AMEBOID MOVEMENT

AMEBOID MOVEMENT

BY
ASA A. SCHAEFFER, Ph.D.
PROFESSOR OF ZOOLOGY, UNIVERSITY
OF TENNESSEE
PRINCETON UNIVERSITY PRESS
PRINCETON
LONDON: HUMPHREY MILFORD
OXFORD UNIVERSITY PRESS
1920
Copyright, 1920, by
Princeton University Press
Published 1920
Printed in the United States of America

PREFACE

Although the subject of ameboid movement is discussed in this book chiefly because of its intrinsic interest, yet the interests of the student of medicine, the psychologist, the physiologist, the evolutionist and the general biologist have constantly been kept in mind. For the medical investigator probably finds no better means of approach to the study of the reactions and especially the movements of the white blood corpuscles, which play such an important part in the economy of the human body, than the ameba; white blood corpuscles and amebas are strikingly similar in many characteristics and in the fundamental processes of the movement they are probably identical. The comparative psychologist is keenly interested in the activities of the ameba because it exhibits to him the operation of the animal mind in its greatest simplicity. To the physiologist ameboid movement has for a long time represented the simplest phase of muscular contraction as it is known in the vertebrates. The philosophical evolutionist sees in the ameba, both in its structure and in its activities, a close approximation to the earliest ancestor of the animals. And the general biologist, aside from his usual interest in the properties of living matter wherever it may be found, is especially interested in discovering how many of the activities of the ameba are common to other organisms.

But in addition to presenting an account of the main facts concerned in the movement of the ameba from the various points of view mentioned above, this book has a second object which is scarcely subsidiary to the main one. This second object is to present the thesis that moving organisms in which orienting organs are absent or not functioning, always move in orderly paths, i. e., in helical or true spiral paths. The movements of the ameba under controlled conditions, which, as the following pages will show, take the form of a helical spiral projected on a plane surface, therefore serve as an introductory study to the movements of organisms generally. For the presumption is strong that there is an innate tendency in all organisms that move which compels them, when free from stimulation, to move in definite predictable paths. This thesis is discussed at some length in [Chapters XII] and [XIII].

In view of the fact that ameboid movement has been considered largely as a theoretical question heretofore, I wish to state at once that my discussion of this subject is based directly on observation and experiment. I have no new theory of ameboid movement to offer; the list of theories is already extensive enough. I am, on the other hand, strongly of the opinion that this fundamental question, if it is to be solved at all, can be solved only by persistent observation and experiment on the ameba and related organisms themselves. “All knowledge is vain and erroneous excepting that brought into the world by sense perception, the mother of all certainty” (Leonardo).

CONTENTS

CHAPTER I
Introduction

The manner of movement common to amebas has attracted the attention of biologists ever since the discovery of ameba by Rösel v. Rosenhof in 1755. In his description of “Der kleine Proteus” he records the observation that the various form changes which the ameba undergoes are associated with the streaming of the endoplasm. This observation marks the very beginning of the investigation of ameboid movement. And this investigation also possesses the distinction of being the most important single observation that has thus far been recorded in this special field, for it is now generally understood that by ameboid movement is meant movement due to the streaming of protoplasm.

The phenomenon of ameboid movement as discovered by v. Rosenhof, was an isolated phenomenon. It attracted attention mainly because of its uniqueness, for it was the only instance of the kind that was then known. It could not be compared with any other form of movement; and the animal itself, considered apart from the streaming of the protoplasm, was unique also, because of its remarkable form changes which it alone, of all the animals then known, exhibited.

But when Corti in 1774 discovered streaming protoplasm in the cells of chara and various other plants, the ameba could no longer be said to occupy this position of isolation. Although streaming is not accompanied by locomotion in chara, it had been observed that movement in the ameba was always accompanied by streaming, so it came to be generally accepted that the really fundamental feature of ameboid movement was the streaming of the protoplasm.

The ameba came to be of especial interest to the physiologists later on when the finer structures of the larger animals were studied more carefully. Thus when the normal movements of the white blood corpuscles were discovered, no one failed to be struck with their ameboid characteristics in almost every detail of movement, feeding habits and gross structure. The great importance of the functions that have been ascribed to leukocytes, and their very widespread occurrence in the higher animals has served to give rise to the belief that ameboid characteristics were not unique among animals, but common to many of them. The discovery of ameboid movements among plant zoospores, among animal ova, in the endoderm cells lining the digestive tract of a great variety of animals, in the nuclei of some animal cells, in the wandering cells of sponges and other animals—all these instances of ameboid movement occurring in such widely different tissues inevitably placed it among the most important phenomena known to occur in organisms.

Out of the discovery that ameboid movement may be exhibited in some form or other in so many different kinds of organisms, grew the theory that even muscular movement as known in man and the higher animals is at bottom a specialized sort of ameboid movement; not merely phylogenetically, but as it is now known. As we shall see however in the following pages, this theory of muscular movement cannot be based specifically on the streaming process per se, but it is very probable, on the other hand, that the same process which underlies contraction of the ectoplasm in the ameba also underlies contraction in muscular tissue.

But this remarkable story of the development of a single unrelated observation into a widespread biological phenomenon is not yet complete. With its further development the following pages are concerned. It will be shown that the movement of the surface film of the ameba is analogous to that of some blue-green algae, diatoms and crawling euglenas, in which organisms the surface film seems to be the vehicle of movement. Thus the ameba finds itself related to these organisms by new ties. More important still is the significance of the wavy path of the ameba, which may possibly be due to the same fundamental mechanism that controls, under suitable conditions, the direction of the path in man and many other animals and motile plant cells. Thus the phenomenon of ameboid movement born in nakedness and utter isolation, has become attired, in a brief space, with the Victorian garb of a Fundamental.

CHAPTER II
Historical Sketch

For the purpose of presenting in brief compass the main published observations and experiments on ameboid movement, we may pass from the observations of v. Rosenhof, mentioned in the introduction, to certain observations which Wallich (’63) recorded. He found that a new pseudopod is usually formed as a small break in the ectoplasm somewhere on the ameba through which the endoplasm then flows. As the endoplasm flows out and the new pseudopod enlarges, the breach in the ectoplasm increases in extent, through a transformation of the ectoplasm in the immediate vicinity of the breach, into endoplasm. But he observed also that some of the endoplasm which flows into the new pseudopod becomes transformed into ectoplasm. Wallich thus demonstrated that ectoplasm and endoplasm are mutually convertible.

The conversion of ectoplasm into endoplasm and vice versa, was regarded by Wallich, however, as a process taking place only occasionally, such as when new pseudopods are formed. It remained for Bütschli (’80, p. 115) to point out that in a moving ameba endoplasm is continually formed from ectoplasm at the anterior ends of all pseudopods, while the reverse process, viz., the conversion of ectoplasm into endoplasm, takes place continually at the posterior end of the ameba. He describes the relation of ectoplasm to endoplasm as a “circulation”; the endoplasm, arriving at the anterior end, becomes changed into ectoplasm, which after remaining relatively stationary for a while on the outer side of the animal, soon finds itself at the posterior end of the ameba, where it is slowly changed into endoplasm. The movement of the endoplasm forward to the anterior end of the ameba completes the cycle.

In 1898 Rhumbler, from observations on several species of amebas, came to the conclusion that in the change from ectoplasm into endoplasm, and vice versa, must be sought the cause of ameboid movement.

Jennings (’04), however, from extended study of the physiology of the ameba, stressing especially movement and feeding, denied that the transformation of endoplasm into ectoplasm, and vice versa, is necessary or even of frequent occurrence during movement. Instead of these transformations occurring regularly, as Bütschli and Rhumbler described them, Jennings concluded that the ectoplasm is more or less permanent, behaving like an elastic skin, which rolls over and over as the ameba moves along. The ectoplasm thus remains ectoplasm, and the endoplasm retains its identity, for considerable periods of time, instead of being continually transformed, the one into the other, as the ameba moves along.

Although observations with regard to movement in ameba have consisted almost wholly of the mutual relations of ectoplasm and endoplasm, it is important to note that the existence of a third layer of protoplasm, outside of the ectoplasm, was foreshadowed by an observation of Bütschli (’92, p. 219) while examining a pelomyxa. To his great surprise he found that there were currents of water, as evidenced by the movement of suspended particles, at the sides and in close contact with the ectoplasm of the pelomyxa, which flowed slowly forwards toward the anterior end. No details were given and no explanation offered for the cause of the currents excepting the suggestion that there might be a thin skin over the animal, which moves slowly forward.

Two years later Blochmann (’94) demonstrated by means of the very fine cilia-like projections which frequently cover the outside of pelomyxas, that the surface of the pelomyxa actually moves forward during active locomotion. He did not state definitely whether or not he considered this surface as a part of the ectoplasm.

This observation of Blochmann was not developed, however, until Jennings (’04), by means of particles attached to the outer surface of amebas, studied the forward movement of this layer. The results of Jennings’ work led him to conclude that the outer surface of amebas, which move forward as demonstrated by attached particles of soot and other substances, is continuous with the ectoplasm, and is really the ectoplasm. The rate of movement of this layer was stated to be about the same as that of the ameba as a whole. He denied the validity of Bütschli’s suggestion that there might be a thin third layer on the outside of amebas or pelomyxas.

But the existence of a third layer of protoplasm as distinct from the ectoplasm, was again maintained by Schaeffer (’17) who found that in some amebas the outer surface moves forward faster than the ameba advances through the water. The third layer was found to be generated over the surface of the ameba, especially in the posterior region of the ameba, and destroyed at the anterior end.

But the purely observational aspect of the problem of ameboid movement has not interested biologists generally as much as the ultimate cause of the phenomenon.

The first attempt that was made to explain ameboid movement in conformity with the demands of modern experimental science, that is, on the basis of physical factors, was made by Berthold (’86). By means of simple experiments with inert fluids (oils, alcohol, water, ether) which were modeled after an experiment described by the physicist Paalzow (’58), Berthold concluded that locomotion in ameboid organisms is due to the physical attraction of the anterior end to the substratum. The ameba was supposed to behave like a drop of fluid which moved towards the point where the tension of the ameba’s surface was decreased by contact with the substratum. The ameba did not push out pseudopods according to Berthold, but they were pulled out because of a difference in surface tension between them and the substratum. But pseudopods which were extended into the water and out of contact with a solid substratum, were said to be extended by a contractile effort of the posterior region of the ameba.

Bütschli (’92, p. 187) pointed out that it was highly improbable that pseudopods in contact with a solid substratum were projected in a fundamentally different way from that in which free pseudopods were extended, as explained by Berthold. Bütschli assumed that all ameboid movement was due to the same fundamental cause. He postulated surface tension as the active agent, as Berthold had done for the extension of pseudopods in contact with a solid substrate; but Bütschli assumed that the decrease in surface tension at the anterior end of the ameba was brought about by the bursting of protoplasmic droplets of a more fluid consistency on the surface of the ameba, the consistency of which was less fluid, thus bringing about a decrease of surface tension and consequent forward streaming of the endoplasm. The necessary migration of the more fluid droplets to the surface was determined by internal conditions. The direction in which an ameba moves was assumed to depend therefore not upon the physical character of the substrate, as suggested by Berthold, but upon such internal changes as control the movement of the more liquid part of the internal protoplasm to the outer surface.

Rhumbler (’98) wrote extensively on the subject of ameboid movement, especially from the point of view of the feeding habits of amebas. He concluded that the flow of protoplasm, while engulfing a food object, was a direct result of the lowering of the surface tension of the protoplasm by contact with the food object (p. 207), thus causing its envelopment. Numerous other writers of the time, including Quincke (’88), Verworn (’89, ’92), Blochmann (’94), Bernstein (’00) and Jensen (’02), agreed in a general way with Rhumbler’s position that surface tension changes are the cause of locomotion in ameba.

In 1904 the general subject of ameban behavior was extensively studied by Jennings, and from his observations he concluded that surface tension cannot account for many of the reactions observed. Other factors, he held, must be at work, such as contractility, which, acting in the posterior region, causes the endoplasm to flow forward. But Jennings found it impossible to explain on the same basis the extension of free pseudopods, and the creeping of a pseudopod, or of the whole ameba, over a solid substratum.

From further observations Rhumbler (’05, ’10) came to modify his earlier views as stated above. The rapid advances in the study of the chemistry of colloids doubtless suggested to Rhumbler that the change from endoplasm to ectoplasm resembled the change from a sol to a gel state, and that in this process of gelation lay the source of energy manifested in ameboid movement. In thus calling attention to, and emphasizing the colloidal nature of, the conversion of endoplasm into ectoplasm and vice versa, the problem of ameboid movement came to be discussed from an entirely new angle. Certain phases of Rhumbler’s theory are developed and elaborated by Hyman (’17) who agrees in general with Rhumbler’s conclusions.

In a series of papers on feeding and other reactions of ameba, Schaeffer (’12, ’16, ’17) concluded that Rhumbler’s general statement, wherein he says that changes in behavior are directly deducible from the action of stimuli in effecting liquefaction or gelation of the ectoplasm, does not hold in many cases of feeding, and that the mechanism controlling locomotion and feeding is not external, as maintained by Rhumbler, but internal.

CHAPTER III
The General Features of Endoplasmic Streaming

The streaming of the endoplasm is the most conspicuous feature of ameboid movement. It is even more noticeable than the movement of the pseudopods themselves, because of its greater speed and because it occurs in all parts of the ameba. Its importance in movement is essential, for no continued locomotion can be observed unless accompanied by streaming. It may be profitable therefore to enquire into the general features of streaming, and to observe some of the necessary consequences streaming imposes upon such an animal as the ameba.

Let us take as an example an Amoeba proteus (Pallas, ’66, emend. Leidy, ’79, emend. Schaeffer, ’16) in characteristic movement (see [Figure 11], p. 37). The main streams of endoplasm are in the same direction as that in which the ameba moves. In the withdrawing pseudopods the current is, of course, toward the main mass of the ameba. The endoplasmic stream is continuous from the posterior end to the tips of the advancing pseudopods. The retracting pseudopods flow into the main stream as tributaries. If, as often happens, the ameba is without pseudopods, there is then a single stream arising in the posterior end and flowing to the anterior end. In such a case it is readily observed how absolutely dependent locomotion is upon endoplasmic streaming.

It often happens, such as when the ameba is receiving a strong stimulus, that streaming is arrested and brought to a stop for a few seconds, more or less. Presently however the endoplasm begins to flow as before. At what point, in such a case, is the first movement of endoplasm detectible? Is it at the free end of the pseudopod, at its middle region, at its base, or at the posterior end of the ameba? Bütschli (’80, p. 116) observed that in a withdrawing pseudopod the streaming begins at the free end of the pseudopod; but his (’92, p. 201) later explanation of ameboid movement seems to require that the endoplasm must begin to move at the base of the withdrawing pseudopod. Jennings (’04, p. 157) observed that in a withdrawing pseudopod the current of endoplasm begins at the base of the pseudopod.

From numerous observations directed toward this point, it appears that the conditions under which streaming is resumed after a pause, whether in the same or in the reverse direction, are of great variety. The shape, size, slenderness, and the position on the ameba of the pseudopod, as well as the strength and character of the stimulus, are among the factors capable of changing in whole or in part the flow of endoplasm in a pseudopod. In an ameba that has been moving along a homogeneous flat surface, as nearly unstimulated as may be, the endoplasm first begins to flow out of the lower half of the retracting pseudopod, if the pseudopod is more or less uniformly conical in shape and rather slender. In such a case it may be said that the retracting pseudopod was withdrawn “by the ameba,” and that it did not itself receive an external stimulus producing retraction. If, however, the tip of a pseudopod as described receives a strong negative stimulus, the endoplasm frequently flows back from the tip while it is still flowing into the pseudopod at the base. But very soon thereafter, in such an event, the streaming becomes unified and the pseudopod is withdrawn. In broad pseudopods about to be withdrawn, the endoplasm may begin to move anywhere along its length. This is undoubtedly due to the continuous local changes in the walls of the pseudopod, which are characteristic of this species of ameba (see p. 20).

In an ameba which has been brought to a standstill, as by a sudden flash of light, the first sign of recurring streaming is in the anterior half, whether the original direction of streaming is resumed or reversed. If the direction is reversed, the active pseudopods retract for a considerable distance before a new one is projected. The endoplasmic stream in a slender withdrawing pseudopod may not reach to the tip for from several seconds to a minute, if the tip is slightly positively stimulated. One may then observe ectoplasm streaming toward the tip and toward the base, in the respective regions, at the same time, with considerable fluctuation back and forth of the neutral zone separating the two streams. The fate of such a pseudopod depends on its size, on its position on the ameba, and the strength of the stimulus affecting it and the rest of the ameba. That is, if the pseudopod is small or on the posterior half of the ameba, or only slightly stimulated, it will be retracted; but if it is large, or on the anterior end of the ameba, or more strongly stimulated than the rest of the ameba, it may again become active.

The fact that protoplasm is practically incompressible makes it clear that if streaming can be observed to begin after a pause at some point after it begins at others, the ectoplasmic walls of the ameba must give way in the region where streaming begins. Since it has been established by observation that the ectoplasm may give way at any point, it follows that one of the principal factors affecting streaming is the elasticity and liquefiability of the ectoplasm.

The streaming in an ameba is coordinated. The direction in which the endoplasm flows in the several pseudopods, when there are no stimuli received externally that produce visible changes in behavior, gives one the impression that there is a “centre” controlling movement. The several pseudopods do not act at all capriciously. The ameba seems to move the pseudopods, not the pseudopods the ameba. If this impression of coordination is correct, it is of the first importance in a study of ameboid movement. Further on, this point will be taken up at length in connection with the character of the path an externally unstimulated ameba describes (p. 109); but there are certain observations which aid in the analysis of the problem of coördination from the point of view of the pseudopod, instead of that of the ameba as a whole, and to these observations we may now direct our attention.

The mass of endoplasm within a pseudopod moves practically always in one direction. In any cross-section of a pseudopod that is more or less cylindrical in shape, the endoplasm in the center moves most rapidly, that near to it less rapidly, while that near the ectoplasm moves very slowly. One never observes a forward stream on one side of the pseudopod and a backward stream on the other. Nor does one observe parallel streams of endoplasm flowing in opposite directions within the same

Figure 1. Illustrating the various directions of endoplasmic streaming in growing and retracting pseudopods. a, two oppositely directed streams in a pseudopod, one directed toward the base and the other toward the tip of the pseudopod, with a neutral zone between. b, two streams flowing toward each other. Cases c to r are self explanatory. s, rotational currents observed occasionally in various species of amebas. t, “fountain currents,” sometimes observed in Amoeba blattae, and rarely in other forms. u and v represent cases of streaming which have not been observed and which probably do not occur. w, similar to v, but with a wide neutral zone between the streams, represents an actual observed case. m and r probably occur only very rarely; no such cases have been seen, but there seems to be no reason why they do not sometimes occur. Excepting m, u, r and v, all these figures were drawn from observed cases of streaming.

ectoplasmic tube, in an ameba of several pseudopods, excepting where there is a wide zone of stationary endoplasm between the streams (Figure 1, v, w). But in “fountain currents,” such as Rhumbler (’98, p. 190) described and figured for Amoeba blattae Bütschli, and which may readily be observed in most species of amebas if immersed in a solution of gelatin thick enough to keep the amebas from sinking, there is a central stream of endoplasm flowing forward, and a peripheral stream of ectoplasm flowing backward, with a thin neutral zone between ([Figure 29], d). As we shall see further on, however, these fountain currents are in principle the same as the currents observed in ordinary locomotion, the apparent difference being due to the fact that there is no locomotion. It is true, then, that within the same pseudopod at any cross section the endoplasm always streams in one direction, and the streaming is unified.

When new pseudopods are formed, or when old ones are retracted, and especially when both these phenomena occur at the same time and close together on a part of an older pseudopod, some of the details of coordination in streaming are readily made out. In [Figure 1] are shown a number of observed cases of pseudopod formation and retraction, with the direction of endoplasmic streams indicated at a given instant. For the purpose of illustration, several (presumably) possible but unobserved cases, m and r, are sketched, and also two cases, u and v, which have not been observed and which probably do not occur. The general conclusion to be drawn from these observations is that, while the endoplasm in the body of an ameba as a whole may be streaming in several different directions at any given instant, that is almost never the case with an individual pseudopod, especially if the pseudopod is of small or medium size and not too flat or otherwise irregular in shape. The pseudopod is therefore the unit of coordinated protoplasmic streaming.

Another general observation which undoubtedly is connected in some way with the problem of coordinated streaming is the following. In externally unstimulated amebas, the new pseudopods are almost without exception directed 60° or less from the direction in which the parent pseudopods are moving.

It is a matter of common observation that an ameba may throw out a pseudopod in any direction whatsoever when stimulated. The ameba may reverse its direction of movement completely, or it may move in scores of different directions at one time for awhile, if properly stimulated. There is no restraint or limit imposed upon the ameba insofar as the direction of movement is concerned. Why then should a great majority of new pseudopods in an unstimulated ameba be projected at an angle of approximately 60° to the parent pseudopod? It might seem at first sight as if the merely physical aspect of the streaming would be a sufficient explanation, in that less resistance would be met with in sending a stream off at a small angle than at a large. But it is probable that inertia plays no part in maintaining the direction of streaming (see p. 123, footnote, for further discussion). It requires perhaps more energy for a pseudopod to flow off from the main stream at an angle of 120° than at an angle of 30°. But it is plain that as many pseudopods are withdrawn as are thrown out, and they are withdrawn at an angle against the main stream of endoplasm in the ameba that is the complement of the angle at which they were projected. Whatever energy might be saved therefore in the projection of a new pseudopod at a small angle with the main stream is lost in withdrawing the pseudopod against the stream at a correspondingly large angle. It is clear therefore that the physics of moving viscous fluids cannot solve the problem. It is probable that the mechanism which controls the direction of locomotion as exemplified in the wavy path of the ameba (see p. 109) is also involved in the direction in which pseudopods are projected.

Some very interesting special cases of endoplasmic streaming are observed during the process of feeding. As is well known, amebas capture their food by the protoplasm flowing around it and engulfing it. If the object is large the protoplasm may flow around it, in contact with it, so that the shape of the object determines the direction in which the enveloping protoplasm flows. If the object is small, particularly if it is a live organism, the behavior of the ameba is quite different (Kepner and Taliaferro, ’13, Schaeffer, ’16). To capture such a food object a cup of protoplasm is gradually formed over it so as to imprison it ([Figure 2]). If the food organism lies against some flat object, the food cup is brought down to the surface of the object all around, thus making escape impossible, before the protoplasm comes into contact with the food organism. Schaeffer (’16, ’18) by experimental methods has shown that the stimulus calling forth the formation of food cups as just described, is the mechanical vibration of the water. At least the same response was produced on the part

Figure 2. Endoplasmic streaming involved in the formation of a typical food cup. a, the ameba is shown moving toward a live food organism that is resting quietly on the bottom. b, the main pseudopod forks, being the first indication that the feeding process has set in. At c the pseudopods have half-way surrounded the prey, but without having come into contact with it. At d the upper sheet of protoplasm, f, (stippled), is flowing dome-like over the prey, while the pseudopods continue to surround it. At e the pseudopods have met and fused with each other and the upper sheet of protoplasm has completely covered the space encircled by the pseudopods, and has fused with the pseudopods. g, sheets of protoplasm which are thrown out along the lower surface under the prey, to form a floor to the food cup. Up to stage e the ameba has not come into physical contact with the prey, but is just about to do so. With the completion of the floor of the food cup, the process of feeding is completed.

of the ameba when the ameba was carefully stimulated by means of very fine clean glass needles. The conclusion is unavoidable therefore that the shape of the food cup and the method of its formation is a racial characteristic and is hereditary. The streaming endoplasm therefore, upon suitable stimulation, takes on a definite form, that of a food cup. This indicates again that the endoplasm is something more than the ordinary fluids of physics, for out of an apparently structureless fluid, organization is effected.

The fact that food cups are formed by amebas implies of course that stimuli are received whose effect cannot be explained as a direct physical reaction. Rhumbler (’10) has attempted to explain the formation of food cups as the direct physical result of the stimulation by the food body; but in recent experiments Schaeffer (’16) has shown that food cups are formed over diffusing solutions of tyrosin, where the solutions were quite as concentrated outside as inside the cup. These results prove convincingly that the shape and size of the food cup are not determined by direct action of the stimulating agent, but by hereditary factors within the protoplasm of the ameba.

Other stimuli also affect streaming characteristically, though not so strikingly perhaps as food stimuli. One of the most widely observed effects on streaming is the momentary pause following stimulation of many sorts. If an ameba that is moving along unstimulated externally, suddenly comes near a food object, it frequently stops forward streaming for about a second, and then begins again, usually at increased speed. The ameba behaves as if it were startled. A similar reaction is observed if a small perpendicular beam of light is flashed near the anterior end of the ameba. Here also streaming is resumed with accelerated speed toward the beam of light. Harrington and Leaming (’00) showed that if strong light, especially at the blue end of the spectrum, is suddenly thrown on the ameba, movement is arrested for a short time. Miss Hyman (’17) has shown recently that if an ameba is strongly stimulated with a glass needle, streaming is arrested momentarily, but the direction of streaming when resumed subsequently, depends partly upon the former direction of streaming and partly upon the location of the stimulus. All of these cases of temporarily arrested movement are strikingly similar to what is observed in the higher animals under similar conditions.

The ingestion of a large food mass produces usually a marked change in streaming. A more or less spherical form is assumed, and if the food mass be a live organism such as a large ciliate, the ameba frequently remains quiet for a considerable interval. If a large amount of food is eaten, as for example a dozen or two colpidia, the ameba may suspend concerted streaming for an hour or more. During this time small pseudopods are projected here and there, but there is no locomotion. But if an ameba eats large masses of carmine, there is usually no pause following ingestion, and the same thing is true when the ameba is induced to eat bits of glass and other indigestible substances. It follows therefore that the interrupted streaming of the endoplasm due to feeding is not caused by the act of ingestion as such, but rather by the onset and continuance of the normal digestive processes on a large scale. These reactions are again strikingly similar to what is observed in many vertebrates, in which a more or less definite body sense, whose sense organs are in the splanchnic region, is supposed to be involved; but what the explanation of similar behavior in ameba is, is not at all clear.

Another factor of great importance in endoplasmic streaming is the nucleus. It was observed by Hofer (’90) that amebas lacking nuclei did not move in a coordinated manner. Štolc (’10) however records a number of observations in which characteristic movement was observed in enucleate amebas ten or more days after the enucleate ameba had been cut off from a normal ameba. Hofer’s amebas died after nine or ten days, while Štolc’s remained alive, some of them for over thirty days. Recently Willis (’16) confirmed Hofer’s findings, but does not discuss Štolc’s results.

The cutting of an ameba into two pieces, one with and the other without a nucleus, is a very simple operation. It is also very easy to observe that within an hour or so the enucleate ameba does not move normally, and that there is no concerted endoplasmic streaming while the nucleate ameba seems to behave normally. But Štolc’s contention that enucleate amebas move characteristically (l. c., p. 159, 160, 167) is not necessarily contradicted by these observations, for Štolc’s observations refer to amebas that lived much longer than the enucleate amebas of Hofer and of Willis. Even if an enucleate ameba is able to recover, after many days, its power of concerted movement, there can be no doubt that enucleate amebas do not move characteristically for a short time after the operation, and that this effect is due to the lack of a nucleus.

Very likely the action of the nucleus on the locomotory processes is neither direct nor specific. The metabolic balance must be disturbed by so fundamental an operation as the removal of the nucleus, and all fundamental activities must in consequence be affected. That food organisms (chilomonas and coleps) may be eaten and digested as Štolc (’10) states, indicates however that the metabolic balance may after a time be regained in some degree, for feeding undoubtedly calls for concerted streaming, and digestion for the formation and transfer of enzymes. Until this point receives further attention therefore, it remains unknown in what way the removal of the nucleus disturbs streaming for some time after the operation; but of the fact that streaming is disorganized for some time, there can be no doubt.

CHAPTER IV
The Transformation of Endoplasm into Ectoplasm

Perhaps none of the factors influencing the streaming of the endoplasm mentioned above exercises as profound and constant an influence as its capacity to form ectoplasm. As has been intimated earlier (p. 3-9) streaming as observed during locomotion is not supposed to be possible at all unless accompanied by the formation of ectoplasm at the forward ends of pseudopods, and its transformation into endoplasm at the posterior end of the ameba. We may therefore next consider the rôle ectoplasm plays in locomotion, and in some other fundamental activities of the ameba.

In the first place it is necessary to define the word ectoplasm, for two entirely different meanings are sometimes given to it. It is used often to designate the clear non-granular layer of protoplasm which thinly covers some of the commoner amebas, and is especially prominent in some of the small species, where the larger part of the anterior end often consists of protoplasm quite free from granules. The other use to which the word is put is to designate the layer of protoplasm on or near the outside of the ameba which is more or less rigid and motionless, resembling the gel state of a colloid. It is the latter meaning that is given the word as used in this discussion, while I shall follow Jennings (’04) and other, earlier, writers in using the word hyaloplasm in speaking of the outer clear layer. It may be necessary to add that neither of these two words is strictly definable, for in some cases, at least, hyaloplasm is not more rigid than the endoplasm, while in other cases it is. Strictness of definition can, of course, come only as investigation proceeds; and these words as well as the word endoplasm, should not be taken as defining the properties of the substances to which they refer, but only as labels.

The demonstration of the most conspicuous and important property of ectoplasm in Amoeba proteus is easily made. With the high power of the microscope one focusses on the upper surface of an active pseudopod, paying especial attention to the small crystals imbedded in the protoplasm. These crystals, although they dance about slightly (Brownian movement) and otherwise change position to a slight extent, nevertheless appear to be held in place by a very viscous medium. Such movement as is observed in these crystals appears more or less erratic; it is not coordinated and it is only by chance in the direction of locomotion of the ameba. While observing the practically stationary crystals of the ectoplasm one can at the same time, though indistinctly, see the forward sweep of the crystals and other granules in the endoplasm below. But observation fails to detect a definite line of separation between the stationary ectoplasm and the mobile endoplasm; the one grades off insensibly into the other.

The formation of ectoplasm in proteus is a much more complicated process than in almost any other ameba, excepting the large species Amoeba carolinensis[1] discovered by Wilson (’00). We shall have occasion however to refer at length to the method of ectoplasm formation in proteus later on, so we may consider proteus first from this point of view, and then take up a few other species in which the process is simpler.

It is a fact more or less familiar to observers of amebas that proteus, as distinguished from the other amebas, has a number of large irregular, roughly longitudinal folds or ridges on its pseudopods and on its main body ([Figure 3]). Under normal conditions these are never absent. They are not found at the free ends of advancing pseudopods, but they take their origin at some little distance from the ends. It is this characteristic of ridge formation that complicates the process of the transformation of endoplasm into ectoplasm; for instead of having to deal with ectoplasm formation at the anterior ends of pseudopods only, we find this process taking place irregularly all over the surface of the ameba.

These folds or ridges were first observed by Leidy (’79) and it is an eloquent tribute to the keen observation of this

Figure 3. Formation of longitudinal ridges and grooves in the ectoplasm of Amoeba proteus. A, B, C, D, showing stages in the development of a single pseudopod. a, b, c, d, d¹, cross sections of pseudopods at the levels indicated. The arrows show the direction of endoplasmic streaming with special reference to the formation of ridges. The numerals 1 to 7 indicate the order in which the ridges were formed. Note the tongues of ectoplasm which extend into the endoplasm, in the cross sections.

sympathetically-minded naturalist, that of the large number of subsequent writers on ameboid movement only one (Penard, ’02, p. 63) seems to have noticed these folds. Leidy says that “ ... the main trunk and larger pseudopods of the same ameba (proteus) assumed more or less the appearance of being longitudinally folded. The endosarc axially flowed as if in the interior of thick walled canals, of which the walls appeared to be composed of finer granular matter with scattered imbedded crystals. In the flow, all the contents did not move with the same rapidity, and usually the smaller particles were swept quickly by the larger ones. Other matter, including some of the largest elements appeared to stick to the inner surface of the extemporaneous tubes, but successively became detached to be carried along with the rest of the contents (p. 46).” “The endosarc appeared to flow within thick walls of ectosarc which often seemed to be longitudinally folded (p. 326).” Penard (’02) confirms Leidy’s observation as to the existence of these folds: “The current (of endoplasm) indeed is not unified, but there exist many currents at the same time because of the fact that the endosarc is divided into a certain number of longitudinal canals or grooves by dense walls, which are of a temporary nature, being broken down and built up from time to time. It is easy to distinguish one canal from the other in this species, the currents being at first more or less parallel, but terminating at the forward end, by their coalescence, as a single mass of liquid (p. 63).” But Penard questions Leidy’s conclusion that the walls are of ectoplasm: “Moreover Leidy deceives himself without any doubt in considering these partitions as folds of the ectosarc. The latter, in the rhizopods, is not a special substance, it is a plasma of surface, specialized for the functions which it has to perform, capable of modification as to its intimate structure, but only so temporarily (p. 63).”

Although it is a very simple matter to prove to one’s satisfaction the mere existence of these folds—a few minutes’ observation under the high power of the microscope will do that—it is a much more difficult matter to observe how these folds originate, because of the incessant changes going on, as recorded by Leidy.

Very young or small pseudopods in proteus have the same general appearance as the pseudopods of other large species (dubia, laureata, discoides, annulata, etc.); that is, there is a central axial stream of endoplasm surrounded by a layer of ectoplasm. But there is one difference even here, and that is the greater thickness of the ectoplasmic walls in proteus in proportion to the diameter of the pseudopod. The ectoplasmic tube however is not solid throughout, but is more or less honeycombed, somewhat like a network, with the spaces filled by endoplasm.

If the ectoplasm is actually endoplasm that has passed into the gel state, then the honeycomb condition just described resembles an intermediate stage where only a part of the endoplasm has been transformed. This network of endoplasm is strong enough however to impede the flow of the main stream of endoplasm along the sides of the pseudopod; but when large objects, such as the nucleus or food masses, too large to be readily carried in the endoplasmic stream, impinge against the imperfectly solidified sides of the tube of ectoplasm, the innermost strands of the spongy network of ectoplasm snap, usually with readiness, allowing the large object to pass by.

The surface of a young pseudopod is smooth, a cross section being oval in shape ([Figure 3], a); but as the pseudopod increases in size, large folds or ridges begin to make their appearance. Usually the first ridges to appear are lateral. They begin as small waves of hyaloplasm which flow out along the sides of the pseudopod for a short distance and then continue to move forward. The endoplasm then flows in a number of small parallel streams amid numerous obstructions through the ectoplasmic tube of the pseudopod into the wave of ectoplasm. After the ridge is well begun, there is frequently observed a slow forward-moving stream of endoplasm within it, but the ridge is never closed from the main endoplasmic stream, as is readily proved by the numerous small streams of endoplasm which continually filter through the ectoplasm into the ridge.

In addition to the lateral ridges, which, as stated, are usually formed first, there appear ridges on the upper side of the pseudopod as well, and presumably also on the under side. So far as could be determined these ridges are all formed in much the same way; that is, by the projection of a small wave of protoplasm from some part of the surface of the pseudopod. The ridges do not always grow by extension at the anterior end as described above. Not infrequently a ridge ten to twenty times as long as wide is pushed out along its whole length at once. This is especially likely to happen in a slender pseudopod that suddenly becomes the main pseudopod. The width of a ridge, especially on the upper surface, does not change much after formation. One can frequently find two or three ridges of about the same width, which run the whole length of the ameba with the exception of a short distance at the anterior end, where, as before stated, there are no ridges.

As the figure indicates, new ridges may be formed from previous ones, either by lateral or endwise extension. In such case the walls of the ridge send out thin waves of hyaloplasm followed by streams of endoplasm, as described above in the formation of the first ridge on a pseudopod. When a pseudopod forms a branch, the ridges on the old pseudopod do not likewise branch, but new ridges are formed which have no connection with old ones, but they may later coalesce with old ridges. Such coalescence is however exceptional. Once a ridge is formed, it retains its identity as a rule; that is, as the ameba moves forward, the ridge in effect moves back over the ameba to lose itself in the wrinkles at the posterior end (See [Figure 11], A). The number of ridges on any random selection of amebas is variable, and is moreover difficult to state. A large ameba may have as many as six or seven side by side on its upper surface. The number on the sides and on the lower surface are difficult to estimate. The space between ridges is about equal to the width of the ridges, but as one passes toward the posterior end, the ridges become more closely crowded together.

From these observations on the formation of ridges it is evident that they do not represent a wrinkling of the surface such as occurs in a semi-rigid curved surface when it is made to occupy a smaller space. The ridges are wrinkles only in appearance, not in origin. The surface of the ridges is younger than the space between them. It appears as if the pseudopod which has to widen as it increases in length, could not liquify the ectoplasm uniformly all around, but only in longitudinal strips here and there, and that through these openings the ectoplasm then flows. There is no question about the greater readiness with which ectoplasm is formed in this ameba as compared with many others, but after a careful comparison of proteus and carolinensis, where ridges are formed, with discoides ([Figure 11], B), dubia ([Figure 11], C), laureata ([Figure 4]) and annulata, where none are formed, the only conclusion presenting itself is that the visible physical properties of the protoplasm of proteus and carolinensis give no hint as to the cause of the presence of ridges in these species. The protoplasm of discoides and laureata is about as viscous as that of proteus, yet in these there is never any ridge formation.

The ridges in proteus recall, of course, the ridges always observed in verrucosa, sphaeronucleosus ([Figure 13]) and their congeners, especially while the latter are in locomotion. A sphaeronucleosus is especially favorable for study in this connection because of its greater activity. This ameba has four or more longitudinal ridges on its upper surface, while in locomotion, which strongly resemble those in proteus and carolinensis. The chief difference lies in the fact that in sphaeronucleosus the ridges are extended at their anterior ends continually, and unless the direction of locomotion is changed, the ridges may retain their identity while the ameba moves several scores of times the length of its body. Along the sides, however, new ridges are continually replacing older ones. When the direction of locomotion is changed, the old ridges usually all disappear into a jumble of ridges and crinkles running in every conceivable direction, and with the reestablishment of locomotion along a more or less straight path, a new set of ridges appears. In sphaeronucleosus and its congeners, the ridges are also not wrinkles, but ridges that are formed later than the surface contiguous to them.

It is interesting to recall also that the ectoplasm in sphaeronucleosus, verrucosa and the rest of this group, is much firmer than in most other amebas.

CHAPTER V
Pseudopods and the Nature of the Ectoplasm

In contrast with the ridge-forming amebas stand those with smooth ectoplasm, such as the common dubia, discoides, villosa, and the rarer laureata and annulata, to mention only a few of the larger forms. In addition to these may be mentioned all the pelomyxas and nearly all the smaller amebas. Much the larger number of species of amebas do not form ridges in the ectoplasm during locomotion.

Figure 4. Amoeba laureata. This ameba is multinucleate, containing a thousand or more nuclei of the shape shown at the right. Ameba 1000 microns long in locomotion. Nuclei 10 microns in diameter.

Of all the amebas with smooth surfaces, the most favorable for observation as to the formation of ectoplasm, is the giant laureata ([Figure 4]), though it is unfortunately of infrequent occurrence. This species is as often found in clavate form as with pseudopods. In cross section it is circular or nearly so. It is often found with zoochlorella growing in it, upon which it seems to depend largely for food, for it seldom has distinctive food masses in it. The nuclei are small and very numerous and the crystals are well formed and numerous, each in a small vacuole, and of a size about two or three times those found in proteus. It will be seen therefore that there are only small bodies in this ameba, none of which (excepting the contractile vacuole) are large enough to change the course of the endoplasmic stream, and streaming is thus reduced to what might be called a typical condition.

In this ameba the endoplasmic stream flows uniformly towards the anterior end where it spreads out slightly so as to preserve the same general diameter of the ameba, for it is a characteristic of this ameba that the anterior end is of about the same diameter as the posterior, when in clavate form. The ectoplasmic tube is built at the anterior end, and remains as constructed until it is drawn in at the posterior end to form endoplasm. It is not all the time undergoing changes such as are observed in proteus. This characteristic is very well shown by focusing with the high power of the microscope on the upper surface of the ameba. The immobility of the ectoplasm is much more readily observed in laureata than in perhaps any other species, a condition that is due chiefly to the large crystals whose displacement is the most convenient criterion of ectoplasmic mobility.

The ectoplasmic tube is not as thick as in proteus, though it appears to be more solid than in that species. It is thrown into folds at the posterior end as it is liquified to form endoplasm, which indicates a firm texture of the ectoplasm. As to the endoplasmic stream, it presents no visible characteristics which set it apart from the fluids of physics; it moves most rapidly in the middle, and gradually less rapidly as the ectoplasm is approached. There is no backward movement of the ectoplasm against the sides of the pseudopod at the anterior end—nothing approaching a “fountain current”—which indicates that the transformation of endoplasm into ectoplasm is rapid and complete. That is, all the endoplasm which reaches the anterior end is turned into ectoplasm. Typically this would result in an ameba of average size, in a layer of ectoplasm of a thickness of about one-seventh of the diameter of the pseudopod (for the area of the cut ectoplasmic tube would equal the area of the endoplasmic stream). But because of friction against the sides of the ectoplasmic tube, there is a layer of endoplasm of appreciable thickness that is practically motionless. This layer of endoplasm therefore makes the diameter of the endoplasmic stream appear smaller than it actually is, and the ectoplasmic tube larger than it is. The actual thickness of the tube of ectoplasm, as distinguished from the flowing endoplasm, is difficult to measure, but it seems to be about one-tenth the diameter of the pseudopod. (Kite (’13) found ameboid ectoplasm to be from eight to twelve microns thick, but he does not state from what part of the ameba nor from what species the ectoplasm was taken.) This would indicate that if the transformation of endoplasm into ectoplasm is as complete as the conditions permit, the thickness of the friction layer would be about one-twenty-third of the diameter of the pseudopod. These observations therefore point to the conclusion that the tendency in laureata is for all the endoplasm to be transformed into ectoplasm at the anterior end, and for the reverse process to occur at the posterior end.

Several of the pelomyxas also move in much the same manner as Amoeba laureata, that is, in clavate form and more or less cylindrical in shape. This is especially the case with Pelomyxa palustris and P. belevskii. But in these species the endoplasm is not completely converted into ectoplasm at the anterior end, as is shown by the fact that there is a slight backward current of endoplasm at the sides near the anterior end (Schultze, ’75). Observation indicates also that the ectoplasmic tube is thinner than would be the case were there complete transformation of endoplasm into ectoplasm at the anterior end. The origin of pseudopods in these pelomyxas is not steady and under control as in laureata, but sudden and eruptive, indicating a less coherent ectoplasm.

The nearest approach to the conditions of streaming as found in Amoeba laureata is found in A. discoides ([Figure 11], B) a species often confounded with proteus. This species is frequently found in clavate form, and the conversion of endoplasm into ectoplasm is complete at the anterior end. In other respects of streaming and pseudopod formation, the two species are also similar.

In another very common species of ameba, Amoeba dubia ([Figure 11], C) the clavate stage of locomotion is comparatively rare, but when it is found it is observed that the transformation of endoplasm into ectoplasm at the anterior end is incomplete, and the endoplasm seems to be of very liquid consistency. This ameba is characterized by the possession, usually, of numerous pseudopods extending from a central mass of protoplasm. In this stage it possesses no main pseudopod as does proteus, discoides, laureata and other species, but there are three or four pseudopods extending actively in the general direction of locomotion. The physical characteristics of these pseudopods, in so far as streaming is affected, are different from those of the clavate amebas. The ectoplasmic tubes are relatively thicker, the endoplasm is less fluid, and new pseudopods are not formed so readily. It appears therefore that an increase of surface in the ameba serves to increase the amount of ectoplasm that is formed during locomotion.

Figure 5. Amoeba limicola, after Penard. Figures a, b, e, illustrate the “eruptive pseudopods” by means of which this ameba moves. f, a variety or separate species whose ectoplasm is somewhat firmer, and whose posterior end possesses a conspicuous uroid. c, the nucleus found in a, b, e. d, the nucleus found in f.

There is another group of amebas in which the endoplasm is much more fluid than in dubia. To this group belong Amoeba limicola ([Figure 5]) and Pelomyxa schiedti ([Figure 6]). The latter never forms pseudopods, and the former does so very seldom. A. limicola is extremely fluid, and in locomotion the flow of the endoplasm can hardly be called streaming, for it rushes about in the body as if it were only partially under control. The ectoplasm does not give way steadily at the anterior end during locomotion, allowing a steady forward flow of the endoplasm, but it breaks away suddenly here or there, allowing the endoplasm to rush through as if it were under considerable pressure. When the endoplasm rushes through these breaches in the ectoplasm, it is usually deflected back along the side of the ameba for a considerable distance, thus leaving a part of the old ectoplasmic wall stand for a few seconds between the reflected wave of ectoplasm and the main body of the ameba. It is then that one can observe especially well the very thin ectoplasm covering the ameba, the thickness of which is about one-fortieth the diameter of the ameba. This ameba is somewhat dorso-ventrally flattened and generally oblong in shape during locomotion.

Figure 6. Pelomyxa schiedti, after Schaeffer. b, bacterial rods characteristic of the genus Pelomyxa. c, v, contractile vacuole. g, glycogen bodies. n, nucleus. u, uroidal projections. At the left is shown a series of outlines of the animal during locomotion. Length, about 75 microns.

Pelomyxa schiedti moves in much the same way that Amoeba limicola does; that is, by eruptive waves of endoplasm which are usually deflected back along the side ([Figure 6], at the left). The endoplasm is likewise of very thin consistency. The thinness of the ectoplasm and the ease with which it may be ruptured, is very well shown by the fact that the large irregular glycogen bodies (Štolc, ’00) which fill it to capacity, lie so close to the surface that it is frequently impossible to see any protoplasm between them and the exterior. The contractile vacuoles which are numerous, also testify in their characteristics, to the ease with which the ectoplasm may be broken. The vacuoles never reach but a very small size (four microns in diameter) presumably because of the thin consistency of the endoplasm and because they can readily break through the ectoplasm. They burst on the surface of the ameba instantaneously, as a small air bubble might burst on pure water. But this ameba differs from limicola in that a cross section of the body is very nearly a circle.

Figure 7. Amoeba radiosa, after Penard. a, the rayed stage. b, the rayed stage in which some of the pseudopods are being withdrawn. One of them is thrown into a spiral as it is being withdrawn. c, the stage preceding the trophic stage shown at d.

Another very interesting feature of Pelomyxa schiedti is the uroid ([Figure 6], u), which in this species consists of a number of very thin projections resembling pseudopods extending from the posterior end. These projections are attached to the substratum and in some way aid in locomotion. These uroidal projections are of considerable length, and may persist for a considerable length of time. Thus while schiedti is unable to form pseudopods at its anterior end, it forms uroidal projections with great ease at its posterior end. But what the conditions are which are necessary for the formation of a uroid, a structure which it may be added, exists in many species of amebas (and perhaps also in Cercomonas), is quite unknown.

In contrast to the amebas thus far discussed from the point of view of the transformation of endoplasm into ectoplasm, there are a number of species in which two distinct methods of endoplasmic transformation occur typically. Among these species are the small Amoeba radiosa ([Figure 7]), A. bigemma ([Figure 8]) and a new species which for convenience will be referred to as bilzi.

Figure 8. Amoeba bigemma, after Schaeffer. a, usual form in locomotion, showing the numerous pseudopods, vacuoles, nucleus and food body. b, rayed stage frequently assumed when suspended in the water. The pseudopods in this stage are clear, slender, and more rigid than those in stage a. c, an excretion sphere attached to a twin-crystal characteristic of this ameba. d, the nucleus, consisting of a clear nuclear membrane and a mass of chromatin granules in the center. e, a small sphere attached to a crystal. f, a twin crystal unattached to a sphere. Length of a, 150 microns; of d, 12 microns; of f, 2 microns.

It is well known that radiosa has two stages: a more or less clavate shaped stage in which the ameba creeps along the surface of some object ([Figure 7], d); and a stage in which a number (eight or less) of long and very slender tapering pseudopods are formed which usually persist for a long time ([Figure 7], a, b). These pseudopods are frequently quite straight and regularly disposed around the central mass of protoplasm (Penard, ’02, pp. 87, 89). In no case are any endoplasmic granules found in these slender pseudopods; they consist entirely of hyaloplasm. In retracting these pseudopods a curious phenomenon is sometimes observed; the pseudopod is rolled up into several (as many as six) turns of an almost perfect helical spiral of a diameter six to eight times that of the pseudopod. But as the process of withdrawal proceeds, the spiral becomes irregular, but parts of some of the turns persist in the last vestiges preceding complete withdrawal ([Figure 7], b). These spirals are also observed in other species besides radiosa (see p. 128 seq.)

Another species of ameba in which a trophic as well as a rayed stage is found, is the recently described species bigemma. In this species the rayed stage is only of occasional occurrence ([Figure 8], b). The larger the ameba is, the rarer is the rayed stage assumed. On very rare occasions one finds a rayed stage in which the pseudopods are long, straight, slender and tapering, and more or less regularly disposed around the central mass of protoplasm. The trophic stage ([Figure 8], a) is much the more common. In this condition pseudopods are formed in large number. They are small, conical or linear, and blunt, and they do not determine the direction of locomotion, as they do in proteus, dubia, or laureata. These pseudopods are often composed only of hyaloplasm, though frequently the basal parts of them consist of endoplasm. When these amebas become suspended in the water, they frequently assume a shape that approaches the rayed condition: six or more long conical pseudopods are run out from the central mass of protoplasm, but the pseudopods are not straight in this case, but irregularly curved and capable of being waved about to a slight extent. The ameba readily passes from this stage to the trophic.

The species Amoeba bilzi ([Figure 9]) has come under my observation on several occasions, and its pseudopodial characters are of considerable interest in this connection. In its usual form this ameba has the general appearance of a sphaeronucleosus.

Figure 9. Amoeba bilzi. a, the ameba in locomotion, showing the ectoplasmic ridges, nucleus, contractile vacuole. b, the transition stage between the rayed stage (which resembles that of radiosa, [Figure 5], p. 30, somewhat) and the stage shown at a. The whole of the ameba flows into the broad pseudopod with the arrow. Length of a, 90 microns.

In size it is about midway between the latter species and striata. It always has a number of prominent longitudinal ridges on its upper surface. Its mode of streaming is essentially like that of striata or sphaeronucleosus. When this ameba is disturbed and left suspended in the water, it throws out four or five or more long slender pseudopods composed entirely of hyaloplasm, excepting a bulbous base which consists of granular endoplasm. The pseudopods are cylindrical with tapering ends. They are very rigid, and once formed, persist for a considerable length of time. When these pseudopods are about to be retracted, the wall weakens at some point and then crinkles while the distal part of the pseudopod bends, often at a decided angle. The crinkling of the wall continues up and down the pseudopod while it is slowly being withdrawn. These pseudopods, as well as those of the rayed state in radiosa and bigemma, are not pseudopods of locomotion but of position; they are not dynamic but static structures. But there are no hard and fast distinctions to be made between these two types of pseudopods, for at least in bigemma and bilzi, there are transitional forms of pseudopods ([Figure 8], b).

The formation of pseudopods and their character depends to some extent upon the firmness and thickness of the ectoplasmic layer; and the character of the ectoplasm in turn depends largely upon the consistency of the protoplasm as a whole. In the following representative list of amebas: limicola, villosa, dubia, proteus, discoides, laureata, bigemma, bilzi, radiosa, sphaeronucleosus, verrucosa, the given order indicates a progressively thicker and firmer ectoplasm as one passes from limicola to verrucosa. But from limicola to bilzi the number of pseudopods directing locomotion increases from one to an average of about twelve in dubia, and then falls gradually to one in bilzi and the others beyond in the list. (See [Figure 10.]) Where the directive pseudopods begin to disappear, the transitional appear, viz., in bigemma and bilzi; but beyond these no transitional pseudopods occur. But along with the transitional there begin to appear also the static pseudopods, which are seen relatively seldom in bigemma and bilzi while in radiosa they occur at almost all times. In sphaeronucleosus and verrucosa no distinctive pseudopods of any kind occur.

If all the known species of amebas in which the necessary characteristics have been recorded, were arranged similarly with respect to the firmness and the thickness of the ectoplasm, the general relations of the various kinds of pseudopods in the list would be approximately the same as in the list given above; but there would appear an exception here and there, indicating the operation of special factors. Such an exception, for example, is seen in proteus in the list of species given, which because of the ridges that it forms (Figure 3) has a smaller number of pseudopods than would be the case if no ridges were formed[2]. It may be concluded, then, that the number and character of pseudopods depends in large part upon the ectoplasm-forming capacity of the ameba; and that this property is intimately associated with the degree of fluidity of the whole mass of protoplasm in the ameba.

Figure 10. Graph representing the relation of firmness and thickness of the ectoplasm with the number and character of the pseudopods in different species of amebas. a, the average maximum number of pseudopods directing locomotion in the different species of amebas. b, the number of transitional pseudopods. c, the number of static pseudopods. d, the estimated degree of firmness and thickness of the ectoplasm of the various species of amebas, grading that of limicola as 1 and that of verrucosa as 6.

That the number and character of pseudopods formed depends in large part upon the firmness and thickness of the ectoplasm was said advisedly. For observations indicate that there are other factors which influence the character of pseudopods besides those which also control the formation of ectoplasm. These other factors indicate their presence readily in the details of structure of the pseudopods. Thus the number of directive, transitional or static pseudopods may be the same in two particular species, yet in their intimate structure and appearance they are always found to differ. In bigemma, bilzi and radiosa, for example, the number of static pseudopods when formed is about the same in the three species, but the similarity ends there. For these species differ in the frequency with which pseudopods are formed, in their persistence when once formed, in the ratio of length to average diameter, in the general shape, in the frequency with which straight pseudopods are formed, in the speed of their formation and withdrawal, in the manner of their withdrawal, in their disposition with respect to geometrical pattern, in the character of the bases of the pseudopods, in the form of the free ends, and so on. Many of these characteristics are still further analyzable into numerous other and more detailed characters. And what is true of the static pseudopods is likewise true of the transitional and the directive. Pseudopod formation is however only a small part of the activity of an ameba. The formation of uroidal projections, of vacuoles of various sorts, of crystals, and so on, are some other general activities that are fully as subject to specific variation as pseudopod formation. Again in behavior to food and various other stimuli, in resistance to various factors in the environment, in reproductive processes, and so forth, there is found similar specific peculiarity. In fact, one looks in vain for similarity between any two species of amebas except in their most generalized characters. From my own experience in extended observation of several dozen species, which included a large number of characters, as pointed out above, I have not found two species of which I can confidently assert that any particular character defined as accurately as possible was present in both. In different words, my experience indicates that no two species are alike in any respect whatsoever. Each species appears unique from every point of view and in the smallest definable detail. The concept of specificity therefore is much more fundamental in amebas than has been believed to be the case hitherto (cf. Calkins, ’12). The intimate structure of amebas is indeed similar to that of higher animals where the precipitin reactions (Richet, ’02, ’12; Reichert and Brown, ’09; Dale, ’12; Nuttal, ’04; also Todd, ’14) have indicated that the various albumins are of specific structure and reaction.

As an example of these specific differences, reference may be made to the three species, protus, dubia and discoides, which have been referred to in the past, almost without exception, by the most experienced teachers of biology, as being one species: proteus. Some investigators of ameboid phenomena have likewise confused these different amebas. Below is given a list of some of the most striking characteristics of these three amebas. This list is of course very sketchy. If the nuclear division phenomena, for example, were well known, which they are not, those character differences alone would doubtless make a list several times as long as this one. Compare with [Figure 11.]

Figure 11. A, Amoeba proteus in locomotion. Note especially the longitudinal ridges. a¹, equatorial view of the discoid nucleus. a², a polar view of the nucleus. a³, equatorial view of a folded or crushed nucleus frequently found in large individuals. a⁴, shape of crystals found in this species.B, Amoeba discoides in locomotion. b¹, b², equatorial and polar views of the discoid nucleus. b³, shape of the crystals found in the ameba. C, Amoeba dubia in locomotion. c¹ and c², equatorial and polar views of the ovoid nucleus. c³-c¹⁰, shapes of crystals found in dubia. In these drawings only such characters as are of special interest for the purpose of this work are emphasized. Dimensions in microns: A, 600; B, 450; C, 400; a¹, 46 × 12; b¹, 40 × 18; c¹, 40 × 32; a⁴, maximum, 4.5; b³, maximum, 2.5; c³-c¹⁰, maxima, 10 to 30.

This fundamental uniqueness of all the characters of the various species of amebas naturally gives rise to the question as to what is the cause of this condition of affairs. Why and how

are the different species of amebas so absolutely different, even to the smallest detail? Why are the apparent resemblances and similarities of their more generalized kinetic characters, such as the formation of pseudopods, of ectoplasm, of crystals, of contractile vacuoles, the general character of endoplasmic streaming, the formation of ectoplasmic ridges, and so forth, found, upon analysis, to resolve themselves into a large number of details which differ more strikingly, the corresponding characters of one from those of the other, than do the generalized characters of which they are composed?

These questions apply, of course, to all other organisms as well as to amebas. Unfortunately, however, these questions are at present unanswerable for all organisms. But for the amebas, at least, the problem of form can be rid of some irrelevant matter which, in numerous instances in the past, has been assumed to be properly included.

In the first place, changing a single character of the protoplasm, such as the degree of viscosity, cannot explain the observed diversity of detail; neither can a variation of a number of the physical characters of fluids produce such differences as are observed in the dynamics of the different species of amebas. Our whole experience with the fluids of physics speaks against such an explanation. But, on the other hand, the invisible details of structure of a fluid may become strikingly manifest under certain conditions, namely, those surrounding the process of crystallization. A slight change in the physical condition may produce a considerable variety of crystal shapes, but this variety of shape has nevertheless very definite limits which cannot be overstepped.

Amebas like crystals are also most rigidly and definitely restricted to a certain range of shape, which must be a direct result of the structure of the protoplasm composing them. Amebas in fact are not any more “shapeless” than crystals are; and it would be quite as exact to say that the crystals of water are shapeless since a great variety of shapes are met with in snow, hoar-frost, etc. The fact that corresponding parts of two species of amebas resemble each other less and less closely as they are analyzed into smaller and smaller details, is in itself conclusive evidence that the protoplasms of the amebas are chemically different; the resemblance between the gross anatomy and physiology between two different species is due to the greater conspicuousness of such characters as are the result of the action of physical processes. That is to say, chemically or molecularly different masses of matter may resemble each other in their molar aspects.

It is to be noted however that the more intimate structure of streaming protoplasm cannot always express itself externally as it can in ameba. As was suggested in the introduction, there is no good reason for supposing that the causes of streaming in the various organisms in which it is observed are fundamentally different. The problem of ameboid movement cannot be considered apart from the streaming of protoplasm in foraminifera, myxomycetes, plant cells, lymphocytes, desmids, diatoms and ciliates. The streaming of endoplasm in some cells, such as in ciliates and plant cells, does not give rise to change of shape of the cell as it does in ameba. In these cases the character of streaming is highly restricted; the unyielding ectoplasm or cell wall as the case may be, prevents any but the most essential features of streaming from occurring. Recalling the analogy of crystallization, streaming in a plant cell or in a ciliate is analogous to crystallization occurring in a tube or vessel too small for the crystals to form properly.

This discussion anent the fundamental chemical uniqueness of each species of ameba is of course not complete without an examination of the views expressed to the contrary. And it is to this side of the discussion that we may now briefly direct our attention.

CHAPTER VI
The Species Question

After the discovery of the ameba by Rösel v. Rosenhof and the introduction of the Linnean system of nomenclature, the number of new species of amebas that were reported increased rapidly. But in 1856 Carter suggested that what had been described as A. radiosa probably was a young stage of A. proteus. With the general acceptance of the Darwinian Natural Selection Hypothesis, the ameba came to be looked upon as standing at the bottom of the scale of organisms, and consequently was supposed to lack generally such characters as the higher forms possessed. The ameba became the representative of the “primordial slime” from which by slow stages the other organisms were evolved. So of the sixty odd species which had been described up to Leidy’s (’79) time, Leidy, following the suggestion of Carter, was inclined to think that the great majority of these represented only changes of shape of about four species (not including the several species that were then known to be parasitic). Since Leidy’s time the prevailing tendency has been to regard most of the “new” species as mere environmental or life cycle stages of a very few species. A very noted exception to this tendency, however, has been Penard’s (’02) great work on the amebas and other rhizopods of the Leman Basin, in which he describes forty-five species of amebas (including Gloidium, Protamoeba, Amoeba, Dinamoeba, Pelomyxa), paying attention mainly to the readily observed ectoplasmic and endoplasmic characters, and the appearance of the resting nucleus.

The remarkable discoveries of Vahlkampf (’05) of the nuclear changes during the division process turned the attention of numerous investigators to this field, and the ectoplasmic and endoplasmic characters thenceforward received scant attention. Thus Calkins (’04) came to suggest as Carter had done many years before, that A. radiosa was merely a young form of A. proteus. And Doflein (’07) intimated that the protoplasmic characters of vespertilio cannot be distinguished from those of verrucosa, radiosa, polypodia, limax and guttula. Schepotieff (’10) in a similar vein, writes: “Wir werden demnach so bekannte und so lange Zeit als selbstständige und typische Amöbenarten aufgefasste Formen wie A. limax, A. polypodia, und A. radiosa nur als Umwandlungsstadien andrer Arten bezeichnen dürfen.” Gläser (’12) remarks: “The most reliable criterion for the classification of the amebas is the division of the nucleus.” Calkins (’12) takes the same view on this point and states that in his opinion the ectoplasmic and endoplasmic characters of amebas conform to four “types,” viz., proteus, verrucosa, vespertilio and limax. The enormous amount of work that has been done on the nuclear division changes as compared with the small amount of work on the cytoplasmic structure has thus naturally tended to an over-estimation of the significance of the nuclear changes.

There are objections to making the nuclear changes the basis of the classification of the amebas.

1. In the first place, to classify the amebas means not only labeling the different species accurately, but also to assign to them their proper place in the system of organisms. All organisms are classified with this purpose in view. This is what is meant by a natural system of classification as contrasted with an artificial system based on only a part, arbitrarily selected, of each of the organisms concerned. In the past all artificial systems have been discarded. It is perhaps unnecessary to say that a classification based on nuclear characters would be a highly artificial system. For in no group of organisms has it been found possible thus far to use the nuclear changes as a basis of classification. The great amount of labor that has been expended by cytologists within recent years on the behavior of chromosomes, and the immense amount of work done by the students of genetics, has failed to show any specific relation whatever between the external characters of organisms and the nuclear behavior.[3] In other words, the peculiarities of mitotic processes have not been found to be correlated with characters in the somatoplasm. It is to be remembered however that all living organisms, with the exception of some of the bacteria, are classified with respect to their external characters, and that in almost all organisms the number of visible and demonstrable specific characters becomes rapidly greater as ontogenetic development proceeds.

2. There is considerable disagreement among the investigators of the nuclear phenomena of amebas as to the actual events occurring during the division process. Cf. Dobell (’14) and Hartmann (’14) in re Amoeba lacertae; Nägler (’09), Gläser (’12) and Wilson (’16) on the presence of a centriole in amebas; etc. The work of Schardinger (’99), Wherry (’13) and Wilson (’16) on the nuclear stages of amebas was done with care, yet Wilson (’16) still remains in doubt as to whether or not these investigators all worked on the same species.

3. Awerinzew (’04, ’06) found that the nuclear changes in Amoeba proteus are similar to those in the heliozoan Actinosphaerium; there being thus greater correspondence in the nuclear changes between species belonging to different orders than there is between species in the same genus. Logically therefore Actinosphaerium would have to be placed in the same genus with Amoeba proteus.

4. There is the great practical objection that in many of the larger species it is extremely difficult to find suitable division stages even though thousands of individuals are at hand, and the search is continued for days and weeks by an experienced investigator (Dobell, ’10). Experimental work, which is usually done with one of the larger species, would thus be greatly handicapped because of the great difficulty in determining the nature of the organism employed.

From these considerations it appears that the attempt to classify the amebas on the basis of the nuclear changes is highly artificial and exceptional, and if we may judge from past attempts to classify organisms on the basis of a single character, is foredoomed to failure. This conclusion does not apply, however, to very minute amebas in which no specific cytoplasmic characters have yet been established, chiefly because of their very minuteness; such amebas could be given specific names for reference but they could not be classified in a natural system excepting perhaps as a group.

But the definiteness and the consistency with which the nuclear division stages occur in any given species of ameba, lends support to the probability that in these animals the relation existing between the chromatin and the cytoplasm are similar to those observed in higher animals; and that the laws governing the transmission of cytoplasmic characters in amebas are quite as inflexible as those governing somatoplasmic characters in the higher organisms. Among the investigators of cytologic and genetic phenomena (among the multicellulars) the belief is practically unanimous that the elaborate mechanism involved in nuclear division is primarily a design for distributing the factors concerned in heredity. Now it would be very strange indeed if a similar and quite as complicated a mechanism in ameba had no function to perform. For what would be the purpose of the complicated nuclear changes in ameba if not concerned with heredity? As has already been seen, however, there are numerous cytoplasmic characters, in the larger amebas at least, that are inherited from one generation to the next with as little variation as is observed in other organisms (Schaeffer, ’16). The recent work of Jennings (’16) on Difflugia and Hegner (’18) on Arcella also indicates that the general processes of inheritance in these organisms which are closely related to amebas, are similar to those observed in higher forms. The conclusion seems justified therefore that the nuclear changes in amebas mean essentially the same thing as in other organisms.

We are now therefore in a position to say that amebas are definitely and thoroughly organized; that they are not really “shapeless”; that they are not more subject to variation than a higher organism is; and that each species differs from all others in probably every visible detail. The large variety of pseudopods observed in different species are seen not to be the result of physical or extrinsic chemical forces acting upon ectoplasms differing in some mere physical character as viscosity. But all these peculiarities are hereditary, and are due to a fundamental chemical structure of the protoplasm which is specific for the species. The highly characteristic nature of the pseudopods formed by the amebas of any species, it is seen, is to be referred to the fundamental structure of the protoplasm, probably its stereochemical structure. And what is of especial importance for this discussion, the character of streaming concerned with pseudopod formation and with movement in general, which is specific for each species, is likewise found, to some extent at least, to be conditioned by the specific structure of the protoplasm.

That the specific character of the pseudopods, and the streaming which of course lies back of it, is not wholly or perhaps even largely, due to the specific structure of the protoplasm, is evident from a consideration of streaming in some other organisms, without a study of which, streaming in amebas can be only imperfectly understood.

The formation of pseudopods is not necessary to streaming. Occasionally one sees internal currents unaccompanied by movement or ectoplasm formation in amebas approximating spherical shape, such as in Amoeba blattae (Rhumbler, ’98) and rarely also in proteus or dubia. But especially well is such streaming seen in a contracted Biomyxa, a naked foraminifer, and in numerous plant cells. In paramecium and other ciliates the continuous circulation of the endoplasm,—a true streaming process,—is an involuntary act. But in Frontonia, another large ciliate, the circulation of the endoplasm is under the control of the animal, that is to say, voluntary, and is set in motion only when feeding, the direction of streaming being away from the mouth so as to drag in the food (see [Figure 32], p. 99). If the food particle is a long filament of Oscillatoria, for example, the endoplasm circulates very much as it does in paramecium, only more rapidly, until the whole filament is wound up into a coil. Then streaming stops. In the second place streaming is not necessarily accompanied by the formation of ectoplasm as observed in ameba. In plant cells the ectoplasm is practically stationary, while the endoplasm is in continual flux. The transformation of endoplasm into ectoplasm and vice versa is therefore not an essential feature of streaming, though it is of locomotion; that is, ectoplasm is always found between endoplasm and water, though it might be possible under certain conditions for endoplasm to come into contact with water without stiffening. And if so, there appears to be no reason why locomotion might not occur. It appears however under normal conditions that a moderate tendency to ectoplasm formation (proteus, dubia) leads to greater efficiency in movement than a very weak (limicola) or a very strong (verrucosa) tendency to form ectoplasm.

In the reticulose rhizopods, as is well known there is no ectoplasm of the kind observed in amebas. The middle of the pseudopod, moreover, is not the region of most rapid streaming as in ameba, but frequently becomes congealed, on the contrary, into a rod-like structure. In general this axial rod has the character of very stiff ectoplasm. The character of streaming in reticulose rhizopods, however, has received very little attention, and detailed comparisons are therefore impossible.

Another interesting property of reticulose pseudopods, which are formed by a streaming process, is their great power in some species, of rapid contraction. If a diatom for example, in its movements breaks loose a pseudopod it is often (though not necessarily) contracted very rapidly, much more rapidly than could be the case if it were accomplished by streaming. It frequently happens that knobs are found on a slender pseudopod. These knobs may move back and forth with great rapidity without visibly affecting the pseudopod ([Figure 12]). The process reminds one of a block sliding on a rope. These observations indicate a very high degree of elasticity in the formed pseudopods of such a rhizopod as Biomyxa as compared with a very low degree of elasticity in the amebas.

It thus appears that the process of streaming is a much more fundamental phenomenon than most of the theories accounting for ameboid movement would lead one to suppose; for these theories concern themselves only with streaming as observed in amebas, and many content themselves with only two or three species. Since the general features of streaming are similar no matter where streaming occurs, no theory is likely to gain acceptance that explains streaming only in one group of organisms. Streaming in rhizopods, myxomycetes, ciliates, plant cells, is most rationally looked upon as caused by the same fundamental process; but the detailed form it takes, especially in freely formed pseudopods, is undoubtedly conditioned by the structure of the protoplasm, both physically and chemically, but more especially the latter.

Figure 12. Illustrating the high degree of elasticity in the pseudopods of Biomyxa vagans. In a and b are shown two stages of a small section of the pseudopodial network, which remained unchanged while a small lump of protoplasm (near the arrow) moved rapidly up and down the slender pseudopod. Movement along the whole length of the pseudopod occupied about half a second. Just exactly what the movement was due to could not be determined, but the distance between the forks in the pseudopod did not change, nor did the thickness of the protoplasmic strand on which the protoplasmic lump moved change noticeably.

CHAPTER VII
Experiments on the Surface Layer of the Ameba

In the preceding chapters we have discussed the streaming of the endoplasm in various representative species of ameba, and its transformation into ectoplasm at the anterior end. We have observed that the details of streaming are not quite the same for any two species of ameba, and that in consequence the character of locomotion also is specific for every ameba. All the observations prove that movement in ameba is always associated with streaming, and streaming (in locomotion) with ectoplasm formation. It follows therefore that the form of movement observed in amebas depends invariably upon the streaming of the endoplasm accompanied by the formation of ectoplasm.

There is however another element which, although it appears to be a consequence of ectoplasm formation, must nevertheless be included in any account of ameboid movement because of the light it is bound to shed on the physical processes concerned in streaming. This element is the thin outer layer which separates the water in which the ameba lives from the ectoplasm. It is the properties of this layer to which we may now direct our attention.

That such a layer exists was indicated by observations of Bütschli (’92) and Blochmann (’94), as already mentioned; but neither of these authors stated definitely whether they considered a third layer actually to exist or whether the ectoplasm as such moved forward. Jennings (’04), as has been pointed out, concluded that no third layer exists and that the particles clinging to the outsides of amebas, which are carried toward the anterior end, are carried by the ectoplasm. Gruber (’12) concluded however that an outer layer exists, composed of gelatinous substance, which moves ahead at about the same rate as the ectoplasm (p. 373). According to Gruber’s view the outer layer is a permanently differentiated layer of material. Schaeffer (’17), on the contrary calls it a layer of protoplasm, which moves forward faster than the forward advance of the ameba.

It is a very simple matter to demonstrate the existence of this layer. Although any insoluble non-toxic substance of low specific gravity such as carmine or soot, when reduced to very small particles and mixed with the water in which the amebas to be examined live, will cling to the outside of the ameba so that the movement of the outer layer can be observed; in my experience the best as well as the most convenient substance to use is the dried flocculent colloidal sediment from ameba cultures, rubbed to powder with the ball of the finger. This powder swells up in water into flocculent masses which are large for their weight and do not show such active Brownian movement as particles of carmine or india ink, and they consequently adhere more easily to the ameba. Moreover no foreign substances are thereby introduced into the water.

Figure 13. Amoeba sphaeronucleosus. In locomotion. Note the nucleus, contractile vacuole, ectoplasmic ridges. This ameba is not known to form pseudopods. Length, 120 microns.

Of the more common species of amebas, those with the firmer ectoplasms are the most favorable for studying the movements of the outer layer. We may therefore first take up several observations on Amoeba sphaeronucleosus ([Figure 13]). This ameba resembles the more common A. verrucosa. It is about 120 microns long and is usually of an oval shape in locomotion. It is more active and less disturbed by jars than verrucosa.

[Figure 14] represents a sphaeronucleosus with a small particle attached to the middle of the upper surface of the ameba. As the ameba moves forward, shown by successive outlines, the particle likewise moves forward, but, as will be observed, at a more rapid rate. Measuring the distance from particle outline 1 to 4, and from ameba outline 1 to 4, it is seen that the rate of movement of the particle compares with the rate of movement of the ameba as 2.48 to 1.

Figure 14. Illustrating the movement of a particle on the upper surface layer of an Amoeba sphaeronucleosus. Length of the ameba, 120 microns.

Figure 15. An Amoeba sphaeronucleosus with two particles attached to its upper surface film, one in the middle and one at the side. a moved 2.6 times as fast as the ameba while b, lying nearer the side, moved only 1.9 times as fast as the ameba. Length, 100 microns.

Particles lying near the side do not move forward as rapidly as those lying in the middle. [Figure 15] shows two particles, one of which, a, lying near the middle of the ameba, moved 2.6 times as fast as the ameba advanced in the region of the particle; while particle b moved only 1.9 as fast as the ameba in front of the particle. The speed ratio of particle a to particle b was as 1.26 to 1.

Figure 16. Illustrating more rapid movement of the surface film in the middle of Amoeba sphaeronucleosus than near the edge. The vertical lines connecting the particle with the ameba outlines were drawn only for convenience of reference. Length of ameba, 120 microns.

[Figure 16] shows a particle lying still more to the side than in the preceding figure. In the first six stages the particle moved 1.85 times as fast as the ameba. The particle then came to the edge. From stage 7 to 10 the particle moved more slowly than the ameba. At stage 11 the particle had come to lie in the posterior half of the ameba, where the tendency of the surface layer is to travel toward the middle of the upper surface. In stage 12 the particle had gotten away from the edge of the ameba and already shows a gain in speed. From stage 13 to 16 the particle moved again about 1.83 times as fast as the ameba. But at stage 16 the edge was reached with a consequent decrease in speed of the particle.

The direction of the path described by a particle carried on the back of an ameba depends upon what part of the ameba is most rapidly forming ectoplasm. That is, the particle tends to

Figure 17. Illustrating the different speeds with which particles move when attached to the surface film of an Amoeba sphaeronucleosus, depending upon their location. Particle a moved 3.5 times as fast as the ameba and b 2.7 times as fast. Length of ameba, 110 microns.

move toward that part of the anterior edge that is advancing most rapidly. Figures 17 and 18 illustrate this point. [Figure 17] shows an ameba with two particles on its back, and with an unequally advancing anterior edge. Particle a moved more rapidly than b because: (1) it was moving away from a more rapidly receding posterior region; (2) the right anterior edge was advancing more rapidly than the left anterior edge; (3) the particle was nearer the anterior edge. The rapidly advancing right edge in stage 4 accounts for the veering of the particle a to the right. The more rapid advance of b from stage 3 to 5 is due to the remoteness of the anterior right edge, which, because of its nearness to particle a pulls on it to a much greater extent than on particle b. That is to say, when a particle lies somewhere between two rapidly growing regions on the anterior edge, leading in different directions, that particle is attracted to the edge less rapidly than a particle lying immediately back of either advancing region. As may readily be observed each change in speed or direction of movement of the particle b finds its explanation in the amount and location of ectoplasm formation at the time. Large particles like a do not so readily reflect changes in the direction of pull of the surface layer.

The rapid rate of movement of particle a—3.5 times as fast as the ameba—finds its explanation in an actively advancing anterior edge that was unusually wide. Particle b moved at a slower rate, 2.7 to 1. It started from near the posterior edge where it moved comparatively slowly for a short distance.

Figure 18. Illustrating the effect on the path of a particle attached to the surface film of an Amoeba sphaeronucleosus when the ameba changes its direction of movement. From stages 3 to 5 the ameba veered to the right, also the particle. From stages 6 to 9 the ameba turned sharply to the left, and this change of direction was reflected in the movement of the particle. Length of the ameba, about 120 microns.

[Figure 18] shows more pronounced changes in the direction taken by a particle attached to the back of an ameba. The change in direction at stage 6 was caused by a wave of ectoplasm thrown out at the left side, and cessation of movement at the anterior edge. At 7 a small wave was thrown out at the anterior edge and a large wave on the left. At stages 8 and 9 the direction of the particle was again a response to the waves of ectoplasm thrown out at the left anterior edge, which thus became the anterior end.

Figure 19. Illustrating the rapid movement of the upper surface of an Amoeba sphaeronucleosus under the most favorable conditions. The particle moved 3.56 times as fast as the ameba. Length of the ameba, 130 microns.

The movement of particles on the under side of an Amoeba sphaeronucleosus depends upon what part of the ameba is attached to the substratum. Where the ameba is attached there is of course no movement of the surface layer and the particles remain stationary. In an ameba attached as shown in figure 20, a, there was a very slow movement of particles forward near the middle of the attached region (x), but whether this was related to the movement of the outer layer of the upper surface was not determined. The movement of these particles was considerably slower than the movement of the ameba. In another ameba attached at the anterior and posterior ends ([Figure 20], b) no movement of particles on the under side could be discerned. The small particles showing Brownian movement, with the surrounding water, are dragged along as a mass. This movement is purely mechanical, and is what would be expected on purely physical grounds, when a more or less cup-shaped object is moved along in water in close contact with a flat surface. Such particles as have become attached to the surface layer on the under side of the ameba, because of their slower movement than that of the ameba, eventually bring up at the sides near the posterior end, as the ameba moves along. From here they are carried forward in the manner already described. Thus there comes about a “rotation” of particles adhering to an ameba as described by Jennings (’04) and Dellinger (’06), though the explanation is different from that given by Jennings (l. c.) as we shall see further on. No case of a similar rotation of larger particles which had sunk into the ectoplasm, as described by Jennings (’04, p. 142), has come under my observation.

Figure 20. Amoeba sphaeronucleosus. a, the under side of the ameba. The part of the ameba attached to the substratum is stippled. Particles attached to the surface film at x moved slowly forward. b, the under side of the ameba, showing the attached parts stippled. The particles suspended in the water at x moved slowly forward with the ameba. c, a cross section of an ameba of shape shown in b, showing the ridges on the surface. Length of the ameba, about 100 microns.

The movement of the surface layer in A. verrucosa is quite like that of sphaeronucleosus. [Figure 21] shows a group of three particles carried by a verrucosa while changing its direction of locomotion. The particles changed position with regard to each other and they moved at different speeds. Particles a, b, c, moved respectively 2.40, 3.26, 2.85 times as fast as the ameba advanced. Other experiments indicate that the outer layer of verrucosa moves at about the same speed, compared with the speed of the ameba, as that of sphaeronucleosus.

Amebas with so-called limax-shaped bodies do not possess surface layers that carry particles forward with the same speed as those amebas with broad bodies. It is only occasionally that large amebas such as proteus are found in a limax or clavate shape. One of the most favorable of the large amebas in this respect is discoides. It is frequently found in clavate shape and it possesses the further advantage in being nearly cylindrical in cross section. It is also more in the habit of loping along the surface in the manner described by Dellinger (’06, p. 57) so that what is observed to take place in discoides in the clavate shape, holds likewise for free pseudopods extended into the water out of contact with a solid support ([Figure 22]).

Figure 21. Illustrating the similarity of the movement of the surface layer of Amoeba verrucosa with that of A. sphaeronucleosus. A group of three particles, connected by dotted lines for reference, change their relative positions as the ameba (verrucosa) changes its direction of movement. Length of the ameba, 150 microns.

Figure 22. Illustrating the movements of an Amoeba proteus, after Dellinger. At c in stage 2 a pseudopod is projected which fastens itself to the substratum as shown at c, 3, while a, 2, is pulled loose. In 4 another pseudopod is projected which fastens itself at d. The ameba is not in contact with the substratum at all points on its under side.

In figure 23 is shown a clavate discoides with a small particle attached to its side. The particle moved forward until it came to lie at the anterior edge, 10. The speed of the particle from 1 to 10 was 1.36 times as fast as that of the ameba, a much slower rate than was observed in sphaeronucleosus. At 6 a new pseudopod was projected for a short distance, thus increasing the amount of new ectoplasm forming in proportion to that of the whole ameba. This change was reflected in the increased speed of the particle, which moved 1.64 times as fast as the ameba from 5 to 6. At 10 the anterior end again spread out and again the particle moved faster—twice as fast as the ameba from 9 to 10. Stages 11, 12, 13 are added to show that the particles do not tend to go to the under surface but remain at or very near the tip. The slight irregularity of the waves of hyaloplasm pushed out at the anterior end accounts for the changing position of the particle after it has reached the anterior edge. The particle remained at the edge of the advancing ameba for several minutes after the stage drawn at 13.

Figure 23. Showing the movement of a particle on the surface layer of an Amoeba discoides. The particle remained on the anterior end of the ameba for several minutes after stage 13. The ameba was about 320 microns long.

In another observation the effect of a narrowing of the advancing tip of the ameba is shown very well. In figure 24 the ameba was advancing with a broad anterior end, as shown at 1 and 2. From 2 to 4, the region where new ectoplasm was made, narrowed down very considerably. These changes in the width of the anterior end are reflected, as in [Figure 17] by a decrease in the relative speed of the moving particle. Thus the particle moved 1.75 times as fast as the ameba from 1 to 2 while from 2 to 4 the particle moved only 1.27 times as fast as the ameba.

Figure 24. Showing the effect of a narrow anterior end on the rate of movement of the surface. Length of the ameba, about 320 microns.

The movement of the third layer in proteus is difficult to study owing to the formation continually of ridges, as explained on page 20. Even in clavate shaped amebas, waves of protoplasm are pushed out on the sides and on the tip with consequent formation of ectoplasm, so that the ameba grows in width slowly at the same time that it grows in length. A typical shape of a proteus in clavate form is slightly tapering toward the anterior end. This shape is maintained by gradual extension of the sides of the anterior half or two-thirds of the ameba as it moves along. These conditions are just the reverse of what was seen to be the case in sphaeronucleosus and verrucosa, where the anterior edge was wider than any other part of the body. But discoides, although free from the ridges and grooves characteristic of proteus, frequently has an anterior edge that is narrower than any part of the body, thus necessitating extension of the sides as the ameba moves forward.

Let us now see what is the effect of ridge formation upon the movement of the surface layer. [Figure 25] shows a proteus and a narrow anterior end in proteus with two pseudopods and a particle attached to the side of the ameba at 1. Both pseudopods advanced until stage 4 was reached, but the particle was not appreciably deflected from an approximately straight path by the small pseudopod at the other side of the ameba. Reference to the figure shows that the particle travelled much faster while the pseudopod on the side was extending than after it began to retract. The particle moved 1.43 times as fast as the ameba from 1 to 4. But from 4 to 7 the particle moved only 1.06 times as fast as the ameba.

Figure 25. Amoeba proteus. Rate of movement of the surface layer as compared with the rate of movement of the ameba. The pseudopod on the right was extended to stage 5; from then on it was retracted, as indicated by the outlines. Length of the ameba, 400 microns.

In the earlier stages the outer layer was pulled toward the tip of both pseudopods, in the later stages only toward one, and in this lies the explanation for a more rapid movement of the particles in the earlier, and a slower movement in the later stages. This effect was also observed in discoides, but the fact that the particle in the later stages moved only very little faster than the ameba is due to a narrow anterior edge and to the formation of ectoplasm in the ridges over the surface of the ameba. The effect of ridge formation on the movement of particles attached to the surface film is well seen when an ameba has two forward moving regions opposite each other. Under such conditions particles located equidistant or nearly so between such regions, move only very slowly or not at all, the pull upon the film being nearly or quite equal. In a similar manner the ridges which are constantly forming on a proteus are continually competing with the anterior end in their pull upon the surface layer, thus preventing rapid forward movement.

Figure 26. Showing the comparative rate of movement of the surface film over the retracting parts of the ameba. In figures 2 to 8 only a part of the ameba is shown. Length of the ameba, 500 microns.

[Figure 26] shows that the surface layer flows away from the tip of a retracting pseudopod that is located near the anterior end. The particle moves slowly until the body of the ameba is reached, when movement becomes more rapid, 8, 9. This proves that the third layer moves away from the retracting parts of an ameba, no matter how large the total area of these parts may be in proportion to the area of new surface that is being made. But whether the speed of the moving third layer changes in correspondence with a larger or a smaller ratio between building and retracting ectoplasm has not been ascertained.

[Figure 27] shows that the relative positions of particles attached to the surface layer may readily change while the ameba deploys its psuedopods. Three particles marked a, b, c and connected

Figure 27. A part of an Amoeba proteus illustrating what is perhaps the most characteristic quality of the surface layer of amebas, its fluid nature. Three particles, a, b, c, were moving forward along an actively growing pseudopod. In stage 2, particles b and c had arrived nearly at the tip of the pseudopod. A pseudopod was then thrown out on the right, which resulted in the movement of a in the same direction, while b and c remained nearly stationary. Later on this pseudopod was retracted. b and c were drawn back toward the main body of the ameba while c remained behind, moving only very slowly. Thus the relative positions of these particles was completely changed.

by a line for convenience of reference, were in the position indicated at 1 when the forward end of the ameba occupied the position indicated by outline 1. As the ameba moved forward the particle c gained slightly on a and b for no ascertainable reason, unless it was on account of the projection of the large pseudopod on the opposite side. At stage 2 a new pseudopod was started on the right, which at stage 3 had grown to large size while streaming in the original pseudopod was arrested. At stage 3 particles a and b retained the same position they had in stage 2, except for a slight turning to the right. Particle c however moved across the base of the original pseudopod and on to the middle of the new pseudopod. At stage 4 a and b had again only slightly moved to the right of the position they occupied in stages 2 and 3, while c moved rapidly toward the tip of the new pseudopod. The new pseudopod was then retracted and at stage 5 the particles had begun to move back toward the main body of the ameba. Particles a and b now gained considerably on c because they were located further away from the tip of the retracting pseudopod. Particles a and b were drawn to the middle of the retracting pseudopod because of the continuous enlargement of the large pseudopod on the right, below, through which the ameba moved on.

The most important feature of this observation is the change in the position of the particle c with respect to that of a and b. The latter particles retained their relative positions with very slight, if any, change, while c swung around a and b nearly 180°, and at the same time changed the distance very greatly between itself and the other particles. Moreover, b, at stage 5 led the procession of particles, while at stage 1, a led. No further demonstration is necessary to show that the surface layer is distinctly fluid and dynamic, and not at all such a static structure as an elastic permanent skin, as Jennings (’04) and Rhumbler (’14) maintained.

CHAPTER VIII
ON THE NATURE OF THE SURFACE LAYER

The observations in the preceding chapters on the general movements of the surface layer of amebas will afford a sufficient basis for an inquiry into the nature of this layer. The mere demonstration of the existence of this layer is, of course, interesting enough, for a number of contradictory statements by various students of the amebas are satisfactorily cleared up by these observations. But the problem of ameboid movement affects other organisms besides amebas, and since the movement of the surface layer is so intimately associated with ameboid movement, it becomes of more than ordinary interest to learn something of the nature and composition of this layer.

In the first place the property of carrying particles toward the anterior end of amebas does not appear to be of any advantage. That is, whatever the movements of the outer layer may be, the ameba does not appear to be better off when particles are carried forward than when none are carried, for such particles are very small and almost without exception devoid of food value. The particles are masses of debris which accidentally adhere to the ameba, and the ameba makes no visible effort to make such particles adhere, nor to get rid of them. The ameba seems to be quite indifferent to the presence of such particles.

On the other hand, as Schaeffer (’17) has pointed out, the capacity for transporting particles cannot but be looked upon as a hindrance to locomotion. As has been stated, the surface film moves in the same direction as the ameba. Whenever the surface film comes against a solid object, it pushes against the object, and nullifies to a certain, though small, extent the energy expended in moving forward. And it will be seen without further argument, of course, that the energy involved in carrying particles forward is not only itself lost but consumes an appreciable part of the energy available for forward movement. This fact, together with the universal occurrence of this phenomenon among amebas indicates beyond question that it is intimately associated with ameboid movement as it is ordinarily understood in amebas, and that it is almost certainly a “necessary” physical consequence of the more fundamental physical processes involved in the movement of amebas.

That the third layer moves in the same general direction as the ameba has already been mentioned. The direction of a moving particle is however not necessarily parallel with the stream of endoplasm below. In a retracting pseudopod that lies nearly parallel to and by the side of the main advancing pseudopod, the particles on the far side and near the base frequently move across the pseudopod at an angle (and therefore also across the endoplasmic stream), and up the active pseudopod on the near side. This shows conclusively that the direction of flowing endoplasm by itself has no direct connection with the direction of flow of the surface layer.

To say that the particles carried by the surface layer bring up at the anterior ends of pseudopods or of the ameba when in clavate shape, admits of further qualification. The advancing edge is not a straight line but an arc, and the sides near the advancing edge are building at a slower rate than the extreme tip. The most rapid formation of ectoplasm is at that point of the ameba that is farthest ahead. At this point all the ectoplasm to be made is still to be made, but as one passes back along the side of the pseudopod more and more ectoplasm is encountered and less and less remains to be made. There is therefore a gradient in the rate and in the amount of ectoplasm formed as one passes back from the forward end of the longitudinal axis of the pseudopod along the side. This is especially the case with certain amebas like Amoeba discoides, A. laureata and others in which the pseudopods are more nearly cylindrical. In such amebas as A. proteus and A. verrucosa, the factor of ridge formation complicates to some extent the longitudinal gradient of ectoplasm formation. But in spite of these specific differences, the general statement still holds that the rate of ectoplasm formation at the extreme anterior end is higher than anywhere else in the ameba, and that the rate gradually falls to zero as the nearly straight and parallel sides of the pseudopod or ameba, as the case may be, are approached.

Now we have seen that if a particle becomes attached to the outer layer of such an ameba as discoides, which has nearly symmetrical pseudopods, at some considerable distance from the tip of the pseudopod, it moves forward until the tip of the pseudopod is reached. It does not tend to come to rest near the tip of the pseudopod, where the rate of ectoplasm formation is much higher than at the sides of the pseudopod, though not as high as at the tip, but it moves on until the tip is reached. That is, the movement of particles on the surface film is toward that small area at the extreme anterior end where the rate of ectoplasm formation is highest.

In such an ameba as verrucosa, however, the highest rate of ectoplasm formation would be, not at a small circular area, but a very narrow strip along the anterior edge; for the rate of ectoplasm formation over a considerable portion of the width of the anterior end of the ameba is practically the same, according to observation. Consequently we do not find particles which are attached to the outer layer tending to move to a point lying on the longitudinal axis, but their paths are found to be straight and parallel with the longitudinal axis, if headed toward any point over a considerable stretch of the anterior edge on either side of the longitudinal axis.

All the evidence that is at hand therefore points to the conclusion that the direction of movement of the surface film in a moving ameba is toward that point where ectoplasm is formed most rapidly.

But where do the particles come from? At exactly what regions of the ameba do they start to travel toward the anterior ends of the ameba? In sphaeronucleosus and its congeners, it is very difficult to determine just when the particles begin to move toward the forward edge. Particles near the posterior end on the upper surface of these amebas moved forward slowly, much more slowly than particles near the middle. Sometimes particles near the posterior end seem to be motionless for some time, but the incessant though slow kneading process going on at the posterior end makes accurate observation difficult. Only in a general way it may be stated that particles begin their forward march at or near the posterior end. In amebas that habitually form pseudopods more accurate information can be obtained.

In proteus or discoides, for example, projecting pseudopods are often suddenly stopped and retracted, with a resultant change of an anterior to a posterior end. Particles attached to the outer surface on such pseudopods move toward the anterior end, of course, as long as the pseudopod is building, in the manner described in the preceding pages. But when the endoplasmic stream is arrested, the forward movement of the particle likewise stops. When the endoplasm starts to flow back into the main body of the ameba, the particle also starts moving back; but there is a period of a few seconds after the endoplasmic stream is reversed during which the particle remains quiet. And when it does start in to move, it moves only slowly. Within a few seconds, however, the average speed of movement is attained. This is true of particles located some distance away from the tip of the pseudopod. If the particle has reached the tip of the pseudopod before reversal of the endoplasmic stream takes place, the particle often remains at the tip until the pseudopod is almost completely withdrawn into the main body of the ameba ([Figure 26], p. 60). At other times such a particle becomes displaced, presumably by irregular retraction of the tip of the pseudopod, and finds itself at the side of the pseudopod. When this happens it moves slowly toward the main body of the ameba, but faster than the tip of the pseudopod does.

It frequently happens, especially in annulata, but also in proteus and other forms with many pseudopods, that when an advancing pseudopod is about to be withdrawn, there intervenes a stage where the endoplasm in the distal part moves away from the ameba, while that in the proximal part moves toward the ameba, with a neutral or motionless zone between. In such case a particle on the distal end moves slowly toward the tip while a particle in the proximal region moves toward the base of the pseudopod. Particles over the neutral zone are motionless. In these cases, however, changes in the direction and speed of the ectoplasmic stream are too frequent and the relative strengths of the distal and proximal currents too variable, to enable one to secure very accurate data by means of camera lucida drawings (a kinematograph is essential for this purpose), so no figures of the speed of movement of such particles are given. Nevertheless the general results of the observations are as stated. It might be added that in some cases the neutral zone for the particles attached to the surface did not coincide exactly with the neutral zone of the endoplasm, but was located a little further distally.

From these observations it appears that a rough index of the direction of movement of the surface film is the direction of the streaming of the endoplasm; and that the surface layer moves away from regions where ectoplasm is in the process of being converted into endoplasm. Since a particle attached to the surface may remain for some time at the tip of a retracting pseudopod, while one that is attached to the sides of a pseudopod moves toward its base, it appears that the speed of the moving surface film is not directly correlated to the rate of transformation of ectoplasm into endoplasm. The slower speed of particles near the posterior end points also in this direction. The formation of ectoplasm at the anterior end seems therefore to be much more intimately connected with the movement of the surface film than the destruction of the ectoplasm, though it is not yet clear that the liquefaction of the ectoplasm is altogether without effect.

Now as to the speed with which the surface film moves. The foregoing illustrations and figures show that the particles attached to a sphaeronucleosus on the upper surface move from 2.5 to 3.6 times as fast as the ameba ([Figure 19]) while particles attached to a discoides move only from 1.2 to 2 times as fast as the ameba moves. In proteus the speed of the particles is still slower, because of the longitudinal ridge-like waves of protoplasm which are continually being thrown out. In this species it frequently happens that because of the numerous ridges, the ameba moves faster than the particles attached to the outer surface; but this is to be looked upon as a mechanical complication, not as indicating a difference in the nature of the surface layer.

How is the difference in the speed of movement of the surface layer between sphaeronucleosus and discoides to be explained? There are no ridges to retard the movement of particles in discoides, while there are ridges in sphaeronucleosus, where the particles move on the average twice as fast as on discoides. In the first place the advancing edge, the edge where ectoplasm is being made, is proportionately much wider in sphaeronucleosus than in discoides as compared with the amount of surface back of it. Figures 23 and 24 show that the rate of movement of the surface film is directly proportional to the amount of new ectoplasm forming. In the second place, the greater part of the under surface in the forward half of sphaeronucleosus is attached to the substrate, so that the surface layer which flows toward the anterior end is derived almost wholly from the upper surface; while in discoides the whole surface in free pseudopods, and nearly the whole surface in attached amebas (cf. Dellinger’s observations described on p. 56) possesses mobile surface protoplasm. Observation of moving particles on these amebas proves this. Then again, the anterior edge of a sphaeronucleosus is not attached at the points farthest advanced, but the point of attachment is some distance back, as indicated in figure 20. The effect of this is to increase the amount of forming ectoplasm in proportion to the surface of the ameba from which surface protoplasm may be drawn. Still one other factor must be considered. As is well known sphaeronucleosus, verrucosa and their congeners possess longitudinal ridges on the upper surface which consist of ectoplasm, covered of course by the surface film. These ridges are formed near the anterior edge, not by wrinkling, but by the construction of new ectoplasm. Once formed, they remain until the ameba, so to speak, flows out from under them. That is, the ridges undergo comparatively slight changes until changed back into endoplasm at the posterior end of the ameba. As the ameba flows ahead the ridges are of course continually being added to or lengthened, by the conversion of some endoplasm into ectoplasm. The ridges may thus retain their identity for a long time although the substance composing them is changed every time the ameba moves the length of its body. It is clear, therefore, that there is more ectoplasm formed at the anterior end of a sphaeronucleosus than would be the case were the upper surface of the ameba plane; and the conclusion therefore is obvious that the formation of ridges, occurring as it does, chiefly at the anterior end, serves further to accelerate the forward movement of the surface film.

If the form of sphaeronucleosus were more regular than it is, the amount of ectoplasm in the process of forming at any given moment could be compared with a similar relation existing in discoides, to see whether these respective ratios were proportional to the speed of the moving surface films in the two amebas. As it is, the irregularity of form of sphaeronucleosus makes such computation subject to the possibility of considerable error. In discoides however the problem is comparatively simple. I therefore did not go into this matter extensively, but merely worked out the relations mentioned in one case, and I mention it here to illustrate the method rather than to record the result, which is not to be taken as very exact.

Figure 28. A clavate Amoeba discoides, showing the amount of ectoplasm that is constantly being made at the anterior end. Length of the ameba, 310 microns.

Since the movement of the surface film is obviously a surface phenomenon, only the surfaces of the amebas need to be taken into account. In [Figure 28] is illustrated a discoides of such a shape as to allow a fairly accurate computation of its surface. Three outlines of the anterior end only are given; the rear portion of the ameba remained approximately the same size and shape in the three outlines. The cross lines at the anterior end divide the forming ectoplasm of the ameba from the formed. As will be noticed the cross lines are drawn through the intersections of two successive outlines. Computing the areas on both sides of the cross lines for the two outlines and averaging them, there is found a ratio of 1 to 10; one-eleventh of the total surface represents forming ectoplasm, and ten-elevenths formed ectoplasm. (One-twenty-second of the total surface was deducted for surface attached to the substratum.) Sphaeronucleosus stands in contrast with discoides for it is attached to the substratum over a much greater area and in consequence only a slight amount of surface is drawn from the under side. This ameba may therefore be regarded in this connection as of only one surface, the upper. That part of outline 1 in [Figure 14] cut by outline 2 indicates, as in discoides, the region of forming ectoplasm, and the space between outlines 1 and 2 may be used as a basis of computation. New ectoplasm is formed in this zone and far enough back to include the tips of the longitudinal ridges, of which we have already spoken ([Figure 13]). The zone of forming ectoplasm would therefore be about twice as wide as the average width of the three zones between the successive outlines in the figure, and of approximately the same shape. On this basis, the surface occupied by forming ectoplasm is ¹/_5.8 of the total surface, and the ratio of formed to forming ectoplasm is 4.8 to 1.

(For the sake of completeness, a few factors whose values cannot easily be computed may be mentioned. 1. The anterior edge is not attached to the substratum at its farthest point, but at some little distance back of the edge, thus increasing the relative amount or forming ectoplasm; but this is offset by the surface of a part of the under side at the posterior end where the surface layer is active. 2. The ectoplasm composing the ridges, which must be added to the formed ectoplasm, would increase the ratio, though only slightly).

Approximately twice as much ectoplasm is therefore in the process of formation in sphaeronucleosus as in discoides when compared with the formed ectoplasm in the respective amebas, over which the surface film is active. This ratio corresponds very well with the rate of movement of the outer surface in these amebas, which as we have seen is about twice as fast in sphaeronucleosus as in discoides.

Where does the surface layer come from and what becomes of it after it arrives at the anterior end? It moves continually forward as long as the ameba moves forward. There would seem to be a tendency therefore for it to accumulate at the end of a free pseudopod in such a form as discoides, and even under ordinary conditions of locomotion where there is occasional attachment to the substratum by very short pseudopods, the surface layer is continually moving toward the anterior end on practically all sides. Every time, therefore, that the ameba moves a little less than its own length, there would accumulate at the tip of the ameba, if it were not removed, an amount of surface layer equivalent to that which covers the whole ameba. No such accumulation can be detected however, from which we infer that it is removed as fast as brought there. And the posterior region of the ameba, which is the main source of the surface film, does not become poorer in this material by reason of its continual flow forward, but new surface is made continually to take the place of that moving forward. This process of destruction and creation of surface is accordingly rapid during active locomotion;—a discoides, moving approximately once its length at room temperature in two minutes, destroying therefore the equivalent of its entire coat of surface in that time; while a sphaeronucleosus, moving once its length in two or three minutes, destroys all its surface every minute.

From what has been said thus far, it must be apparent that there is striking resemblance between the general movement of the surface layer of the ameba, and of a surface tension layer in a drop of fluid in which the tension is changed at some point. Let us now inquire briefly into this resemblance.

As is well known the surface of a liquid in contact with another liquid, solid or gas, with which it does not mix, behaves like a stretched membrane, so that when the tension is reduced at any point the surface layer moves away from that point. A good illustration of the effect of a decrease of surface tension is found in a drop of clove or other oil with which some substance that reduces the surface tension, such as alcohol or soap, is brought into contact at one side. If previously some dust particles have been placed on the surface of the oil drop, it will be easy to see that the surface of the oil moves to the opposite side from where the alcohol or soap solution touched the oil. In practice it is a very simple matter to lower the surface tension of a drop of fluid as described, so as to show the movement of particles on the surface. Almost any liquid may be used for this purpose. But it is comparatively very difficult to increase the surface tension at some point of a drop of fluid in such a way as to cause particles on the surface to move toward that point. The principle underlying the movement of the surface film in both cases is however exactly the same; so, although it would be more desirable to compare the surface movements in a drop of fluid in which the surface tension is increased at some point, because this is what happens in an ameba during locomotion, we shall nevertheless find it necessary to consider a drop of fluid in which the surface tension has been lowered. The application of the illustration is readily made.

When the surface tension of a drop of fluid is lowered by bringing into contact with it some other substance that possesses this power, the surface rushes away at great speed in all directions from the point where the tension is lowered, because usually the tension is reduced very considerably. In this surface movement it is found that new surface is made where the tension is lowered and old surface is destroyed, that is, pulled into the interior over a large part of the surface opposite to where the tension is lowered. The speed of the surface movement is most rapid near the point where the tension is lowered and becomes gradually slower as the opposite side of the drop is approached, where there is no movement. This variation in speed of the moving surface seems to be due largely to the small area in which the tension is lowered as compared with the whole surface of the drop.

In the ameba the conditions are reversed. The surface layer moves toward a point with increasing speed, instead of away from a point. In both the ameba and the drop the greatest speed is attained near the small area where the change in surface tension occurs.

The behavior of large and heavy particles on the surface of a drop of fluid and on an ameba are similar. A heavy particle as of sand, or a small glass rod, laid on the ameba, is not moved by the surface layer. It forms an island of surface matter around which the moving surface layer flows. Precisely the same thing happens in surface layer movements in inanimate fluids.

Again in point of thinness there is no disagreement so far as microscopic observation goes. Neither the surface film on an ameba nor the surface film on a fluid can be directly observed microscopically to be different from the fluid below it. The surface layer is, as is generally believed, of molecular dimensions, and its thickness is beyond the limits of vision. Unless some special means is discovered therefore for making visible the surface film, such as a process of staining, it may be impossible to ascertain its ultimate structure directly, for it overlies a mass of heterogeneous fluid whose composition is constantly changing.

It seems to follow from what is observed of the surface tension layers of the fluids of physics that such layers must be of the same constitution as the body of the fluid over which the layer is formed, although, as is well known, the proportion of the ingredients in the surface layer is different from that in the body of the fluid. Now since the resemblance between the surface layer of an ameba and a surface layer on a drop of fluid has thus far been found to be complete, it is pertinent at this point to discuss Gruber’s (’12) suggestion that the movement of particles forward on an ameba is due to the forward movement of an inert layer of mucus or gelatinous material secreted by the ameba.

To begin with, observation does not support Gruber’s suggestion. No such layer can be seen. Such a layer, since it is shown to persist for several minutes at least, should remain after an ameba bursts, under experimental conditions, but no such remains can be seen. Its existence should be demonstrable by the use of dyes, but the evidence is negative. Indeed there is not any direct evidence that can be brought in support of the suggestion that this surface layer is gelatinous in composition. Moreover, as we have seen, the layer on the ameba that carries particles forward seems to be destroyed at the anterior end, for in what other way would particles remain at the anterior end after being brought there? But the supposition that a gelatinous layer might be drawn into the interior at the anterior end is also negatived by observation, for no very small particles clinging to or imbedded in the surface substance are ever drawn into the ameba, as would almost certainly be the case if the substance composing the layer were gelatinous. And as to supposing that this layer, if gelatinous, might behave essentially as a surface tension layer and therefore be drawn in at the anterior end of the ameba, this is contrary to the experience of physics; for the physical nature of the ameba would make it impossible for the ameba to have a surface layer of gelatinous matter. There do not seem to exist any grounds therefore for supposing that the outermost layer of an ameba, the layer that carries particles as described in the preceding pages, can consist of an inert substance as Gruber suggests.[4]

From these considerations, then it appears that all the evidence available, both direct and indirect, points to the conclusion that the behavior of the surface layer on the ameba resembles in general and in detail the behavior of a surface tension layer in an inert drop of fluid, and that we must regard the surface layer on the ameba as a true surface tension layer. This layer is therefore a dynamic layer, containing free energy, and capable of performing work. It is physiologically distinct from ectoplasm, as ectoplasm is distinct physiologically from endoplasm. But the distinctive properties which the surface layer possesses are functions of its position. These properties clearly indicate that its constitution is protoplasmic, corresponding to the fluid parts of the internal protoplasm.

The surface layer of the ameba is probably identical with what is commonly called the plasma membrane or semi-permeable membrane as postulated by Overton (’07). The peculiar structure supposed to be possessed by plasma membranes are held to be due chiefly to surface forces. The fact that the surface layer of the ameba is continually being destroyed and re-created during locomotion does not support the view that the plasma membrane is of inert composition, as for example, lipoidal, as has been suggested. The observations, on the contrary, confirm Höber’s (’11) view that the plasma membranes generally are living structures. But it may be regarded as certain that if lipoids are present in the protoplasm of the ameba, these substances, according to the principle of Willard Gibbs, will be found in higher concentration in the surface film than in the body of the ameba.

Perhaps the most important question that arises in connection with the surface layer of the ameba is: What causes it to move in the manner described? But we can do little more than ask the question. It has been seen that the surface film moves toward an area of increased tension rather than from an area where the tension has been lowered. However, since we are completely in the dark respecting the composition of the surface layer or of the fluid parts of the ameba, it is exceedingly hazardous to venture an explanation. If the surface layer should have its tension lowered by a concentration of lipoids in it, we would be faced by the necessity of explaining their removal at the anterior end. If we turn to electrical causes we meet again with great difficulties. An ameba moves with the electric current, when a current is passed through the water. The surface layer under these conditions behaves normally, as may be inferred from Jennings’ (’04) figure on page 198. That is, the current controls the direction of the movement of the ameba, with the current leaving the ameba at the point of highest surface tension. This is contrary to the action of the mercuric capillary electrometer, in which the mercury column also moves with the current, but because of lowered surface tension where the current leaves the mercury. The conditions surrounding these cases are so different however, that very little can be gained by setting them in contrast to each other.

CHAPTER IX
The Surface Layer and Theories of Ameboid Movement

The observations recorded in the preceding two chapters, while they do not tell us anything about the direct cause of the movement of the surface layer, nevertheless indicate clearly enough that the area where ectoplasm is made is the area toward which the surface film flows. There is no question therefore of the intimate relation between the transformation of endoplasm into ectoplasm and the movement of the surface layer.

The apparent absence of movement in the surface film of the pseudopods of Difflugia (Schaeffer, ’17) and the definitely proved absence of movement in the surface layer in the foraminiferan Biomyxa and myxomycete plasmodia, where no ectoplasm is formed in the manner observed in amebas, also indicates a causal relation between movement of the surface layer and ectoplasm formation. The relation moreover seems to be a necessary one for the movement of the surface layer is contrary to the processes involved in locomotion. In other words, from the standpoint of the ameba, it is a “necessary evil,” so far as locomotion is concerned.

The transformation of endoplasm into ectoplasm is unfortunately not understood, though from what we know in a general way of the behavior of colloidal solutions it seems to be a surface tension effect due to (or accompanying) a change of phase. Something akin to gelation occurs as Kite (’13) has shown. It is a problem in the chemistry of colloids. But the structure or composition of the protoplasm is complex and practically unknown, and it is quite open to criticism whether analogies from the behavior of pure solutions of colloids, such as gelatin, afford any real basis for an explanation.

Although a knowledge of the movements of the surface layer is interesting enough by itself, it will achieve its true importance only when it can be related to other processes in the ameba in a causal manner. It does not at present give us any greater insight into the ultimate cause of ameboid movement, although it is clear that an important step in this direction has been taken. But no theory of ameboid movement can be accepted that demands conditions in the ameba that are contrary to those described in the preceding chapters, in connection with the surface layer. From this point of view therefore the discovery of the true nature of the outside surface of the ameba is of importance, for it widens to a very considerable extent the observational basis with which any theory of ameboid movement must conform. Since the properties of the outer layer are here described in detail for the first time, it becomes necessary to enquire to what extent the more commonly held theories of ameboid movement conform with the observed behavior of the surface film. Although the surface tension theory was the first detailed theory proposed toward an explanation of ameboid movement, I shall discuss Jennings’ (’04) observations on the movements of the ameba first, because a great part of his work deals with the movements of the surface film, although he did not recognize it as distinct from the ectoplasm in its movements.

It is generally recognized that Dellinger’s (’06) work proved that Jennings’ conception of the ectoplasm as a permanent skin in which the ameba rolls along, is probably inadequate for such amebas as proteus; though singularly enough it is still supposed that verrucosa and its congeners move in the way described by Jennings (Hyman, ’17, p. 83).

Jennings (’04) describes the movements of amebas, both proteus and verrucosa “types,” as follows:

“In an advancing Amoeba substance flows forward on the upper surface, rolls over at the anterior edge, coming into contact with the substratum, then remaining quiet until the body of the ameba has passed over it. It then moves upward at the posterior end, and forward again on the upper surface, continuing in rotation as long as the ameba continues to progress. The motion of the upper surface is congruent with that of the endosarc, the two forming a single stream (p. 148).

“We have demonstrated above, for Amoeba at least, that the forward movement is not confined to a thin outer layer, but extends from the outer surface to the endosarc; in other words that the outer surface moves in continuity with the internal substance (p. 150).

“There is no regular transformation of endosarc into ectosarc at the anterior end. On the contrary the ectosarc here retains its continuity unbroken, moving across the anterior end in the same manner as across other parts of the body. In the same way the ectosarc is not regularly transformed into endosarc in the hinder part of the body.... Such transformation is by no means a regular accompaniment of locomotion” (p. 174).

According to Jennings, locomotion is aided by the projection of waves of hyaloplasm at the anterior edge, “an active movement of the protoplasm of a sort which has not been physically explained.” These waves, attaching themselves to the substratum, enable the ameba to pull itself along by a rolling movement as described in the quotation above.

As to the rate of movement of the outer surface as compared with that of the endoplasm, Jennings concluded:

“The direction of movement of particles on the outer surface is the same as that of the underlying particles of endosarc. The rate is also about the same as for the endosarc, though often, or perhaps usually, the outer particles move a little more slowly than those in the endosarc” (p. 142).

In view of the observations recorded in the preceding pages it is clear that Jennings’ statement that substance after moving forward on the upper surface, rolls over the anterior edge is quite erroneous. The attached particles, if heavy, may do so, but the surface film itself does not. It is, on the contrary, taken into the interior at the anterior edge.

The statement that the movement of the outer surface is congruent with that of the ectoplasm can likewise not be substantiated by observation, as has been demonstrated in the preceding pages. It is difficult to distinguish between the ectoplasm and the surface layer in such amebas as sphaeronucleosus and verrucosa, for there are no large crystals or other bodies which get caught in the ectoplasm as it is formed from endoplasm at the anterior end. But attentive observation will demonstrate very definitely that the ectoplasm here is stationary to the same degree as in proteus. The stationary properties of the ectoplasm are however not properly a matter for discussion; for five minutes’ observation of a proteus, discoides, annulata, particularly a laureata, under 300 diameters magnification, will convince anyone that the ectoplasm is stationary while the surface film, with attached particles, moves over it. No one can possibly come to any other conclusion. Jennings’ conclusion was due undoubtedly to an error of observation.

Jennings’ statement that the rate of movement of the outer surface is the same as that of the endoplasm (p. 142) when taken in connection with his other statement that the ectoplasm is a more or less permanent skin, presents a mechanical impossibility; for unless the outer surface moves twice as fast as the endoplasm, no rolling movement would be possible. Several of Jennings’ figures (especially Figures 38, 39, and 41) indicate in fact that he conceived of the outer surface as moving faster than the ameba advances, or that the upper surface moves over the ameba as the ameba moves over the substrate. Jennings’ theory requires that the surface layer move twice as fast as the ameba advances. Hyman (’17) also makes a similar mistake in referring to the rate of movement of the outer surface (p. 85).

Lest the discussion of this point be suspected of being merely verbalistic, it should be recalled that the surface layer of proteus often moves at about the same rate as the ameba; that the surface layer of discoides moves about twice as fast as the ameba; that the surface layer of verrucosa and sphaeronucleosus moves about three times as fast as the ameba; and that the ectoplasm does not move at all. It is of course incumbent on one to discuss what is stated; one is not at liberty to select one of several possible interpretations.

To illustrate this point graphically so as to avoid as far as possible future confusion [Figure 29] is appended. In a is shown a particle traveling on an ameba at the same rate of speed as the ameba; at b is shown a particle that moves twice as fast as the ameba; at c the attached heavy particle does not move at all. For the sake of completeness d, [Figure 29], is added here. It illustrates the backward moving ectoplasm in an ameba that is suspended in a jelly medium that prevents the ameba from sinking to the bottom. It must be admitted that in thus considering the rate of movement of the various tissues of the ameba from a single standpoint, a point outside of the ameba, little room is left for confusion.

Figure 29. a, a particle attached to an ameba and moving at the same rate as the ameba. This condition is often observed in proteus where the surface film, owing to its destruction during the formation of the longitudinal ridges, retards the forward movement of this layer. b, a particle attached to the surface film of an ameba moving twice as fast as the ameba. This condition is seen in discoides, verrucosa, sphaeronucleosus, etc. c, a particle on an ameba that does not move at all although the ameba does. This is seen when a heavy particle is laid on an ameba, too heavy for the surface film to move. d, movement of ectoplasm in an ameba suspended in a jelly medium. The vertical lines are to be considered as stationary.

There is comparatively little friction, if any at all, between the upper surface and the endosarc, according to Jennings’ view, since both these layers move at the same rate and as a single stream. On the other hand there must be very considerable friction between the endoplasm and the lower ectoplasm, which does not move at all. This difference in the amount of friction must show itself in the different speeds of the endoplasm near the upper ectoplasm and near the lower ectoplasm. Observation indicates however that the most rapid streaming of the endoplasm is in the middle of the ameba or pseudopod and that it gradually becomes slower as the ectoplasm is approached on all sides.

We said above that if the ectoplasm were a more or less permanent skin in which the ameba rolled as described by Jennings, the upper surface (=ectosarc, Jennings), according to a well known mechanical principle, would have to move ahead about twice as fast as the ameba advances. Now the upper surface of sphaeronucleosus and of verrucosa in locomotion was found to move from three to three and a half times as fast as the ameba (Chapter VII). In discussing movement in “verrucosa and its relatives” Jennings says “the essential features of the movement seem to be (1) the advance of a wave from the upper surface at the anterior edge; (2) the pull exercised by this wave on the remainder of the upper surface of the body, bringing it forward. Most of the other phenomena follow as consequences of these two” (p. 146). Thus the amount of stretch of the upper surface would exceed the amount of pull on it from 50% to 75%!

Jennings’ explanation of ameboid movement in which the important factor is a more or less permanent ectoplasm in which the ameba rolls along, would unquestionably produce rotary currents. Rhumbler (’98) recognized this and after full consideration rejected the idea that the ectoplasm is a permanent skin in which the ameba rolls along in locomotion, because rotary currents are not observed in a moving ameba. Anyone who doubts that rotary currents would be produced under these conditions can convince himself by putting a quantity of glycerine and some shavings in a large transparent rubber balloon or celloidin bag and letting it roll slowly down an incline in front of a strong light. If not too much glycerine is placed in the balloon, the shape of an ameba is closely enough approximated, and the rotary currents—down at the posterior and up at the anterior end—are well shown.

From all these considerations it is quite clear that Jennings’ explanation of ameboid movement as a rolling movement can not any longer be maintained. His “discussion of this matter (the rolling movement hypothesis) is an excellent example of the fact that acumen and excellent reasoning may lead one astray in scientific matters when the observational basis for the reasoning is not secure.” (These are Jennings’ own words in criticism of Rhumbler on the same subject!)

The surface tension theory, with its many modifications, has had a great many more adherents than any other theory that has been advanced to explain ameboid movement. It represents the attempt of biologists to explain a vital phenomenon on physical grounds. The fact that it has been held to go further in this direction than any other, and the fact of its greater simplicity, doubtless are responsible for its wider acceptance. The recent criticism to which this theory has been subjected, however, indicates clearly enough that this theory does not really give a very adequate idea of the processes involved in ameboid movement after all, and in so far as feeding processes are concerned, the theory does not seem to apply at all according to Schaeffer (’16, ’17). But it could hardly have been anything more than excellent guesswork if the surface tension theory as advanced by a number of writers had been found adequate, for the observational basis was very narrow, as the preceding pages have shown, and as the succeeding pages further show.

Not anything like a complete historical account of this theory with its numerous modifications will be attempted here. It would be a large undertaking, for nearly every biologist and biochemist has expressed himself on the subject. It does not appear that much is gained by merely recording the opinions, even of biologists, unless they are based on experimental or observational data, preferably their own. Scientific questions are not decided by ballot vote, and it is not apparent what value such a record of opinions would have except the doubtful one of showing whether the persons involved declared for or against the surface tension theory. Moreover such an account would not be interesting reading for those who want to know first of all what amebas can do. Only the more important modifications of the surface tension theory as applied to ameboid movements will therefore be discussed and these modifications will be considered important in proportion to the amount of observation or experiment on which they are based.

Attention has already been called in Chapter II to Berthold’s (’86) theory of ameboid movement, which was the first attempt to explain this phenomenon on physical principles. As will be remembered, Berthold thought that the nature of the ameba’s immediate environment determined when and in what direction it should move, the source of the energy of movement being supposed to be a decrease in the tension of the surface film of the ameba, brought about by some factor in the ameba’s immediate environment.

One of the most elaborate attempts that has been made toward explaining ameboid movement on the basis of surface tension phenomena was that of Bütschli (’92). From his extensive knowledge of the lower organisms, especially the protozoa, he concluded that protoplasm is an emulsion of two fluids: a more concentrated “plasma,” insoluble in water; and a thinner fluid, “enchylema.” Ameboid movement was brought about by migration of enchylema droplets to the surface of the ameba at the anterior end, where they burst and spread over the surface, lowering its tension. The effect of this change in tension was held to be a flowing backward of the surface of the ameba and a flowing forward of the endoplasm. This is what happens in a drop of fluid, such as oil, on water to one side of which is brought a soapy solution. Bütschli described many experiments with fluids on which the surface tension was changed by appropriate means to simulate the process of movement. After Bütschli had developed his surface tension theory of movement, he discovered, as has already been noted, that in a pelomyxa the surface layer moves forward instead of backward as required by the surface tension theory. In spite of this however he still maintained that his theory of movement could be modified to apply to amebas generally, although so far as I have been able to find, he did not then or subsequently state how. From this we may infer that Bütschli himself probably concluded that the surface tension theory of movement as he developed it, is not of general application or is nothing more than a step in the development of such a theory.

Rhumbler has written a number of papers on the mechanics of ameboid movement, most of which are concerned with elaborations and modifications of a surface tension theory very similar to Bütschli’s. Rhumbler published a general outline of his theory in 1898. The transformation of endoplasm into ectoplasm at the anterior end, and the reverse process at the posterior end, was stated to be an important part of his theory of movement, but just how this was necessary to surface tension effects was not explained in physical terms. Feeding was assumed to be caused by the direct action of the food body on the surface layer (ectoplasm) of the ameba. The presence of the food body, he held, produced a lowering of the surface tension of the ameba thus causing the ameba to flow around it (’98, p. 207). Subsequently, however, he (’14) came to the conclusion that many amebas cannot have fluid surfaces as usually understood, since they do not spread as a film over water when they come into contact with the surface. From this and other observations Rhumbler concluded (’14, pp. 501-514) that the surfaces of amebas are not to be compared with surface tension films on drops of inert simple fluids; but with the surface films of emulsions which take on the properties of a solid. Since the question of ameboid movement is not especially discussed in this later paper, it may be assumed that in this respect his (’98, ’10) earlier views have not been materially modified. Rhumbler has suggested a great many physical models for the explanation of various ameboid activities such as feeding, defecation, movement and so forth.

In general agreement with Bütschli and Rhumbler were Verworn (’92), Blochmann (’94), Bernstein (’00), Jensen (’01, ’02), and recently Hirschfeld (’09) and McClendon (’12). All these authors held that ameboid movement is a surface tension phenomenon. The application of the surface tension theory in explaining ameboid movement demands a fluid surface and a fluid interior and it is perhaps unnecessary to add that Bütschli, Rhumbler and the others mentioned held that the protoplasm is fluid. The question as to whether protoplasm is a fluid or possessed of an internal structure was however hotly debated and we find Fleming (’96), Heidenhain (’98), Klemensievicz (’98), Dellinger (’06) and others opposing the group of authors just mentioned, by contending that the streaming protoplasm must have some kind of structure. This question no longer concerns us however, owing to our rapidly increasing knowledge of colloidal solutions, for it is undoubtedly correct to hold that protoplasm is colloidal.

We have already insisted (p. 46) that the problem of ameboid movement is made more difficult by narrowing it down to the movements of ameba, and that to see the problem in its fullest aspect requires consideration of streaming protoplasm wherever found. Now it happens that there is in certain respects greater diversity of streaming to be found in plant cells than in animal cells, and it is not surprising therefore that explanations of streaming and ameboid movement have taken a different direction among botanists than among zoologists. It is for this reason doubtless that Ewart (’03), while espousing the surface tension theory as explaining streaming, does not look to the superficial surface of a plant cell as the source of the necessary energy, but to the interior of the protoplasm. This idea is, of course not entirely original with Ewart, for Bütschli, as we saw, believed that protoplasm has an emulsoid structure; but according to Bütschli’s hypothesis, the surface forces were not brought into play in movement until the droplets of enchylema spread over the surface and so reduced the tension. Ewart, however points out that there is very much more surface energy present in the interior of streaming protoplasm than is required for all the movements known to protoplasm, including muscular contraction. According to Ewart’s hypothesis the emulsion globules (disperse phase) have their surface tension lowered at corresponding points by electrical currents traversing the endoplasm, the electrical currents themselves originating in chemical actions.

While all available evidence from the study of colloidal solutions and from observation from protoplasm confirms Ewart’s statement that more than sufficient energy is available in the interior of colloids for all purposes of movement, there is little or no evidence that the proper electrical currents are present to release or transform the surface energy into that of movement. This step in his explanation is therefore highly hypothetical and at present unconvincing. Moreover, this step in the theory would not be applicable to streaming as observed in amebas, without very considerable modification.

Recently Hyman (’17) has developed the surface tension theory of movement in the direction indicated by Ewart. The motive power is supposed to have its source in the contractility of the ectoplasm. The endoplasm is held to be a passive stream, not an active stream as Ewart supposed to be the case in plant cells. The power of contractility is held to be due to the process of gelation of endoplasm into ectoplasm, which is due to a change of phase, the fluid part of the endoplasm becoming dispersed and thereby developing surface energy in proportion as the amount of surface of the fluid is increased. This increase of surface produces the phenomenon of contractility.

Miss Hyman is wrong however when she says (p. 90) that the withdrawal and contraction of pseudopods are processes of gelation. This is clearly a physical impossibility, for the ectoplasm of the withdrawing pseudopod must become liquified into endoplasm, before it can be withdrawn. All writers excepting Jennings and Hyman are agreed on the continual transformation of ectoplasm into endoplasm at the posterior end while the reverse process goes on at the anterior end; and Hyman herself states (p. 89) that new ectoplasm is formed as the growing pseudopods extend into the water. So there must be liquefaction of the ectoplasm in withdrawing pseudopods, or very soon the whole ameba would be transformed into ectoplasm. As was shown in the preceding pages, liquefaction of the ectoplasm at the posterior end goes on at the same rate as gelation of the endoplasm at the anterior end. But at another place Hyman says:

“In fact according to Jennings, Dellinger, Gruber, and Schaeffer the surface of the ectoplasm actually flows forward at about the same rate as the forward advance, and this indicates that the advancing ectoplasm at the tip of the pseudopodium is derived from the surface ectoplasm and not from a transformation of endoplasm into ectoplasm at the end of the pseudopodium as Rhumbler supposed” (p. 89).

This quotation is not strictly accurate. Jennings says: “The pseudopodium grows chiefly from the base, so that any part of the surface retains nearly its original distance from the tip” (p. 156). Dellinger in a general way confirmed Jennings’ conclusions. Gruber concluded that the outer layer was gelatinous, not protoplasmic. Schaeffer held the third layer to be extremely thin, “too thin to be seen easily,” so it is impossible that the ectoplasm at the tip of a pseudopod, the thickness of which is readily seen, can be derived from the surface film.

The main conclusion however in Miss Hyman’s paper is that there exists a metabolic gradient in the pseudopods of advancing amebas, the highest rate of metabolism being at the tip and the lowest at the base, for any one pseudopod. This conclusion is bound to be of the first importance in the explanation of ameboid movement. It will give our first real insight into the chemistry of ameboid movement. The fact that her method of demonstrating gradients has yielded uniform results in the hands of Child (’15), who originated it, as well as in her own when applied to a great many different organisms, entitles her conclusions to careful examination.

Figure 30. Disintegration of an ameba in ¼ molecular KNC. After Hyman. a, ameba flowing in the direction of the arrow. b, the ameba has abandoned pseudopod 1 and flows into pseudopod 2, which has become reactivated. The ameba was exposed to KNC at this stage and, as is usual in such experiments, the posterior end at x becomes active. c, the youngest pseudopod, at x, disintegrated first. d, the next youngest pseudopod, 2, disintegrated next. Pseudopod 1, the oldest, disintegrated last.