Fine Structure and Surface Topography of Endospores - CiteSeerX

Journal of General Microbiology (I972), 71,367-381

367

Printed in Great Britain

Fine Structure and Surface Topography of Endospores of Thermoactinomyces vulgaris By A N N E M c V I T T I E , H. W I L D E R M U T H " A N D D. A. H O P W O O D John Innes Institute, Colney Lane, Norwich, NOR 70 F (Acceptedf o r publication I 5 February 1972) SUMMARY

Thin sections, freeze-etch preparations and carbon replicas of Thermoactinomyces vulgaris sporangia and endospores were examined in the electron microscope. The outer spore coat contained parallel arrays of long fibrous striations spaced at about 5 nm; there appeared to be several layers with a different orientation of the fibres in each. The outer coat was polyhedral with slightly concave pentagonal and hexagonal faces. A combination of geometrical considerations, analysis of the patterns of faces seen in the replicas and freeze-etchings, and model building suggested that the polyhedron had I 2 pentagonal and approximately I 2 hexagonal faces.

INTRODUCTION

The thermophilic actinomycete Thernzoactinomyces vulgaris produces thick-walled, heatresistant endospores with morphological and biochemical properties very similar to those of the endospores of the eubacterial genera Bacillus and Clostridiunz (Cross, Walker & Gould, 1968 ; Dorokhova, Agre, Kalakutskii & Krassilnikov, 1968, 1970; Cross, Davies & Walker, 1971). Our results from thin sections have confirmed this earlier work while freeze-etchings and carbon replicas have provided new information on the fine structure of the outer spore coat and on the overall polyhedral shape of this layer. The strain of Thermoactinomyces vulgaris used in this study has been shown to have a system of genetic exchange involving transformation (Hopwood & Ferguson, I 970 ; Hopwood & Wright, 1971, 1972). Genetic analysis of endospore formation in T. vulgaris should therefore be possible. The cortex and spore coat layers of T. vulgaris appear identical to those of Bacillus species and, as far as their formation is concerned, genetic analysis would be expected merely to duplicate studies already well developed in Bacillussub tilis(Schaeffer, I 969). On the other hand, because of the mycelial nature of actinomycete growth, details of the initiation of sporangium and forespore formation may well differ from those in the unicellular endospore-forming genera. Secondly, the polyhedral shape of the outer layer in T. vulgaris spores presents an interesting problem which may be open to genetic analysis by the isolation and examination of mutants. METHODS

Organism and medium. The studies reported here were carried out on Thermoactinomyces vulgaris strain ~9 grown on MMC minimal medium. Details of the origin and growth of this strain are in the accompanying paper by Hopwood & Wright (1972). Liglzt microscopy. Coverslip cultures (Williams & Davies, 1967) were examined and photographed in a Zeiss Photomicroscope.

*

Present address : Zoologisches Institut der Universitat, Kunstlergasse I 6, Zurich, Switzerland. 24-2

368

A. M C V I T T I E , H. WILDERMUTH A N D D. A. H O P W O O D

Electron microscopy. Coated grids were laid on sporulating cultures, pressed down gently and evenly and removed for direct viewing of whole, untreated spores in the electron microscope. For thin sectioning, ' spore suspensions' in distilled water were prepared from slopes after 3 days of incubation (Hopwood & Wright, 1972). Fixation and embedding were carried out according to the procedure of Wildermuth (1970) with the following modifications. The spore suspensionswere centrifuged at iooog for 10min and the pellets resuspended in 5 to 6 ml of glutaraldehyde fixative. After fixation, washing in cacodylate buffer and subsequent centrifugation the pellets were embedded in 2 % agar and the resulting blocks cut into pieces approximately I mm3 which were transferred to osmium tetroxide for postfixation. After subsequent dehydration the material was left overnight in a mixture of 3 parts 1,2-epoxypropane to I part Araldite (without accelerator) and then transferred to a I to I mixture, then to a I to 3 mixture and finally to pure Araldite on successive days. After overnight soaking in Araldite plus accelerator, polymerization in fresh Araldite mixture was carried out at 60 "C. Thin sections were cut on a diamond knife, stained with lead citrate and viewed in a Siemens Elmiskop ra. Freeze-etching. Spores were collected from mass cultures grown in Petri dishes by scraping with a loop, suspended in 25 yo glycerol, centrifuged and the pellet frozen-etched according to the method of Moor (1964). Carbon replicas. Spores were harvested as before and washed once in distilled water. Drops of a dense suspension were placed on grids freshly coated with a thick formvar film. After drying in air replicas were made according to the method of Bradley & Williams (1957) omitting the metal shadowing. Thick replicas were made to reduce breakage during etching and washing. RESULTS

General features. General features of the aerial mycelium were shown by examining sporulating coverslip cultures under phase-contrast. Mature spores appeared spherical and highly refractile (Fig. I, 3). Non-refractile, presumably immature, spores were also seen (Fig. I, 2). Some spores were borne directly on the main hyphae (sessile spores) but others were attached to the hypha by a very short stalk (Fig. 3). In the electron microscope the spores had a polygonal outline; in Fig. 4 they are shown in situ still attached to the lysed hypha. Fine structure of spores and sporangia. Examination of thin sections in the electron microscope yielded further information on the arrangement of spores on the aerial hyphae. Each spore developed singly within a sporangium which appeared as a lateral extension of the parent hypha (Fig. 5). The spore was thus surrounded by sporangial cytoplasm which appeared identical to the hyphal cytoplasm. The sporangium was limited by a plasma membrane and wall continuous with the hyphal membrane and wall (Fig. 5 to 7). A narrow electron-transparent zone was sometimes seen between the outer surface of the spore and the surrounding cytopIasm (Fig. 6). In Fig. 5, sectioned parallel to the long axis of the hypha, the hyphal segment adjacent to the one bearing the sporangium has undergone lysis. The plane of section in Fig. 6 and 7 is roughly perpendicular to the long axis of the hypha. Fig. 5 and 6 show fully mature spores still within the sporangium, while Fig. 7 shows an immature stage prior to the development of the spore coats. At this stage these outer layers are represented only by the thin electron-dense outer membrane. The material harvested from sporulating cultures consisted largely of mature spores which had been released from their sporangia (Fig. 8,9). The various layers surrounding the spore core were clearly seen. The innermost of these is the cortex with a broad electron-

Structure of Thermoactinomyces vulgaris spores

Fig. I to 3. Phase-contrast photomicrographs of sporulating aerial mycelium. Note mature, refractile spores and non-refractile, presumably immature, spores. Bars represent 10pm. Fig. 4.Electronmicrograph of whole, untreated spores attached to a lysed hypha. Bar represents I pm.

369

370

A. MCVITTIE, H. W I L D E R M U T H A N D D A. H O P W O O D

Fig. 5 to 7. Thin sections showing mature (Fig. 5 , 6) and immature (Fig. 7) endospores within sporangia. Bars represent 0.25 pm.

Structure of Thermoactinomyces vulgaris spores

371

translucent inner zone and a narrow more electron-dense outer zone (also referred to as the diffuse layer by Dorokhova et al. ( I 970)). Surrounding this is the inner spore coat consisting of several concentric electron-dense layers. Between these and the thick electron-dense outer spore coat there is a less dense intermediate coat. Shrinkage frequently caused the cortex to separate from the inner coat as in Fig. 8. Mature spores usually had a polygonal outline in cross-section, the polygonal shape being due to the outer coat. The edges, corresponding to sections of the faces of a polyhedron in three dimensions, tended to be concave (Fig. 8). Sections usually showed eight to ten edges. Freeze-etchings provided little information about the internal structure of spores and sporangia since cross-fracturing was rare, presumably because of the hardness of the spore coats. The cross-fractured object shown in Fig. I I may be a sporangium and if so, the object in its centre is presumably a spore. Its exposed surface has a pattern of intermeshed bands differing from that seen on the mature spore (Fig. 13, 14); it may be the outer layer of an immature spore or a subsurface layer of a mature spore, e.g. a layer of the multilayered inner coat. The strands of material within the cytoplasm are presumably membranes and their presence may indicate that the cell is lysing. The pitted plasma membrane can be seen together with ‘connecting strands’ running into the relatively thick cell wall. Two cross-fractured spores are shown in Fig. 10. Their appearance is similar to that of sectioned spores, but it is more difficult to distinguish the boundaries of the cortex and coat layers. Fig. 12 shows a mature spore partially released from its sporangium. Remnants of the sporangium can be seen adhering to the spore. F h e structure of the spore surface. Frozen-etched preparations of unfractured spores showed that mature spores were polyhedral in shape with pentagonal and hexagonal faces (Fig. 14, 17, 18). Details of the fine structure of the spore surface were also well shown. The most striking feature was a pattern of striations with a centre-to-centre spacing of about 5 nm between the dark lines (Fig. 13, 14). The outer layer thus appeared to consist of sheets of long fibres arranged in parallel arrays, these arrays being sometimes slightly curved (Fig. 13). As shown in Fig. 13, there were at least three of these fibrous sheets one above the other with a different orientation of the fibres in each. If, as seems likely, these sheets constitute the outer spore coat, up to six might be expected since the outer coat has a thickness of up to 30 nm. The direction of the fibres did not necessarily change at the edge bounding two faces; conversely, changes of direction were seen within faces but these were probably where breaks occurred exposing an underlying sheet (Fig. 14). It is of course possible that the breaks were introduced during the freeze-fracturing procedure. Polyhedral geometry of the mature spore. Regular polygonal faces were not a l w y s seen in freeze-etchings (e.g. Fig. 13), but in all instances where they were clearly visible they were pentagons (Fig. 17) or hexagons (Fig. 18) and appeared to be concave. Occasionally the three faces meeting at a common vertex were unambiguously visible. Only two of the four possible patterns were seen: two pentagons and one hexagon ( ‘ 5 , 5 , 6’ vertex, Fig. 14) and one pentagon and two hexagons (‘5, 6, 6’ vertex, Fig. 17). The freeze-etchings presented an intriguing problem: the spores seemed to have a definite polyhedral shape but how could this shape be determined when at most one could see clearly only three or four of the faces of a given spore? Carbon replicas provided the means for a partial solution to this problem. In carbon replicas the surface of the spore was often traversed by sharply defined ridges which tended to lie at the edges of the faces, thus outlining them more clearly than in the freeze-etchings (Fig. 19). As in the freeze-etchings ‘5, 5, 6’ (Fig. 20) and ‘ 5 , 6, 6’ (Fig. 19) vertices were present while ‘ 5 , 5, 5’ and ‘6, 6, 6’ vertices were not seen.

372

A. MCVITTIE, H. W I L D E R M U T H A N D D. A. H O P W O O D

Fig. 8, 9. Thin sections of mature spores showing spore core (SC) surrounded by cortex (C), inner multi-layered coat (IMC), intermediate coat (IC) and outer coat (OC). Fig. 8 shows the polygonal shape of the outer coat. Bars represent 0-25 pm. Fig. 10.Freeze-etch preparation showing two cross-fractured spores. The spore core (SC) and outer coat (OC) are clearly visible but the boundaries of the other layers are indistinct. Bar represents 0.25 p m ; encircled arrow indicates shadowing direction.

Structure of Thermonctinomyces vulgaris spores

Fig. I I , 1 2 . Freeze-etchings. Bars represent 0.25 pm; encircled arrows indicate shadowing direction. Fig. I I . Possibly a cross-fractured sporangium containing a spore (SP). The plasma membrane (PM) is pitted and there are short ‘connecting strands’ at the membrane-cell wall boundary (small arrows).

Fig.

12.

Mature spore showing surface fibrous pattern and remnants of the surrounding sporangium (SPM).

373

374

A. M C V I T T I E , H. W I L D E R M U T H A N D D. A. H O P W O O D

Fig. I 3. Freeze-etch preparation; the outer spore coat shows fibrous striations running in different directions in different layers (small arrows). Bar represents 0.25 ,urn; encircled arrow indicates shadowing direction.

It was possible to examine a large number of spores since many fields similar to the one shown in Fig. 19 were photographed. Information on the following points was obtained: the relative numbers of pentagonal and hexagonal faces; the relative numbers of ‘ 5 , 5’’ ‘ 5 , 6’ and ‘6,6’ edges (i.e. edges between two pentagons, one pentagon and one hexagon, or two hexagons); the relative numbers of ‘ 5 , 5, 6’ and ‘ 5 , 6, 6’ vertices; and the average edge length of the polygons. The photographs of the replicas were examined independently by the three authors and the results are given in Table I . The observers differed in the number of faces which they considered scorable but there was good agreement that pentagons and hexagons occurred in approximately equal numbers ; moreover, the relative frequencies of ‘ 5 , 5’, ‘ 5 , 6’ and ‘6, 6’ pairs scored by the three observers were not significantly different (x: = 3-09;P 0.6) and were therefore averaged (Table I).

-

Structure of Therniouctinomyces vulgar is spores

Fig. 14. Freeze-etch preparation showing polyhedral shape and fibrous pattern in the outer spore coat. A ‘ 5 , 5 , 6 ’ vertex is shown. There is a discontinuity in fibre orientation within the lower pentagonal face. Bar represents 0 . 2 5 ,urn; encircled arrow indicates shadowing direction. Fig. 1 5 , 16. Photographs of a 24-face cardboard model made with equal numbers of regular pentagons and hexagons. Two different orientations are shown, both apparently compatible with the view of the spore in Fig. 14.

375

376

A. MCVITTIE, H. WILDERMUTH A N D D. A. H O P W O O D

Fig. 17, 18. Freeze-etchings of spores showing a ‘5, 6, 6’ vertex (Fig. 1 7 ) and a hexagonal face (Fig. I 8). Bars represent 0.5 ,urn; encircled arrows indicate shadowing direction. Fig. 19,20. Carbon replicas of spores showing pentagonal and hexagonal faces outlined by ridges. Fig. 19 shows ‘5,6,6’vertices (arrows), and Fig. 20 shows a ‘5,5,6’vertex. Bar represents 0.5,um.

Structure of Thermoactinomyces vulgaris spores Table

I.

Numbers of single faces, pairs of faces with a common edge and sets of three faces around a common vertex seen in carbon replicas of spores Totals scored

Observer I 2

3

Average

377

Single faces

y h - - 7 &r

Spores

Faces

92 91 99 -

163 I33 172 -

(5*’

‘6-/-’

Two faces with common edge

Three faces with common vertex

A

‘55’

‘5,6’

‘6’6’

‘5,5,6’ ‘5’6,6’

83 80 9 57 I7 4 8 65 68 6 35 8 2 5 I7 55 16 7 8 88 84 79(51%> 77(49 %> I 1(14%) 49(67 %> 14(19%> 4(38 %> 7(62 %>

*

Pentagon. f- Hexagon,

From a knowledge of the geometrical properties of polyhedra it was now possible to determine approximately the total number of faces since, for polyhedra with trihedral vertices, 4f2+3f3+2f4+f5+of6-f,-2f8 . . . = 12, where ‘f’ stands for face and the subscript indicates the number of edges of the face (Thompson, 1952). Thus for a polyhedron consisting only of pentagons and hexagons there must be 12 of the former but there may be any number of hexagons. If the spore has 12 pentagons and, as was found experimentally assuming no bias due to any favoured orientation of spores on the grid, an approximately equal number of hexagons this gives an approximate value of 24 for the total number of faces. It was also possible to derive an estimate of the number of faces by a second approach involving model building. Measurements on five pentagons and five hexagons from the replicas gave average edge lengths of 0.19 pm for pentagons and 0.18 ,urn for hexagons; the weighted average for all 10 polygons was 0.18 pm. The spores seen in the replicas were approximately spherical with a mean diameter of 0.81 pm. The models consisted of pentagons and hexagons outlined on the surface of a spherical ball of plasticine using short lengths of plastic-coated wire cut to correspond to the average edge length. They were built up vertex by vertex, starting at one point, until the whole surface was covered with polygons; the three angles at each vertex were made approximately equal. Details of the five models made are given in Table 2 . In most cases restrictions were imposed as shown in the Table. The first model could be rejected since several of the polygons had very unequal angles. Moreover, there were six ‘ 5, 5, 5 ’ or ‘6, 6, 6 ’ vertices out of a total of 50. The shape of this polyhedron would be somewhat irregular because ‘5, 5, 5’ vertices would result in protruding bumps and ‘6, 6, 6’ vertices would create flattened areas on the surface. The numbers of different kinds of edges and vertices in the remaining four models were then considered to see whether it would be possible to distinguish between the models by comparing these frequencies with those found in the replicas. It was found, however, that the 24- and 25-face and one of the two 23-face models had very similar edges and vertices (Table 3). The data available were incapable of distinguishing between them, although analysis of a larger number of faces, edges and vertices could, in principle, differentiate between certain models. But these four models do not exhaust the possibilities even for 23-, 24- and 25-face polyhedra. There is, for instance, at least one other 24-face polyhedron lacking ‘5,5, 5 ’ and ‘ 6,6,6’ vertices and differing only in the arrangement of pentagons and hexagons from the plasticine model made, and there are presumably others containing such vertices. 22- and 26-face models were not investigated, but it is likely that they too would be

A. MCVITTIE, H. W I L D E R M U T H A N D D. A. H O P W O O D

378

Table

Details of spore models made by outlining pentagons and hexagons on the surface of spherical plasticine balls

2.

Number of polygons A

r

Total

\I

Pentagons Hexagons

Restrictions imposed

27

I2

15

None

24

I2

12

‘ 5 , 5 , 5 ’ and ‘6,6,6’

23

I2

I1

23

I2

I1

25

12

13

‘5,595’ and ‘6,6,6’ vertices 2 ‘595’5’

4 ‘6,6,6’

-

vertices excluded Pentagons in three None rows of four 2 ‘6,6,6’ Pentagons in two rows of six (a) Started with one 2 ‘6,6,6’ pentagon surrounded by five hexagons (b) ‘ 5 ’ 5 5 ’ vertices excluded

compatible with the existing data. The q f a c e model with pentagons arranged in two pairs and two rows of four adjacent ones was made by sticking together regular polygons (i.e. with equal angles and edges) cut out of thin card; Fig. 15 and 16 show two views of this model compatible with the pattern of faces seen in the freeze-etching in Fig. 14. It was thus possible to conclude only that the spores had an external polyhedral shape with 12 pentagonal faces and approximately 12 hexagonal ones, making a total of approximately 24. It was impossible to determine by the methods applied so far whether there was a unique polyhedral shape or whether two or more different polyhedra were present. DISCUSSION

The appearance in thin sections of Thermoactinomyces vulgaris spores was similar to that reported by Cross et al. (1968), Dorokhova et al. (1968, 1970), and Cross et al. (1971). The formation of endospores within a sporangium attached to the parent hypha was especially clearly shown here (Fig. 5). Additional details of the sporangium revealed by freeze-etching were the small particles on the plasma membrane and the short ‘connecting strands’ of fibrous material which were seen running across the boundary between the membrane and wall (Fig. 11). Both features were also seen in freeze-etchings of vegetative cells of Bacillus species by Holt & Leadbetter (1969) and of Streptomyces by Wildermuth (1971). The fibrous structure shown in freeze-etchings is clearly present on the outer surface of the spore and may be continuous throughout the outer spore coat. Spores which had not

Structure of Thermoactinomyces vulgaris spores

379

emerged fully from the sporangium showed that this structure was not present on the sporangial wall surface (Fig. 12; Cross et al. 1971). That the fibrous layer was not present in the intermediate or inner spore coats was inferred from the fact that it was this layer which also showed the polyhedral shape known from sections to be confined to the outer coat (see Fig. 14). Fibrous layers were also seen in the spore coats of Bacillus fastidiosus, B. subtilis and B. Iicheniforrnis by Holt & Leadbetter (1969). As in Thermoactinomyces vulgaris the fibres in these bacilli were spaced at about 5 nm and the fibres in a given layer were orientated in different directions with respect to those in the layers above and below. In B. coagulans Gould, Stubbs & King (1970) showed fibres with a 5-7 nm spacing on the outer surface of the spores. The fibrous pattern was correlated with the presence of an alkali-soluble protein in the outer coat. Freeze-etchings provided very clear evidence for the polyhedral nature of Thermoactinornyces vulgaris spores (Fig. 14, 17, 18). The polygonal faces had a concave curvature as could also be seen in sections (Fig. 8). Since there were no obvious ridges between the faces it is likely that the ridges seen in carbon replicas (Fig. 19, 20) were to some extent artifacts resulting from the formation of folds or pleats in the outer layer of the spore integument when the underlying layers shrank during drying. (A shrinkage is indicated by the fact that the diameter measured in carbon replicas (0.81,urn) was less than that determined by light scattering of fresh spores in aqueous suspension, which was 1.05 pm, assuming the spore to be spherical - M. J. Daniels, personal communication.) Similar ridges were seen in scanning electron micrographs of spores of the related species T. sacchari (Lacey, 1971) and of both T. vulgaris and Actinobzjida dichotomica (Williams, I 970). Scanning electron microscopy may, however, also involve some distortion of shape due to desiccation under vacuum, and spores fixed and dehydrated in an attempt to preserve their natural shape showed less pronounced ridges than untreated ones (Williams, I 970). With the techniques used here it was possible to conclude only that the shape of Thermoactinomyces vulgaris spores corresponded to a polyhedron with approximately 24 faces, I 2 being inevitably pentagonal and the remaining ones hexagonal. Scanning electron microscopy would reveal a greater proportion of the surface of a given spore. It would therefore be possible to examine a large number of vertices and perhaps to determine also whether pentagons occurred singly, in pairs, or in rows of three or four. It would then be easier to distinguish between some of the possible polyhedra (see Tables 2 and 3). But since certain pairs of polyhedra are so closely related it is uncertain whether it would be possible, even with this technique, to determine whether or not the spores had a unique polyhedral shape. The polyhedra discussed above are clearly different from the only regular polyhedron consisting entirely of pentagonal and hexagonal faces, the truncated icosahedron. This is an isometric structure with 12 pentagons and 20 hexagons; the vertices are all of the ‘5, 6,6’ kind and there are no adjacent pentagons (see Cundy & Rollet, 1961). The polyhedra with 23’24 or 25 faces can only be isometric, as in the plasticine models, if the faces are not regular polygons but have small differences in angles and edge lengths; the cardboard models made from regular polygons were not isometric (Fig. 15, 16). Since all the electron microscope evidence suggested that the spores were approximately isometric, the faces are probably not completely regular polygons. This is in contrast to the situation in the ‘spherical’ viruses where the virus shape corresponds exactly to a polyhedron with regular polygonal faces. Virus capsids are, however, only about one tenth or less of the diameter of Thermoactinoinyces vulgaris spores and are formed by the assembly of one or a few kinds of protein subunit. The much larger outer polyhedral shell of T. vulguris spores is more complex, being

380

A. MCVITTIE, H. WILDERMUTH AND D. A. HOPWOOD

composed of superimposed fibrous sheets which appear to be added to a basically spherical structure during spore maturation. In view of these differences in size and development a close similarity between the structure of virus capsids and spore outer coats is not expected. An analysis of mutants with altered spore shape, if they can be isolated, should reveal to what extent the polyhedral shape is under genetic control, while further study of the fibrous component of the outer coat should show if there is any regular relationship between the molecular architecture of this layer and its overall polyhedral shape. A comparative study of Thermoactinomyces vulgaris and Actinobijida dichotomica spores should also prove rewarding. The latter are clearly polyhedral (Williams, 1970)but, although they are similar in size to those of T. vulgaris, the shape appears different; the spore seems to have many fewer faces and may be a dodecahedron with 12 pentagonal faces. The existence of such alternative architecture in itself indicates an element of genetic control. It will be interesting to know how the detailed structure of the outer spore coat of A . dichotomico compares with that of T. vulgaris. The significance of the polyhedral topography of the spores of these thermophilic actinomycetes is not certain. However, it is conceivable that the polyhedral layer plays a role in preventing deformation of the spores in the same way as comparable structures in architecture (Marks, 1960). It may be significant that the spores of Thermoactinomyces vulgaris have been shown to survive for over IOO years in nature (Cross & Johnston, 1971). We are grateful to E. Wehrli, Department of General Botany, Swiss Federal Institute of Technology, Zurich, who kindly made the freeze-etchings, and to Professor R. W. Horne for helpful criticism of the manuscript. REFERENCES

BRADLEY, D. E. &WILLIAMS, D. J. (1957). An electron microscope study of the spores of some species of the genus Bacillus using carbon replicas. Journal of General MicrobioZogy 17,75-79. CROSS,T., DAVIES, F.L. & WALKER,P.D. (1971). Thermoactinomyces vulgaris. I. Fine structure of the developing endospores. In Spore Research - 1971, pp. 175-180. Edited by A. N. Barker, G. W. Gould & J. Wolf. London: Academic Press. CROSS, T. & JOHNSTON, D. W. (1971). Thermoactinomyces vulgaris. 11. Distribution in natural habitats. In Spore Reseavch-Ig71, pp. 315-330. Edited by A. N. Barker, G. W. Gould & J. Wolf. London: Academic Press. CROSS,T., WALKER, P. D. & GOULD, G. W. (1968). Thermophilic actinomycetes producing resistant endospores. Nature, London 220, 352-354. CUNDY,H. M. & ROLLET,A. P. (1961). Mathematical ModeZs, 2nd edn, p. 1 1 0 . Oxford: Clarendon Press. DOROKHOVA, I. A., AGRE,N. S., KALAKUTSKII, L. V. & KRASSILNIKOV, N. A. (1968). Fine structure of spores in a thermophilic actinomycete, Micromonospora vulgaris. Journal of General and Applied Microbiology 14,295-303. DOROKHOVA, L. A., AGRE,N. S., KALAKUTSKII, L. V. & KRASSILNIKOV, N. A. (1970). Electron microscopic study of spore formation in Micromonospora vulgaris. Microbiology (a translation of Mikrobiologiya) 39, 589-593. GOULD, G. W., STUBBS, J. M. & KING,W. L. (1970). Structureand composition of resistant layers in bacterial spore coats. Journal of General Microbiology 60,347-355. HOLT,S. C. & LEADBETTER, E. R. (1969). Comparative ultrastructure of selected aerobic spore-forming bacteria: a freeze-etching study. Bacteriological Reviews 33, 346-378. HOPWOOD, D. A. & FERGUSON, H. M. (1970). Genetic recombination in a thermophilic actinomycete, Thermoactinornyces vulgaris. Journal of General Microbiology 63, I 33-1 36. HOPWOOD, D. A. & WRIGHT,H. M. (I 97 I). Genetic transformationin Therrnoactinomycesvulgaris. Heredity 27, 483. HOPWOOD, D. A. & WRIGHT,H. M. (1972). Transformation in Thermoactinomyces vulgaris. Journal of General Microbiology 71,383-398.

Struciure of Thermoactinomyces vulgaris spores

387

LACEY,J. ( I 97 I). Thermoactinomyces sacchari sp.nov., a thermophilic actinoniycete causing bagassosis. Journal of General Microbiology 66, 327-338. MARKS,R. W. (1960). Dymaxion World of Buckminster Fuller. New York : Reinhold. MOOR,H. (1964). Die Gefrier-Fixation lebender Zellen und ihre Anwendung in der Elektronenmikroskopie. Zeitschrijit fur Zellforschung 62, 546-580. SCHAEFFER, P. (1969). Sporulation and the production of antibiotics, exoenzymes and exotoxins. Bacteriological Reviews 33, 48-71. THOMPSON, D. W. (1952). Growth and Form, 2nd edn (reprinted), p. 737. Cambridge University Press. WILDERMUTH, H. (1970). Development and organization of the aerial mycelium in Streptomyces coelicolor. Journal of General Microbiology 60, 43-50. WILDERMUTH, H. (1971).The fine structure of mesosomes and plasma membrane in Streptomyces coelicolor. Journal of General Microbiology 68, 53-63. WILLIAMS, S. T. (1970). Further investigations of actinomycetes by scanning electron microscopy. Journal of General Microbiology 62, 67-73. WILLIAMS, S. T & DAVIES,F. L. (1967). Use of a scanning electron microscope for the examination of actinomycetes. Journal of General Microbiology 48, I 7 I - I 77.