ... ... ... ... ... 4-51.9 ... ... ... ... ... 4.52.27 Beats are slower ... ... 2-82.30 ... ... ... 2-7

2.33 .. ... 2-1Change to Ca mixture, beats quicker at once.

2.35 ... ... ... 5.72.37 ... ... ... 6-22.39 ... ... ... ... ... 6-3

Change to mixture without Ca or Sr.2.41 ... ... ... 0.82.43 ... ... ... ... ... 00

In this experiment the heart continued to beat regularly for an hourand a half with Sr substituted for Ca, though the beats were lessfrequent and less forcible than when Ca was employed.

As in the frog's heart, magnesium is quite unable to replace calciumin the elasmobranch heart.

By an oversight, experiments were not carried out on the elasmo-branch heart with barium.

Qualitatively, the effect of increasing the concentration of magnesiumis the same on the elasmobranch hearts as on the frog's heart. Theeffects of alteration of the hydrogen ion concentration and the effectsof the simple trivalent ions are qualitatively the same as on the frog'sheart.

In so far as they have been studied, then, the relations of theelasmobranch hearts to electrolytes resemble strikingly the relations ofthe frog's heart to electrolytes and there is therefore reason to thinkthat the essential character of the intimate mechanism of these heartsis similar.

Yet since the total saline content of the fluids required to maintainthe heart-beat of the frog and of these fishes is so different, it is not asimple matter to make any quantitative comparisons between them.

G. R. MINES.

But comparison of the elasmobranch hearts with one another ispossible and in fact the hearts are admirably suited to the purpose.They are closely alike in structure and are well suited by the same" normal" solution.

Investigation of the action of Mg and of simple trivalent ions onthe hearts of the ray and dog-fish by quantitative methods brought outcertain unexpected differences of considerable interest.

Action of magnesiutm. Removal of the small concentration ofmagnesium present in Knowlton's solution causes a slight increase inthe rate of the heart-beat with no other modification. Increasing theconcentration of Mg caused slowing of the beat. These effects were

7__

Id,~~~~~~~~~~~I.

Fig. 8. Raia clavata. Effect of various concentrations of Mg.

first studied in the ray and were found to be of very regular occurrence.If the concentration was raised above a certain value there was completearrest of the beat, but on removing the excess of magnesium recoverysoon took place. Fig. 8 is from an experiment on a ray. It will benoticed that doubling the ordinary concentration of Mg in the perfusionfluid produced very marked slowing, while trebling it caused completediastolic arrest.

The heart of the dog-fish was found to be much less sensitive tomagnesium than the heart of the ray, concentrations some six or sevenor even ten times as great being needed to produce equal effects. Fig. 9is from a typical experiment on ScyllUum. Since the effect of Mg on.he rate of the beat is so very clearly defined, it was employed for the

480


quantitative expression of the difference between the hearts of the twospecies in relation to this kation.

A number of tracings similar to those reproduced in part in Figs. 8and 9 were treated as follows. The contractions per minute werecounted. The rate during the perfusion with the solution containing005 M Mg for the two minutes immediately preceding the change of

Fig. 9. Scyllium canicula. Effect of various concentrations of Mg.

solution was called 100: the first minute after the change to anotherconcentration of Mg had been made was neglected and the rate duringthe next two minutes was expressed as a percentage of the original rate.

The results of a number of experiments on rays and on dog-fishdealt with in this way are plotted in Fig. 10. The experiments onrays included specimens of Raia clavata and Raia blanda, some largeand some very small. No difference could be made out between thetwo varieties of rays and the size of the specimens had very littleinfluence. If anything the smaller specimens were slightly moresensitive than the larger. But it is obvious that the experiments onS&yllium stand right away from those on Raia, and in spite of the

PH. XLmII. 32

481

482 G. R. MINES.

considerable individual differences there could be no confusion betweenthe least sensitive ray and the most sensitive dog-fish heart.

A few experiments made on the heart of the angel-fish, Rhinasquatina, showed that in its sensitiveness to Mg this heart lies in aposition intermediate between that of the ray and the dog-fish, but muchnearer the latter than the former. Thus the values for concentration ofMg '025 M, 05 M and 075 M were 50, 32 and 25 respectively. Morenumerous experimients are necessary to define the relations precisely.

130.

120.

110-

100._

90s

fv.

c0040o

0

a 20.00.10.

.0111 .0511 91MConoentration of Magnesitum.

Fig. 10. Described in text. The dots represent observations on rays, thecrosses, observations on dog-fish.

It is of interest to note that in morphological classification'S&yllium canicula and Rhina squatina belong to different families ofthe same sub-order, while rays belong to a different sub-order ofPlagiostome fish.

1 See Bridge, Cambridge Natural History.

"A.

ELECTROLYTES ON THE HEART.48Rhina squatina, the angel-fish, monk-fish or angel-shark, is more

ray-like than shark-like in its babits, but is intermediate between raysand sharks in its external and internal structure.

The facts suggest that the method I have outlined may be of valuein tracing out phylo-genetic relations. I propose to return to thissubject after the completion of experiments on other groups ofanimals.

Before considering the interpretation of the action of magnesiumon these hearts it will be best to describe the effect on them of certaintrivalent ions.

Action of trivalent kations. The heart of the ray under theconditions described proved fully as sensitive to the simple trivalentkations of the rare earths in neutral solution as did the heart of the

II :I lfI

Fig. 11. Small ray. Heart perfused with Knowlton's solution. At first signalNd *00001 M. At second signal Knowlton's solution. Interval of six minutesbetween the two parts of the tracing.

Fig. 12. Raia clavata. Heart perfused with Knowlton's solution without Mg.Effect of Ce 000002M in same solution.

frog. Fig. 11 is a typical instance of the action of a solution containing*00001 M of a trivalent rare earth-in this case neodymium . The shortapplication indicated by the signal line sufficed to keep the auricle atrest for a long time. The tracing again illustrates a phenomenon to

1 A pure specimen of Neodymium bromate, kindly given to me by Prof. C. James ofNew Hampstead College, U.S.A.

32-2

483

G. R. MINES.

which reference was made in an earlier section of this paper. Theoutburst of contractions which interrupts the diastolic quiescence ofthe ventricle originated in the bulbus aortse. In this case they did notget through to the auricle. I have seen the beat stopped by a con-centration of *000001 M. The effect on the rate is uncertain; some-times there is marked slowing, in other cases the arrest occurswithout previous alteration in frequency. Fig. 12 gives an instancewhere well-marked slowing precedes arrest. The results of experimentson six specimens may be summarised as follows.

Hearts of Raia clavata.No. Ion and conc. Effect

1. 'Ce *0001 M Beats cut down enormously, rate reduced to 28Q0/0.Ce *000001 M Arrest in 11', no previous change in rate.

2. Nd *00001 M Arrest in 4 or 5 beats.3. Nd *00001 M Arrest in less than 2'.4. Nd -000001 M Rate reduced to 54 0/0.5. Ce *000002 M Arrest in less than 1'.

Repeated with same reaiilt.6. Ce -000002 M Rate reduced to 55 0/0.

We may regard *000001 M as about the liminal concentration inwhich the simple trivalent ion produces a marked effect on the rayheart within a few minutes. To produce arrest within a few minutes*00001 M is as a rule amply sufficient.

The hearts of Scyllium and Rhina are considerably less sensitive tothese ions than is the heart of the ray as the following experimentsshow.Scyllium 1. Nd -0001 M Stopped heart in less than 3' (Fig. 13).

2. Nd -0001 M Rate reduced to 29 e/0 in 2'. Arrest in 41'.Nd -00005 M Rate reduced to 880/0 in 2'. No reduction in

size of beat in 5'.Nd -00005 M Rate reduced to 90 °01 in 2'. Size of beat

markedly reduced in 5'.

Rhina 1. Nd -00001 M No slowing in 3'.2. Nd *0001 M Arrest in 2j' (Fig. 14).

Nd *00005 M Arrest in 2'.Nd -000025 M Arrest in 1'.

The last experiment mentioned shows the increased sensitiveness.which appears on repeated treatment with the trivalent kation. Inthis respect as in many others the effects of the trivalent ions on theseelasmobranch hearts closely parallel their effects on the frog's heart.

1 Cerous chloride puriss. Merck.

484


I have not found the alteration in rate of the frog's heart (exceptwhere this is the obvious result of " missed beats ") so marked a featureof their action as in the case of the fish hearts. The tendency of the.effect of the rare earth to persist during perfusion of a soluition freefrom it, also the prompt removal of the effect of the rare earth by theaddition to the perfusion fluid of a small amount of alkali (that is tosay by the reduction of the hydrogen ion concentration of the solution),is clearly shown.

Fig. 13. Seyliilmn canichli. Effect of Nd 0001 M.

Fig. 14. Rhinia squaItinia. Effect of Nd *0001 M.

Another point to which I called attention in describing the actionof simple trivalent ions on the frog's heart and which I have en-countered in these experiments is illustrated by Fig. 15. After theheart had been acted on by a relatively high concentration of simpletrivalent ions and the beat restored by means of a slightly alkalinesolution, if the alkali is removed from the solution after a short timeit not infrequently happens that the heart-beat falls off again very

485

G. R. MINES.

quickly, although the solution perfusion is one which before main-tained the heart-beat perfectly. It appears as though the alkali doesnot immediately remove the effects which the rare earth has produced,though it masks them; to obliterate the effects altogether a moreprolonged treatment or a more decidedly alkaline solution is required.The significance of this point will be dealt with in a future communica-tion.

Fig. 15. Rhina squatina. Described in text.

Action of luteo-cobalt chloride. The complex trivalent ion [Co (NH3)6Ij.is much less active than the simple trivalent ions we have considered inaffecting the hearts of the elasmobranchs. The experiments showedconsiderable divergence among themselves as will be seen from thefollowing.Ray hearts Conc. of [Co (NHa)6] C1,

No. 1 *0001 M*0001 M

No. 2 *001 M

No. 3

No. 4

*0001 M*0001 M*0005 M.0005 M

Effect

Rate reduced to 770/0 in 2'.Rate reduced to 52 0/0 in 2'. (Fig. 16.)Rate reduced to 83 0/0 during 2nd minute: perfusion

not continued.Arrest in i'.Very gradual slowing, then missed beats.Rate reduced to 66 0/0 in 2'.Rate reduced to 73 0/0 in 2'.

Some experiments in which a solution of the salt which had beenstanding for some weeks during the hot weather was used are notincluded, since it was found that the substance had undergone partialdecomposition, with liberation of ammonia and consequent disturbanceof the CH. of the solution.

486

ELECTROLYTES OV THE HEART.

The irregularity of these results forbids any numerical deductionfrom them, but they suggest that the ray heart is more sensitive thanthe frog heart to the complex trivalent ion in question.

.~~~~~P

Fig. 16. Raia clavala. Effect of luteo-cobalt chloride *0001 M.

INTERPRETATION OF THE ACTION OF POLYVALENT KATIONS ON THEELASMOBRANCH HEART.

We have seen that the elasmobranch hearts are affected by thesimple trivalent kations quite in the same -way as the frog's heart: thehypothesis advanced to explain the action of these ions on the frog'sheart, which was supported by much experimental evidence, may there-fore be applied without modification to the present case. To re-statethis hypothesis in a few words: Some essential part of the excitatoryor contractile mechanism of the heart involves certain surfaces ormembranes whose differential ionic permeability must be kept near aparticular value. Since, other things being equal, the ionic permeabilityof a membrane depends upon its electric charge, the electric charge ofthese surfaces must be kept at or near a particular value.

As the result of experiments by a variety of methods on manydifferent substances, it is found that the electric charge of a surface inIcontact with an aqueous solution depends on the nature of the surfaceitself and on the ions present in the solution in contact with it.

Wbatever the nature of the surface, any positive ion tends to renderthe charge more positive, any negative ion tends to render the chargemore negative, but the relative effects of different ions vary enormouslyboth with the nature of the ions and with the nature of the surface.With some ions and some surfaces the effect will be far too small to bedetected.

487

G. R. MINES.

All negative surfaces which have been tested have been found verysensitive to simple trivalent kations. They are all less sensitive (asregards their electric charge) towards simple divalent kations; but herea great divergence appears and surfaces tend to take their places in twoseparate groups. For while some are only from five to thirty times assensitive towards simple trivalent as towards simple divalent kations,others are thousands of times more sensitive towards the simpletrivalent than towards the simple divalent kations. The first group ofsurfaces may be referred to as suspensoid, the second as emulsoid.The suspensoid surfaces are appreciably sensitive to simple monovalentions, though from twenty-five to perhaps a thousand times less so (H-and OH' excepted) than to the simple trivalent ions, while the emulsoidsurfaces are not appreciably affected by monovalent ions (H and OH'excepted) and are probably many million times less sensitive to themthan to the simple trivalent kations. The emulsoid differ from thesuspensoid surfaces as regards their relations towards electrolytes in yetanother respect. For while the latter are at least as sensitive towardscertain complex trivalent kations as towards the simple trivalent kations,emulsoid surfaces are far less sensitive towards the complex thantowards the simple trivalent kations.

As regards the relation of the charge of surfaces to the hydrogen ionconcentration of the aqueous solutions in contact with them (and withthis of course the hydroxyl ion concentration is inevitably linked), onecan only say that both suspensoids and emulsoids are known which areextremely sensitive to alterations in the CH. of the solution, while othersiispensoids exist whose sensitiveness towards H- is only twice as greatas towards other monovalent kations. Special sensitiveness towardsthe CH. is not a specific emulsoid or suspensoid character; it dependson a feature which may be possessed by suspensoids and by emulsoids.Although most materials may be classed as suspensoid or as emulsoid,it is certain that all transitions between the one and the other stateare possible, the character depending immediately on some suchphysical property as viscosity or dielectric capacity.

The reasons for believing the surfaces in the heart, the electriccharge of which is a matter of importance, to be, like protein substanceswhich have been investigated, of the emulsoid type and very sensitiveto the CH. of the solution have been detailed in previous papers towhich reference may be made(, ).

We have now to add evidence from the study of the action ofmagnesium. Mg is the only divalent kation the action of which on the

488


heart can be studied without disturbance by other factors. The calciulngroup of divalent ions enters into special chemical relations with somepart of the heart mechanism, as they do with certain other tissues, whilethe remaining divalent ions except MgM when added to neutral solutionscombine with OH' and render the solutions acid.

As has been shown in a preceding section (p. 480) Mg in sufficientconcentration produces diastolic arrest of the heart in just the same wayas do the simple trivalent ions. The points of difference are in theconcentrations needed to produce the effect and in the rate at which theeffect is removed on perfusion with a solution free fromi the ions whichprovoked it. The great difference in the concentrations of simpletrivalent ions and of Mg- needed to produce the same effect agrees wellwith the former conclusion that the surfaces in the heart are emulsoidsurfaces. The greater readiness with which the Mg-- effect is removedaccords also with results obtained on non-living materials; its explanationis easy to follow. There is reason to think that the alteration of chargeof a surface by ions is due to the adsorption by the surface of those ions.The adsorption of the trivalent ions is due to an attraction far morepowerful than that which determines the adhesion of the divalent ions.The latter adsorption-complex will exist only in the presence of a con-siderable concentration of the free divalent ions in the surrounding fluid,the former will be in equilibrium with an infinitesimal concentration offree trivalent ions. Thus, on perfusion with the solution free from thepolyvalent ions, the divalent adsorption-complex will break down muchmore rapidly than will the trivalent adsorption-complex.

If the action of Mg- on the heart is indeed an action on the electriccharge of surfaces in the heart, we should find the action of a givenconcentration of Mg- greatly affected by the state of other conditions ofthe solution which influence the electric charge of these surfaces. Theapparent effect of Mg* on the heart should be increased by thesimultaneous action of other ions tending to confer a positive charge; itshould be diminished by altering the solution so as to make it, apartfrom the Mg-, tend to confer a more negative charge. Such alterationsin the solution can be brought about most simply by altering slightlyits CH.. The action of Mg- is, in fact, closely dependent on this.

Fig. 17 shows the record of a ray's heart beating at first underperfusion with a solution containing the usual concentration of magnesiumchloride -005 M and of CH. about 10-5. At the first signal mark thesolution is changed to one differing from it in containing a higher con-centration of magnesium-*025 M. The beat is quickly arrested. At

489

490 G. R. MINES.

the second signal the solution is changed to one still containing *025 Mmagnesium, but of CH. abouit 109. As is evident from the tracing, thereis quickly a recovery of the beat and the heart continues to contractas strongly in -this solution as in that which contained no excess ofmagnesium '.

Fig. 17. Raia clavata. Relation of action of Mg to CH.. See text.

We have here therefore additional reasons for believing that in theelasmobranch hearts as in the frog's heart there are surfaces of anemulsoid character whose electric charge is a matter of much importancefor the performance by the organ of its function of rhythmic contraction.

It remains to consider what is the cause of the differences whichhave been described in the degree of sensitiveness of the differentelasmobranch hearts towards di- and tri-valent kations. That thedifference between the ray and the dog-fish hearts in their relations tomagnesium and to the rare earths is not due to a difference in the easewith which the essential region is reached, seems to me to be assuredby the facts that the difference is not related to the relative sizes of thehearts nor to the time during which perfusion is continued. Thus thedog-fish heart will come into equilibrium and settle down to a steadyrate of beat for an indefinite time with a solution of magnesium perfusingit which would stop a ray heart within a few seconds.

Our hypothesis provides a simple explanation. We know that theelectric charge of a surface depends not only oni the electrolytes incontact with it, but also on the nature of the surface itself. No fact incolloid chemistry is better established than that two different surfacesin the same solution may acquire different electric charges. From theexperimental evidence at present before us we have no absolute demon-stration that the surfaces in the ray heart and those in the dog-fishheart when confronted with the same " normal " solution bear differentcharges. It is possible that they may, and I think it may be worth

I It is to be noted that there was no precipitate formed on this alteration of the CH.of the magnesium containing solution.

ELECTROLYTESA ON THE HEART.

while to' point out here that there is nothing in what is knowninconsistent with such an idea, and that the facts about the differencein sensitiveness of the hearts to di- and tri-valent ions could be explainedon the simple asstumption that these hearts differed slightly in theiriso-electric points.

Iso-electric points are commonly defined in terms of hydrogen ionconcentration. If the CH. were kept constant they might be defined interms of a selected polyvalent ion, positive or negative according towhether the surface in neutral solution without polyvalent ions acquireda negative or a positive charge. The polyvalent ions being keptconstant, every change in the CH. of the solution in contact with asurface of such mxiaterial as protein will alter its electric charge. Nowin experiments which I have carried out on the frog's heart (and whichI propose to publish later in conjunction with parallel experiments onthe ionic permeability of protein surfaces), it is very clear that while theCH. is a matter of very great importance for the activity of the heart, ityet may vary within measurable limits without preventing the functionof the heart.

One could not tell for instance by mere inspection of its behaviourwhether a frog's heart was being perfused with a solution of CH. 1065or of CH. 10-7.

Yet the electric charge of the membranes must differ, though slightly,in the two- cases. In order to upset the ionic permeability of thesesurfaces sufficiently to disturb the working of the heart, the electriccharge of the surfaces must differ more widely from a certain meanvalue.

Now it is easy to imagine two hearts possessing membranes whose.iso-electric points differ, so that when in contact with the same solution-the one set of membranes has a charge like that of the frog's heart insolution CH. 10-65 while the other set has a charge like, that of the frog'sheart in solution CH. 10-75. The behaviour of these two hearts in the" normal " perfusion fluid will not be noticeably different, but it will befound easier to arrest' the first heart than the second heart by increasing-the " positive-charging power " of the solution. in contact with it. Forthe fir8t heart has membranes already more nearly "too-positive " thanthose of the second. (See also p. 504.)

If this is the explanation of the case in point it should be found thatthe heart of Scyllium is more easily made "too-negative " than is thehear.t of Raia by polyvalent. anions. I hope to investigate this pointnext summer.

491

G. R. MINES.

There is, however, another way in which different surfaces maydisplay differences in their relations to electrolytes; it may be expressedas follows: Two surfaces possessing the same electric charge in someparticular solution may require different coincentrations of the sameelectrolyte to produce equal changes in their charges. It is clear thatthe existence of such a difference in inertia between the surfaces in thehearts of the different species would result in differences in the behaviourtowards electrolytes of exactly the kind we have noted. The underlyingcause of such differences in inertia is at present obscure, but willdoubtless be revealed as a result of further research in colloid chemnistry.

It is worth while recalling that the properties of a colloidal substanceat any time depend not only on the chemical nature of the substanceand on the electrolytes in contact with it, but also on the previoushistory of the substance.

Now the history of the surfaces in the hearts of rays on the one handand of dog-fish and angel-fish on the other, differs in this significantrespect. They have been laved for years, or one might more properlysay for generations before the experiment, with solutions of differenthydrogen ion concentration. For there is a small but distinct differencein the CH. of the blood of Selachii and Batoidei, as appears fromexperiments given in the following sectiotn.

HYDROGEN ION CONCENTRATION OF THE PLASMATAOF ELASMOBRANCHS.

The object of these experiments was to determine whether anydifference in the C.H of the plasmata of the different species existed.Each experiment consisted in a comparison of the plasmata of two fishes,treated as far as possible side by side. The plasmata were prepared forcomparison thus. A large ray, for example, was killed by pithing, themiddle branch of the ventral aorta exposed, clamped with a Spencer-Wells forceps and a glass cannula introduced. A centrifuge tube wasfilled from the cannula and placed in the centrifuge. While this wasturned by an assistant, the same operations were performed on a dog-fish.The collected blood was placed in the centrifuge, and the spinningcontinued until both samples were free of corpuscles. The samples werethen poured off' and tested with indicators. The order in which the

I It is of interest to note that the plasma thus obtained remained fluid for from oneto three hours, sometimes longer; it clotted in a minute or two on adding a little crushedmuscle exactly as in the well-known experiment with bird plasma.

492

ELECTROLYTES ON THE HEART. 493

animals were bled was of course varied in different experiments. Thesemay be summarised as follows.

1. Seyllium canicula. Raia clavata.Equal quantities of the plasmata placed in similar watch-glasses.

Indicator Seyllium Raia

Di-nitro-hydroquinone Orange-red Orange-redp-nitro-phenol ... Yellow Yellowc-naphthol-phthalein Greenish BrownishRosolic acid ... Pink Yellowish pink

After shaking both samples thoroughly with air in tubesRosolic acid ... Pink Yellowish pink

(Difference increased)

The difference in tint was more striking still on employing Walpole's device; viewingthe Scylliumn plasma+indicator through a layer of Baia plasma and the Raia plasma+ indicator through a layer of Scyllium plasma.

The difference between the plasmata was also well shown by adding a few drops ofeach to 5 c.c. distilled water in each of two similar test-tubes and testing with rosolicacid. Adding two, four or eight drops of the plasma did not further influence the tintobtained with rosolic acid.Such tubes matched Sorensen's) 9 0 sec. 8-2 sec.phosphate standard mixtures)

I.e. C= 1O-7 7 C = 10-7.4

2. Raia clavata Scylliunt caniculaPlasma of both faintly straw-coloured

Rosolic acid ... Orange-pink Pink

Comparing these with the standard tints given in Klincksieck et Valette, Code desCouleurs

Between 61 & 66 Between 41 & 46Rouge orang6 Rouge

Titrating with N/100 HCI to a definite tint with methyl orange8-8 c.e. 14-5 c.c.

3. Baia blanda Bhina squatinaRosolic acid ... Yellowish PinkMatches Code des Couleurs 136 Between 71 & 66

Titrating with N/100 HNO3 to methyl orange8,3 c.¢. 11-0c.c.

4. Rhina squatina Baia blandaRosolic acid ... Pink Orange-pink

(By Walpole's device)

The CH. of the plasma of Scyllium and of Rhina is therefore rathersmaller than that of Raia, and although this difference is slight, in viewof the long time it has probably had in which to act,. it cannot be

4G. R. MINES.

neglected as a possible factor in causing the difference between thesurfaces in the hearts of these species, the existence of which we deducedfrom the experiments recorded on previous pages, whether we conceiveof this as a difference in inertia or as a difference in iso-electric points.The latter hypothesis has much to recommend it; the further factorswhich might determine a difference in iso-electric points of the surfaceswill be discussed in connection with the experiments on the heart of theinvertebrate Pecten, which differs from vertebrate hearts in its relationstowards electrolytes in a manner explicable most readily by the existenceof a considerable difference in their iso-electric points-a differencemuch wider than that we suppose to exist between the hearts of thesenearly related fishes.

EXPERIMENTS ON THE HEART OF PECTEN.In the following experiments medium-sized or large specimens of

the lamellibranch Pecten maximus, the common scallop, were used.Operation. Shortly after removal from the aquarium the valves of

Pecten gape, and a piece of wood can be inserted to keep them apart.The mantle is then detached from the flat, upper valve with the handleof a scalpel, the attachment of the adductor muscle cut through andthe valve removed. The concave shell containing the animal is thenfixed in a dissecting dish, a hole is drilled in the shell to allow of theescape of fluid and the pericardium is freely opened up. A silkthread is passed round the thin transparent "auricle" and the nozzleof a cannula (of the form shown in Fig. 1) tied in. The efferentvessels of the heart are cut open to allow free egress to the perfusionfluid. A mintute platinum hook is placed in the superficial part of theventricle and connected by a fine thread to a lightly poised recordinglever.

The perfusion fluid. The blood of Pecten is said to have practicallythe same composition as the sea-water in which the animal lives(8).In attempting to determine whether the perfusion method wasapplicable to the Pecten heart, I first tried sea-water as a perfusionfluid. Directly the sea-water reached the heart the latter stopped infirm contraction. The result was the same whether the sea-water usedwas taken from the aquarium tanks or on the same day from theEnglish Channel. The behaviour of the heart when perfused withsea-water reminded me strongly of the behaviour of the frog's heartwhen treated with an alkaline solution. I therefore compared the CH'of the blood of Pecten with that of sea-water. Since the blood is

494


colourless, such a comparison is readily made with indicators. Rosolicacid shows the difference best. While the sea-water gives a full pinkcolour with this indicator, the blood of Pecten gives a yellowish tint. Adifference of this kind was found in every case tested. The blood hasa higher CH. than the sea-water in which the Pecten is living. Whilesea-water is faintly alkaline in reaction the blood of Pecten ispractically neutral. Comparing with Sorensen's standard mixtures, Ifound the CH. of the water taken from outside the breakwater atPlymouth to be about 10-8.2, that of Pecten blood was about 10-7.

That the difference in CH. was indeed the reason why sea-waterwould not maintain the heart-beat was shown conclusively by the dis-covery that sea-water neutralized by the addition of a little hydrochloricacid forms an excellent perfusion fluid. The addition of HC1 to sea-waterintroduces no strange ions, but increases the CH.. To bring sea-waterto the same CH. as Pecten blood required about 1'5 c.c. N/10 HC1 in100 c.c. of freshly collected sea-water. So large an amount of acid isneeded because the sea-water contains phosphates and carbonateswhich act as " buffers." (See Palitzsch(9).) Fig. 18 illustrates the effectof changing the CH. of sea-water perfusing the heart.

Fig. 18. Pecten. Heart perfused at beginning of traoing with sea-water + HCI N/10 1 5 c.c.in 100,c.c. CH.10-7. At signal change to sea-water+HCl 1 c.c. in 100 c.c. CH. 10-786.The second piece of tracing is after return to the solution of CH. 10-7.

Sea-water, then, of CH. 10-7 will allow the heart to beat. Sea-water of CH. 10-8 stops the heart in systole; sea-water of CH1. 10 6 stopsit in diastole.

I next tried perfusion with mixtures of M/2 solutions of NaCl,KCI and CaC12, well aerated. It is not possible to determitne accu-rately the CH. of such solutions by the colorimetric method, since inthe absence of " buffers " the indicator added itself appreciably disturbsthe reaction. But owing to the CO2 in the air, such solutions of thesalts of strong acids and bases deviate slightly from neutrality on theacid side. [Their CH. is probably about 10-6 5.]

495

G. R. MINES.

To my surprise I found that these solutions also stopped the heartin systole, that is to say the heart behaved as though the CR. of thesesolutions was too low. A plain solution of NaCl in distille(d water alsocaused systolic arrest. And yet the CH. of these solutions wascertainly higher than that of the neutralised sea-water or the Pectenblood itself.

The explanation is quite simple. Both the blood and the sea-water contain considerable quantities of magnesium. I found that the'addition of MgCl2 to the Na,Ca,K mixture enabled it to maintain theheart-beat very efficiently. Two experiments on the heart of Sepiaofficinaltis showed that its behaviour in these respects is similar to thatof Pecten. The presence of K was found to be not essential, and sofor the sake of simplicity it was omitted. The concentration of Mgemployed varied from *01 M to 1 M. The last-named concentration wasfound most satisfactory and the solution generally employed for per-fusion after the preliminary experiments was prepared thus:-

CaCl2 M/2 1 c.c., MgCl2 M/2 20 c.c., NaCI M/2 to 100 c.c.

Before entering on the explanation of the action of Mg we shallgive the results of experiments on the alterations in the composition ofthis fluid.

Fig. 19. Pecten. Effect of removal of Ca.

Substitution of Li for Na. The substitution of an equi-molecularconcentration of LiCl for NaCl in the solution caused a pronouncedrise of tone together with increased frequency and incomplete relaxationof the contractions. These symptoms repemble those produced by thesame substitution in the elasmobranch heart, but the beat is bettermaintained in the Pecten heart. I have seen the Pecten heart continue

496


beating for over 10' while perfused with a solution containing lithiumiD place of sodium. In this case the return to the sodium mixture wasfollowed by increased rapidity of beats and then by stoppage. Returnto lithium again started some feeble beats. That the perfused Pectenheart comes into equilibrium with the fluid perfusing it quite rapidlyappears fronm what follows.

Importance oj Ca and the substitution of Sr and Ba. The effect ofremoving Ca from the perfusion fluid is exactly like that producedunder the same conditions on the heart of elasmobranch or amphibian.There is an immediate falling off in the amplitude of the contractions,some quickening and arrest in relaxation. In the instaulce given inFig. 19 the falling off is more gradual than usual. Restoration of theCa is followed immediately by rise of tone and increased force ofbeats.

Fig. 20. Pecten. Replacement of Ca by Sr. The lower is the directcontinuation of the upper tracing.

Substitution of Sr for Ca causes rise of tone and often irregularity;the contractions become much slower. This is exactly the effect pro-duced in the frog's heart(2). In the concentrations tried the substitutionappears less perfect than in the frog's heart. Fig. 20 illustrates thesepoints.

497

33PJI. XJ,III.

G. R. MINES.

The replacement of Ca by Ba causes still more marked rise of tone,and ultimately arrest in systole. In two out of three experiments onthe substitution of Ba for Ca, I noticed the curious phenomenon whichis recorded in Fig. 21. The very slow rhythmic contractions recall thecurious oscillations of tone which Fano, Botazzi and others havedescribed in the auricles of the tortoise.

Since the solutions used here all contained a high concentrationof Mg, it is obvious that Mg is totally unable to take the place ofCa.

In order to settle the relations of these ions to the heart muscle(Ca, Sr, and Ba) it will be necessary to test them in various concentra-tions-to see if, for instance, it would be possible to get a better effectby the substitution of a much smaller concentration of Sr or of Ba for

Fig. 21. Pecten. Replaoement of Ca byyBa.

a certain concentration of Ca, instead of using the chemically equivalentconcentrations as one does naturally in experiments where the generalrelations of these things is first sought.

At present I have not attempted to work this out. I have merelymade experiments enough to convince myself that some degree ofsubstitution at all events is possible in the case of these three closelyrelated elements: that Sr behaves more exactly like Ca than does Ba;that Mg is totally unable to replace Ca; and that these facts hold goodfor the heart of the mollusc Pecten as they do for the vertebrate heartswhich have been investigated.

Importance of Mg. I have already mentioned that the addition ofMg to a solution, which without it arrested the heart at once in systole,enabled this solution to maintain the beat in a satisfactory way. Theremoval of Mg from this solution, as might be expected, is followed bysystolic arrest of the heart. How quickly this effect may be broughtabout is shown in Fig. 22. Keeping the C0. constant, increase inthe concentration of Mg causes slowing of the heart. Conversely, reduc-tion of the Mg concentration is immediately followed by quickening.

498


Increase in the Mg concentration from 1 M to 15 M stops the heartin diastole. In a heart which has been beating for some while in thepresence of 05 M Mg, diastolic arrest was caused by 1 M Mg. Theinfluence of concentration of Mg on the rate of beat is similar to thatdescribed for the elasmobranch hearts. One instance will suffice toillustrate the point.

Pecten maximus, heart perfused.

Composition of perfusion fluid Number of beats per 65"

Ca 1, Mg 20, Na to 100 14, 13, 12Ca 1, Mg 15, Na to 100 17, 17, 15Ca 1, Mg 20, Na to 100 10, 12, 11

Fig. 22. Pecten heart. At A the fluid perfused was the following mixture of M12solutions: MgCl 2 c.c., CaCl -8 c.c., KCI -8 c.c., NaCi to 100 c.c. At B the samebut without Mg.

The relations of the Pecten heart to magnesium thus resemblethose of the vertebrate hearts studied towards the same io.n, in thatincreasing concentrations of Mg reduced the frequency of the beats andlead uiltimately to diastolic arrest. They differ in the highly importantrespect that while removal of Mg from a neutral perfusion fluidmaintaining the beat of the vertebrate heart causes only a littlequickening of the beat, the heart of Pecten under similar treatmentstops in systole. The heart behaves when perfused with such a solutionexactly as if the solution were too alkaline. And in fact the addition ofa very small concentration of acid to the solution devoid of magnesiumremoves the systolic condition and sometimes re-starts the beat. It isdifficult to hit off the exact concentration of acid needed to keep theheart going in such a solution, but I have obtained distinct evidence ofthe successful substitution of an increased CH. for Mg.

499

G. R. MINES.

The explanation of these facts is assisted by the observations on theaction of simple trivalent ions which follow.

Action of trivalent kations. The addition of 00001 M Ce to theusual perfusion fluid causes slowing of the beat and fall of tone. With*0001 M there is diastolic arrest in two or three minutes, the effect beingquite similar to that described in the various vertebrate hearts. As arule recovery was much more prompt on removal of the rare earth fromthe solution than in the case of the vertebrate heart. Ce-** and Nd***were the ions tested. The complex trivalent ion Co (NH3)6- was muebless active than the simple trivalent ions, though as with the elasmo-branch hearts the amount of effect produced by it was different indifferent cases. Fig. 23 gives an instance of comparison of the actionsof Co(NH3)8a and Ce .

It is very interesting to find that it is possible to replace Mg in theusual perfusion fluid with a very small concentration of Ce or Nd. Theconcentration with which the best results were obtained was -00003 M.An instance is given in Fig. 24. With higher concentrations, such as*0001 M, the systolic condition following the removal of Mg wassucceeded by a state of diastolic arrest, just as when the solution wasmade too acid.

In the case shown in Fig. 24 a concentration of 00003 M Ce iseffectively replacing a concentration 1 M Mg. In other words one cerousion is doing the work of some 3000 divalent ions.

As with acid, so with the trivalent ion it is very easy to over-shootthe mark. The exact concentration needed appears to vary somewhatwith different hearts, and in a number of experiments negative resultswere obtained. But in four cases, as in the instance given in Fig. 24,the addition of a very low concentration of the rare earth solution tothe perfusion fluid produced recovery of the beat which was maintainedfor some minutes.

HIere again then we find Mg producing effects like those producedby acid and by simple trivalent ions. The difference between the Pectenheart and the vertebrate hearts, whose relations to these ions have beenexamined, is that the Pecten heart, in order that it may beat, requiresthe presence of positive ions in such excess with respect to their powerof conferring positive charges as inevitably to stop the vertebrate heart.

The general relations to electrolytes of these different hearts is quitesimilar. We may assume then, for the present, that the essentialarrangement of those parts of the mechanism which we are attacking isthe same in the various hearts. If then we suppose that in order that

500


Fig. 23. Fig. 24.

501

oo

CD

-4

04o b

0o -

0.o

Ca

o

0

_3 0o

oq Z

33-3

G. R. MINES.

the Pecten heart may beat its "membranes" muns t have the same electriccharge as those of the vertebrate hearts, we come to the simple con-clusion that in the neutral solution, practically free from electrolyteshaving any notable effect on the electric charge of surfaces-in solutionswhich would produce the same electric charge on pieces of the samematerial (iso-polarising solutions)-the membranes of the Pecten heartwill acquire a charge much more negative than the membranes of anyof the vertebrate hearts.

The simplest way of expressingf such a difference is to say that therespective membranes have different iso-electric points. In order to,bring them to the same electric potential as the solution surroundingthem, the membranes of the Pecten heart must be in a solution moreacid or richer in polyvalent kations than that required to bring themembranes of the other hearts to the same condition. In neutralsolutions free from excess of polyvalent ions. we know the membranesof the vertebrate hearts possess a negative charge. The membranes ofthe Pecten heart in the same solution would possess a more intenselynegative charge. We have been led to the conclusion that thesemembranes or surfaces are " emulsoid" in character, and there is everyreason to suppose that they are of protein miaterial. Now it is a well-known fact that various proteins have different iso-electric points. Itis well seen in their relation to dyes. Some proteinis are strongly basicin their dyeing reactions, others have an acidic character. The formerof course have affinity for acidic dyes, the latter take up basicdyes.

It is not necessary to suppose that the material of which the surfaces,whose ionic permeability is so important a matter for the functionalactivity of the heart, are composed, should form any large proportion ofthe total mass of the muscle: but on the other hand it is quite possiblethat it may.

Pieces of the heart muscle of Pecten and of Raia were soaked inNaCI M/2, teased out and placed on a slide, and then treated withmethyl green and orange G, washed, dehydrated, cleared and mounted.The material from Pecten was found to stain strongly with the orange,while that from the ray took up the methyl green strongly. This wasconfirmed on repetition. Methyl green is a basic substance, orange Gan acidic substance, so these oliservations would indicate that thematerial of which the pecten heart is.composed is more basic, that is tosay has a more negative charge than the material of which the ray heart.is composed, when they are treated in the same way.

502


Such differences almost certainly depend on differences in chemicalconstitution. Proteins, like their constituent amino-acids, are amphoteric:whether their acidic or basic properties become predominant depends ontheir constitution. For example, it is well known that the protaminesand histones, which are composed chiefly of di-amino acids, are stronglybasic substances. Here the basic - NH2 groups greatly out-number theacidic - COOH groups.

The presence of an unusually large proportion of di-carboxy mono-amino acids would render a protein more acidic in nature. It is knownalso that the presence of phosphorus in the molecule tends to conferacidic properties on the protein, as for example in caseinogen.

It is evident from these considerations that relatively slight differencesin the proportions of the various amino-acids going to build up theproteins of which the different hearts are composed would suffice toproduce differences in the iso-electric points of these proteins.

In the bio-chemistry of species no fact is nmore generally recognisedthan the specificity of proteins. Many lines of work converge to showthat "all flesh is not the same flesh," and it has been regarded asaxiomatic that differences in the chemical composition of specimens ofliving niatter will determine differences in the behaviour of thesespecimens.

The considerations which I have here advanced, based on the factsrecorded in this and previous papers, indicate more definitely one methodby which differences in chemical constitution-and possibly quite smalldifferences-of certain essential constituents of the living machine.maybecome responsible for differences in the relations of the machine asregards the influence of chemical environment on its performances.

GENERAL REMARKS ON THE ACTION OF IONS ON THE HEART.

The observations recorded in this and in previous papers togetherwith others as yet unpublished lead to the following general conclusions.as to the mode of action of ions on the heart. They may be grouped.provisionally under three headings:-

1. "Nomadic" ions: Li, Naw, K@, Rb, Cs-, (Hi); CI', NO'3, (OH).2. "Combining" ions: Ca-, Sr-, Ba".3. "Polarising" ions: H', Mg-, Ce"' and other simple trivalent

ions, complex trivalent kations; OH', citrate"', phosphate"'.1. These ions produce their effects bv passing from one region to

another, carrying their charges and so setting up differences of potential

503

G. B. MINES.

between the various parts. Their mobilities and volumes are probablythe principal factors in determining their actions.

2. These act by forming chemical compounds with some essentialconstituent of the heart muscle.

3. These modify the electric charge and thus the ionic permeabi-lity of certain surfaces or membranes in the heart muscle, thus affectingin a differential manner the passage through these membranes of the" nomadic " ions.

The relations of the " polarising" ions to the surfaces in differenthearts, as also their relations to sur-faces of different emulsoid materials, B'may be put mnore clearly with the C.'aid of the diagram in Fig. 25.

The line AB is pivoted at Aaround which point it can swing, forinstance, into the position AB'.The charge of the surface is ill-dicated on the scale S where theline AB crosses it. This line canbe swung to the right to the sameextent by a very small concentra-tion of a simple trivalent kation,by a relatively very large concen-tration of a divalenit kation, or by .a small increase in the C01.-thatis to say, by an increase in thehydrogen ion concentration ac-companied, as it must inevitablybe, by a simultaneous decrease inthe hydroxyl ion concentration.Now in the absence of any poly- L-c,l .,,i,II,III,IcI Ivalent ions, and when the concen- Rtration of hydrogen and hydroxyl cw; Xions is equal, that is to say when Athe solution is neutral(Ci . 10-7)w Fig. 25. Diagram to explain the relationthe s t iof the electric charge of surfaces tothe charge on an immersed surface certain ions. The scales P and Ris not as a rule zero. In most cases represent concentrations of Mg andthe surface is negative to the solui- Ce respectively, the zeros being crossedtion. But the nature and intensity by AB. Divisions on sliding scale Stion. But the nature and intensity are arbitrary. Further description inof the charge depend on the nature text.

504

EL.ECTROLYTES ON THE HEART.50

of the sturface, as we have seen. This is represented in the diagramby making S a sliding scale. For most surfaces it will have to beshifted to the right, as it is shown in the diagram. In order to makethe charge on the surface zero, there must be added to the solutiona certain very small concentration of trivalent ions, a large con-centration of divalent kations, or the CH. must be increased.

Now with regard to the surfaces in the heart whose electricalcharge we believe to be of a matter of importance, their charge mustlie within certain limits if the activity of the heart is to continue.Let us call these limits on the diagram -3 and + 1. If the charge israised above + 1 the beat stops; if the charge is reduced below -3 thebeat is also stopped. In the case of the Pecten heart, the scale isshifted so far to the right that -3 lies slightly to the right of the linewhen it is in the position AB. Thus in order to allow the heart tobeat the line mlust be deviated towards the position AB' by one of themeans indicated. With the various vertebrate hearts, the scale lies sothat with the line vertical the charge indicated lies between -3 and + 1.But with the ray heart we suppose that the vertical line AB passesover say - 1, while with the scale set for the membranes of the dog-fishheart it passes over -2. Clearly on this hypothesis, although bothhearts will beat in the solution indicated by the vertical line, a greaterdeviation of this line to the right will be necessary to stop the dog-fishheart than the ray heart.

SUMMARY.

The hearts of the elasmobranch Scyltium, Baia and Rhina and thatof the invertebrate Pecten show many close resemblances to the heart ofthe frog in their relations to electrolytes. The action of magnesium onthese hearts accords with the hypothesis that this substance acts byaltering the electric charge of the surfaces which are affected by thehydrogen ion and by trivalent kations.

Certain interesting differences in the quantitative relations of thehearts to those electrolytes which act by altering the electric charge ofsurfaces in the heart, indicate the existence of differences in the iso-electric points of the materials of which the different hearts arecomposed.

I wish to thank Miss D. Dale of Newnham College for her valuableassistance in many experiments, and Professor Stanley Gardiner for

505

506 G. R. MINES.

the use of a University Table at the Plymouth Marine BiologicalLaboratory during the greater part of the long vacation.

The expenses of this research were defrayed in part by a grant from the GovernmentGrant Committee of the Royal Society.

REFERENCES.

1. Mines. This Journal, XL. p. 327. 1910.2. Mines. Ibid. xLi. p. 251. 1911.3. Mines. Ibid. x2ii. p. 309. 1911.4. M. Fischer. Oedema, Wiley and Sons. 1910.5. Sorensen. C. R. Carlsberg Lab. viii. p. 1. 1909.6. Overton. Pfluger's Archiv. XcII. p. 346. 1902.7. Ringer. Practitioner, xxxi. p. 81. 1883.8. Dakin. Liverpool Marine Biology Committee Memoir, xvii. p. 73.9. P a I i t z sch. Publ. de Circonstance, No. 60, Conseil perm. internat. pour l'explor.

de la mer, Copenhagen, June 1911.

CAMBRIDGE: PRINTED BY JOHN CLAY, M.A. AT THE UNIVERSITY PRESS

exert their action on the tissue, and through this of learning more to ...

Documents

Transcript of exert their action on the tissue, and through this of learning more to ...