The effect of growth in the presence and absence of 86 mM glucose on fruiting-body number and size, fruiting-body proportions and morphogenesis in the slime mould Dictyostelium discoideum, strain Ax-2, has been investigated. Cells grown in the absence of glucose (N.S. cells) formed 2·1 times more fruiting bodies than glucose-grown cells allowed to differentiate under the same conditions and at the same cell density. Glucose fruiting bodies were 2·6 times larger than N.S. fruiting bodies.

During aggregation, N.S. aggregation streams generally broke up into numerous secondary aggregation centres. Glucose streams generally did not break up but moved into the initial aggregation centre. Each secondary centre formed one small grex, whereas initial centres fragmented into several grexes which were larger than those formed from secondary centres. The preponderance of secondary centres in N.S. aggregation and of initial centres in glucose aggregation accounts for the difference in size and number of fruiting bodies. We speculate about the mechanism giving rise to the morphogenetic difference.

The spore: stalk ratio was 3·95:1 in glucose fruiting bodies and 2·70:1 in N.S. fruiting bodies. This difference is not related to the difference in fruiting-body size because proportions are size invariant for both fruiting-body types. Some difference in the physiological mechanism which determines proportions is suspected.

The life-cycle of Dictyostelium discoideum consists of two distinct phases which are mutually exclusive in the sense that cell growth and almost all cell multiplication are confined to the solitary feeding phase, whereas morphogenesis and differentiation occur in the absence of growth during the subsequent multicellular phase. (For a detailed description of the life-cycle see Bonner, 1967.) However, conditions pertaining during the growth phase can influence differentiation. Variations in the composition of the growth medium of the axenic strain, D. discoideum Ax-2, cause alterations in the levels of key metabolites and enzymes in the growing cells (Weeks & Ashworth, 1972; Ashworth & Quance, 1972). These differences in metabolite and enzyme levels are frequently preserved during differentiation (Hames, Weeks & Ashworth, 1972). Further, variation in the glucose content of the axenic growth medium affects the size and number of fruiting bodies formed. At the same cell density, cells grown in the presence of 86 niM glucose form fewer, larger fruiting bodies than cells grown in the absence of added carbohydrate (Quance & Ashworth, 1972). Thus it is possible to study the morphogenetic and biochemical behaviour of two physiologically distinct but genetically identical cell populations.

In this paper we describe the difference in size and number of fruiting bodies formed by cells grown in the presence and absence of glucose, the proportions (i.e. the spore: stalk ratios) of the two types of fruiting body, and the morpho-genetic event which gives rise to the difference in fruiting-body size and number.

Growth of cells

D. discoideum Ax-2 cells were grown at 22°C in suspension in 2 1 flasks containing approximately 700 ml of medium. The flasks were shaken on a rotary shaker at 140 rev/min with a radius of rotation of 2·75 cm. The media contained peptone, 15·2 g/1.; yeast extract, 7·6 g/1.; Na2HPO4, 5 mg/1.; KH2PO4, 5mg/l.; pH 6·7. Cells grown in this medium, which contains no added carbohydrate, will be referred to as N.S. cells. Glucose cells were grown in the same medium with the addition of glucose to a final concentration of 86 HIM.

Flasks were inoculated from stock cultures of cells (not spores), the inoculation densities being 7× 103/ml for N.S. and 1·4×104/ml for glucose. Growth was continued for approximately 65 h, after which time cells of each type reached a concentration of about 2 × 106/ml.

Differentiation

Cells were harvested from growth flasks and put on to the surfaces of black Millipore filters (catalogue no. AABP 04700) to allow them to differentiate. These filters contain a filtration aid (probably a detergent) which may affect morphogenesis and which must be removed before use by extensive washing in tap and distilled water.

Harvesting was done by spinning the cell suspension in culture medium at 350 g for 15 min at 4°C. The cells were then washed once with cold distilled water and resuspended in cold distilled water at 5 × 107/ml; 0·5 ml of this suspension was spread over the entire surface of each 47 mm Millipore filter, giving an average cell density of 1·76 × 106/cm2. The Millipore filters were placed in 60 mm plastic Petri dishes on support pads saturated with 1·6 ml of pad diluting fluid (PDF). The PDF contained in 11 of 50 mM-Na/K phosphate buffer at pH 6·5; KC1, 1·5 g; MgCl2.6H2O, 0·5 g; streptomycin sulphate 0·5 g. Dishes were incubated at 21°C in a dark, humid container. Under these conditions aggregation began 9 h after the beginning of incubation and fruiting-body construction was complete after 24 h (Sussman, 1966).

Assessment of fruiting-body population density

This was done with a binocular microscope by counting the number of fruiting bodies in four fields, chosen at random from each Millipore filter. The area of each field was 15·9 mm2.

Preparation of fruiting bodies for measurement

Fruiting bodies were fixed by exposure to the vapour of 40% formalin solution for 1 h and then to the vapour of 2% OsO4 for a further hour. The Millipore with attached, fixed fruiting bodies was then immersed in 50% ethanol for 10 min, then in distilled water. The spore heads of many fruiting bodies were damaged during immersion in ethanol, but it was easy to select only undamaged fruiting bodies at the next stage of preparation.

Next, a few drops of 1% agar in distilled water were put on to a warm micro-scope slide and the agar kept liquid by placing the slide in the beam of an infrared lamp supported about 2 ft (0·6 m) above the bench surface. Using a pair of needle forceps, undamaged fruiting bodies were dislodged from the Millipore surface and placed in the agar on the slide. When about 20 fruiting bodies had been transferred to the slide it was removed from the infra-red beam to allow the agar to solidify. A few more drops of liquid agar were then added and a coverslip placed over the preparation. When the second lot of agar had solidified, the preparation was sealed with vacuum grease. A glucose fruiting body prepared by this method is shown in Fig. 1.

Fig. 1

A glucose fruiting body prepared for measurement as described in Materials and Methods. The arrowed region is where the stalk was slightly damaged during removal from the Millipore filter.

Fig. 1

A glucose fruiting body prepared for measurement as described in Materials and Methods. The arrowed region is where the stalk was slightly damaged during removal from the Millipore filter.

Measurement of fruiting bodies and volume calculation

A Projectina microscope was used to measure the fruiting bodies. At × 60 magnification the fruiting bodies were projected on to the instrument’s screen, which has a graduated linear scale. Four measurements were taken on each fruiting body (Fig. 2): (a) the major axis of the spore head; (b) the minor axis of the spore head; (c) the diameter of the base of the stalk; (d) the length of the stalk between points A and B in Fig. 2. As stalks were invariably curved they were divided by eye into a number of straight sections the lengths of which were added to give the total stalk length. The basal disc of the stalk was not measured.

Fig. 2

Diagram showing the measurements taken from each fruiting body. Stalk length was measured between points A and B as described in the text.

Fig. 2

Diagram showing the measurements taken from each fruiting body. Stalk length was measured between points A and B as described in the text.

The volumes of the spore heads were calculated by treating them as prolate spheroids (Bonner & Slifkin, 1949). The volume of a prolate spheroid is given by the formula
formula
where a is the major semi-axis and b the minor semi-axis (Turrell, 1946). The volumes of the stalks were calculated by treating them as cones, using the formula
formula

where r is the radius of the base and h the height. Some comments on the use of these methods are given in the Discussion.

Photography of developing cultures

Photographs were taken at 15 min intervals from the beginning cf aggregation to the early standing grex stage, using a 24 mm Repro Sum mar lens attached to a 35 mm camera body. Illumination was provided by an electronic flash lamp (intensity 400 J) supported at an angle 30° to the horizontal.

During an experiment the culture was kept at 21°C in a dark, humid container which was opened for the minimum possible time for each exposure. Each culture was allowed to complete development after photography to ensure that its differentiation was normal.

Fruiting-body density

The average fruiting-body density obtained with N.S. cells was 247·6/cm2. (S.E. ±13·8, range 170·6–407·7), whereas for glucose cells it was 118·5/cm2 (S.E. ± 7–9; range 51·3–181·2). While the difference between N.S. and glucose was statistically significant (t = 8·12 for 41 degrees of freedom; P < 0·001), the extreme variability of fruiting-body density within each group was noteworthy. The average ratio of number of N.s. to glucose fruiting bodies was 2–1:1. Cultures with N.S. and glucose fruiting bodies are shown in Fig. 17. Assuming that aggregation was 100% efficient, these results would mean that these were on average approximately 7 × 103 cells/N.s. fruiting body and 1·5 × 104 cells/glucose fruiting body.

Fruiting-body size and proportions

One hundred and seventeen fruiting bodies of each type were measured. These were taken from two cultures of each type. The average volume of N.s. fruiting bodies was 1·3 × 106μm3 (S.E. ± 1·2 × 105; range 5·3 × 104 to 8·1 × 106), whereas for glucose fruiting bodies it was 3·2 × 106μm3 (S.E. ± 2·0 × 105; range 3–4 × 105 to 1·5 × 107). Here too the variability was noteworthy. The ratio of average volumes was 2·6:1 (glucose:N.S.). This is slightly different from the ratio of fruiting-body numbers, but because of the range in fruiting-body numbers between cultures of cells of the same type represents reasonably good agreement. Histograms showing the distribution of fruiting-body volumes for N.S. and glucose cells are given in Fig. 3. It will be noted that the ranges of fruiting-body volumes were almost identical.

Fig. 3

Distribution of fruiting body total volumes (a) for N.S. and (b) for glucose. Arrows indicate average values.

Fig. 3

Distribution of fruiting body total volumes (a) for N.S. and (b) for glucose. Arrows indicate average values.

The ratios of spore volume to stalk volume for N.S. and glucose fruiting bodies were obtained from the slopes of the respective regression lines (Figs. 4, 5). A marked difference in the proportions of the fruiting bodies was found. For N.S. fruiting bodies the spore: stalk ratio was 2·70:1, whereas for glucose fruiting bodies it was 3·95:1. Two things stand out from these figures. First, the scatter of points for glucose fruiting bodies was greater than that for N.S. fruiting bodies, as is indicated by the respective correlation coefficients, r = 0·75 for glucose and r = 0·89 for N.S. Both N.S. and glucose fruiting bodies had greater spore: stalk ratios than the fruiting bodies measured by a similar technique by Bonner & Slifkin (1949), who obtained a spore:stalk ratio (taken from the slope of their line) of about 1·75:1.

Fig. 4

Graph of spore head volume against stalk volume for N.S. fruiting bodies. The solid line was obtained by regression analysis of the data. Its slope is 2·70. The correlation coefficient (r) was 0-89 for 115 degrees of freedom (P < 0·001). For comparison, the lines for glucose fruiting bodies (G) and for wild-type fruiting bodies (Bonner & Slifkin, 1949) (PS) are shown.

Fig. 4

Graph of spore head volume against stalk volume for N.S. fruiting bodies. The solid line was obtained by regression analysis of the data. Its slope is 2·70. The correlation coefficient (r) was 0-89 for 115 degrees of freedom (P < 0·001). For comparison, the lines for glucose fruiting bodies (G) and for wild-type fruiting bodies (Bonner & Slifkin, 1949) (PS) are shown.

Fig. 5

Graph of spore head volume against stalk volume for glucose fruiting bodies. The solid line was obtained by regression analysis of the data. Its slope is 3·95. The correlation coefficient (r) was 0·75 for 115 degrees of freedom (P < 0·001). For comparison, the lines for N.S. fruiting bodies (NS) and wild-type fruiting bodies (Bonner & Slifkin, 1949) (BS) are shown.

Fig. 5

Graph of spore head volume against stalk volume for glucose fruiting bodies. The solid line was obtained by regression analysis of the data. Its slope is 3·95. The correlation coefficient (r) was 0·75 for 115 degrees of freedom (P < 0·001). For comparison, the lines for N.S. fruiting bodies (NS) and wild-type fruiting bodies (Bonner & Slifkin, 1949) (BS) are shown.

There seemed to be two possible reasons why the proportions of glucose and N.s. fruiting bodies were different. First, there might be some physiological difference between the two cell populations which affects the mechanism responsible for determining proportions. Secondly, proportions might in some way be determined by the absolute size of fruiting bodies so that glucose fruiting bodies may have a greater spore:stalk ratio than N.S. fruiting bodies because they are larger, and vice versa. In Fig. 6, spore:stalk ratio (calculated directly by dividing spore volume by stalk volume) has been plotted against fruiting-body volume for both glucose and N.S. cells. Regression analysis of these data gave, in both cases, lines which were almost parallel to the volume axis: the slope for N.S. was — 0·01 and the slope for glucose was + 0·03. Thus it appeared that, in both cases, spore: stalk ratio was constant over the entire range of volumes. Further, the correlation coefficients were small (r = −0·08 for N.S., r = 0·22 for glucose), indicating no dependence of proportions on total volume.

Fig. 6

Graphs of spore:stalk ratio against fruiting body total volume. The lines were obtained by regression analysis, (a) N.S. fruiting bodies: slope, 0·01; correlation coefficient, -0·08; P > 0·1. (b) Glucose fruiting bodies: slope, 0·03; correlation coefficient, 0·22; P− 0·02.

Fig. 6

Graphs of spore:stalk ratio against fruiting body total volume. The lines were obtained by regression analysis, (a) N.S. fruiting bodies: slope, 0·01; correlation coefficient, -0·08; P > 0·1. (b) Glucose fruiting bodies: slope, 0·03; correlation coefficient, 0·22; P− 0·02.

Morphogenetic origin of size difference

It seemed likely that the morphogenetic event giving rise to the difference in fruiting-body size and population density between N.S. and glucose might occur at one of two stages in the life cycle, either (a) at the aggregation stage or (b) at the stage the aggregate breaks up into individual grexes. Accordingly, we investigated these stages by photographing developing cultures at 15 min intervals, starting at the beginning of aggregation and continuing until the standing grex stage.

A distinct difference between the patterns of aggregation of N.S. and glucose cells was discovered. Early aggregation of N.s. cells was characterized by highly branched streams around an aggregation centre (Fig. 7). The streams then developed nodular swellings (Fig. 8). Stream breakage then took place at the sides of swellings proximal to the original aggregation centre. Each swelling then acted as a secondary aggregation centre and absorbed distal streams (Fig. 9). Secondary aggregation centres were usually smaller than the initial centre.

Fig. 7–10

Aggregation of N.S. cells.

Fig. 7. Early aggregation showing highly branched streams.

Fig. 8. One hour later, showing nodular stream swellings.

Fig. 9. Seventy-five minutes after Fig. 8, showing break-up of streams and formation of secondary centres.

Fig. 10. Two hours after Fig. 9, showing early standing grexes.

Fig. 7–10

Aggregation of N.S. cells.

Fig. 7. Early aggregation showing highly branched streams.

Fig. 8. One hour later, showing nodular stream swellings.

Fig. 9. Seventy-five minutes after Fig. 8, showing break-up of streams and formation of secondary centres.

Fig. 10. Two hours after Fig. 9, showing early standing grexes.

Early glucose aggregation streams were less branched than those of N.S. cells (Fig. 11). Moreover, what branches there were coalesced to form broad streams which tapered evenly towards the aggregation centre (Fig. 12). Stream swelling was rare with glucose cells. The streams were usually absorbed by the initial aggregation centre, which either remained as a single aggregate until all the streams had been absorbed or itself fragmented into a number of smaller aggregates of roughly equal size (Figs. 13, 14).

Fig. 11–14

Aggregation of glucose cells.

Fig. 11. Early aggregation.

Fig. 12. Seventy-five minutes later, showing broad streams. Note that the two initial centres roughly in the middle of the picture have become much larger.

Fig. 13. Two hours after Fig. 12. The streams have almost been absorbed by the centre which is beginning to break up.

Fig. 14. One hour after Fig. 13, showing aggregates just before standing grex formation.

Fig. 11–14

Aggregation of glucose cells.

Fig. 11. Early aggregation.

Fig. 12. Seventy-five minutes later, showing broad streams. Note that the two initial centres roughly in the middle of the picture have become much larger.

Fig. 13. Two hours after Fig. 12. The streams have almost been absorbed by the centre which is beginning to break up.

Fig. 14. One hour after Fig. 13, showing aggregates just before standing grex formation.

Standing grex formation from aggregation centres of N.S. and glucose cultures are shown in Figs. 15 and 16. These photographs demonstrate the difference between the two very clearly. In both cases there were large centres which broke up into several standing grexes. Standing grexes from large centres appeared roughly the same size in both cases. The difference between the two was in the number of small centres between the large ones, each of which gave rise to only one grex. Such centres were numerous in N.S. cultures and almost non-existent in glucose cultures. The grexes formed singly from small centres were, almost without exception, smaller than those formed by the break-up of large centres. The small centres originated from swellings on aggregation streams.

Fig. 15

N.S. standing grexes. There are groups of large grexes formed from initial centres (arrows I, for example), and numerous small grexes formed from secondary centres (arrows 5).

Fig. 15

N.S. standing grexes. There are groups of large grexes formed from initial centres (arrows I, for example), and numerous small grexes formed from secondary centres (arrows 5).

Fig. 16

Glucose standing grexes. Almost all are large and grouped together. Figs. 15 and 16 are different preparations from those shown in Figs. 714.

Fig. 16

Glucose standing grexes. Almost all are large and grouped together. Figs. 15 and 16 are different preparations from those shown in Figs. 714.

Fig. 17

Millipore filters showing NS and glucose (G) fruiting bodies. The black filters are 47 mm in diameter.

Fig. 17

Millipore filters showing NS and glucose (G) fruiting bodies. The black filters are 47 mm in diameter.

Proportions of fruiting bodies

It is important to know whether addition of glucose to the growth medium of the cells affects the physiological mechanism for determining proportions at the differentiation stage. Compared with the method of Bonner & Slifkin (1949), our method of proportion measurement was slightly less laborious -enabling more fruiting bodies to be examined -but also less accurate. The inaccuracy lay in our treating stalks as cones, whereas Bonner and Slifkin treated them as tapered cylinders. In general, stalks seemed to taper slightly less rapidly than would cones of similar height and basal radius, so that our measurement gave an underestimate of stalk volume. However, since there was no obvious difference in the shapes of glucose and N.s. stalks, the method seemed reasonable as a means of comparing fruiting bodies of the two types. We conclude that there was a distinct difference in the spore:stalk proportions of glucose and N.s. fruiting bodies, the ratios being 3·95:1 for glucose and 2·7:1 for N.S.

Because glucose fruiting bodies were on average larger than N.S. fruiting bodies, it was possible that the proportions of fruiting bodies were in some way dependent on the size of the grexes from which they were formed. This would be contrary to the view (Bonner & Slifkin, 1949; Bonner, 1957) that proportions are constant over a wide range of fruiting-body sizes. Our analysis, by plotting spore:stalk ratio against fruiting-body volume, strongly supports the views of previous workers. The slopes of our lines from these plots are very nearly parallel to the volume axis, indicating that proportions remained the same irrespective of volume, and the regression coefficients are small, indicating no dependence of proportions on fruiting-body volume. In view of this, it seems that the different proportions of glucose and N.s. fruiting bodies could not be related to the difference in absolute volume. Hence, we suggest that some physiological difference between the two affects the mechanism responsible for determining proportions, but we have no indication from biochemical studies of what this difference might be.

The fact that both N.S. and glucose fruiting bodies appeared to have greater spore:stalk ratios than the fruiting bodies measured by Bonner & Slifkin (1949) deserves specific comment. Part of the reason for this was undoubtedly that we underestimated stalk volume, but in order to bring our glucose fruiting bodies into line with Bonner & Slifkin’s, their stalks would have to be roughly 167% larger than our measurements suggested. So large an error could only have arisen if the stalks were approximately cylindrical (a cylinder is 200% greater in volume than a cone of equal height and basal radius), whereas in fact all stalks tapered considerably (Fig. 1). We conclude that our fruiting bodies did, indeed, have greater spore:stalk ratios than those of Bonner & Slifkin. This may have been for several reasons: (1) We used a different strain of D. discoideum. (2) Our strain was grown axenically in suspension and not on an agar surface with a bacterial food supply. (3) Differentiation took place on the surface of a Milli-pore filter, not on agar. (4) Differentiation took place at 22°C, compared with 17°C in the study by Bonner and Slifkin, who produced evidence that temperature affects proportions. All of these factors may affect proportions to some degree. Indeed, our results appear to demonstrate that conditions pertaining during the growth phase of the life-cycle affect the proportions of differentiated fruiting bodies.

Difference in size of glucose and N.s. fruiting bodies

During aggregation the initial streams which formed around an aggregation centre were either absorbed into the intial centre or broke up, forming numerous secondary centres. This breaking up of the stream was extensive during aggregation of N.S. cells and uncommon during aggregation of glucose cells. Secondary centres were generally smaller than initial centres and gave rise to only one grex. Initial centres broke up into several grexes which were larger than these formed from secondary centres. We found no evidence for break-up of grexes at later stages of development. The smaller average size of N.S. fruiting bodies was due to their formation predominantly from small grexes from secondary centres, whereas the larger glucose fruiting bodies were formed predominantly from larger grexes from initial centres. The greater number of N.S. than glucose fruiting bodies arose because the difference in aggregation pattern resulted in the formation of more grexes per unit area, i.e. per unit number of cells.

Although the above argument seems plausible, it was not immediately obvious that aggregation was the only developmental stage at which the size of grexes and hence of fruiting bodies was determined. In particular, was it true that initial centres of glucose and N.S. broke up to form grexes of roughly equal size? Hohl & Raper (1964) showed that there was an upper limit, the ‘critical mass’, to the size of grexes formed from aggregates. If an aggregate was smaller than the critical mass, one grex formed from it, but if it exceeded the critical mass, two grexes formed. Was the critical mass for glucose and N.S. the same? Because the ranges of fruiting-body size for glucose and N.S. were almost identical, we feel that the critical masses were probably the same. If the critical mass for glucose had been greater than that for N.S., we would have expected the upper size limit of glucose fruiting bodies to be greater than that for N.S. In fact only one glucose fruiting body was found which was larger than the upper N.S. size limit. We conclude that initial centres of both glucose and N.S. broke up to give grexes of about the same size. Secondary centres were smaller than the critical mass and thus gave only one grex. The smaller average size of N.s. fruiting bodies may be accounted for by the preponderance of secondary centres in N.S. cultures, and vice versa for glucose.

Hohl & Raper reported that the critical mass for ‘wild type’ fruiting bodies was 15 × 106μm3. Taking the volume of a wild-type cell as about 7 × 102μm3 (Garrod & Born, 1971), this critical mass would represent about 2× 104 cells. (This is probably an over-estimate because the figure of 15 × 106μm3 for critical mass must include some extracellular space.) We have estimated that the average size of glucose fruiting bodies was about 1·5 × 104 cells. Since this average must be somewhat below the critical mass, we conclude that the critical mass for strain Ax-2 was of roughly the same order as that of the wild type in terms of numbers of cells involved. A calculation based on our volume measurements reinforced this conclusion, suggesting that the critical mass did not exceed 4× 104 cells, except in one case which may have been aberrant. Because our experiments were not designed to test them specifically, all conclusions relating to critical mass must be regarded as provisional.

Shaffer (1957) described in detail the formation and behaviour of swellings on aggregation streams of D. discoideum. This behaviour was extremely variable, but in itself constituted only one aspect of the even greater variability of stream development. It seems that the different behaviour of both N.s. and glucose aggregation streams may each represent a slight reduction in overall variability, and therefore a simplification, in that each shows a preponderance of one type of stream behaviour. However, the wide ranges of fruiting-body population density for both glucose and N.S. indicates considerable variability in the degree to which their aggregation streams exhibit the different types of behaviour. Shaffer suggested that variability is occasioned by the simplicity of the mechanisms controlling development (Shaffer, 1957, p. 405). This may well be correct, but it also seems possible that developmental controlling mechanisms may be complex in that there may be many controlling factors, variation in any one of which may alter the behaviour of aggregation streams in one or a number of different ways. It may be more accurate to say that controlling mechanisms may be imprecise or imprecisely co-ordinated, rather than simple.

Clearly, speculation as to the physiological difference which is responsible for the difference in behaviour of N.S. and glucose aggregation streams must be cautious, N.S. streams develop swellings which become secondary aggregation centres much more commonly than glucose streams. This could be because N.S. centres are less efficient inhibitors of centre formation than glucose centres. Inhibition of centre formation by existing centres has been demonstrated in Polysphondylium violaceum (Shaffer, 1963) and in D. discoideum (Bonner & Hoffman, 1963). Theoretically, inhibition of centre formation within the aggregation territory of an existing centre may occur if the centre efficiently entrains the surrounding cells, i.e. if it maintains the highest frequency of emission of acrasin pulses (Cohen & Robertson, 1971). So it is possible that glucose centres entrain the surrounding cells more efficiently than N.s. centres. It is also possible to put forward an explanation of the different behaviour in terms of cellular cohesiveness (cell-cell adhesion) or cellular adhesiveness (cell-substatum adhesion), or both. Briefly, either a decrease in cohesiveness or an increase in adhesiveness might in principle, account for the reduced tendency of glucose streams to form swellings. Both effects would tend to make the cells less likely to round up into aggregates. On the other hand, a decrease in cohesiveness might cause glucose streams to fragment more easily, which they clearly do not do. We hope that further experimentation will allow us to assess the relative importance of the factors mentioned in the control of aggregative behaviour.

It is clear that conditions pertaining during the growth phase of the slime mould life-cycle can influence both morphogenesis and pattern formation (spore:stalk ratio) during the differentiation phase. Because the slime mould, grown axenically, is amendable to biochemical analysis, it provides a system in which morphogenesis and pattern formation may be correlated with biochemical variables.

We thank Mr G. Asquith for advice on photography and for preparation of the figures. The work was supported by the Science Research Council.

Ashworth
,
J. M.
&
Quance
,
J.
(
1972
).
Enzyme synthesis in myxamoebae of the cellular slime mould Dictyostelium discoideum during growth in axenic culture
.
Biochem. J
.
126
,
601
608
.
Bonner
,
J. T.
(
1957
).
A theory of the control of differentiation in the cellular slime moulds
.
Q. Rev. Biol
.
32
,
232
246
.
Bonner
,
J. T.
(
1967
).
The Cellular Slime Moulds
, 2nd ed.
New Jersey
:
Princeton University Press
.
Bonner
,
J. T.
&
Hoffman
,
M. E.
(
1963
).
Evidence for a substance responsible for the spacing pattern of aggregation and fruiting in the cellular slime moulds
.
J. Embryol. exp. Morph
.
11
,
571
589
.
Bonner
,
J. T.
&
Slifkin
,
M. K.
(
1949
).
A study of the control of differentiation: the proportions of stalk and spore cells in the slime mould Dictyostelium discoideum
.
Am. J. Bot
.
36
,
727
734
.
Cohen
,
M. H.
&
Robertson
,
A.
(
1971
).
Chemotaxis and the early stages of aggregation in cellular slime moulds
.
J. theoret. Biol
.
31
,
119
130
.
Garrod
,
D. R.
&
Born
,
G. V. R.
(
1971
).
Effect of temperature on the mutual adhesion of preaggregation cells of the slime mould, Dictyostelium discoideum
.
J. Cell Sci
.
8
,
751
765
.
Hames
,
B. D.
,
Weeks
,
A.
&
Ashworth
,
J. M.
(
1972
).
Glycogen synthetase and the control of glycogen synthesis in the cellular slime mould Dictyostelium discoideum during differentiation
.
Biochem. J
.
126
,
627
633
.
Hohl
,
H. R.
&
Raper
,
K. B.
(
1964
).
Control of sorocarp size in the cellular slime mould Dictyostelium discoideum
.
Devl Biol
.
9
,
137
153
.
Quance
,
J.
&
Ashworth
,
J. M.
(
1972
).
Enzyme synthesis in the cellular slime mould Dictyostelium discoideum during the differentiation of myxamoebae grown axenically
.
Biochem. J
.
126
,
609
615
.
Shaffer
,
B. M.
(
1957
).
Variability of behaviour of aggregating cellular slime moulds
.
Q. JI Microsc. Sci
.
98
,
393
405
.
Shaffer
,
B. M.
(
1963
).
Inhibition by existing aggregations of founder differentiation in the cellular slime mould Polysphondylium violaceum
.
Expl Cell Res
.
32
,
603
606
.
Sussman
,
M.
(
1966
).
Biochemical and genetic methods in the study of cellular slime mould development
.
In Methods in Cell Physiology
, vol.
2
(ed.
D. M.
Prescott
), pp.
397
410
.
New York
:
Academic Press
.
Turrell
,
F. M.
(
1946
).
Tables of Surface and Volumes of Spheres and of Prolate and Oblate Spheroids, and Spheroidal Coefficients
.
Berkeley, California
:
University of California Press
.
Weeks
,
A.
&
Ashworth
,
J. M.
(
1972
).
Glycogen synthetase and the control of glycogen synthesis in the cellular slime mould Dictyostelium discoideum during the growth (myxamoebal) phase
.
Biochem. J
.
126
,
617
626
.