The Dynamics of Living Matter

* Engelmann, Ueber den Ur sprung der Muskclkraft, Leipzig, 1893. from objections such as we have mentioned ; namely, by assuming that chemical, not thermal conditions determine the absorbtion of fluid by the anisotropic substance. The action of the nerve upon the muscle might consist in facilitating a chemical change which increases the absorption of water by the anisotropic substance of the muscle. The thermodynamic conception of muscular contraction has been abandoned by many authors, and the surface energy has been con- sidered in its stead as the cause of muscular contraction and work. D'Arsonval, Imbert, and more recently Bernstein, have tried to offer a hypothesis of this kind. We shall understand these hypotheses better if we first consider Quincke's theory of protoplasmic motion.*

When a drop of oil is put on the surface of water which is in contact with air, the oil spreads in an extremely thin layer at the limit between water and air. This process continues until a film of oil exists between water and air. The conditions for the spreading of the oil on the sur- face of the water are as follows: the particle of oil O at the left end of the oil drop (Fig. 7) is under the influence of three surface tensions directions, OA, OB, and OC, and with different force. One is the

surface tension between air and water, which tends to pull the particle from O in the direction OA. The second is the surface tension OB at the limit of oil and air, which tends to pull the particle O in the direction of the tangent OB from O. The third force is the surface tension at the limit of oil and water, which tends to pull the particle in the direction of the tangent OC from O. The surface tension at the limit of water and air is greater than the sum of the surface tensions at the limit between oil and air, and oil and water. The surface tension between air and water is 8.25 mg., between oil and air 3.76 mg., and between oil and water 2.73 mg. The particle O will therefore be pulled toward the left ; and the same will happen with the next particle of oil, until the surface water air is substituted by the surface oil air.

These phenomena of spreading are accompanied by motions in the neighboring particles of liquid. If oil spreads at the surface of water, * Quincke, Sitzungsbtrichtc der Berliner Akadtmic der Wissensch., p. 791, 1888. the moving oil will, through friction, set the adjoining particles of water also in motion. The superficial layers of water, therefore, will move away from the center of spreading, and water will move toward the center from the interior, and from below. The arrows in Fig. 8 represent the currents in the water caused by the spreading of the oil. WATER Quincke holds that such

streaming. Such a streaming occurs constantly in the cells of Chara or Nitella. Quincke gives the following explanation for this process : all protoplasm contains oil or fat, and the surface layer of each cell must, therefore, be surrounded by a film of oil or fat. The oil will form through hydrolysis traces of fatty acid. The protoplasm contains substances which form soaps with the fatty acid. A soap solution must spread at the limit between oil and water, as the surface tension between oil and water is greater than the sums of surface tension between oil and soap solution, and between water and soap solution (which is zero). When the soap solution spreads, it must pull with it the adjacent particles of protoplasm. In this phenomenon of spreading, new particles of the surface of oil come in contact with protoplasm, new soap is formed, and the process is repeated. These phenomena of spreading, which constantly repeat themselves, furnish the energy for the constant streaming of protoplasm. The protoplasmic streaming occurs in the case of Chara or Nitella in one direction only. Quincke believes that this is due to an asymmetry in the structure of the cell, which renders the resistance to the streaming greater in one direction than in the opposite direction. Hence, the streaming occurs in one direction only; namely, that of least resistance.

The motion of an Amoeba can be imitated by bringing a drop of olive oil which contains a trace of fatty acid upon a one half to two per cent solution of Na2CO3. The oil, in this case, forms at its surface a film of solid soap. As soon as this dissolves at one spot, soap solution must spread at the surface of water and oil, and the moving soap solution must set also the neighboring layer of oil into motion. In the oil drop, therefore, two movements must occur, one at the periphery, which is directed away from, and one in the center, which is directed toward, the center of spreading. The arrows in Fig. 9 indicate these streams. The particles that flow from the interior toward the periphery produce a bulging out, and this is the analogue of the formation of a pseudo-

podium. According to Berthold,* the phenomena of streaming in the interior of an Amceba in the process of the formation of a pseudo- podium are such as to agree with the ideas of Quincke. Biitschli has come to the same conclusion. It seems to me, however, that if it is true that the Amceba is covered with a solid surface film, one condition for the formation of a pseudopodium must be a local liquefaction of protoplasm. In consequence of such a liquefaction, new protoplasm must flow out, which, subsequently, will form a new solid film at its surface. This may again be liquefied, and a new streaming may occur, etc. Such liquefactions can be caused by lack of oxygen, as we saw in a previous lecture; but they may also be caused by other chemical changes. I am inclined to believe that phenomena of liquefaction play at least some role in these processes of protoplasmic motion.

Imbertf published several years ago a hypothesis concerning the contraction of smooth muscle fibers, which assumes that the "stimulus" which causes the contraction of smooth muscles produces an increase in the surface tension between the longitudinal fibrils and the surround- ing liquid of the muscle cell. These fibrils are long and thin cylinders ; every increase in surface tension must have a tendency to make these fibrils more spherical, i.e. thicker and shorter. Such a change of form occurs indeed during contraction, but it is difficult to understand why the fibrils do not assume this form under the influence of surface tension alone, without stimulation. To meet this difficulty, Imbert assumes that smooth muscle fibers cannot contract unless they are stretched passively. He presupposes that their arrangement in the body is such that this prerequisite is generally fulfilled.

Bernstein has tried to explain away some of the weak spots in this hypothesis. t The surface energy at the limit between two media is equal to the product of surface tension into the surface. The work which surface tension can do is measured by the product of the decrease in surface, times the surface tension. From this it follows that the sur- face energy can do considerable work only, when the decrease in surface is large. From this Bernstein justly argues that Imbert's assumption is incorrect, inasmuch as the area required for the work the muscle really does, must be much larger than that between the fibrils and the neighboring liquid. Bernstein therefore assumes that the fibril consists of a row of quite small ellipsoids, whose long axis is in the direction of the fibrils, and whose form is determined by elastic forces. An increase in the surface tension must make these ellipsoid elements more spherical, and thus the fibril becomes shorter and thicker.

How can the nerve impulse or an artificial stimulation of the muscle increase the surface tension? There are several possibilities. Sub- stances might be formed in this case which increase the surface tension at the limit between Bernstein's hypothetical ellipsoids and the surround- ing liquid. Another possibility might be that, through the process of innervation or stimulation, an existing difference of electrical potential between the ellipsoids and the surrounding liquid might be diminished. D'Arsonval explains the efficiency of electrical stimuli in this way.* Hermann has offered another hypothesis ; namely, that the contrac- tion is a process of coagulation, and the relaxation, a process of lique- faction.f He was led to this idea by the fact that the change in form which the muscle undergoes in the case of rigor mortis is similar to that in contraction, and that moreover a number of other features are common to both. Although all these hypotheses concerning muscular contraction have been known for a number of years, none has led to a new discovery. The reason lies possibly in the fact that one or more links in the catenary series of processes which underlie muscular con- traction have been ignored in these hypotheses. It is well known that a muscle gains in mass through " contractions, and that it undergoes atrophy when it remains at rest. This fact indicates, in my opinion, very clearly that phenomena or reactions which directly or indirectly lead to growth form a part in the process of muscular contraction. I consider it quite possible that no hypothesis concerning muscular contraction will prove fertile until this relation between activity and growth of the muscle is recognized.

Were scientists with a purely physical training to be asked to give a hypothesis concerning cell division, I believe that their hypothesis would not take into consideration the phenomena of growth. Nevertheless, these phenomena form an obvious link in the catenary series of processes which result in the division of the nucleus and the cell. It had been tacitly recognized by botanists that the growth of a cell precedes its division and is possibly the cause of the division. The botanist J. Sachs was the first to definitely state that in each species the ultimate size of a cell is a constant for each organ, and that two individuals of the same species but of different size differ in regard to the number, but not in regard to the size of their cells.* Amelung, a pupil of Sachs, determined the correctness of Sachs's theory by actual counts. Sachs, in addition, recognized that wherever there were large masses of protoplasm, e.g. in Siphonese and other cceloblasts, many nuclei were scattered throughout the protoplasm. He inferred from this that "each nucleus is only able to gather around itself and control a limited mass of protoplasm." f He points out that in the case of the animal egg the reserve material - fat granules, proteins, and carbohydrates — are partly transformed into the chromatin substances of the nuclei, and that the cell division of the egg results in the cells reaching that final size in which each nucleus has gathered around itself that mass of protoplasm which it is able to control. Morgan J and Driesch§ tested and confirmed the idea of Sachs for the eggs of Echinoderms. Driesch produced artificially larvae of sea urchins of one eighth, one fourth, and one half their normal size by isolating a single cleavage cell in one of the first stages of seg- mentation of the fertilized sea-urchin egg. He counted in each of the dwarf gastrulse resulting from these partial eggs the number of mesen- chyme cells and found that the larvae from a ^ blastomere possessed only £, those from a \ blastomere only ^, and those from a | blasto- mere only •§• of the number of cells which a normal larva developing from a whole egg possessed.

Moreover, he could show that when two eggs were caused to fuse so as to produce a single larva of double size, the gastrulas of such larvae had twice the number of mesenchyme cells. Driesch drew from his observations the conclusion that each morphogenetic process in an egg reaches its natural end when the cells formed in the process have reached their final size. Gerassimowll found that by exposing dividing cells of Spirogyra to a low temperature the division became irregular, and it happened that the nuclear material instead of being divided between the two

§ Driesch, "Von der Beendigung morphogener Elementarprocesse," Arch, fur Enhvicke- Inng'smechanik, Vol. VI, 1898. "Die isolirten Blastomeren des Echinidenkeims," ibid., Vol. masses of protoplasm remained in one of the two daughter cells ; some- times all the chromosomes were united into one nucleus and sometimes he obtained two nuclei. He found that cells with an increased mass of chromatin only began to divide after their protoplasm had reached a much greater mass than that found in the normal cells with half the mass of the chromosomes. This fact seems to indicate that cell division is determined by the ratio of the mass of the chromosomes to that of the protoplasm. If the mass of the chromosomes in a cell is increased, the tendency for cell division does not develop until the mass of protoplasm is increased also. It is the merit of Boveri to have found the law which governs this condition for cell division.* He compared the process of segmentation in normally fertilized fragments of sea-urchin eggs, which contained the normal number of chromo- somes, with that of enucleated fragments which contained only the sperm nucleus, and whose mass of chromatin was only one half of that of the normal fertilized egg. In order to understand his results, the reader's attention should be called to the following fact: in the process of cell division each daughter nucleus of the egg contains just as many chromosomes as the mother nucleus, but the mass of chromatin (and possibly the other constituents of the nucleus) of each chromosome in the daughter nucleus is only one half of that of the corresponding chromosome of the mother nucleus. The next phase -- the resting stage between two divisions — consists in the growth of the chromosomes of the two daughter nuclei until they have reached the mass of the original chromosomes and then a new nuclear and cell division begins.

The material for the growth of the chromosomes is furnished by the protoplasm and according to the above-quoted idea of Sachs by the reserve material included in the protoplasm and not by the "living" part of the latter itself. It is, however, questionable whether this latter discrimination has any real basis. In this way the process of cell division in the egg consists in the gradual transformation of proto- plasmic into chromatin material of the nucleus until a definite ratio between the mass of the chromosomes and the protoplasm is reached. When this is established, no new cell divisions are possible until the mass of the protoplasm is increased again through the absorption of food stuffs on the part of the cell. The process of the transformation of protoplasm into chromatin is necessarily rendered discontinuous through the fact that the chromosomes cannot grow indefinitely, but that growth will stop as soon as they have reached a certain size, and this fact leads apparently to the process of nuclear divisions. I

* Boveri, Zellen-Studien, Heft 5. Ueber die Abhangigkeit der Kertigrosse und Zellen- zahl der Seeigel-Larven -von der Chromosomenzahl der Ausgangszellen, Jena, 1905. have thought of the possibility that the continuation of the synthetical process which leads to the formation of nuclear material from certain constituents of the protoplasm gives rise to the formation of astrospheres as soon as the maximal growth of the chromosomes is reached. But" this does not need to enter into our consideration for the present.

Since each daughter nucleus of a dividing blastomere has the same number of chromosomes as the original nucleus of the egg, it is clear that in a normally fertilized egg each nucleus has twice the mass of chromosomes as the nucleus of a merogonic egg, i.e. an enucleated frag- ment of protoplasm which has only the sperm nucleus. Boveri has not only ascertained this fact but he has also ascertained the further fact that the final size of the cells after the cell divisions have come to a standstill is always in proportion to the original mass of the chromatin contained in the egg ; the cells of the merogonic embryo, e.g. the mesen- chyme cells, are only half the size of the same cells in the normally fertilized embryo. Driesch has just furnished a further proof of Boveri's law, that the final ratio of the mass of the chromatin substance in a nucleus to the mass of protoplasm is a constant in a given species. He compared the size of the mesenchyme cells in a sea-urchin embryo produced by artificial parthenogenesis with those of a normally fertilized egg and found them half of the size of the latter. When the fertilized eggs and the parthenogenetic eggs are equal in size from the start, - which is practically the case if eggs of the same female are used, --the process of the formation of mesenchyme cells comes to a standstill in the normally fertilized eggs when the number of mesenchyme cells is half as large as the final number of mesenchyme cells found in the parthenogenetic egg.* As a matter of fact, Boveri's results as well as those of Driesch were obtained by counting the cells formed by eggs of equal size and not by only measuring the size of the cells. It is most remarkable that certain apparent exceptions to Boveri's law which Driesch has actually found have been predicted by Boveri.

The fact that the process of cell division comes to a standstill when the ratio of the mass of the chromosomes in the nuclei of an egg or an organ to that of the surrounding protoplasm reaches a certain limit, suggests in my opinion the possibility that this ratio is determined by the laws of mass action and chemical equilibrium. If this is correct, the synthesis of nuclein compounds from the protoplasmic constituents must be a reversible process. This suggestion would gain in probability if it could be shown that a reduction of size in protoplasm in the case of starvation is also followed or accompanied by a reduction in the size of the nuclei.

The fact that the cell division is as a rule preceded by a synthetical process explains possibly the fact mentioned in the second lecture that the phenomena of cell division in a fertilized egg come soon if not immediately to a standstill when the atmospheric oxygen is with- drawn from the egg. We have mentioned Schmiedeberg's view in regard to the role of oxygen in synthetical processes. But even if this view were not correct, we can understand that lack of oxygen might indirectly interfere with the synthesis of the nuclein compounds.

E. P. Lyon has shown that the chemical conditions and processes in the cell differ in the various phases of cell division. He found that during different stages of cell division the egg of the sea urchin shows a different resistance to the effects of HCN, and this difference repeats itself during each of the successive segmentations. More recently he has added the important fact that the production of CO2 on the part of the egg of the sea urchin also undergoes periodic variations during segmentation.

If we now turn to the physical side of the phenomena of cell division, we shall meet almost the same uncertainty which confronted us in the case of muscular contraction. We shall therefore confine ourselves to the enumeration of a few facts, with occasional hints for possible further work. When we watch the process of cell division in an egg, we can dis- criminate at least three distinct phases of this process: first, the ap- pearance of systems of radiation — the so-called astrospheres — in the protoplasm of the cell. The second phase is the disappearance (liquefac- tion?) of the nuclear wall, and the division and migration of certain constituents of the nucleus, namely, the chromosomes toward the centers of the astrospheres, and the formation of two new nuclei. The third phase consists in the separation of the protoplasm into two pieces in a plane, which, from the position of the astrospheres, as a rule, can be predicted. This latter separation is the process of cell division proper.

If newly fertilized eggs of the sea urchin are put into sea water, whose osmotic pressure has been adequately raised by the addition of some salt, e.g. NaCl, or sugar, no segmentation occurs as long as the eggs remain in this solution. If they are brought back from this solu- tion into normal sea water, they will segment, provided they have not been left too many hours in the hypertonic sea water. There is, how- ever, a characteristic difference between this segmentation and the normal segmentation. If the eggs are brought back into normal sea water after two hours, they do not divide, as a rule, first into two, and then into four cells, but into three or four cells simultaneously. If they are left for three or four hours in the hypertonic solution, and then

brought back into normal sea water, they break apart into from six to sixteen cells simultaneously in about ten or twenty minutes after being put back into normal sea water. If they remain for five or six hours in the hypertonic solution, many eggs suffer. If they do not suffer they soon break up, when put into normal sea water, into a still larger number of cells than if they remained only for three or four hours in the hypertonic sea water. I have seen such eggs divide simultaneously into about forty cells, or more, in from ten to twenty minutes after being put back into normal sea water. These phenomena of segmen- tation are accompanied by violent phenomena of streaming or proto- plasmic motion at the surface of the egg. From these facts I concluded that while the hypertonic sea water inhibits the cell division, it allows the division of the nucleus, which precedes the segmentation of the protoplasm.* W. W. Normanf undertook a histological examination of the eggs under these conditions. He found that if the concentra- tion of the sea water be adequately, but not excessively, raised through the addition of a definite amount of NaCl, KC1, or MgCl2, the nuclei of the eggs divide in the hypertonic sea water karyokinetically, into two, four, and eight successively, while no cell division occurs. When such an egg with eight nuclei is put back into normal sea water, it divides as a rule into more than eight cells simultaneously. If a slightly too high concentration is used, the distribution of the nuclei in the egg does not become so regular; if the concentration is still a little higher, an excessive number of astrospheres is formed, as Morgan and Norman found.

In this case, the nuclear material is often not scattered in the egg, although the nucleus seems to be broken into smaller fragments, for if brought back into normal sea water such eggs break up rapidly into a larger number of cells. R. Hertwig had already observed the formation of astrospheres in the unfertilized eggs of the sea urchin when he added a little sulphate of quinine to the sea water, and Morgan applied the method used by myself and Norman to the unfertilized egg and found an excessive number of astrospheres, just as Norman had observed in the fertilized egg.J

It is obvious from these and other experiments not mentioned here that the loss of water on the part of the fertilized egg ultimately retards all the phases of nuclear and cell division, but not all quantitatively alike. It seems that the chemical process of transformation of protoplasmic into chromatin material is less interfered with than the cell division proper. This follows from the fact that the chromosomes may divide without cell division. Under normal conditions, the growth of the chromo somes is followed by a formation of astrospheres and division of the nuclei, and this in turn is followed by a cell division. The loss of water which the egg undergoes in the hypertonic sea water seems to inter- fere mostly with the cell division. It is possible that the viscosity of the protoplasm is increased by the loss of water, and that this condi- tion interferes somewhat with the migration of chromosomes after they have divided and still more with the segmentation of the protoplasm. The growth of the chromosomes and the subsequent formation of as- trospheres seem, however, to continue for some time in the hypertonic sea water.

O. and R. Hertwig (as well as Roux) have noticed that as a rule the plane of division of a non-spherical cell is at right angles with the direction of the greatest diameter, or extension of the cell. Driesch has given a nice experimental proof for this rule. If the newly fer- tilized egg of the sea urchin be gently pressed under a cover glass, so that it is slightly flattened, the plane of division is at right angles to the slide. The position of the plane of cleavage is determined by the position of the nuclear spindle, and the latter depends upon the position of the centrosomes or astrospheres. The question hence arises, How does it happen that in most cases the common diameter of the two astrospheres coincides with the longest diameter of a cell? This posi- tion of the astrospheres or centrosomes becomes comprehensible on the assumption that these organs not only repel each other, but are also repelled by the external surface of the nuclei and the inner surface of the cell limit. The forces involved in this repulsion must be forces such as occur in liquids, as the contents of the egg of the sea urchin is mainly liquid. It must, moreover, be taken into consideration that the space in which the process of cell division occurs, is generally of microscopic and always of capillary dimensions. It is therefore quite possible that the repelling forces in this case are capillary forces. There is, however, another fact to be considered ; namely, that in the process of cell division the egg of some animals becomes elliptic, with its long axis falling in the direction of the common diameter of both astro- spheres. This has given rise to the idea that the spindle or the astro- spheres elongated the egg.

I have often noticed - - as others have undoubtedly done before me — an elongation of the egg of the sea urchin in the direction of the spindle, but this always occurred imme- diately before the cell division. It gives easily the impression as if contractile forces were active in radial directions in the astrospheres and that these forces had something to do with the process of cell divi- sion. Certain deviations from Hertwig's law may be only apparent.

Such exceptions occur in epithelial cells, but it is quite possible, that not the whole cell but only one part of it is in this case the seat of the processes which cause the orientation of the astrospheres or the cen- trosomes. R. Lillie has expressed the idea that electrical forces play a role here. - Were this idea correct, it should be an easy matter to control the orientation of the plane of cleavage in the cell by means of a gal- vanic current; such is, however, not the case, at least as far as our present experience goes. If hydrodynamic and "contractile" forces are responsible for the orientation of the astrospheres or centrosomes in the cell, it should be expected that these latter organs are solid or at least more viscous than the rest of the liquids of the cell. If the centro- somes are fixed organs of the cells, and multiply by division, they must naturally be solid or at least possess a solid surface. Alfred Fischer assumed that the formation of astrospheres depended upon a process of coagulation. This has not yet been proved, although this author has imitated the well-known figures of astrospheres in coagulated proteins.

If the process of nuclear division in transparent cells, e.g. egg cells, is observed, the impression is easily gathered that the astrospheres cause a liquefaction of the nuclear membrane and an emulsification of certain constituents of the nucleus. If this observation be correct, the phenomena of spreading and the phenomena of streaming con- nected with such a process might be the forces which carry the chro- mosomes of the nucleus toward the center of the astrospheres. This assumption is in harmony with the fact that the withdrawal of water from the egg cell diminishes the velocity of the nuclear division,* inas- much as the loss of water may easily increase the viscosity of protoplasm, and thus diminish the velocity of the process of streaming, finally ren- dering it entirely impossible.

The phenomena of streaming can be demonstrated most beauti- fully in the experiment described above in which eggs of the sea urchin were put into hypertonic sea water, whose concentration was just adequate to prevent the cell division, without preventing the nuclear division. When such eggs are put back into normal sea water after about three hours, the most powerful phenomena of streaming may be witnessed, resulting in the formation of knobs. The streaming seems to occur around the chromosomes or fragments of nuclear mate- rial as a center. Afterwards, each such knob, or projection, formed by the streaming becomes a separate cell.f

t This amceboid, character of cell division had been observed and described before by O. and R. Hertwig and called " Knospenfitrchung" In my earlier experiments on artificial parthenogenesis, I frequently had opportunity to observe cell divisions of a character which made it clear that phenomena of streaming underlie cell division, at least, in these cases. Figures 10-13 give an illustration of such a case. The egg had been treated with hypertonic sea water, and when put back