Earlier workers examined the behaviour of foreign particles placed as markers on aggregates of D. discoideum that were migrating over the surface of the culture plate (Bonner, 1959; Francis, 1959, 1962). Comparable observations, made on aggregates in other conditions and at other stages, have now provided further information about the movement of individual cells within the aggregates. Before reporting them, the course of development must be described in some detail.

Development of the grex

During aggregation on an ordinary culture plate, D. discoideum amoebae crawl towards centres, in which they pack themselves together, forming rounded aggregates of no fixed shape. Papillae develop on the side of the aggregates away from the agar, and by extension, roughly perpendicular to the substratum, transform them into cylindrical multicellular organisms with tapered tips (Text-fig. 1, A-E). Such an organism, which contains from a dozen to a few hundred thousand cells, has been named a grex (Shaffer, 1962) because ‘aggregation’ is derived from the Latin aggregare, to form a grex. A fully formed D. discoideum grex loses its perpendicular posture and migrates over the surface of the agar, at up to 2 mm./hr., lying either flat along its entire length, or with a variable amount of its anterior section sticking up into the air (Text-fig. IF, G). The migrating grex does not make a stalk, and is often called a slug. Eventually migration comes to an end, and the grex starts to culminate (Text-fig. 1H-M): it rounds up, and its axis once more becomes perpendicular to the substratum. This is accomplished by the tip ceasing to move forward, and the rest of the mass gathering underneath it. The first sign of the stalk is a thin-walled, axial, cellulose tube, or stalk sheath, within the tip. This moves towards the substratum and soon makes contact with it; meanwhile the grex temporarily decreases in height and increases in girth. The stalk sheath is continuously extended at its apical end, and the cells trapped inside it become vacuolated, cellulose-walled stalk cells. The grex elongates again, though to a lesser extent than before, and soon, as it travels up the stalk that passes through the middle of it, its rear or basal end leaves the ground. The posterior cells differentiate into spores, and when all the anterior cells have entered the stalk, the fruiting body is complete (Raper, 1935; Bonner, 1944, 1952, 1959; Raper & Fennell, 1952; Bonner, Koontz & Patton, 1953).

TEXT-FIG. 1.

The development of the grex. A: A hemispherical aggregation centre, now without aggregation streams, or with none in the plane of the outline. B-E: It forms a polar papilla and elongates away from the substratum to turn into a grex—at this stage, a slug. F, G : The fully formed slug bends over till it lies on the ground and then migrates, leaving the collapsed slime sheath behind it. H-I : At the end of migration, the grex shortens and becomes erect again. J : The grex decreases in height, and the stalk reaches the ground. K-L: The grex elongates again and climbs off the ground. M : The resultant fruiting body.

TEXT-FIG. 1.

The development of the grex. A: A hemispherical aggregation centre, now without aggregation streams, or with none in the plane of the outline. B-E: It forms a polar papilla and elongates away from the substratum to turn into a grex—at this stage, a slug. F, G : The fully formed slug bends over till it lies on the ground and then migrates, leaving the collapsed slime sheath behind it. H-I : At the end of migration, the grex shortens and becomes erect again. J : The grex decreases in height, and the stalk reaches the ground. K-L: The grex elongates again and climbs off the ground. M : The resultant fruiting body.

A grex is surrounded by a delicate membrane, the so-called slime sheath, which is less than 0·1 μ thick (Raper, 1935; Francis, 1962). It is particularly obvious during the migration stage because the slug creeps along inside it, extending it at the front end and vacating the older part, which remains behind, collapsed like empty dialysis tubing, as a slime track. The slug does not have to lie on a continuous surface to advance : it can travel through the air supported for only a few per cent, of its length by contact with fine fibres, such as fungal hyphae. The vacated slime sheath that is left spanning the gaps between the supports is then not a flat ribbon, but a very fine thread.

Using Lycopodium spores and carbon as external markers, Bonner (1959) found that when a slug was migrating on agar, the slime sheath remained stationary relative to the ground. He concluded that it provided the necessary mechanical support for the cells to advance inside it. Francis (1959, 1962), too, found that foreign spores placed anywhere on a slug but on the tip remained stationary relative to the ground. Those put on the tip were carried forward for a brief period, but they advanced less rapidly than the tip did, and came to rest a short distance behind it. I confirmed these results (Text-fig. 2A).

TEXT-FIG. 2.

Diagrammatic comparison of the behaviour of particles adhering to the slime sheath, when the slug is migrating over the agar (A), and when it is hanging down from it (B). In each case, successive stages are shown from left to right.

TEXT-FIG. 2.

Diagrammatic comparison of the behaviour of particles adhering to the slime sheath, when the slug is migrating over the agar (A), and when it is hanging down from it (B). In each case, successive stages are shown from left to right.

For the following work, Dictyostelium discoideum was grown on Escherichia coli on non-nutrient agar or on phosphate-buffered agar containing 0·5 per cent, glucose and 0·5 per cent, peptone. The markers placed on the slime sheath were usually particles of carmine, some about the same size as the cells, others larger. Lycopodium spores were also used, with similar results. When necessary, the markers’ movements were measured with a graduated eyepiece, or recorded with a camera lucida, or camera; but, in general, they could be appreciated without these aids. Aggregates of all sizes behaved in the same way. In experiments in which it was desirable to minimize orientation responses to light and temperature gradients, the culture plates were kept, between observations, in cardboard boxes in black envelopes in a drawer lined with expanded polystyrene.

Hanging slugs

Although culture plates of D. discoideum have often been incubated upside down, no one has remarked on the problems of an inverted life, and this is perhaps because the organism succeeds in producing entirely normal fruiting bodies in these conditions. It succeeds because the developing fruiting body is not oriented by the force of gravity (Raper, 1935). Nor does gravity have any effect during feeding or aggregation, when the cells are crawling about on the agar surface. It obtrudes appreciably only on the elongated slug.

In an inverted culture place, the slugs hung straight downwards when first formed. Although most of them did manage to become horizontal again and to migrate along the agar surface—just how the tips were guided back to the agar will be considered in another paper—here we are concerned with the fate of those slugs that either remained hanging vertically from the agar ‘ceiling’ for an hour or two, and sometimes considerably longer, after their initial elongation, or returned to this position during migration.

Even when the plate was not inverted, slugs when first formed, regularly raised themselves straight up into the air till only their rear ends rested on the agar. The fact that these stalkless bodies could not lift themselves entirely off the agar surface was not surprising, because their sheaths alone were obviously not sufficiently resistant to longitudinal compression for the slugs to climb them. But one might perhaps have supposed that a hanging slug would have had no difficulty in climbing down from the agar ‘ceiling’ by descending the slime sheath it was ‘spinning’. In reality, not one hanging slug was seen to do this, unless its tip first managed to adhere to some other fixed surface, nor in inverted cultures that had produced a total of > 104 slugs, were any found to have done it.

What, then, happened to the cells in a hanging slug? They were certainly not completely immobilized. In the first place, the slugs often developed curvatures, though this could have been due to very minor changes in cell position. Also, they shortened in response to slight desiccation, and in a nearly saturated atmosphere they could become very long and narrow: among 350 hanging slugs examined in inverted cultures that had been incubated in darkness and contained all stages between aggregation and mature fruiting bodies, one slug had a length as great as 5 mm., and several specimens had attained a length about forty times their breadth. What was not determined was the extent to which these slugs had been lengthened by cells arriving after they had been formed. There was no doubt that hanging slugs that were already sausage-shaped could elongate considerably without any further influx of cells. Yet it was also clear that elongation did not continue indefinitely, for slugs that had hung for many hours had relatively thick bodies, with many kinks and ‘warts’. We may note that slugs in an uninverted plate, when they had just been formed and were still erect, tended to be rather narrower than those migrating flat on the agar surface; so, too, did slugs creeping over aerial fungal hyphae. What these narrow slugs had in common was their minimal contact with the substratum.

Was it true, then, that in so far as a hanging slug was changing its length, some of its cells did move, but that otherwise its cells did not show locomotory activity ? This question was examined by using the behaviour of the slime sheath as an indicator of cell movement.

Lycopodium spores and carmine particles placed on the tip or anywhere on the sides of a hanging slug (more than thirty slugs observed) travelled towards its base and then came to rest on the agar beside it (Text-fig. 2B). Clearly, the slime sheath was still being produced at the tip, and it was then moving towards the base and piling up there. This movement was approximately as rapid as the slug could migrate when lying on the agar. In very old hanging slugs, the slime sheath had accumulated at the base in sufficient quantity to be readily visible as a solid, gnarled, peduncle, nearly as wide as the slug. Thus, whereas in normal migration the lateral slime sheath adhered to the ground and remained stationary while the slug advanced inside it, a hanging slug remained almost stationary while its sheath travelled backwards along it. There could be little doubt that this retrogression was due to the locomotory activities of the cells.

To see whether the cells next to the sheath were being carried back with it or travelling along some other regular path, minute drops of a dilute solution of Nile-blue sulphate were deposited with a fine pipette on various regions of ten hanging slugs. These stained some of the underlying cells in very limited areas. Markers were placed on each slug near the stained cells. By the time the markers reached the base, the stain had become slightly more widely distributed, but the stained cells had not moved backwards or forwards as a group.

Normal development

The foregoing results raised the question of the extent to which similar concealed locomotory activity was a feature of normal development in an un-inverted plate at stages when the longitudinal axis of the aggregate was per-pendicular to the agar.

The first ‘perpendicular’ stage was the formation of the grex. The details varied slightly with the conditions of culture. When there was plenty of space between the cells before aggregation, a centre was approximately hemispherical, except where aggregation streams were flowing into it. A papilla developed at its pole and extended at right angles to the agar until it had incorporated all the cells in the centre (Text-fig. 1A-E); and during this process the rest of the aggregate not only decreased in volume, but also, to a variable extent, increased its height relative to its width, so that it approximated to a more complete sphere.

Within about hr., markers placed on the tip of a young papilla were carried down it on to the spheroidal part of the aggregate, and then down the side of this till they reached the agar (Text-fig. 3B). Similar results were obtained when the transformation into a slug was more advanced (Text-fig. 3C, D). It was more of a surprise that markers deposited on the pole of a hemispherical aggregation centre were carried down to the agar even before a papilla was visible (Text-fig. 3A). These results could all be obtained in the absence of any further influx of cells into the centres.

TEXT-FIG. 3.

The behaviour of a marker placed on the apex of: A, a hemispherical aggregation centre; B, a centre that has formed a papilla; C, an aggregate well-advanced towards becoming a slug; D, a fully formed slug that has not yet begun to migrate over the sub-stratum. Diagrammatic.

TEXT-FIG. 3.

The behaviour of a marker placed on the apex of: A, a hemispherical aggregation centre; B, a centre that has formed a papilla; C, an aggregate well-advanced towards becoming a slug; D, a fully formed slug that has not yet begun to migrate over the sub-stratum. Diagrammatic.

When the nutrient was sufficiently rich for the pre-aggregation cells to form a continuous layer, the initial aggregates were flattened masses with founded but irregular perimeters. If not too large, such a mass would produce a single papilla roughly in the middle of its free surface and transform itself into a single slug. The surface of a larger aggregate developed two or more hillocks, and the valleys between them gradually deepened. At a variable stage of separation, each hillock produced a papilla at its apex and then turned into a slug.

When markers were scattered over the whole of a large, smooth aggregate before this had developed any papillae, they were all carried down to the agar directly if only a single slug resulted; but, in those cases where the aggregate broke up into a number of hillocks, some of the makers piled up in the clefts between them and reached the agar only later when cleavage was completed (Text-fig. 4; Plate 1).

TEXT-FIG. 4.

The pattern of movement of markers placed on different parts of the surface of a large aggregate at two stages of its transformation into three slugs. Diagrammatic.

TEXT-FIG. 4.

The pattern of movement of markers placed on different parts of the surface of a large aggregate at two stages of its transformation into three slugs. Diagrammatic.

It may be added that markers placed on a large aggregation stream soon came to rest on the agar at the edge of the stream. However, their actual displacement relative to the ground was very slight. What happened was that the course of the stream shifted laterally until it was clear of them. Presumably the cells found it more difficult to crawl forward under the foreign bodies than on the bare agar alongside them. In contrast, when markers were placed on the middle of large, old Polysphondylium violaceum aggregation streams, neither the markers nor the streams changed their positions. The reason why these streams continued to flow forward underneath the markers was because they, unlike those of D. discoideum, were encased in slime sheath. (These results, which refer to streams in air, are to be distinguished from those obtained using very much finer markers and under-water cells (Shaffer, 1963).)

The second ‘perpendicular’ period began when migration ended. Throughout the stages when the grex was re-erecting itself and shortening (Text-fig. 1H-J), particles placed on any part of the sheath, including that covering the tip, were carried down to the agar; they could travel the entire length of the grex in about 1 hr., and sometimes took less. It was expected that once the base of the stalk had been grounded, the stalk would provide an adequate surface over which the grex could advance, and that therefore the slime sheath would no longer be transported towards the agar. However, to get better records, experiments on early culmination were repeated using grexes culminating horizontally from vertical agar surfaces; and it was then discovered that markers placed on the tip continued to be transported to the agar even after a grex had passed the phase of minimal height—when the bottom of the stalk reached the agar (Text-fig. 1 J)— and had begun to elongate again (Text-fig. 5). (In earlier papers, the word ‘grex’ was used unchanged in the plural. However, it now seems preferable for the plural to be grexes.)

TEXT-FIG. 5.

Behaviour of markers added at intervals in early culmination, while the grex first decreased in height and then began to rise again. Numbers indicate minutes between successive drawings.

TEXT-FIG. 5.

Behaviour of markers added at intervals in early culmination, while the grex first decreased in height and then began to rise again. Numbers indicate minutes between successive drawings.

It seemed just possible that at this stage the stalk sheath might be behaving like the slime sheath—being carried away from the tip, and piling up at the base. If so, it would have had to slip over the stalk cells inside it. This appeared to be possible, since these were still little differentiated, and Whittingham & Raper (1960) had shown that immature stalk cells could readily be squeezed out of the tip of the stalk sheath, and Gezelius (1959) had discovered that it was only as the stalk cells matured that they became bound to the sheath by a secondary deposit of cellulose. As the primary sheath was strongly positively birefringent, if it accumulated at the base, either folded or merely crumpled, it should have been easily detectable there with a polarizing microscope. In fact, at the extreme base the sheath decreased rather than increased in quantity, and remained positively birefringent with respect to the stalk’s axis. The final blow to the hypothesis of a moving stalk sheath came when later stages of culmination were investigated. It was discovered that particles were still carried back to the agar when the grex had fully elongated again, by which time the more basal part of the stalk was well differentiated (Text-fig. 6).

TEXT-FIG. 6.

Behaviour of markers placed on the same grex as in Text-fig. 5 during later culmination. Before sporulation, markers on all parts of the grex were carried towards the substratum and deposited wherever the rear end of the grex happened to be when they reached it. After sporulation, markers on the pre-stalk cells still moved towards the sub-stratum until they encountered the rising spore head; they were then lifted up on it. Numbers indicate minutes between successive drawings.

TEXT-FIG. 6.

Behaviour of markers placed on the same grex as in Text-fig. 5 during later culmination. Before sporulation, markers on all parts of the grex were carried towards the substratum and deposited wherever the rear end of the grex happened to be when they reached it. After sporulation, markers on the pre-stalk cells still moved towards the sub-stratum until they encountered the rising spore head; they were then lifted up on it. Numbers indicate minutes between successive drawings.

As a grex climbed the stalk, it continued to make slime sheath, which initially was left behind as an outer envelope around the stalk and was later severed by the mass of spores (Raper & Fennell, 1952). Markers on the grex were still trans-ported some distance towards the agar, even after the base of the grex had begun to move up the stalk (Text-figs. 6, 7A). The particles always became stationary when they reached the base of the grex, and they were then deposited on the stalk (or basal disc) at that level. When the spores differentiated, the rear of the grex became opaque, and markers on its outside were no longer carried towards the agar. The precise stage at which this happened, in terms of the relative distance that the grex had climbed the stalk, was rather variable. Sometimes it was when only the basal disc had been uncovered by the receding grex; sometimes not till there was also a short length of stalk exposed. This may account for the controversy between Raper & Fennell (1952) and Bonner (1944, 1959) as to the precise timing of sporulation. We may presume that the slime sheath stopped moving when a sufficiently large proportion of the cells in contact with it had turned into spores. Eventually the mass of spores became lemon-shaped. The adoption of a form determined by surface tension showed that the slime sheath had disappeared. Markers that were on the outside of the spore head at this stage then retained an approximately constant position on it as it was carried up the stalk; they therefore now moved away from the agar (Text-fig. 7B). Particles on the outside of the small cylinder of the remaining prestalk cells, which was still clad in slime sheath, were not only left behind the apex as it advanced, but continued to be transported very slightly towards the agar. When the rising spore head made contact with them, they began to be carried upwards (Text-fig. 6, 7B). They remained near the stalk, unless some more apical particles displaced them from this position and pushed them towards the equator of the spore head.

TEXT-FIG. 7.

A, markers placed on a grex before sporulation were carried towards the sub-stratum and left on the stalk. B, Markers on another grex after sporulation were carried towards the substratum when on the pre-stalk region, away from the substratum when on the spore head. Time intervals in minutes.

TEXT-FIG. 7.

A, markers placed on a grex before sporulation were carried towards the sub-stratum and left on the stalk. B, Markers on another grex after sporulation were carried towards the substratum when on the pre-stalk region, away from the substratum when on the spore head. Time intervals in minutes.

It was thus clearly established that the slime sheath continued to be moved towards the agar until the approximate moment when the underlying cells differentiated into spores or entered the stalk. .

Except in so far as the anterior cells were recruited into the stalk, small numbers of superficial cells, vitally stained by applying droplets of Nile-blue sulphate to various regions of culminating grexes, remained in approximately the same position within a grex, though they gradually became more scattered, just as in hanging slugs.

An incidental observation may be recorded here. In stained cells anywhere along a grex but at the apex, the stain was concentrated in large, dark-blue granules (or possibly vacuoles), and the rest of the cytoplasm was nearly colour-less. When a cell reached the apex, where the stalk sheath was secreted, the granules disappeared and the whole cell became a pale greenish blue. Another curious feature was that the apex, once stained, tended to retain its diffuse colour even after only unstained prestalk cells were reaching it.

One hypothesis that could account for the continued transport of the slime sheath towards the agar during culmination was that the grex was essentially climbing a ‘greasy pole’, or, in other words, that it was not that the stalk sheath slipped on the cells inside it, but that the cells outside it slipped on the stalk sheath. If this really did happen, a grex cell could adhere only very weakly to the stalk. This could indeed be demonstrated very strikingly in D. mucoroides, a species in which the grex produced a long stalk before the spores differentiated. If one transversely severed the slime sheath over the rear end of a grex that was lying or had been laid on its side on the agar, and then pulled on the exposed stalk, the rest of the stalk readily came out from inside the grex without a single cell sticking to it externally, except at its tip. There it was slightly expanded, and the pre-stalk cells might have been expected to be able to hang on simply by adhering to the cells inside the open end of the sheath. Although the rate of movement in culmination was enormously less than the rate at which the cells were artificially made to slide along the stalk, this operation did show that the cells adhered to one another and to the slime sheath much more strongly than to the stalk sheath. The same result could be obtained with D. discoideum; but because the operation had to be performed before the spores differentiated, this species provided less suitable material, since there was then only a very short length of stalk exposed, and one had to sever the slime sheath while gripping this stalk without carrying any loose cells along it.

In the original experiments, only one or two markers were placed on each grex. When more were used, the results were rather surprising. The grex tip being the site where the slime sheath was continuously being extended, it was expected that if a number of markers were placed on the extreme tip, they would gradually separate from one another as the underlying sheath increased in area, and that this displacement would be radially symmetrical about the axis of the grex, with the markers spreading over all sides of the tip on their journey towards the base of the aggregate. The first hint that this might not happen came from the observation that if several markers were placed on the tip of an incipient grex, or of one that was hanging or in early culmination, when they reached the agar, they very commonly did not lie all around the circumference of the case, but were all, or nearly all, clustered at one side of it. Further investigation showed that if markers were dotted about over the whole of the leading face of the tip, they were some-times carried backwards in a fairly symmetrical fashion; yet, in the majority of cases, the entire field of them was displaced to one side of the tip and then travelled back along one side of the grex (Plate 2). However, even when transported unilaterally the individual markers could still move apart from one another during the course of the journey (Plate 2, Figs. B, C) (though those on the same ‘longitude’ were brought together at the end of it, if the base was still resting on the agar). Between observations, these cultures were kept in the dark.

When the grex was elongated, it was difficult to analyse this asymmetrical behaviour because of the slight sideways shifts the tip was prone to undergo even in the absence of markers. But when the grex was compact, it was clear that the tip, though it continued to point in much the same direction as before, was itself displaced relative to the rest of the aggregate and to the ground (Plate 2). This displacement was in the opposite sense to that of the group of markers. In other words, the advancing tip outflanked the markers, though this was disguised by their own movement away from the axis of the aggregate. This phenomenon was further investigated with grexes culminating horizontally and kept in darkness between observations; it was then observed that even relatively small single markers deposited on one side of the tip could deflect it to the opposite side (Text-fig. 8).

TEXT-FIG. 8.

Single markers placed laterally on the tip of a grex that was elongating again after grounding the stalk deflected the tip to the opposite side, while themselves being transported to the substratum. Time intervals in minutes.

TEXT-FIG. 8.

Single markers placed laterally on the tip of a grex that was elongating again after grounding the stalk deflected the tip to the opposite side, while themselves being transported to the substratum. Time intervals in minutes.

Several workers have discussed the locomotory behaviour of cells in the interior of a slug migrating over the ground (Raper, 1956; Bonner, 1959; Francis, 1959). It is generally believed that all the cells crawl forward actively, although the possibility that, in the main, only the cells in contact with the slime sheath are active and that the interior cells are carried along passively, has not been rigorously excluded. Labelling the cells with pigment reveals that the great majority of them retain their relative positions, at least for several hours (Raper, 1940). The fact that a small proportion of them advance relative to the others (Bonner, 1952) shows only that slow movement within the aggregate is possible. Also a few of the interior cells move towards the rear of the slug (Bonner, 1952); but only if it could be shown that these were not crawling in that direction would this prove that all the other cells were actively advancing. The best evidence that the internal cells are active is that the larger a slug is, the faster it moves (Bonner, Koontz & Paton, 1953; Francis, 1959, 1962).

If, in fact, all the interior cells do crawl forward on average at the same speed as the slug advances, it would be simple to account for this behaviour if the lateral surface of each cell was stationary and remade each time the cell advanced its own length, being created at the cell’s front end and resorbed at its back end. Conversely, if, as a certain amount of evidence suggests (Shaffer, 1963,1964a—d), the cell surface does behave in this way, it is difficult to explain how any cells not actively advancing could be carried forward rather than left behind. The alternative possibility is that the cell surface is much more permanent and subjected to some pattern of deformation during locomotion; but if it is, it is somewhat harder to visualize how cells could rapidly move forward when lying on top of one another.

In normal migration the lateral slime sheath lies on the ground and remains stationary while the slug crawls forward inside it. In a hanging slug it is carried back to the rear end, while this scarcely advances at all. A whole hanging slug thus somewhat resembles a single Amoeba prevented from getting sufficient traction on the substratum. Instead of advancing while the ectoplasm remains stationary, such an Amoeba remains stationary while its ectoplasm moves back-wards, the motor mechanism presumably being unchanged (Schaeffer, 1920). We may reasonably assume that the sole difference between a hanging and a migrating slug is the behaviour of the lateral sheath as determined by whether enough of it is anchored to the substratum. If so, we may take the displacement of the hanging slug’s sheath as an indication that all its cells are undergoing normal locomotory activities—if we accept that all the cells are active during migration.

The remaining question about hanging slugs is why they cannot escape from the agar ceiling by climbing down their slime sheaths. Presumably the back end of a slug adheres too strongly to the part of the slime sheath that covers it, and the latter to the superstratum. There seems to be no theoretical objection to the rear surfaces of the hindmost slug cells adhering to the sheath and remaining quiescent while the lateral surfaces of these same cells, whether in fact temporary or permanent, take part in normal motor activity. Adhesion between the rear end of a slug and the sheath covering it must presumably be a significant braking force also during migration. This brake will be additional to the brake at the anterior end, previously postulated in analyses of migration (Francis 1959,1962; Shaffer, 1964e), that results from the necessity of expanding the slime sheath there.

Yet why should the area of adhesion at the rear of a hanging slug not be progressively reduced to vanishing point as the hindmost cells, advancing at slightly different rates over their neighbours, one by one detached themselves from the critical part of the slime sheath ? This would be but a continuation of the process by which the broad base of an initially hemispherical aggregation centre, which adheres everywhere to the substratum, is greatly reduced in area as the mass turns itself into a slug. Undoubtedly hanging slugs continue to elongate for a considerable time, and can become proportionately much narrower than those migrating flat on a surface. Nevertheless the maximum ratio of length to breadth observed was still several orders of magnitude less than it would have been had the cells drawn themselves out into a single file. It would seem that there must be some additional factor opposing elongation. Such a factor has already been postulated to act in normal migration, because previous models of slug organization apparently led to the conclusion that a slug should have become ever longer and thinner till it was a single line of cells (Shaffer, 1964e); and the existence of a brake acting on the rear cells of the slug aggravates this problem. But it has not yet been experimentally determined whether or not a normal slug does tend to an equilibrium length in given conditions. It would also be interesting to study individual hanging slugs in an environment free from disturbing gradients to find out at what rate they can lengthen, and whether there really is any limit to their ultimate extension.

The retrogression of the slime sheath wherever it occurs may be taken as an indication of an equivalent amount of concealed motor activity in the opposite direction by the underlying cells. This concealed movement is to be distinguished from that reported by Bonner (1952): the changes in relative cell position re-vealed by vital staining. Bonner & Adams (1958) found that the most extensive cell redistribution in development occurred during the slug’s initial elongation (a stage they referred to as ‘the end of aggregation’). Their results were complicated by the fact that the cells in question were entering the slug from aggregation streams, and that it was just possible that the new arrivals might simply have been collecting as an outer layer around the earlier ones. In any case, even if the cells were sorting themselves out by competitive overtaking, the length of the aggregate was apparently the maximum distance any cell could have travelled. The motion during this stage revealed by slime-sheath markers is equivalent to a distance many times greater. It is not dependent on the influx of new cells.

The initial changes in the shape of the aggregation centre are extremely slow compared with the rate of cell movement in aggregation, and there was previously the problem of accounting for the apparent quiescence of the cells on arrival at the centre. It now appears that the movement of the cells, either separately or in streams, along the agar towards the centre, is replaced by concealed motor activity directed away from the agar and towards the apex of the aggregate. Subsequently, elongation results from some of the cells moving forward on the others, but the impression one receives that the cells in the spheroidal part of the mass remain very much less active until their turn comes to enter the base of the cylindrical body that is drawing itself out from its pole, must again be incorrect. This concealed motor activity, which continues throughout slug formation, clearly increases the cells’ opportunity of sorting themselves out by competitive overtaking at this stage. It partly or completely replaces actual forward progress in all later stages when the grex axis is perpendicular to the substratum.

At the end of migration the grex rounds up, and the tip, which in the extreme case has been lying flat on the agar surface, rises to the apex of the mass and, in doing so, acquires a posture perpendicular to the substratum. So far as I am aware, the mechanism of this change in orientation has not previously been discussed. At the outset of this work, it seemed possible that the tip moved or was moved over the surface of the rounded, major part of the aggregate, while this remained relatively static; or even that all the cells in the major part of the aggregate started rotating about a horizontal axis that passed transversely through it. With both these possibilities, there was the problem of how such movement could be related to the normal cell movement involved in the locomotion of the grex, and of what halted it once the tip had reached the erect position. The cause of re-erection has become less obscure now that we know that throughout the ′ process the grex cells remain oriented towards the tip and in rapid motion relative to the sheath. All that is really required for re-erection is that the cells, in crawling a distance roughly equivalent to the length of the grex, should eliminate radially asymmetrical features in the environment of the radially symmetrical aggregate, the significant environmental asymmetry perhaps being chemical or even largely mechanical.

A single relatively small particle placed on the side of the tip can deflect it; this shows the sensitivity of the reaction (Text-fig. 8). But what may well be more revealing about the nature of the reaction is the behaviour of a number of fine carmine particles scattered over the whole of the leading face of a grex tip. Instead of being carried back on all sides of the tip, they commonly travel as a group down one side of it, while the tip itself is displaced in the opposite direction. The fact that the particles need actually cover only a small fraction of the area over which they are initially distributed suggests that the effect is mediated through the slime sheath. Raper (1940) was the first to suggest that the slime sheath might be important in governing migration and the onset of culmination, and Francis (1959,1962) assigned it a major rôle in his quantitative theory of slug locomotion, and postulated that alterations in its strength influenced a slug’s shape and orientation.

If the sheath became somewhat less extensible within a short distance of each marker placed on the grex tip, it would favour the advance of cells at the periphery of the area in which a group of markers was deposited—the sector in which the advance did occur presumably being determined by some minor asymmetry. The effect of the sheath could be due purely to contact with the markers, or perhaps to slight desiccation acting directly on the sheath or on the secretory activity of the underlying cells. The application of the markers might be thought to be traumatic. The fact that they can move apart from one another on their journey to the base does not really contradict this, as the damage might have been repaired. However, a grex can be loaded with much greater numbers of particles without producing any other developmental effect; and, of course, a grex may itself collide with an obstacle or, without experimental interference, abruptly come in contact with the substratum. It is therefore likely that the fate of groups of markers does throw light on the mechanism of a grex’s normal orientation responses to the substratum.

It has also been noticed that contact with the substratum may considerably affect a slug’s length : slugs that have minimal contact, because they are projecting straight out from it or are supported by only a few fine filaments, being regularly long and narrow, unless desiccated. Again it may well be by an effect on the slime sheath that more extensive contact can produce a fatter, shorter slug.

However, there are considerable difficulties in trying to asses the slime sheath’s rôle in development (Shaffer, 1964e). We still have almost no information about its mechanical properties when it is in contact with the cells that are making it. One relevant new observation is that it can be considerably altered when a long way behind the tip; thus, in early culmination, its diameter increases several times while it is being passed back over the pre-spore cells. It would be easier to in-vestigate its importance if there were some means of removing limited areas of the sheath without serious mechanical disturbance of the underlying cells. It may be worth suggesting that this might possibly be effected by applying drops of the fluid in which the spores are suspended; for it is presumably this fluid, rather than the cellulose-walled spores, that dissolve the overlying sheath in the normal course of culmination. In this, it might act enzymatically or perhaps by virtue of its being highly surface active. This latter property does not seem to have been remarked, but it can be easily demonstrated by depositing a drop on the moist surface of the culture plate : a wave of fluid is driven outwards at its expanding margin, leaving the agar in the middle visibly drier.

How does a grex rise up its stalk? This is a problem on which divergent opinions have been expressed. Raper & Fennell (1952) concluded that the pre-spore cells could climb the stalk by their own locomotion, but that the anterior cells could not do so, and that the vacuolation of the stalk cells provided the main lifting force throughout culmination and, after the spores had differentiated, the sole one. Bonner (1959), on the contrary, concluded that the locomotion of the pre-stalk cells provided the only significant lifting force throughout culmination, and that the pre-spore cells were essentially inactive. By extension, he suggested that the young stalk initially was thrust down through the passive pre-spore mass by a force generated at the tip.

Before the current investigation was begun, various theoretical arguments and some experimental evidence already challenged both these interpretations (Shaffer, 1964e). It was concluded, in the first place, that the stalk could descend through a solid mass of cells only if all of the cells, including the pre-spores, actively crawled towards the apex of the grex and thus climbed the stalk. It was also reasoned that the simultaneous flattening of the grex—the shortening of the antero-posterior axis—was due to the rear of the grex advancing more rapidly than its front and, in fact, almost swallowing it before this relationship was reversed and the grex elongated again. Flattening was therefore considered to be merely a continuation of the process involved in rounding up at the end of migration; but whereas the latter could easily be seen to be due to the rear end of the grex continuing to move forward relative to the ground after the tip had stopped doing so (Raper, 1935), it was only a hypothesis that in an early stage of culmination, when none of the grex was visibly climbing away from the agar and the tip was actually sinking a short distance towards it, the cells themselves were crawling perpendicularly away from it. If they were, the slime sheath over both the pre-stalk and pre-spore cells would presumably have been moving towards the base of the grex throughout this period. The behaviour of surface markers has now shown that it does, in fact, do this.

It was expected that as soon as the base of the stalk made contact with the ground, the stalk would serve as an adequate ‘substratum’ for the grex’s advance, and that the retrogression of the slime sheath would stop, just as it does when a slug starts to migrate over the agar. However, retrogression continues. The simplest explanation of this is that the stalk is a ‘greasy pole’ which the cells slide down almost as fast as they climb up it. This may seem rather improbable in that one might have expected the organism to have been able to develop a non-slip surface for its stalk. Yet we may note that an unstated assumption apparently implicit in Raper & Fennell’s belief (1952) that after sporulation the sole lifting force was the vacuolation of cells inside the stalk, was that the remaining grex cells could slide up the stalk. Although it has not been directly demonstrated that the cells do slip on the stalk, what has been shown is, first, that the grex cells adhere relatively poorly to the stalk sheath, and secondly that the adhesion between the rear face of the grex and the slime sheath acts as an important braking force. It is this brake that prevents a hanging slug from climbing down from the agar. This rear brake, operating during culmination, must hinder the cells from climbing the stalk, and so can reasonably account for their slipping on it. There is believed to be an anterior brake also, resulting from the necessity for expanding the slime sheath at the tip, that opposes the cells’ advance (Francis, 1959, 1962; Shaffer, 1964e), but it cannot make the cells slip on the stalk, since the slipping we are postulating can take place only in so far as new slime sheath is actually produced.

If the grex does not, in fact, slip on the stalk, the interior cells must be less active than the outer ones and, when the stalk is first grounded, almost completely quiescent. But, if so, one would expect the outer cells to advance over the inner ones rather than carry back the slime sheath, at least if the lateral cell surface is temporary. If the cell surface is permanent, it is conceivable that the sides of the outermost cells that are in contact with the sheath could be much more active than their other sides. But if the opposite sides of a cell could operate at such different rates, it would be hard to account for the fact that minute markers on the upper surfaces of separate, monopodial, underwater cells remain stationary relative to the substratum (Shaffer, 1963).

The behaviour of the sheath markers has also demonstrated that the underlying pre-spore cells continue to crawl upwards as the grex rises until approximately the moment when they differentiate into spores. As for the pre-stalk cells, the fact that markers deposited on the tip at any time during culmination, including the period after sporulation, do not remain there, but are left behind as it advances, shows that these cells actively climb the stalk to its full height, and that they are not carried passively up it. On the contrary, if, as is probably the case, the cells do slide along the stalk, as they must do if they are to be moved passively along it, the transport of markers towards the agar reveals that the passive movement must be downwards, and that consequently the cells must be forced to climb actively that much further upwards.

After the globular spore head has formed, the markers on the remaining pre-stalk cells travel only a short distance towards the agar. Presumably at this stage it is no longer the slime sheath but the spores that act as a brake on the rear of the pre-stalk-cell mass and cause what slipping there is. When the markers meet the advancing spore head, they, together with any markers already on it, are then 8 carried up with it, maintaining their position on it because the spore head is being lifted passively. We may conclude that a grex rises because all its cells crawl away from the substratum until they differentiate into spores or stalk.

Recently (Shaffer, 1964e) I have drawn attention to the considerable difference between the published velocities of migrating and of culminating grexes. This is probably too large to be due to experimental conditions, yet cell locomotion is supposed to be basically the same in both stages. I suggested that a grex might be slowed down in culmination by having to construct a stalk. But this is unlikely to be a limiting factor, since there would seem to be nothing to stop the grex from advancing faster than the stalk was extended, until the rear of the grex had reached the end of the stalk. It now appears that although grex velocity is much reduced at the start of culmination, average cell velocity, relative to the slime sheath, is not. In so far as the cells manage to extend the stalk, this limits the amount the grex can slip on it; hence the rate of stalk extension should, in fact, set a minimum rather than a maximum for the velocity of culmination. A D. purpureum grex may make stalk intermittently when lying on the ground (Raper & Thom, 1941); it should therefore be possible to determine the effect of stalk manufacture on velocity in the absence of all retrogression.

Lastly, the transport of the slime sheath may well be relevant to a curious problem in the morphogenesis of D. polycephalum (Shaffer, 1962, pp. 176–8). In this species, several grexes may culminate in contact with one another. They start to diverge soon after culmination begins, yet in the final fruiting body their stalks diverge only at a much greater height, being stuck together below this by their enveloping slime sheaths (Raper, 1956). The problem is what ‘zips’ the slime sheaths together above the point of initial grex divergence. Assuming that the individual grexes behave as they do in D. discoideum, we may hazard that the backward transport of the slime sheaths and their accumulation in the angles between the grexes plays an important part in this process.

  1. In an inverted culture plate, a migrating grex, or slug, may hang vertically downwards from its rear end for several hours while its base maintains nearly the same position in space. Instead of advancing inside its external slime sheath, while this remains stationary, as happens during normal migration when lying flat on a substratum, the slug succeeds only in carrying its slime sheath back from the tip where it is made, to the base, where it piles up.

  2. The slime sheath of a rounded aggregation centre, of a centre elongating to form a grex, and of a grex when decreasing in height in early culmination, is also rapidly transported from the apex to the base.

  3. It is concluded that the cells in all these stages are carrying out normal locomotory actions perpendicularly away from the agar, while changing their position in space much more slowly, and in some stages, on average, not at all.

  4. This concealed oriented locomotion increases the cells’ opportunity of sorting themselves out; it helps to explain the re-erection of the grex at the end of migration; and it accounts for the descent of the stalk through the basal pre-spore cells and the simultaneous flattening of the grex in early culmination.

  5. The inability of a hanging slug to climb down from the agar ‘ceiling’ unless its tip first makes contact with a substratum, implies that its rear face is anchored by its adhesion to the slime sheath. This adhesion at the rear presumably acts as a braking force in normal migration and in culmination.

  6. Hanging slugs become unusually long and narrow. Nevertheless there are new indications that some unknown factor opposes elongation in all slugs.

  7. At all levels of a grex, the slime sheath continues to be carried towards the substratum throughout culmination, while undergoing considerable changes of shape, until the underlying cells differentiate into spores or stalk. It is concluded that, contrary to previous views, both the pre-spore and pre-stalk cells actively climb the stalk until they so differentiate, and that they are not pushed up it by the vacuolation of the stalk cells.

  8. Probably a culminating grex continuously slides back down its stalk, at times almost as fast as the cells climb it. Backsliding can be attributed to the slime sheath acting as a brake on the rear, or base, of the grex, and to the cells adhering poorly to the stalk.

  9. The retrogression of the slime sheath may largely account for the difference between the published velocities of migrating and culminating grexes, and for the delay in the complete separation of the component grexes in a D. polycephalum fruiting body after their initial divergence.

  10. A number of particles scattered over a D. discoideum grex tip tend all to be carried down to the base on one side of it, while the tip itself is displaced to the opposite side. This reaction is probably mediated by a modification of the slime sheath and may well be relevant to the problem of grex orientation.

Mouvements cellulaires à P intérieur d′agrégats du Myxomycète Dictyostelium discoideum, révélés par des marques superficielles

  1. Dans une boîte de culture inversée, un agrégat migrateur, en forme de limace, peut pendre verticalement vers le bas, de son extrémité arrière, pendant plusieurs heures, tandis que sa base garde à peu près la même position dans l’espace. Au lieur d’avancer à l’intérieur de son enveloppe externe de mucus alors que celle-ci demeure stationnaire—ainsi qu’il advient pendant une migration normale sur un substrat plat—la ‘limace’ réussit seulement à repousser vers l’arrière sa gaîne de mucus depuis l’extrémité où elle se forme jusqu’à la base où elle s’accumule.

  2. L’enveloppe de mucus d’un centre d’agrégation arrondi, d’un centre s’allongeant pour former un ‘troupeau’ (= ‘grex’, mot latin proposé par l’auteur pour désigner l’organisme intégré que constitue l’agrégat allongé capable de se déplacer), et d’un ‘troupeau’ dont la hauteur diminue au début de la formation du sporocarpe (‘culmination’), se trouve rapidement transportée de l’apex à la base aussi.

  3. On conclut que les cellules à tous ces stades effectuent des mouvements de locomotion normaux, perpendiculairement en s’éloignant de l’agar, tandis que leur position dans l’espace change beaucoup plus lentement et, à quelques stades, en moyenne pas du tout. Cette locomotion orientée et cachée accroît les chances des cellules de se ségréger ellesmêmes; elle aide à expliquer la ré-érection de l’agrégat à la fin de sa migration; et elle rend compte de la descente du pédicelle à travers les cellules de la base qui donneront les spores, et de l’aplatissement simultané du ‘troupeau’ an début de la formation du sporocarpe.

  4. L’inaptitude d’un agrégat allongé, pendant, à descendre par reptation du ‘plafond’ d’agar, à moins que son extrémité ne prenne d’abord contact avec un substrat, implique que sa face postérieure soit ancrée par son adhésion à l’enveloppe visqueuse. Cette adhésion à l’arrière agit probablement comme un frein au cours de la migration normale et de la ‘culmination’.

  5. Les agrégats pendants en forme de limace deviennent exceptionnellement longs et étroits. Il y a néanmoins des indices nouveaux de l’existence d’un facteur inconnu s’opposant à l’élongation chez toutes les’limaces’.

  6. A tous les niveaux d’un ‘troupeau’, l’enveloppe muqueuse continue à être repoussée vers le substrat tout au long de la culmination, tout en subissant des modifications considérables de sa forme, jusqu’à ce que les cellules sous-jacentes se différencient en spores ou en pédicelles. On conclut que, contrairement aux idées antérieures, les cellules initiales des spores et du pédicelle grimpent activement au tube jusqu’à ce qu’elles se différencient ainsi, et qu’elles ne sont pas poussées à son sommet par la vacuolisation des cellules du tube.

  7. Un agrégat culminant glisse probablement vers le bas de son pédicelle de façon continue, par moments presque aussi vite que les cellules grimpent le long du pédicelle. Ce glissement en arrière peut être attribué au rôle de frein que joue l’enveloppe muqueuse à l’arrière, ou à la base, de l’agrégat, et au fait que les cellules adhèrent mal au pédicelle.

  8. Le retour en arrière de la gaîne visqueuse peut largement rendre compte des différences entre les rapidités d’agrégats en migration et en érection, et du retard dans la séparation complète des divers agrégats d’un sporocarpe de D.poly-cephalum après leur divergence initiale.

  9. Des particules parsemées sur l’extrémité d’un agrégat de D. discoideum tendent toutes à être transportées vers la base, d’un seul côté, tandis que la pointe elle-même se déplace du côté opposé. Cette réaction est probablement assurée par une modification de l’enveloppe visqueuse et peut bien se rapporter au problème de l’orientation du ‘troupeau’.

Bonner
,
J. T.
(
1944
).
A descriptive study of the development of the slime mold Dictyostelium discoideum
.
Amer. J. Bot
.
31
,
175
82
.
Bonner
,
J. T.
(
1952
).
The pattern of differentiation in amoeboid slime molds
.
Amer. Nat
.
86
,
79
89
.
Bonner
,
J. T.
(
1959
).
The Cellular Slime Molds
,
150
pp.
New Jersey
:
Princeton University Press
.
Bonner
,
J. T.
,
Koontz
,
P. G.
&
Paton
,
D.
(
1953
).
Size in relation to the rate of migration in the slime mold Dictyostelium discoideum
.
Mycologia
45
,
235
40
.
Bonner
,
J. T.
&
Adams
,
M. S.
(
1958
).
Cell mixtures of different species and strains of cellular slime molds
.
J. Embryol. exp. Morph
.
6
,
346
56
.
Francis
,
D. W.
(
1959
).
Pseudoplasmodial movement in Dictyostelium discoideum
.
M.S. Thesis
,
Wisconsin University
.
Francis
,
D. W.
(
1962
).
The movement of pseudoplasmodia of Dictyostelium discoideum
.
Ph.D. Thesis
,
Wisconsin University
.
Gezelius
,
K.
(
1959
).
The ultrastructure of cells and cellulose membranes in Acrasiae
.
Exp. Cell Res
.
18
,
425
53
.
Raper
,
K. B.
(
1935
).
Dictyostelium discoideum, a new species of slime mold from decaying forest leaves
.
J. agrie. Res
.
50
,
135
47
.
Raper
,
K. B.
(
1940
).
Pseudoplasmodium formation and organization in Dictyostelium discoideum
.
J. Elisha Mitchell sci. Soc
.
56
,
241
82
.
Raper
,
K. B.
(
1941
).
Developmental patterns in simple slime molds
.
Growth
,
5
,
41
76
.
Raper
,
K. B.
(
1956
).
Dictyostelium polycephalum n. sp.: a new cellular slime mould with coremiform fructifications
.
J. gen. Microbiol
.
18
,
16
32
.
Raper
,
K. B.
&
Fennell
,
D. I.
(
1952
).
Stalk formation in Dictyostelium
.
Bull. Torrey bot. Club
,
79
,
25
51
.
Raper
,
K. B.
&
Thom
,
C.
(
1941
).
Interspecific mixtures in the Dictyosteliaceae
.
Amer. J. Bot
.
28
,
69
78
.
Schaeffer
,
A. A.
(
1920
).
Amoeboid Movement
,
156
pp.
New Jersey
:
Princeton University Press
.
Shaffer
,
B. M.
(
1962
).
The Acrasina
.
Adv. Morphogen
.
2
,
109
82
.
Shaffer
,
B. M.
(
1963
).
Behaviour of particles adhering to amoebae of the slime mould Polysphondylium violaceum and the fate of the cell surface during locomotion
.
Exp. Cell Res
.
32
,
603
06
.
Shaffer
,
B. M.
(
1964a
).
Intracellular movement and locomotion of cellular slime-mold amebae
.
In Primitive Motile Systems in Cell Biology
(Eds.
R. D.
Allen
&
N.
Kamiya
), pp.
387
405
.
New York
:
Academic Press
.
Shaffer
,
B. M.
(
1964b
).
Pseudoplasmodia and intracytoplasmic displacements of the collective amoeba Polysphondylium violaceum
.
Exp. Cell Res
.
(in press)
.
Shaffer
,
B. M.
(
1964C
).
Antistrophic pseudopodia of the collective amoeba Polysphondylium violaceum
.
Exp. Cell Res
.
(in press)
.
Shaffer
,
B. M.
(
1964d
).
Mechanical control of the manufacture and resorption of the cell surface in collective amoebae
.
J. Theor. Biol
,
(in press)
.
Shaffer
,
B. M.
(
1964C
).
The Acrasina
.
Adv. Morphogen
.
3
,
301
22
.
Whittingham
,
W. F.
&
Raper
,
K. B.
(
1960
).
Non-viability of stalk cells in Dictyostelium
.
Proc. nat. Acad. Sci. Wash
.
46
,
642
59
.
Plate 1

A : Lycopodium spores dotted about the surface of a large aggregate at the end of aggregation and before any sensible elongation into grexes. B : 20 min. after deposition the aggregate showed clear signs of cleaving in three. Only in the middle section could an incipient central papilla now be detected, but already nearly all the markers had been carried to the periphery of the aggregate. Most of the others had accumulated in the shallow cleavage furrows. Some of those that had reached the agar no longer touched the perimeter because the upward movement of the cells had slightly reduced the aggregate’s width. C : After a further 25 min. one cleavage had been completed and the papilla of the smaller daughter was just becoming visible. More of the markers had reached the agar, including those that had been held up in the>cleavage furrows.

Plate 1

A : Lycopodium spores dotted about the surface of a large aggregate at the end of aggregation and before any sensible elongation into grexes. B : 20 min. after deposition the aggregate showed clear signs of cleaving in three. Only in the middle section could an incipient central papilla now be detected, but already nearly all the markers had been carried to the periphery of the aggregate. Most of the others had accumulated in the shallow cleavage furrows. Some of those that had reached the agar no longer touched the perimeter because the upward movement of the cells had slightly reduced the aggregate’s width. C : After a further 25 min. one cleavage had been completed and the papilla of the smaller daughter was just becoming visible. More of the markers had reached the agar, including those that had been held up in the>cleavage furrows.

Plate 2

Carmine particles were scattered over the leading face of the tips of three culminating grexes soon after they had reached their minimal height. The particles on each grex, instead of being dispersed radially, were all carried in one direction, while the tip was displaced in the opposite one. Nevertheless, the markers did move apart during their journey, as the underlying sheath expanded. For photography, a heavier load of carmine was used than was needed to produce this behaviour. B: 30 min. later than A; C: 35 min. later than B.

Plate 2

Carmine particles were scattered over the leading face of the tips of three culminating grexes soon after they had reached their minimal height. The particles on each grex, instead of being dispersed radially, were all carried in one direction, while the tip was displaced in the opposite one. Nevertheless, the markers did move apart during their journey, as the underlying sheath expanded. For photography, a heavier load of carmine was used than was needed to produce this behaviour. B: 30 min. later than A; C: 35 min. later than B.