High-resolution autoradiography has been used to establish that during the incompatibility reaction that follows fusion between plasmodia of a ‘killer’ and a ‘sensitive’ strain of the myxo-mycete Physarum polycephalum, the nuclei of the sensitive strain are selectively damaged, enclosed in vacuoles and eliminated from the cytoplasm. This damage is visible as increased chromatin condensation and nucleolar segregation. Nuclear envelopes of both strains show blebbing, and there is an increase in the size and frequency of cytoplasmic vesicles of endoplasmic reticulum. Multiple nuclear fusions are seen between all combinations of genetically like and unlike types of nuclei throughout the course of the incompatibility reaction. After the reaction, mean nuclear diameters increase over 2–3 days to give nuclei an order of magnitude greater in volume than the controls; the population size range returns to normal in 4–5 days. Fusions between incompatible plasmodia carried out when the killer strain is at or very near to mitosis do not produce an immediate incompatibility reaction, but give plasmodia that are neutral and act as neither killers nor sensitives; these heterokaryons convert to killer phenotypes after a few days.

Plasmodia of the myxomycete Physarumpolycephalum Schweinitz are syncytia which may cover many cm2 in area and contain millions of synchronously dividing nuclei; except in the haploid Colonia strains these nuclei are predominantly diploid, with a tendency for some of them to be polyploid. Plasmodia of the same strain normally fuse when they come into contact, but plasmodia that differ genetically may fail to fuse (fusion incompatibility) or if fusion does occur may undergo post-fusion incompatibility reactions (Carlile, 1973). In a given pair of plasmodia which undergo an incompatibility reaction after fusing, the ‘killer’ phenotype is defined as that which survives the reaction to the exclusion of the ‘sensitive’ phenotype. Detailed study of the reaction between a pair of strains, the killer 15 and the sensitive 29, showed that in the presence of nutrients incompatibility took the form of an extensive lethal reaction occurring ca. 5 h after fusion and destroying strain 29, but that under starvation conditions the 29 phenotype was more slowly eliminated without any effects visible to the naked eye (Carlile, 1972). Genetic analysis demonstrated that the basis of lethal reactions among the progeny of strain 29 was allelic incompatibility between 3 dominant alleles, let A, let B and let C, and the corresponding recessive alleles let a, let b and let C (Carlile, 1976), such that a recessive phenotype with respect to any of these loci predictably sustained damage following fusion with another plasmodium which was dominant at that same locus.

Border & Carlile (1974) carried out ultrastructural investigations on plasmodia of strains 15 and 29 fixed shortly after lethal reaction damage had been observed, and concluded that the lethal reaction involved the destruction of nuclei. That study has now been extended using high-resolution autoradiography to distinguish the nuclei of the killer and sensitive strains from each other, and observing the ultrastructural changes occurring before, during and after the visible incompatibility reaction.

Since there is an absence of gross damage in the reaction under starvation conditions, it was initially intended to carry out these studies on water agar, until preliminary investigations showed that nuclear destruction is a frequent event in starving plasmodia, so that any analysis of nuclear degeneration was likely to be ambiguous. The present studies were therefore carried out in rich nutrient conditions where spontaneous nuclear degeneration was rare (less than 0·2%).

Strains

The plasmodial strains used in these experiments were strain 1 × A7029 (i.e. 29) and strains obtained by the sexual fusion of haploid amoeba clones derived from the spores of strain 29 (Carlile, 1976). All are mt1mt2 with respect to mating type. Since strain 29 is homozygous (/in A2A2, fus B2B2) with respect to the genes controlling plasmodial fusion (Cooke & Dee, 1975), all’these progeny strains fuse readily. Work was concentrated on the fusion between strain 29 (phenotype ABC) and strain 29·02 × 29·19 (i.e. strain 02 × 19, phenotype aBC): the reaction here results from incompatibility between a single pair of alleles, let A and let a, with 29 acting as the killer strain.

Routine culture

Strains were maintained as microplasmodia in shaken liquid culture on semi-defined liquid medium (Carlile, 1971) containing glucose (SDG medium) and as macroplasmodia on semidefined agar medium containing starch (SDSt agar).

Production of labelled incompatibility reactions

For each experiment, the macroplasmodia used were synchronous plasmodia produced from liquid cultures as described by Mohberg & Rusch (1969) and transferred to SDSt agar 0·5 h after pipetting onto filter paper to allow fusion of the microplasmodia. It was necessary to use cultures with known times of mitosis so that any confusion with mitosis-associated nuclear changes could be avoided. The stage of the nuclear cycle of plasmodia involved in an experiment was checked by making smear preparations with up to 1 mm 3 of material fixed in glycerol/ ethanol (1:1), and examining them under phase-contrast microscopy. Except in some control experiments, nuclei of either the killer or the sensitive strain were labelled with the radioactive DNA precursor [6–3H]thymidine (Radiochemicals Centre, Amersham). The thymidine solution (40 μM, sp. act. 23 Ci/mol) was added to SDSt agar immediately before pouring into 3-cm tissue culture dishes, to give a final activity of 230 × μCi/ml. Plasmodia were grown on this medium for 24 h, i.e. through at least 2 division cycles. Parallel controls were maintained on medium containing unlabelled thymidine or distilled water instead of radioactive thymidine to monitor for any deterioration due to the radioactivity. To minimize dilution of or contamination with the radioactive agar, all transferring of plasmodial material was done on sterile pieces of Millipore filter; the inoculum on the radioactive agar plate was surrounded with 0 · 5-cm1 pieces of filter so that the growth of 24 h covered several pieces.

Reactions were initiated by inducing a killer and a sensitive strain plasmodium to fuse in proportions of approximately 1:10 respectively; a piece of killer plasmodium-covered filter was inverted on top of a rapidly growing sensitive plasmodium on SDSt agar, at a time when both plasmodia were in early S-phase as determined by smear preparations. This procedure was the same whether killer or sensitive plasmodium had been labelled, but in the latter case an 18–24 h growth period after the labelling was necessary so that the recipient sensitive plasmodium could reach a practical size. When studying a time sequence it was desirable to start with an organism large enough to survive successive samplings, since there was always some residual variation in time between duplicate plasmodia.

Autoradiography

Samples for autoradiography were taken at hourly intervals and fixed in 2 · 5 % glutaraldehyde in 0 · 1 M phosphate buffer containing 2 mM CaCl2, pH 7·2, for 45 min at 4 °C, and postfixed in 1 % osmium tetroxide (same buffer, time and temperature). Any attached agar was removed before dehydration in graded ethanols, and specimens were finally embedded in Araldite via propylene oxide.

Resin sections (ca. 1 μm thick) for light microscopy autoradiographs were collected on gelatine-subbed slides and covered with Kodak ARto stripping film, and developed in D 19b developer (Rogers, 1973) after 7 days’ exposure; they were then examined under phase contrast or stained with Azure B (Stevens, 1966) and viewed in bright-field illumination.

Ultrathin sections for electron microscopy were collected on palladium-coated copper grids, stained with uranyl acetate and lead citrate, carbon-shadowed over the stain, and then coated with Ilford L4 emulsion using the method of Caro & van Tubergen (1962). The specimens were incubated in light-proof dry, cold and vacuum conditions for 9–13 weeks before development (in Kodak Microdol-X), which was carried out on drops of the solutions on waxed staining dishes. The sections were examined in AEI 6, AEI 6B or AEI Corinth 275 electron microscopes.

Comparison of control plasmodia grown on radioactive or non-radioactive thymi-dine-containing media with those grown on thymidine-free SDSt showed that neither the increased thymidine concentration nor the radioactivity appeared to have any ultrastructural side effects. The very thin agar used (2–3 mm) in the culture dishes resulted in morphological indications of starvation after about 48 h, but experimental plasmodia had been transferred to thick SDSt agar long before this occurred.

Examination of light microscopy autoradiographs indicated that over 98% of all nuclei became labelled in 24 h, whichever strain was used. In electron microscopy autoradiographs an unequivocal density of silver grains was produced after 12 weeks’ exposure to the emulsion, irrespective of the growth period on cold agar subsequent to the labelling. Labelled nuclei showed at least 25 silver grain units per thin section after processing, frequently 75–100; background labelling of cytoplasm devoid of nuclei was virtually non-existent by these criteria. Occasional nuclei seen with less than 5 grains were regarded as being unlabelled and may have arisen as a result of fusing labelled and unlabelled plasmodia during 5-phase (immediately after mitosis-plasmodia have no G1 phase), when DNA synthesis is maximal and any cytoplasmic thymidine could have been rapidly assimilated by nuclei of the ‘unlabelled’ strain. The resolution of the EM autoradiographs was limited by the extra thickness of the emulsion covering the specimen and sometimes details were obscured by a high density of silver grains; some micrographs are therefore shown from sections that were not processed for autoradiography although they are from the same specimen blocks as sections that were. About 500 grids were examined, of several samples from each of some 70 plasmodia, of which half were radioactively labelled fusions.

In plasmodia formed by the fusion of killer and sensitive strains, no ultrastructural changes were seen in samples taken up to 90 min after a successful fusion had been initiated; the nuclei appeared identical to those seen in control (Fig. 1) plasmodia. At 2–2 · 5 h after fusion nearly all nuclei, whether derived from the killer or sensitive strain, had developed small dilations of the nuclear envelope, and an increase in the size of rough endoplasmic reticulum vesicles was observed (Fig. 2); no other changes were detected at this time. By 3–3 · 5 h after fusion a range of ultrastructural alterations were seen (Figs. 3-12). Amongst all the experimental fusions carried out, 23 specimens were examined between 3–5 h after fusion, and similar ultrastructural changes were found in all of them. In any one sample 65–75% of all sensitive strain nuclei (as identified by the presence, or absence, of silver grains) were in a comparable state, with the other sensitive nuclei showing greater or lesser divergence from the control condition. By comparison of the extent of changes seen in different nuclei a probable sequence of events could be constructed, as illustrated by the order in which the figures are presented.

The earliest effect on the nucleoplasm of sensitive strain nuclei appeared to be a reduction in the size of the nucleolus, with the appearance of an adjacent patch of finely fibrous material, shown in Fig. 3 (autoradiograph) and Fig. 4 (untreated section from the same block). Further damage appeared as an increase in chromatin condensation until clumped chromatin occupied 25-30% of the extra-nucleolar volume of the nucleus, as compared with only 10% in the controls, estimated by tracings; following this the nucleolus loosened out and disintegrated so that extra-nucleolar chromatin clumping could no longer be assessed (Fig. 10). Aggregates of condensing chromatin became increasingly associated with the nuclear envelope and with the remains of the nucleolus, whilst the granular and fibrillar components of the nucleolus became partially segregated, forming alternating patches of different textures in the nucleus. At this stage breaks in the nuclear envelope were occasionally observed.

In these same samples the killer nuclei were meanwhile little changed in appearance, except for an increase in size and number of blebs or dilations of the nuclear envelope, giving these nuclei a beaded appearance in section (Figs. 6, 7). Their nucleoli were apparently unaltered, and the nucleoplasmic chromatin was no more condensed than in the controls. In contrast with the killer strain nuclei, the nuclear envelope blebs of the sensitive ones had not substantially increased since the 2-h samples (see Figs. 6, 7), yet there was a further increase in the number and size of vesicles of rough endoplasmic reticulum (Fig. 8).

Two other phenomena were seen simultaneously with this increasing disruption of the sensitive strain nuclei, i.e. extensive nuclear fusions (Fig. 5; see below) and a cytoplasmic reaction which resulted in the elimination of the 02 × 19 strain nuclei from the plasmodium. ‘Cracks’ began to appear in the cytoplasm around the sensitive strain nuclei, usually at a distance of 70–100 nm from the nuclear envelope, as seen in Figs. 7, 9 and 10—12. Each crack was first seen as an electron-dense line about 10 nm thick, which then split down the middle to separate the bulk of the cytoplasm from the degenerating nucleus; the vacuoles so formed ultimately ejected their contents, nucleus and attached fragments of cytoplasm, at the plasmalemmal surface. Autoradiographic evidence firmly established that only sensitive, and not killer, strain nuclei were being eliminated in this manner (Figs. 13, 14). Although this isolation process was specifically seen around the sensitive strain nuclei, it appeared not to be dependent on the degree of internal damage sustained by the nucleus: cracks were seen around sensitive strain nuclei as yet undamaged, except for the presence of one or two membrane dilations, whilst in the same plasmodium some sensitive strain nuclei were highly degenerate and still not enclosed in vacuoles. By 4–5 h after fusion this vacuolization of sensitive strain nuclei became increasingly widespread; eventually plasmodial damage was so widespread as to become unspecific.

The dense lines in the cytoplasm seen to precede the cracks are not the membranous product of vesicle fusion, neither did we find any evidence of the vacuoles being associated with plasmalemmal invaginations. We were also not able to demonstrate a clear triple-layered ‘unit membrane’ structure for the vacuole wall, as was usually visible in nearby nuclear and mitochondrial membranes. As would be expected from the large internal surface area of the plasmodium most of the rejected nuclei are found in the invaginating channels, and the persistence of diffuse radioactivity frequently observed many hours after a reaction which involved the elimination of the radiolabelled nuclei suggests that nuclear material ejected into the channels may be at least partially recycled. In the later stages of the reaction many nuclei are ejected directly into the slime layer on the dorsal surface of the organism; this slime is left on the surface of the agar as the slime mould retracts, and it is then that the nuclei become visible in the light microscope as the particulate debris so characteristic of the post-fusion incompatibility reaction. Following the reaction the plasmodium remains immobile for several hours with little protoplasmic streaming occurring; streaming and movement are usually resumed after 12–16 h.

The other notable characteristic of the post-fusion reaction observed here was the frequent occurrence of nuclear fusion, both between nuclei of unlike strains and between like strains of both types (Figs. 5, 9). In plasmodia examined 3–4 h after initiation of the reaction up to 25 % of all nuclei seen were involved in such fusions, and the total number of fusion events must have been much higher. In dissimilar fusions the nuclei were even seen to be reseparated by the vacuolization of the sensitive strain nucleus. Nuclei in close proximity to one another first came into contact at the points of membrane dilation; the outer membranes then fused and this fusion spread until the nuclei shared a common outer membrane. The inner membranes then interacted to form rows of vesicles with nucleoplasmic continuity between them.

Although the plasmodia appeared macroscopically normal after 24 h, smear preparations revealed that they still contained far fewer nuclei than normal, and that those they did contain were grossly enlarged with an average diameter of 7–8 μm (Fig. 15) as opposed to the normal range of sizes of control nuclei of 3·5—5 μm, depending on the stage of the nuclear cycle. This size increase was first detectable at 12–16 h after fusion and continued for a couple of days until maximum mean diameters of 10–11 μm were reached. The nucleoli in these nuclei were frequently highly vacuolated or ring-shaped, and often multiple. Nuclear sizes within the plasmodia returned to normal after 4–5 days.

Since plasmodia with given synchrony were being used for all these experiments, tests were carried out to determine whether the incompatibility reaction was influenced by the stage in the cell cycle reached by killer and sensitive plasmodia at the time of fusion. It was discovered that fusion between a killer plasmodium at or very near to mitosis with a sensitive plasmodium at any stage in the mitotic cycle did not result in an incompatibility reaction. When the heterokaryotic plasmodia so produced were subjected to further experimental fusions with the 2 strains, they were found to behave as neither type, i.e. they were neither sensitive to killing by 29 nor capable of inducing a reaction with 02 × 19. This neutral condition was not indefinitely stable; all the heterokaryons finally converted to a killer phenotype within a few days. This was often seen to result from a typical, but delayed, post-fusion reaction.

The present results confirm the earlier indications (Border & Carlile, 1974) that the destructive effects of the post-fusion incompatibility reaction are directed primarily against the nuclei of the sensitive strain. Furthermore, details have been obtained as to the nature of this damage and the mode of elimination of the rejected nuclei from the plasmodium. A new phenomenon, that of somatic nuclear fusions accompanying the incompatibility reaction, has been observed, as have circumstances under which the reaction can be bypassed to produce a transient neutral phenotype.

Damage to sensitive strain nuclei

Following the beginning of the formation of blebs or dilations of the nuclear envelope on nuclei of both kinds, specific damage to the sensitive strain nuclei is first visible as an increase in the degree of chromatin condensation. Breaks in the nuclear envelope are also seen occasionally, but most consistent is the disruption and disintegration of the nucleolus, in which the fibrillar and granular components segregate separately.

Chromatin condensation is a common indicator of nuclear damage, which may be as unspecific as physical rupture of the nuclear membrane or general cell necrosis (e.g. Trump, Goldblatt & Stowell, 1965). Segregation of nucleolar components however appears to be more specific. In mammalian cells it has been induced by actinomycin D inhibition of RNA synthesis (Reynolds, Montgomery & Hughes, 1964) or by disrupting DNA synthesis (Lapis & Bernhard, 1965; Bernhard, 1971). Gontcharoff & Rao (1972) mention briefly that actinomycin D will cause segregation of nucleolar components in Physarum polycephalum, although specific interference with DNA eventually produced ring-shaped nucleoli as seen by Sachsenmaier & Rusch (1964) and as we frequently saw in the 24 h after the reaction. Nucleolar segregation (in mammalian cells) has been shown to require protein synthesis as well as a continued energy supply (Monneron, 1971); under conditions of non-specific necrosis the nucleolus shrinks without notable separation of its components. Nuclear breakdown can take place normally at several points in the life cycle of myxomycètes, but where electron microscopy has been done of such events (e.g. Schuster, 1969; Charvat, Ross & Cronshaw, 1973; Bechtel, 1976) none of them appear to involve nucleolar segregation such as was seen here. The extent of the nucleolar segregation seen in these lethal reactions is not as complete as that shown by Reynolds et al. (1964) and Bernhard (1971), but this may be due to differences in chromatin organization between mammals and myxomycètes-for instance it is known that the nucleolar DNA of Physarum is largely extra-chromosomal (Bohnert, Schiller, Bohme & Sauer, 1975). Taken in the context of other published reports therefore a reasonable hypothesis would seem to be that the incompatibility reaction damage is the result of some assault on the nuclear chromatin structure, rather than a metabolic inactivation of the affected nuclei.

The significance of the nuclear envelope dilations is not clear. Nuclear blebs have previously been reported in a variety of cell types, including Physarum (Goodman & Ritter, 1969). They may be an indication of hyperactivity which may be responsible for the increase in vesicles of rough endoplasmic reticulum; in the case of the sensitive strain nuclei the vesicle proliferation may be halted by the internal damage to the nuclei, giving rise to the observed difference in number of blebs around killer and sensitive nuclei. The relationship of the vesicles to the reaction is also undetermined, but they may contain enzymes for the degradation of rejected material.

The absence of a prompt incompatibility reaction when fusion takes place with the killer plasmodium at mitosis may be associated with the low metabolic activity of nuclei at this time: close to mitosis there is little or no RNA synthesis (Kessler, 1967) although RNA synthesis rises again to its maximum rate in about 10 min after karyokinesis (Davies & Walker, 1978). Further work will be needed to explain the neutral state of such plasmodia.

Elimination of nuclei

The sensitive strain nuclei are rejected from the fusion plasmodium in the course of the post-fusion incompatibility reaction along with varying amounts of adjacent cytoplasm, the nucleus and cytoplasm first being delimited by apparently newly formed membrane. A similar process of membrane delimitation of damaged areas has been observed during the fast post-fusion reaction in strains of Didymium iridis (Upadhyaya & Ling, 1976), and micrographs of nuclear abortion in sporulating plasmodia (Schuster, 1969; Bechtel, 1976) suggest that the same process may be operating in all cases. To account for the lack of synchrony between visible nuclear damage and vacuole formation in the present somatic incompatibility reaction, it is necessary to suppose that the initiating signal for the elimination process occurs some time before chromatin damage is seen, and that the intermediate stages in the 2 processes are independent.

The source of the crack/vacuole membrane has not been determined; it does not arise from the fusion of existing membranous vesicles, nor does it appear to be derived from surface invaginations, for 3 reasons. First, connexions between cracks and surface were never seen; secondly, there is no evidence of cortical differentiation of the cytoplasm adjacent to the vacuoles; and thirdly, the plasmalemmal membrane is a strongly developed one that typically shows clear trilaminar structure, and is quite different in appearance from the membrane around the vacuoles. It may be that the membrane forming the elimination vacuole forms de novo in the cytoplasm.

Nuclear fusion

Perhaps the most surprising consequence of fusing these incompatible strains was the induction of frequent and repeated nuclear fusions. Plasmodial adherence to the diploid condition is known to be not very consistent, but actual nuclear fusion was only known to take place in a sexual context between myxomycete amoebae of different mating types. Multiple sexual fusions have however been recorded, at a specific growth stage, with amoebae of another myxomycete, Didymium iridis (Ross & Cummings, 1970). Further studies will be needed to establish whether there is any genetic association between determinants for somatic incompatibility and those for nuclear fusion. It is interesting to note that a close association does exist in several fungi between sexual compatibility and somatic incompatibility: for instance in Neurospora crassa the mating type locus A/a controls both simultaneously (Metzenberg & Ahlgren, 1973; Newmeyer, Howe & Galeazzi, 1973), and in Podospora anserina the mod genes which suppress somatic incompatibility also bring about sexual sterility (Boucherie & Bernet, 1974; Boucherie, Bégueret & Bernet, 1976).

Whatever the nature of the genetic control of the high-frequency somatic nuclear fusions, their occurrence between killer nuclei and sensitive nuclei raises the possibility that chromatin from the sensitive strain may survive the incompatibility reaction and undergo parasexual recombination with that from the killer strain. The use of genetic markers will be required to establish whether this does occur, since the specificity of the [3H]thymidine label is too short-lived.

Ultrastructural studies of nuclear fusion are surprisingly few, but the details of these events that we have observed are broadly similar to those in gamete karyogamy in the fungus Allomyces (Pommerville & Fuller, 1976).

Repeated nuclear fusion is the most likely explanation for the existence of the giant nuclei seen 1–4 days after fusion of the plasmodia, rather than inhibition of nuclear division which could also give large nuclei (Gull & Trinci, 1974). Nuclear enlargement in Physarum has also been seen in senescent plasmodia by McCullough, Cooke, Sudbery & Grant (1969), but such plasmodia died soon afterwards, whereas ours all recovered and returned to normal. The mechanism by which the population of plasmodial nuclei was restored to normal size ranges is not known, but experiments are in progress to determine this.

This work was undertaken with the financial support of the Science Research Council; we would also like to thank Professor P. Gahan and Dr J. Gay for helpful discussions.

Bechtel
,
D. B.
(
1976
).
Nuclear degeneration in the myxomycete Physarella oblonga
.
Protoplasma
90
,
179
187
.
Bernhard
,
W.
(
1971
).
Drug-induced changes in the interphase nucleus. Adv
.
Cytopharmac
.
1
,
49
67
.
Bohnert
,
H. J.
,
Schiller
,
B.
,
Bohme
,
R.
&
Sauer
,
H. W.
(
1975
).
Circular DNA and rolling circles in nucleolar rDNA from mitotic nuclei of Physarum polycephalum
.
Eur. J. Biochem
.
57
,
361
369
.
Border
,
D. J.
&
Carlile
,
M. J.
(
1974
).
Somatic incompatibility following plasmodial fusion between strains of the myxomycete Physarumpolycephalum: the effect on their nuclei
.
J. gen. Microbiol
.
85
,
211
219
.
Boucherie
,
H.
,
Bégueret
,
J.
&
Bernet
,
J.
(
1976
).
The molecular mechanism of protoplasmic incompatibility and its relationship to the formation of protoperithecia in Podospora anserina
.
J. gen. Microbiol
.
92
,
59
66
.
Boucherie
,
H.
&
Bernet
,
J.
(
1974
).
Protoplasmic incompatibility’ and female organ formation in Podospora anserina: properties of mutations abolishing both processes
.
Molec. gen. Genet
.
135
,
165
174
.
Carlile
,
M. J.
(
1971
).
Myxomycètes and other slime moulds
.
In Methods in Microbiology
, vol.
4
(ed.
C.
Booth
), pp.
237
265
.
London
:
Academic Press
.
Carlile
,
M. J.
(
1972
).
The lethal interaction following plasmodial fusion between two strains of the myxomycete Physarum polycephalum
.
J. gen. Microbiol
.
71
,
581
590
.
Carlile
,
M. J.
(
1973
).
Cell fusion and somatic incompatibility in my’xomy’cetes
.
Ber. dt. bot. Ges
.
86
,
123
139
.
Carlile
,
M. J.
(
1974
).
Incompatibility in the myxomycete Badhamia utricular is
.
Trans. Br. mycol. Soc
.
62
,
401
402
.
Carlile
,
M. J.
(
1976
).
The genetic basis of the incompatibility reaction following plasmodial fusion between different strains of the myxomycete Phy sarum polycephalum
.
J. gen. Microbiol
.
93
,
371
376
.
Caro
,
L. G.
&
Van Tubergen
,
R. P.
(
1962
).
High-resolution autoradiography. I. Methods
.
J. Cell Biol
.
15
,
173
188
.
Charvat
,
L
,
Ross
,
I. K.
&
Cronshaw
,
J.
(
1973
).
Ultrastructure of the plasmodial slime mould, Perichaena vermicularis. I. Plasmodium
.
Protoplasma
76
,
333
351
.
Cooke
,
D. J.
&
Dee
,
J.
(
1975
).
Methods for the isolation and analysis of plasmodial mutants in Physarum polycephalum
.
Genet. Res
.
24
,
175
187
.
Davies
,
K. E.
&
Walker
,
I. O.
(
1978
).
Control of RNA transcription in nuclei and nucleoli of Physarum polycephalum
.
FEBS Letters, Amsterdam
86
,
303
306
.
Gontcharoff
,
M.
&
Rao
,
B.
(
1972
).
Dependence of the nuclear structure on DNA and RNA synthesis
.
Chromosoma
38
,
441
457
.
Goodman
,
E. M.
&
Ritter
,
H.
(
1969
).
Plasmodial mitosis in Physarum polycephalum
.
Arch. Protistenk
.
111
,
161
169
.
Gull
,
K.
&
Trinci
,
A. P. J.
(
1974
).
Ultrastructural effects of griseofulvin on the myxomycete Physarum polycephalum. Inhibition of mitosis and the production of microtubule crystals
.
Protoplasma
81
,
37
48
.
Kessler
,
D.
(
1967
).
Nucleic acid synthesis during and after mitosis in the slime mould, Physarum polycephalum
.
Expl Cell Res
.
45
,
676
680
.
Lapis
,
K.
&
Bernhard
,
W.
(
1965
).
The effect of mitomycin-C on the nucleolar fine structure of KB cells in cell culture
.
Cancer Res
.
251
,
628
645
.
Mccullough
,
C. H. R.
,
Cooke
,
D. J.
,
Foxon
,
J. L.
,
Sudbery
,
P. E.
&
Grant
,
W. D.
(
1973
).
Nuclear DNA content and senescence in Physarum polycephalum
.
Nature, Neto Biol
.
245
,
263
265
.
Metzenberg
,
R. L.
&
Ahlgren
,
S. K.
(
1973
).
Behaviour of Neurospora tetrasperma matingtype genes introgressed into N
.
crassa. Can. J. Genet. Cytol
.
15
,
571
576
.
Mohberg
,
J.
&
Rusch
,
H. P.
(
1969
).
Growth of the large plasmodia of themy’xomy’cete Physarum polycephalum
.
J. Bact
.
97
,
1411
1418
.
Monneron
,
A.
(
1971
).
Action of some drugs on liver nuclei and polysomes
.
Adv. Cytopharmac
.
1
,
131
144
.
Newmeyer
,
D.
,
Howe
Jr.,
H. B.
&
Galeazzi
,
D. R.
(
1973
).
A search for complexity’ at the miting-type locus of Neurospora crassa
.
Can. J. Genet. Cytol
.
15
,
577
585
.
Pommerville
,
J.
&
Fuller
,
M. S.
(
1976
).
The cytology of the gametes and fertilization of Allomyces macrogynus
.
Arch. Microbiol
.
109
,
21
30
.
Reynolds
,
R. C.
,
Montgomery
,
P.
O’B. & Hughes
,
B.
(
1964
).
Nucleolar ‘caps’ produced by actinomycin D
.
Cancer Res
.
2411
,
1269
1277
.
Rogers
,
A. W.
(
1973
).
Techniques of Autoradiography
, 2nd edn.
Amsterdam
:
Elsevier
.
Ross
,
I. K.
&
Cummings
,
R. J.
(
1970
).
An unusual pattern of multiple cell and nuclear fusion in the heterothallic slime mould Didymium iridis
.
Protoplasma
70
,
281
294
.
Sachsenmaier
,
W.
&
Rusch
,
H. P.
(
1964
).
The effect of 5-fluoro-2’-deoxyuridine on synchronous mitosis in Physarum polycephalum
.
Expl Cell Res
.
26
,
124
133
.
Schuster
,
F. L.
(
1969
).
Nuclear degeneration during spore formation in the true slime mould, Didymium nigripes
.
J. Ultrastruct. Res
.
29
,
171
181
.
Stevens
,
A. R.
(
1966
).
High resolution autoradiography
.
In Methods in Cell Physiology
, vol.
2 (ed
.
D. M.
Prescott
), pp.
255
310
.
London and New York
:
Academic Press
.
Trump
,
B. F.
,
Goldblatt
,
P. J.
&
Stowell
,
R. E.
(
1965
).
Studies of mouse liver necrosis in vitro. Ultrastructural and cytochemical alterations in hepatic parenchymal cell nuclei
.
Lab. Invest
.
14
,
1969
1999
.
Upadhyaya
,
K. C.
&
Ling
,
H.
(
1976
).
Ultrastructure of clear zones exhibiting cytoplasmic incompatibility following somatic cell fusion in the myxomycete Didymium iridis
.
Ind. J. exp. Biol
.
14
,
652
658
.