A NOTE IN REPLY TO CRITICiSMS OF MY THEORY OF MUSCULAR CONTRACTION. By WILLIAM MCDOUGALL. (Received for publication 30th September 1909.) SOME twelve years ago I published a description of the structure of cross-striated muscle,l based on a study of muscle of several different types by the aid of the highest powers of the microscope then available. A year later, I followed up this paper with one propounding a new hypothesis of the mechanism of the contraction of cross-striated muscle,2 based on the account of its structure given in the former paper. Even before the publication of the second paper, my energies were wholly called off to other branches of science, and I have found no opportunity of following up these physiological studies; nor, being committed to the exacting study of the mind, do I see any prospect of returning to them. For some years after their publication my papers appeared to have been still-born, and I had resigned myself to the hope of seeing my facts independently rediscovered and my hypothesis independently restated. But recently several physiologists have criticised my hypothesis, and one, Mr E. B. Meiggs, has adopted it, and has made a number of new observations and ingenious experiments in support both of the hypothesis and of the description of structure on which it is based.3 The criticisms directed against my hypothesis have implied serious misunderstandings of it-misunderstandings which in large part, no doubt, have been due to lack of clearness in the hurried exposition of an unpractised writer. I believe that if these misunderstandings can be removed it will appear that the criticisms of my hypothesis have been misdirected, that no serious objection to it has been pointed out, and that some of the observations and reasonings of my critics afford it strong support. I have therefore sought and obtained the permission of the editor of this Journal to occupy a few of his pages with this explanatory note. I propose first to state very briefly the principal conclusions of my two papers, leaving aside all finer details; and then to consider the bearing upon them of the observations of my critics. 1 " On the Structure of Cross-striated Muscle," Journal of Anatomy and Physiology, Vol. Xxxi., 1897. 2 "A Theory of Muscular Contraction," Journal of Anatomy and Physiology, vol. xxxii., 1898. 3 "The Structure of the Element of Cross-striated Muscle, and the Changes of Form which it undergoes during Contraction," Zeitschrift ffir allgemeine Physiologie, Bd. viii., 1908.
54 McDougall In the paper on the structure of muscle I tried to show- 1. That from the wing-muscles of certain insects (bee, wasp, house-fly, Dytiscus) the contractile elements can readily be isolated as slender filaments or sarcostyles; that these can be observed under the microscope while still living and capable of contraction; that the study of these sarcostyles affords the key to the understanding of the structure of all (or of many) types of cross-striated muscle, because, in spite of certain differences, the muscles of the legs of various insects, of the rat and frog, and of the crab and crayfish, are all essentially of the same type, and, when contracting, go through the same series of changes of shape. 2. That the sarcostyle of the insects' wing-muscle is, in its resting or uncontracted condition, a slender cylindrical tube a little more than 2MA in diameter, consisting of a very delicate inextensible membranous wall containing a fluid or viscid substance; that the cavity of this tube is divided into sections by transverse inextensible septa at regular intervals of 4MA, so that the whole sarcostyle is made up of a series of cylindrical chambers or sarcomeres each 4MA in length and a little more than 2Mu in diameter. 3. That the inextensible membrane which forms the lateral curved wall of each cylindrical sarcostyle is folded or crinkled longitudinally in such a way as to permit of lateral distension or bulging of the sarcomere in spite of the inextensibility of the bounding membrane. (This folding is clearly shown in my published photographs of transverse sections of uncontracted muscle.) 4. That each sarcomere during contraction bulges laterally more and more as its ends are brought nearer together; but that this bulging is partially restrained by an elastic and extensible membrane which divides the cavity of the sarcomere into two equal chainbers. The elastic tension of this membrane, attached to the lateral wall of the sarcomere at the middle of its length, causes a constriction in the bulging wall (which therefore appears as a double bulge, an appearance which is clearly reproduced in several of my photographs). 5. That the proportions of the sarcomere (both in this and in all the other types of muscle examined by me, whose absolute dimensions present large differences) are such that the change of shape it undergoes during contraction (namely, a shortening or approximation of its discoid ends with a bulging of the side wall) involves but little change of its volume or cubic capacity; though it is impossible to measure its dimensions so accurately as to show that some change of its volume is, or is not, produced. At the conclusion of my paper describing the structure of muscle, I threw out the suggestion that the change of shape of the sarcomere during contraction may be caused by the passage of water through the membrane forming the lateral wall of the sarcomere from the sarcoplasm into the interior of the sarcomere. I suggested also that this passage of water into the sarcomere may be determined by an increase of the osmotic equivalent
Reply to Criticisms of my Theory of Muscular Contraction of its fluid contents, resulting from the breaking down of a large molecule into a number of smaller molecules. Before publishing my second paper I made a large number of experiments with mechanical models, designed to reproduce the essential features of the structure of the sarcostyle. It was easy to make a model that contracted about 30 per cent. of its length, when fluid was pumped into it. For this purpose I used tubes made from oiled silk, pieces of intestine, and tubes of parchment paper, such as are used for making sausage-skins. If any such flexible, but inextensible, cylindrical tube is constricted by inextensible annular threads at regular intervals greater than the radius of these annular bands, and fluid is then pumped in, it shortens by about 30 per cent. of its length, each section bulging until its wall in longitudinal section is approximately semicircular. This fact is, however, so obvious that I did not think it necessary to describe these experiments in my second paper. I also made a number of determinations by the freezing method of the osmotic equivalent of finely chopped muscle, before and after long-continued strong stimulation. As I found that the osmotic equivalent in the second case was but little greater than in the former, it seemed improbable that the flow of water into the sarcomere could be due to endosmosis only; I therefore inclined to the view that it might be due to the production of acid in the fluid contents of the sarcomeres and a consequent attraction of water from the sarcoplasm by the acidulated colloids within the sarcomere. I proceeded to examine under the microscope the isolated sarcostyles of the house-fly's wing-muscles, when submitted to the action of weak solutions of lactic acid, and of hypertonic and hypotonic solutions of common salt. I found and reported in my second paper that, if the living sarcostyles are teased out in white of egg to which the acid or salt solutions are added, they respond to these influences exactly in the way demanded by my hypothesis; that, namely, very dilute solutions of lactic acid and hypotonic saline solutions cause them to contract; that hypertonic saline solutions cause them to remain extended. I showed also that the presence of lactic acid in the fluid surrounding the sarcomere certainly causes water to pass into the sarcomere; for, on increasing the percentage of acid beyond the point necessary to cause contraction (namely, to about 4 parts in 10,000 of water), the whole sarcomere becomes enormously distended; and further, that washing these distended sarcomeres in saline solution will reduce them to their normal proportions, presumably through the removal of the acid. I showed that, while it is possible to observe under the microscope the actual contractions of isolated sarcostyles produced by weak acid and hypotonic salt solutions only in the case of those of the insects' wing-muscles (for the simple reason that the living sarcostyles could not, in my experience, be separated from muscles of other types), yet other muscles, notably those of the insect's legs and those of the claw of the crayfish, respond to similar influences in an exactly similar way en masse. I therefore formulated my hypothesis anew in the following words:-" I 55
56 McDougall would suggest, then, that contraction is the result of the passage of fluid into the sarcomeres from the sarcoplasm, determined by the setting free of lactic acid in the fluid contents of the sarcomere, aided perhaps by an increase in the osmotic equivalent of these fluid contents through an increase in the number of molecules in solution. Then, so long as the acid remains present in the fluid of the sarcomere, the additional fluid absorbed will be retained and the state of contraction will continue. But as soon as the acid escapes from the sarcomere, the additional fluid will also escape with it into the sarcoplasm, and allow relaxation to take place." ' I concluded my paper by showing that plausible explanations in terms of the hypothesis may be given of a number of the secondary phenomena of muscular contraction, namely, the forms of the curves of contraction and relaxation, the greater force and economy of the early stage of contraction, the summation of stimuli-effects in tetanic contraction, the staircase effect, fatigue effect, "die Verkurzungs-Riickstand," " contracture," rigor mortis, and water rigor. In the year 1905, Professor J. Bernstein published a paper 2 which in the main is an examination of my hypothesis. He describes experiments with simple models like those mentioned above made by Reuleaux in the year 1900 and by himself at a later date; from these experiments he arrives at the conclusion, first, that if the wall of the chambers be extensible (like india-rubber; he uses the word elastic in this popular sense) distension will not produce contraction; secondly, that if the membranous wall is inextensible, it must be folded longitudinally. He reckons roughly the mechanical force that might be developed by osmotic pressure if the muscle consists of such chambers, i.e. chambers of inextensible but folded sidewalls; and he concludes: " Der osmotische Druck welcher in den kontraktilen Elementen der Muskelfaser bei der Kontraktion anwachst wurde ausreichen die Muskelkraft zu erzeugen, wenn die hierzu notwendige Bedingungen erfullt waren, welche darin bestande, dass diese Elemente aus kleinen mit langs-gefalteten Wandungen versehenen Blaschen in bestimmter Anordnung gebildet waren." And again he writes: " Es kann also keinem Zweifel unterliegen dass eine Zunahme des Quellungsdruckes in den angenommenen kontraktilen Elementen die Muskelkraft erreichen kann." " Die unerlassliche Bedingung aber fur das Zustandekommen der Kontraktion auch unter dieser Annahme besteht darin dass die kontraktilen Elemente die oben angegebene Struktur und Anorduung besitzen. Bevor daher weitere Konsequenzen an eine solche Hypothese geknupft werden, halte ich es fur nothwendig die Frage mit histologischen Mitteln zu behandeln. Obgleich ich weiss, dass nach allen bisherigen histologischen Untersuchungen eine solche Struktur, wie sie nach jener 1 C"A Theory of Muscular Contraction," Journal of Anatomy and Physiology, vol. xxxii. p. 200. 2 Cc Zur Theorie der Muskelkontraktion. Kann die Muskelkraft durch osmotischen Druck oder Quellungsdruck erzeugt werden? " Pfluger's Arch., Bd. 109.
Reply to Criticisms of my Theory of Muscular Contraction Annahme verlangt werde, nicht bemerkt worden ist und mir dieselbe nach eigener Anschauung hochst unwahrscheinlich vorkornmt, mochte ich doch nicht versatimen, die Herren Histologen von Fach zu einer Entscheidung aufzufdrdern." In short, Bernstein ignores completely my description of the structure of muscle, making no reference to my earlier paper devoted to that subject. He assumes that I had assumed, in a manner displaying gross ignorance of the most elementary mechanical principles and without any histological justification, a structure of the contractile elements which would not yield the effects desired; and then proceeds to show that, if the structure which I had laboriously and minutely described were present, my hypothesis would afford an adequate explanation of the contraction of muscle. I am compelled to believe that both of my papers were inaccessible to Professor Bernstein; that he knew of the second one only, and of that only through some most inadequate and misleading summary or reference by some other author. It is to me very satisfactory to find that Professor Bernstein's conclusion is wholly favourable to my hypothesis, and I shall hope that he will some day do me the honour of reading my original papers and that he will then accept my view as at least a tenable and well-based working hypothesis. I will add that nothing would please me more than that " die Herren Histologen von Fach" should take notice of my description of the structure of muscle and either confirm it or show how and why my account is unacceptable. In the present year Professor K. Hurthle has published a long paper1 describing a new and seemingly very valuable method for the investigation of the structure of muscle. He freezes the muscle, and dries it while still frozen. On then adding glycerine he obtains a fixed picture, which he believes is truer (i.e. less changed from that presented by the living tissue) than that obtainable by any other method of fixation. But from the point of view of my hypothesis his most interesting observation is that, if muscle be thus dried while frozen in the uncontracted state, and Ringer's solution be then added to it, it undergoes a contraction equal to at least 50 per cent. of its length. Here, then, we seem to have very good evidence that the contraction of muscle is a process determined by its mechanical structure, and that its structure is capable of determining contraction after the death of the tissue, if only the structure be preserved unchanged. I submit that, while this result is strictly compatible with my hypothesis, it is very difficult to reconcile with any of the other hypotheses at present in the field. Nevertheless, HUrthle rejects my hypothesis on the following grounds. After describing my hypothesis in a few words as a variety of Engelmann's hypothesis, he writes: " Gegen beide Untersuchungen, von McDougall und von Meiggs, ist aber einzuwenden dass sie auch nicht 1 "Ueber die. Struktur der quergestreiften Muskelfasern von Hydrophilus im ruhenden und tatigen Zustand," Pfluger's Arch., Bd. 126. 57
58 McDougall am frische Objekt oder wenigstens nicht ohne Zusatz von Reagentien angestellt worden sind, sondern an fixierten oder warmestarren Fasern. Am lebenden Objekt kann die Hypothese auch in dieser Form, wenigstens an den Beinmuskeln von Hydrophilus, nicht bestatigt werden, da bei der Kontraktion weder eine Volumzunahme der Fibrille im ganzen noch ihrer doppeltbrechenden Abschnitte festzustellen ist." 1 On this I would remark, first, that my hypothesis is in no sense a variety of Engelmann's; secondly, that my observations were in very large part made on living muscles, and that a considerable proportion of my published photographs were made from living muscles and from muscles fixed only by the application of steam to the cover-glass of the preparation, which process, as I have found again and again by careful comparison, produces no perceptible change of the appearances; thirdly, that it is quite impossible to measure the sarcostyles before and after contraction with such accuracy as would rule out the passage into them from the sarcoplasm of such a quantity of water as would suffice to bring about contraction; for, as I showed by many measurements, the proportions of the sarcomeres are such that the entrance of a very small quantity of water must bring about the change of shape which occurs during contraction; fourthly, against Hurthle's statement that he could observe no increase of volume of the sarcostyles during contraction, we may set Meiggs' conclusion that the quantity of sarcoplasm diminishes during contraction.2 Lastly, in view of Professor Hurthle's misrepresentations, I am compelled to suppose that my papers have not had the good fortune to be read by him. Professor J. S. Macdonald has criticised my hypothesis in the course of his article "The Structure and Function of Striated Muscle." 3 I feel the liveliest gratitude to Professor Macdonald, if only because he seems to have read my papers before criticising them. Nevertheless, I must point out that his criticisms are based, in part at least, on misunderstanding. After handsomely accepting my demonstration of the membranous wall of the sarcomere, Macdonald misrepresents my hypothesis as assuming that the passage of water into the sarcomere is caused by an increase of the osmotic pressure produced by increase of the number of molecules in solution in the fluid contents of the sarcomere; and he goes on to argue that the force that might be developed in this way is altogether inadequate, and decides " that the total chemical change of a contraction is quite incapable, merely by its addition to the number of molecules within the muscle, of giving rise to a sufficient amount of molecular motion to explain the force developed. No osmotic pressure hypothesis can rest solely on a basis of chemical change."4 In reply to this criticism, I would say, first, that if I rightly understand Macdonald's argument in regard to this point (and of this I am by no 1 Op. cit., S. 151. 2 op. cit., S. 108. 3 This Journal, vol. ii., No. 1. 4 op. cit., p. 11.
Reply to Criticisms of my Theory of Muscular Contraction means sure), he leaves out of account the fact that in any transverse section of a muscle or a muscle-fibre there is a very large number of sarcomeres, and that the forces developed in these must be added together in reckoning the total force exerted by the muscle. (This seems to me the explanation of the fact that Macdonald reaches in this respect a conclusion the opposite of Bernstein's). Secondly, even if I am wrong in this interpretation, and if Macdonald has not made this oversight, his conclusion is not, as he supposes, fatal to my hypothesis, and for this reason-i do not assume that the passage of water into the sarcomere is due to osmotic pressure developed by increase of number of molecules. That was the suggestion made in my first paper; but, as I have pointed out above, the hypothesis was changed in my second paper in the light of experiments made; and in place of osmotic pressure I substituted the power of the acidulated colloid contents of the sarcomere to attract water. I submit that the force so developed may be of very considerable magnitude. Unfortunately, I have not been able to follow up this question; but, since my papers were published, I have consulted the highest authority on osmotic pressures, and I am told that this power of imbibition of water by acidulated colloids is certainly not to be identified with osmotic pressure, and that nobody understands the nature of the process. Nevertheless, the process is a very real one; and the fact that lactic acid but very little stronger than that which suffices to cause contraction of the isolated sarcostyles produces a swelling in all dimensions, which goes on to complete rupture and destruction of the sarcostyles, proves that the force so developed is a very considerable one, much greater probably than any that eould be developed by osmotic pressure in the sarcomeres; for the soaking of isolated sarcomeres in distilled water produces no such swelling and disruption of them. Professor Macdonald thinks that my observation of the swelling of the sarcostyles when immersed in weak lactic acid solution does not support my hypothesis; for, as he says, how should the presence of lactic acid outside the sarcomere produce increase of osmotic pressure within it? This criticism, however, does not touch the second form of my hypothesis. I assume, perfectly legitimately, I submit, that the colloid contents of the sarcomere imbibe water or attract it through the membranous wall, whether the acid, as normally, I believe, is set free within the chamber, or is introduced into the fluid bathing it. Further, it remains open to me to vary my hypothesis by assuming that the lactic acid which determines the passage of water into the sarcomere is normally set free in the sarcoplasm; and, if that is the case, my production of contraction by immersion of the isolated sarcostyles in weakly acidulated white of egg is strictly comparable to the process of normal contraction. Professor Biedermann discusses theories of contraction at the conclusion of his long review of the "Vergleichende Physiologie der irritabeln Substanzen." 1 He gives a very brief, though fair, statement of my hypothesis 1 " Ergebnisse der Physiologie," viii. Jahrgang, 1909. 59
60 McDougall and of my view of the structure of muscle and of the changes of shape undergone by the elements in contracting. The latter he seems inclined to accept as not altogether baseless, especially as regards those points in which it has been verified by Meiggs; and the hypothesis he treats as deserving of serious consideration, pointing out the unfounded character of Bernstein's criticisms of it. In only one respect must I ask to be allowed to rectify Professor Biedermann's account of my view; like Macdonald, he represents me as holding that osmotic pressure is the cause of the passage of fluid into the sarcomere. I have already pointed out that in my second paper I suggested a modification of this view, to the effect that the principal cause, aided perhaps by increased osmotic pressure, is the attraction of the water by the colloid contents of the sarcomere in the presence of a very small percentage of lactic acid. Meiggs has confirmed, especially by means of photographs taken by ultra-violet rays, the principal points of my account of the structure of striated muscle. He, however, has not satisfied himself of the existence of the membranous lateral wall of the sarcomere and is sceptical as to the evidential value of staining in this respect. I would therefore emphasise the fact that, with both haematoxylin and gold chloride staining, it is possible to secure by differences of preliminary treatment of the tissue either the walls and transverse membranes of the sarcostyles alone stained, or the contents of the chambers alone stained, the transverse membranes being left as delicate clear lines between the darkly stained blocks of the coagulated contents of the chambers. The pictures of both kinds that I have frequently obtained have been so clear and sharp as to leave no doubt in my mind as to the reality of the transverse discs or membranes of a different constitution from the intervening substance, and similar as regards staining reactions to the lateral walls of the sarcomeres. I may add that my eyesight has suffered so much deterioration under the strain of these observations and of later work, that I cannot hope to see these delicate pictures again; and that no one should expect to see these finer details unless he enjoys perfect eyesight and is prepared to spend many hours a day for many weeks in the use of the highest powers of the microscope. But the evidence of the reality of the membranous wall does not consist only in its appearance to the eye. It consists largely in the observable peculiarities of the sarcostyle under a great variety of conditions, especially its behaviour when disrupted in various ways; and Meiggs is entirely in agreement with me in holding that the sarcostyle behaves just as though it were a membranous tube filled with liquid. To my mind three principal difficulties remained in the way of my views at the time of publication of my second paper. Two of these have been diminished, if not abolished, by the work of Meiggs. One was the generally accepted statement that the sarcostyles are made up of alternating bands of doubly and singly refracting substance of approximately equal thickness. My account of their structure was incompatible with this old-
Reply to Criticisms of my Theory of Muscular Contraction established dogma, and this fact probably accounts in large measure for the almost complete ignoring of it by physiologists for so many years. M eiggs has paid special attention to this point. He rightly insists that it is impossible to settle the question by the observation of the muscle-fibre en masse under polarised light. And, examining the isolated uncontracted sarcostyles of the insects' wing-muscles, he finds that " it can at least be positively asserted that the immensely greater proportion of the substance of the living sarcostyles, from Ad1 to O-, is evenly doubly refractive," and that the transverse membranes alone are possibly singly refractive.' This being so, the explanation of the production of the appearance of alternate light and dark bands when polarised light traverses the thickness of the muscle-fibre, in which it is subjected to immensely complex influences, remains of minor importance from my point of view. The second difficulty was that it seemed impossible to bring the contraction of unstriated muscle into line with my hypothesis. This difficulty also has been diminished by Meiggs' ingenious suggestion that the contraction of plain muscle is an elastic reaction, and that the process in the plain muscle-fibre corresponding to contraction of striated muscle is its lengthening, caused by imbibition of water: a suggestion supported by observations that seem to prove that " the contraction of smooth muscle is accompanied by a passage of fluid from the fibres to the interstitial spaces," and that those reagents which cause swelling and contraction of the sarcostyles of striated muscle cause relaxation or elongation of plain muscle-fibres.2 The third difficulty is that most striated muscles seem capable of shortening during contraction to an extent equal to 50 per cent. or more of their length. Now, a simple cylinder will shorten when laterally distended with fluid about 33 per cent. of its length only, its lateral wall becoming in longitudinal section approximately semicircular. As I mentioned above, it is easy to make models which will contract on distension to this extent. I believe that the presence of the extensible transverse septa that I have described in the sarcomeres will account for a greater degree of shortening than 33 per cent.; but it is not easy to prove this. Since publishing my second paper, I have submitted the mechanical problem to three able mathematicians, but none of them has been able to give me a positive answer to my question, whether adverse or favourable to my view. In the year 1900 I therefore devoted considerable time and money to the construction of a model which should reproduce the mechanical conditions I imagine to obtain in the sarcomere, in the hope that I should be able to cause it to contract on distension by more than 33 per cent. of its length. I have to admit that my attempt was unsuccessful; but the difficulties of constructing such a model are very great, and I believe that the failure of 1 op. cit., S. 105. 2 "The Application of McDougall's Theory of Contraction to Smooth Muscle," the American Journal of Physiology, vol. xxii., No. 4. 61
62 Reply to Criticisms of my Theory of Muscular Contraction my attempt was due to the imperfection of my models rather than to unsoundness of the principle assumed. In conclusion, I submit, then, that the criticisms hitherto made of my work invalidate neither my account of the structure of cross-striated muscle nor my hypothesis as to the cause of its contraction; but that the work done since the publication of my papers tends on the whole to bear out my conclusions, especially the work of Meiggs, the calculations of Bernstein based on the working of his models, and the demonstration by Hurthle that the muscle, after being dried while frozen, will contract on immersion in Ringer's fluid.