/ . Embryol. exp. Morph. Vol. 40, pp. 215-228, 1977
Printed in Great Britain © Company of Biologists Limited 1977
215
Pattern formation in Dictyostelium discoideum
I. Development of prespore cells and its relationship to the
pattern of the fruiting body
By D. FORMAN 1 AND D. R. GARROD 2
From the Department of Biology, University of Southampton
SUMMARY
Immunofluorescent staining of the prespore cells of the cellular slime mould Dictyostelium
discoideum was carried out using a heterologous spore antibody. The highly specific staining
of the prespore vesicles (PSVs) within the prespore cells enabled quantitative determinations
to be made of the rate and extent of development of these cells throughout the life cycle.
The results showed that PSVs first appeared in a large proportion of the cells shortly after
the cells had chemotactically aggregated into multicellular masses. During the later phases
of the life cycle, the proportion of cells containing PSV increased, as did the fluorescent
intensity of their PSVs, until the early culmination stage of development when 85-90 % of
the total cell population contained PSVs. Lowering the temperature of development delayed
the onset of vesicle formation and decreased the proportion of prespore cells in the total
cell population. Changing the growth conditions of the cells prior to multicellular development also had a significant effect on the proportions of prespore cells, as did the use of a
mutant known to give rise to fruiting bodies with a reduced number of spores.
The comparability between these estimates of prespore cell proportions at culmination
and previously reported spore:stalk ratios within fruiting bodies confirms the view that
PSVs are reliable indicators of prespore cells. The finding that temperature and growth
conditions and the use of mutants all of which are known to affect spore:stalk ratios, also
all affected prespore proportions in the expected direction, adds further weight to this argument.
The fact that prespore cells are beginning to differentiate early in the multicellular phase
of the life cycle and the related finding that such differentiation always precedes formation
of the grex tip are results of considerable importance to the development of a model for
pattern formation in D. discoideum.
INTRODUCTION
The fruiting body of the cellular slime mould, Dictyostelium discoideum,
consists of a mass of spores on top of a cellular stalk and thus represents a
simple linear pattern of differentiation. The pattern is size-invariant, the ratio
of spore to stalk being roughly constant irrespective of the size of the fruiting
body. The fruiting body is formed from a polarized mass of cells, the grex,
1
Author's address: Department of Bacteriology and Immunology, University of Glasgow,
Western Infirmary, Glasgow, Gil 6NT, U.K.
2
Author's address (for reprints): Department of Biology, University of Southampton.
Medical and Biological Sciences Building, Southampton, SO9 3TU, U.K.
216
D. FORMAN AND D. R. GARROD
there being correspondence between a cell's position within the grex and
whether it forms spore or stalk. Cells at the anterior end of the grex form the
stalk, while cells from the posterior end form the spores. These precursors of
spore and stalk have been called 'prespore' and 'prestalk' cells, respectively
(Raper, 1940; Bonner, 1944).
The terms 'prespore' and 'prestalk' can be justified by more than cell position within the grex since a number of differences between the two cell types
have been discovered (listed by Garrod & Ashworth, 1973). The major distinguishing feature is the exclusive possession by the prespore cells of vesicles
containing material which is thought to be extruded from the cells at culmination to form the coat of the spores (Hohl & Hamomoto, 1969; Takeuchi, 1972).
These prespore vesicles (PSV) have been identified by electron microscopy
(Maeda & Takeuchi, 1969; Gregg & Badman, 1970), and by the powerful
technique of immunofluorescent staining with an antibody against the spore
coat (Takeuchi, 1963).
Even in the absence of additional knowledge, the immunological evidence
would seem to many to argue most persuasively that the prespore cells are the
precursors of the spores of the fruiting body. This in turn would imply that the
anterior-posterior prestalk-prespore pattern which has been shown to exist in
the grex (Takeuchi, 1963; Gregg, 1965) is the precursor of the stalk-spore
pattern of the fruiting body. However, it has been suggested recently that the
fruiting body pattern may be formed only during the actual process of fruiting
body construction and the presence of prespore cells in the grex bears no relation to the final pattern (Farnsworth, 1973; Farnsworth & Loomis, 1976).
The reason why such a controversy can exist is that the mechanism of pattern
formation in D. discoideum is not understood. Although numerous hypotheses
have been put forward (reviewed by Loomis, 1975) we strongly felt that insufficient experimental work had been done on the descriptive basis of pattern
formation to provide a firm basis for any of these hypotheses. Our first intention
in this study has been, therefore, to provide more descriptive information on
the development of prespore cells in D. discoideum. Secondly, we have sought
evidence on whether the prespore and prestalk cells in the grex are related to
the final pattern of the fruiting body.
MATERIALS AND METHODS
Growth and development of cells
Dictyostelium discoideum cells (strain Ax-2) were grown vegetatively in
axenic conditions (Watts & Ashworth, 1970) on a rotary shaker at 22 °C and
140 rev./min (radius of rotation 2-75 cm). For most experiments the growth
medium was supplemented with 86 mM glucose. Cells were harvested in the
exponential phase of growth when they were at a density of 2-4 x 106 per ml.
Cells were harvested by centrifugation at 350 g for 5 min at 4 °C. They were
Pattern formation in D. discoideum
217
then washed twice in cold distilled water and resuspended in distilled water at
a density of 1 x 108 per ml.
Ax-2 cells supplied with bacterial nutrients, D. discoideum cells of the mutant
strain P-4 (kindly provided by Dr J. Gross), and Dictyostelium mucoroides
cells were grown by inoculating spores with E. coli B/r on a standard agar
medium (Sussman, 1966) and incubating at 22 °C. Vegetative cells were harvested in cold distilled water after 40 h of growth and centrifuged at 35Og
for 5 min at 4 °C to remove excess bacteria. The cells were washed three times
in cold distilled water and resuspended at 2 x 108 per ml.
Vegetative cells were allowed to develop by placing 0-5 ml of the cell suspension over the surface of washed 47 mm Millipore filters (AABPO4700)
resting on absorbent pads saturated with 1-6 ml of a buffered salt solution at
pH6-5 (1 litre 50 mM phosphate buffer with 1-5 g KC1, 0-5 g MgCl 2 .6HO 2 ,
0-5 g streptomycin). The Millipore filters in Petri dishes were incubated in
dark humid conditions at 22 °C or other experimental temperatures. In such
conditions cells develop with a high degree of synchrony (Sussman, 1966).
Preparation of single cell suspensions
In most experiments single cell preparations were required from the multicellular cell masses formed after aggregation. Such cell masses were dissociated
by the method of Takeuchi & Yabuno (1970). Cells were washed off a Millipore
filter at the requisite time in 50 mM Tris-HCl buffer pH 7-0 and centrifuged at
700 g for 10 min at 4 °C. They were then resuspended in Tris-HCl containing
25 mM dimercaptopropanol (BAL) and 0-1 % pronase. Cell masses were maintained in this solution for 15 min during which time they were triturated with
a Pasteur pipette. This treatment reliably dissociated aggregates and grexes
into single cell suspensions. The cells were centrifuged at 700 g for 10 min
and resuspended in standard salt solution (Bonner, 1947) at 2 x 105 per ml. This
suspension was poured on top of glass coverslips contained in a Petri dish
and the cells were allowed to settle out and adhere to the coverslips for 30 min
at room temperature. The salt solution was decanted and the cells fixed for
3 min in ice-cold methanol. The cells were then washed in phosphate buffered
physiological saline (PBS) at pH 7-0 and kept in PBS until required for staining.
Preparation of spore specific antibody
As specificity is known to be enhanced by utilizing spore antigens from a
different species of Dictyostelium (Takeuchi, 1972), it was decided to produce
antibody from spores of D. mucoroides. Such spores were collected from fruiting
bodies of D. mucoroides, formed after 2-3 days development on agar at 22 °C,
using a wire loop and suspended in PBS. The saline was lightly centrifuged
(100 g for 1 min) in order to pellet out any contaminating stalk. The spores
were then spun out of the supernatant at 700 g for 10 min and resusperided in a
small amount "of PBS. They were then frozen and thawed in order to lyse any
218
D. FORMAN AND D. R. GARROD
remaining vegetative cells, and washed in PBS. Finally the spores were suspended in distilled water, lyophilized and stored at 4 °C until required.
Spores were inoculated into New Zealand White rabbits, from which a preinjection sample of blood had been obtained from the marginal ear vein. Ten mg
of freeze-dried spores suspended in 1 ml of physiological saline were injected
intravenously into the marginal ear vein. This was followed by a series of five
subcutaneous injections each containing 5 mg of spores into the neck and back
on days 5, 23, 30, 40 and 52. In total 35 mg of spore material was injected which
contained 2-5 mg of soluble protein (Lowry, Rosebrough, Farr & Randall, 1951).
Two days after each of the four booster injections, trial bleedings were made
from the ear vein and the serum antibody level was assessed using precipitin
and Ouchterlony double diffusion tests (Ouchterlony & Nilsson, 1973). Seven
days after the final injection, the rabbits were bled out by heart puncture.
Blood was left to clot for 1 h at room temperature and, after ringing, overnight
at 4 °C. The serum was then taken off and centrifuged to remove any remaining
cells. It was stored in aliquots at - 27 °C.
Testing the serum antibody
For the precipitation and Ouchterlony tests, an antigen solution was prepared
by sonicating 10 mg lyophilized D. mucoroides spores in 1 ml PBS (0-7 mg
protein per ml) containing 0-5% sodium deoxycholate. Fragments were discarded by centrifugation at 29000g for 30 min. After the final booster injection, dilutions of the antigen to 0-62 jug protein/ml and to 175/^g protein/ml
gave positive reactions in the precipitin and Ouchterlony tests respectively.
Using undiluted antigen solution on the Ouchterlony double diffusion plates
with the antiserum caused the formation of three precipitation lines. After
absorption of the serum with vegetative cells from axenically grown D. discoideum (Takeuchi, 1963) a single specific precipitation line was produced.
For immunofluorescent use it was found appropriate to dilute such absorbed
antiserum by 1 in 100.
Staining of single cell preparations
The dissociated cells, methanol fixed on coverslips, were stained by adding one
drop of the diluted D. mucoroides absorbed antispore serum. This was allowed to
react for 20 min at room temperature in moist conditions before being washed
off in PBS for 10 min. One drop of anti-rabbit immunoglobulin (sheep) conjugated with fluorescein isothiocyanate (Wellcome Reagents Ltd) was then applied
to the cells for 10 min at room temperature. A further 20 min wash in PBS was
then given and the coverslips mounted in 90% glycerol: 10% PBS.
Control staining was carried out using non-immune pre-injection serum
followed by the commercial conjugate. Controls were also made using the
conjugate alone. At no time was any specific staining observed with such
controls.
Pattern formation in D. discoideum
219
Microscopy and counting
Cell preparations were observed on a Zeiss Universal microscope using either
phase contrast or illumination with an HBO 200 mercury arc lamp, BG 38
and BG 12 exciter filters, and Zeiss barrier filters 44 and 50.
When it was necessary to assess the proportion of specifically stained cells,
a field of cells on the coverslip was selected at random and the total number of
cells counted under phase contrast illumination. The same field was then
observed under u.v. light and the number of cells containing the specifically
stained fluorescent vesicles was counted. Where possible counting was carried
out using a 'blind' procedure.
RESULTS
Development ofPSV during the life cycle
The time course of appearance of PSV in log phase cells allowed to develop
on Millipore filters is shown in Fig 1. The first distinctive appearance of the
fluorescent vesicles was evident in a large number of cells (65 ±1-4% S.D.) after
12 h of development. This was 3-4 h after the chemotactic aggregation stage
and corresponded approximately to the time of first appearance of the grex
tip. At an even earlier time, from about 10 h of development onwards, a diffuse
specific staining could be detected but fluorescence was not localized in distinct
vesicles (Fig. 2) making it difficult to estimate in what proportion of the cells
it appeared. Specific staining was never observed in cells prior to aggregation.
As development proceeded through the standing, migrating and culminating
grex stages there was a gradual increase in the proportion of cells with PSV.
The rate of increase in prespore cell numbers varied between experiments but
by 18-19 h of development (early culmination) there was stabilization of the
proportion of prespore cells at 85-90 % of the total cell population. After the
early culmination stage, prespore determinations were of little relevance without
an additional estimate of the number of mature spores and stalk cells, neither
of which could be readily assessed.
Besides these quantitative changes in the number of cells containing prespore vesicles there was also a marked increase in the fluorescent intensity of
vesicles during development from aggregation to early culmination (Figs. 2 and
3). This was accompanied by a decline in background cytoplasmic staining so
that by 14 h of development it was relatively easy to distinguish between prespore cells and unstained cells, presumably prestalk cells (Fig. 4).
Effect of temperature on PSV development
The temperature at which cells develop is known to affect the spore: stalk
ratio in fruiting bodies of D. discoideum. Lowering the temperature increases
the proportion of stalk cells (Bonner & Slifkin, 1949; Farnsworth, 1975).We
felt that if the prespore: prestalk ratio in the grex is relevant to the pattern of
220
D. FORMAN AND D. R. GARROD
100 r-
90
80
70
60
10
11
12
13
14
15 16
Time (h)
17
18
19
20
21
a
Fig. 1. Percentage of prespore cells (cells containing prespore vesicles) out of total
cell population counted against time of development on Millipore filters. Bar lines
represent ± standard error of the means from separate experiments. Drawings
represent stage of development. This is a summary of four experiments in which
proportions of prespore cells dissociated from cell masses, were counted at two
hourly intervals throughout the multicellular phases of the life cycle. Each point on
the graph is the mean percentage calculated from two separate 500 cell counts for
each of two separate experiments. Thus 2000 cells were counted at each of the
hourly intervals. The cells were of the strain Ax-2, grown in glucose medium, and
developed at 22 °C.
the fruiting body, this ratio should be affected by temperature in the same way
as the proportions of the fruiting body.
Figure 5 shows the results of experiments in which Ax-2 cells were allowed
to develop at different temperatures between 15 and 22 °C and the proportion
of prespore cells in the population counted. Since the speed of development
varied with temperature prespore cells were counted at two-hourly intervals
from the time of their first appearance until early culmination at each temperature. Mature fruiting bodies were formed 2-3 h after the final reading at each
temperature. •
Between 15 and 22 °C there was an almost linear relationship between temperature and the number of prespore cells in the population at the early culmination stage. Above 22 °C we found no further increase in the proportion of
prespore cells.
Low temperature delayed the onset of vesicle formation by slightly less than
1 h for every 1 °C drop in temperature below 22 °C. At all temperatures the
Pattern formation in D. discoideum
221
Fig. 2. Cells dissociated from early aggregates, after 10 h of development on
Millipore filters, and stained with absorbed anti-Z). mucoroides spore serum with
F.I.T.C. observed under conditions as stated in methods. Note the faint, diffuse
nature of the staining and the lack of obvious vesicles, x 760.
first appearance of vesicles coincided approximately with the onset of grex tip
formation. Thereafter the proportion of prespore cells increased for 6-8 h when
culmination began at all temperatures. Thus the time between tip formation
and the onset of culmination was roughly constant at different temperatures.
This was because the period of migration was shortened at low temperature.
At 15 °C for example there was little or no migration.
Effect of growth conditions on PSV development
It has been shown that cells of D. discoideum Ax-2 grown under different
conditions form fruiting bodies with different spore: stalk ratios. Thus fruiting
bodies of log-phase cells grown in axenic medium supplemented with 86 mM
glucose (glucose cells) had a spore: stalk ratio of 3-95:1, whereas fruiting bodies
of log-phase cells grown in unsupplemented medium (N.S. cells) had a
spore:stalk ratio of 2-70:1 (Garrod & Ashworth, 1972). (These spore:stalk
ratios were determined by volume measurements on fixed fruiting bodies formed
after development on Millipore filters.) If the prespore :prestalk ratio in the grex
is relevant to the spore: stalk pattern of the fruiting body the proportion of
prespore cells in glucose grexes should be greater than that in N.S. grexes.
Figure 6 shows the prespore: prestalk ratios in early culminating grexes of
cells grown under these different conditions. The proportions of prespore cells
in the populations was 89 % for log phase glucose cells and 79 % for N.S. cells.
Clearly the prespore:prestalk ratios of the populations differ in the same way
as the spore: stalk ratios of their fruiting bodies.
15
EMB 40
222
D. FORMAN AND D. R. GARROD
Fig. 3. Cells dissociated from early culminants, after 18 h of development on Millipore filters, stained and observed as in Fig. 2. Note the large number of vesicles
with bright fluorescence, x 760.
Fig. 4. Cells dissociated from grexes, after 14 h of development on Millipore filters,
stained and observed as in Fig. 3. Note the distinction between stained and unstained cells, x 760.
Pattern formation in D. discoideum
223
100
22 °C
90
80
70
60
50
12
13
14
15
16
17
18 19
Time (h)
20
21
22
23
24
25
26
Fig. 5. Percentage of prespore cells in cell populations grown at 22 °C and allowed
to develop at different temperatures plotted against time of development on Millipore filters. All the cell populations were of the strain Ax-2 grown in glucose
medium.
Development ofPSV in mutant P-4
D. discoideum P-4 is a mutant which has a reduced proportion of spores in
the fruiting body (Hohl & Raper, 1964) and an increased susceptibility to
cyclic-AMP induction of stalk cells (Chia, 1975). If the prespore :prestalk
ratio in the grex is relevant to the pattern of the fruiting body, the proportion
of prespore cells in P-4 grexes should be similarly reduced.
Figure 7 shows P-4 cells which were allowed to develop on Millipore filters
at 22 °C exhibited a marked reduction in the proportion of prespore cells at
the early culmination stage compared with Ax-2 cells. A reduction in the proportion of prespore cells in P-4 was found at all stages of the life cycle subsequent
to the first appearance of PSV. The development of P-4 was delayed by 3-4 h
compared with Ax-2 glucose cells. On Millipore filters, P-4 formed a large
number of fruiting bodies with thickened stalks. We did not obtain a substantial
number of normal fruiting bodies as have been found with P-4 developing on
cellophane membranes (Chia, 1975; Hamilton & Chia, 1975).
15-2
224
D. F O R M A N A N D D. R. G A R R O D
100 r
90
Glucose cells
80
NS cells
70
60
50
9
10
12
13
14
15 16 17
Time (h)
18
19
20 21
Fig. 6. Percentage of prespore cells in populations of Ax-2 glucose and Ax-2 N.S.
cells plotted against time of development on Millipore filters. The cells were grown
and developed at 22 °C. The first point on each line corresponds to the time of
initial appearance of prespore cells; the last point to the early culmination stage.
DISCUSSION
Relationship to pattern formation
These results enable us to make two main points in relation to pattern
formation. Firstly, all of the conditions affecting the spore:stalk ratios of
fruiting bodies which we have used, affect the proportion of prespore cells
during the grex stage of the life cycle in the same direction. It seems unlikely
that the material in the prespore vesicles which our antibody binds is a nonspecific sorus binding material as has recently been suggested (Farnsworth &
Loomis, 1976). This is because thorough washing of the spore with saline
before inoculation completely disrupted spore heads into single spores and
should have removed any such material. Rather, we feel that the antibody is
against the spore coat and the antigen in the PSV therefore represents spore
coat material. Even if the suggestion of Farnsworth & Loomis were correct,
staining with antibody would still indicate which cells were involved in forming
spores, for the sorus is that part of the fruiting body which consists of spores.
We conclude that the prespore proportion in the grex is directly related to the
spore: stalk ratio of the fruiting body and that there is a direct casual relationship between possession of PSV and differentiation into spores. We therefore
Pattern formation in D. discoideum
100
225
r
Ax-2 glucose cells
70
60
50
40
30
12
13
14
15
16
17 18 19
Time (h)
20
21
22
23
24
Fig. 7. Percentage of prespore cells in Ax-2 and P-4 cells grown and developed at
22 °C, against time of development on Millipore filters. Stages of development refer
to P-4. Stages for Ax-2 as in Fig. 1.
disagree with the model of Farnsworth (1973) which proposes that pattern
formation occurs only during the process of culmination.
Secondly, prespore cells can first be identified by means of fluorescent PSV
at the stage of the late aggregate, approximately at the time when the grex tip
first begins to form. Both Miiller & Hohl (1973) and Hayashi & Takeuchi
(1976) report identification of PSV at about this time but in a smaller proportion of cells. Since we could detect specific cytoplasmic fluorescence which was
not collected into vesicles up to 2 h before the first positive identification of PSV,
it seems likely that differentiation begins even earlier than the late aggregate
stage. This view is reinforced by the finding that two cell populations which,
if mixed, sort out to the prestalk and prespore regions of the grex respectively
226
D. FORMAN AND D. R. GARROD
can be separated by isopycnic centrifugation from populations of preaggregation cells (Takeuchi, 1969; Bonner, Sieja & Hall, 1971; Maeda & Maeda, 1974).
Therefore, much evidence now suggests that the first stage in pattern formation,
the diversification of prespore and prestalk cell types, probably begins early in
the life cycle, possible before the multicellular stage is reached and certainly
before the grex develops overt polarity by forming a tip. This point is of major
importance in relation to the model for pattern formation which we develop
in the following paper (Forman & Garrod, 1977).
Relationship to other work
Our time course of accumulation of prespore cells is intermediate between
those of Miiller & Hohl (1973) and Hayashi & Takeuchi (1976). The former
found a gradual linear increase in numbers of prespore cells from late aggregation to early culmination, while the latter found a sudden increase at late
aggregation with no further increase in the later stages of the life cycle. Factors
which may have contributed to discrepancies between the three studies are
firstly, the use of different strains of D. discoideum, secondly the use of different
vegetation growth conditions and thirdly the use of different conditions for
development. Both Miiller & Hohl and ourselves allowed cells to develop on
Millipore filters, a condition which shortens the time of grex migration compared with development on non-nutrient agar which was used by Hayashi &
Takeuchi. It will be very important to establish whether there is a gradually
increasing proportion of prespore cells or whether the full complement of prespore cells appears soon after aggregation.
It is useful to comment on the merits and disadvantages of our techniques
for detecting prespore cells compared with the techniques of other workers.
The main advantage of our procedure of staining cells which have been dissociated from grexes is that it enables us to deal with very large numbers of cells.
Techniques involving the detection of PSV by electron microscopy (Miiller &
Hohl, 1973; Farnsworth & Loomis, 1976) are necessarily more laborious and
therefore deal with smaller numbers of cells. A disadvantage of our technique
is that it does not enable us to count prespore cell numbers in individual grexes.
We therefore cannot yet comment on the curious result obtained by Farnsworth & Loomis (1976) by counting PSV with the electron microscope that
the number of prespore cells is constant irrespective of the size of the grex.
Our figures for the number of prespore cells in the population at early culmination (between 75 % and 89 % for all Ax-2 cell populations developed at
22 °C) agree well with spore:stalk ratios in fruiting bodies obtained by other
workers (Bonner, 1967; Farnsworth, 1975). It should be pointed out that
spore:stalk ratios obtained by making volume measurements on fruiting
bodies (Bonner & Slifkin, 1949; Garrod & Ashworth, 1972) invariably give
lower spore:stalk ratios than counting prespore cells or spores. Volume
measurements make no pretence at being able to estimate cell number. The
Pattern formation in D. discoideum
227
discrepancy with ratios obtained by cell counting techniques arises because
the formation of spores from prespore cells involves a decrease in cell volume
while the formation of stalk cells from prestalk cells involves an increase in
volume. There are thus many more spore cells per unit volume of the spore
head than there are stalk cells per unit volume of stalk.
We would like to thank Lynette Banham, Alistair Nicol and Sherilee Taylor for excellent
technical assistance. This work was supported by the Science Research Council. D. Forman
was in receipt of a S.R.C. Studentship.
REFERENCES
J. T. (1944). A descriptive study of the development of the slime mould Dictyostelium
discoideum. Am. J. Bot. 31, 175-182.
BONNER, J. T. (1947). Evidence for the formation of cell aggregates by chemotaxis in the
development of the slime mould D. discoideum. J. exp. Zool. 106, 1-26.
BONNER, J. T. (1967). The Cellular Slime Moulds, 2nd ed. Princeton, New Jersey: Princeton
University Press.
BONNER,
BONNER, J. T , SIEJA, T. W. & HALL, E. M. (1971). Further evidence for the sorting out of
cells in the differentiation of the cellular slime mould D. discoideum. J. Embryol. exp. Morph.
25, 457-465.
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 D. discoideum. Amer. J. Bot. 36, 727-734.
CHI A, W. K. (1975). Induction of stalk cell differentiation by cyclic AMP in a susceptible
variant of D. discoideum. Devi Biol. 44, 239-252.
FARNSWORTH, P. (1973). Morphogenesis in the cellular slime mould D. discoideum; the formation and regulation of aggregate tips and the specification of developmental axes. /.
Embryol. exp. Morph. 29, 253-266.
FARNSWORTH, P. (1975). Proportionality in the pattern of differentiation of the cellular slime
mould D. discoideum and the time of its determination. /. Embryol. exp. Morph. 33,
869-877.
FARNSWORTH, P. & LOOMIS, W. F. (1976). Quantitation of the spatial distribution of prespore
vacuoles in pseudoplasmodia of D. discoideum. J. Embryol. exp. Morph. 35, 499-505.
FORMAN, D. & GARROD, D. R. (1977). Pattern formation in D. discoideum. II. Differentiation
and pattern formation in non-polar aggregates. /. Embryol. exp. Morph. 40, 229-243.
GARROD, D. R. & ASHWORTH, J. M. (1972). Effect on growth conditions on development of
the cellular slime mould D. discoideum. J. Embryol. exp. Morph. 28, 463-479.
GARROD, D. R. & ASHWORTH, J. M. (1973). Development of the cellular slime mould D. discoideum. Symp. Soc. gen. Microbiol. 23, 407-435.
GREGG, J. H. (1965). Regulation in the cellular slime moulds. Devi Biol. 12, 377-393.
GREGG, J. H. & BADMAN, W. S. (1970). Morphogenesis and ultrastructure in Dictyostelium.
Devi Biol. 22, 96-111.
HAMILTON, I. D. & CHIA, W. K. (1975). Enzyme activity changes during cyclic AMP induced
stalk cell differentiation in PA, a variant of D. discoideum. J. gen. Microbiol. 91, 295-306.
HAYASHI, M. & TAKEUCHI, I. (1976). Quantitative studies on cell differentiation during morphogenesis of the cellular slime mould. Devi Biol. 50, 302-309.
HOHL, H. R. & HAMAMOTO, S. T. (1969). Ultrastructure of spore differentiation in Dictyostelium'. the prespore vacuole. /. Ultrastruct. Res. 26, 442-253.
HOHL, H. R. & RAPER, K. B. (1964). Control of sorocarp size in the cellular slime mould D.
discoideum. Devi Biol. 9, 137-153.
LOOMIS, W. F. (1975). Dictyostelium discoideum - A Developmental System. New York:
Academic Press.
228
D. FORMAN AND D. R. GARROD
D. R., ROSEBROUGH, N. J., FARR, A. L. & RANDALL, R. J. (1951). Protein measurement with the Folin phenol reagent. /. biol. Chem. 265-275.
MAEDA, Y. & MAEDA, M. (1974). Heterogeneity of the cell population of the cellular slime
mould D. discoideum. Expl Cell Res. 84, 88-94.
MAEDA, Y. & TAKEUCHI, I. (1969). Cell differentiation and fine structures in the development
of the cellular slime mould. Dev. Growth & Differ. 11, 232-245.
MULLER, V. & HOHL, H. R. (1973). Pattern formation in D. discoideum: temporal and spatial
distribution of prespore vacuoles. Differentiation 1, 267-276.
OUCHTERLONY, O. & NILSSON, (1973). Immunodiffusion and immunoelectrophoresis. In
Handbook of Experimental Immunology, 2nd ed. (ed. D. M. Weir). Oxford: Blackwell.
RAPER, K. B. (1940). Pseudoplasmodium formation and organisation in D. discoideum. J.
Elisha Mitchell Scient. Soc. 56, 241-282.
SUSSMAN, M. (1966). Biochemical and genetic methods in the study of cellular slime mould
development. In Methods in Cell Physiology, vol. 2 (ed. D. Prescott). New York: Academic Press.
TAKEUCHI, I. (1963). Immunochemical and immunohistochemical studies on the development
of the cellular slime mould D. discoideum. Devi Biol. 8, 1-26.
TAKEUCHI, I. (1969). Establishment of polar organisation during the slime mould development. In Nucleic Acid Metabolism, Cell Differentiation and Cancer Growth (ed. E. V.
Crowdy & S. Seno). Oxford: Pergamon Press.
TAKEUCHI, I. (1972). Differentiation and dedifferentiation in cellular slime moulds. In
Aspects of Cellular and Molecular Physiology (ed. K. Hamoguchi). Tokyo: University of
Tokyo Press.
TAKEUCHI, I. & YABUNO, K. (1970). Disaggregation of slime mould pseudoplasmodia using
EDTA and various proteolytic enzymes. Expl Cell Res. 61, 183-190.
WATTS, D. & ASHWORTH, J. M. (1970). Growth of myxamoebae of the cellular slime mould
D. discoideum in axenic culture. Biochem. J. 119, 171-174.
LOWRY,
{Received 4 January 1977)