New Phytol. (1997), 137, 373–388

Tansley Review No. 96 Structural diversity in (vesicular)–arbuscular mycorrhizal symbioses B  F. A. S M I T H"*    S. E. S M I T H# " Department of Botany, The University of Adelaide, SA 5005, Australia # Department of Soil Science, University of Adelaide, SA 5005, Australia (Received 11 February 1997)  Summary I. Introduction : Arum-types and Paris-types II. Possible functional implications III. Extent of the two classes in the plant kingdom 1. Bryophytes and Pteridophytes 2. Gymnosperms 3. Angiosperms

373 374 375 377 377 379 379

IV. V. VI. VII. VIII.

Is the distinction between classes useful ? The structural basis The role of the fungal genome Physiology revisited Conclusions Acknowledgements References

 This review describes diversity in the structure of (vesicular)–arbuscular (VA) mycorrhizas, i.e. endomycorrhizas formed by Glomalean fungi. In particular, we consider the extent in the plant kingdom of the two classes first described by Gallaud (1905). These are : (1) the Arum-type, defined on the basis of an extensive intercellular phase of hyphal growth in the root cortex and development of terminal arbuscules on intracellular hyphal branches ; (2) the Paris-type, defined by the absence of the intercellular phase and presence of extensive intracellular hyphal coils. Arbuscules are intercalary structures on the coils. However, there have been many reports that in Paris-types arbuscules are relatively few in numbers, small, or absent altogether. A survey of the literature has revealed that Paris-types occur more frequently in the plant kingdom than Arumtypes and predominate in ferns, gymnosperms and many wild angiosperms. The cultivated herbs that are the subject of much experimental work are mostly Arum-types. Although evidence is still limited, there are differences at the family level. In 41 angiosperm families there are records of only Paris-type VA mycorrhizas and in 30 families records of only Arum-types. Another 21 families have examples of both classes, or intermediates between them. Accordingly, we consider whether the original division into two classes is still useful. We conclude that it is when considering the physiology of the symbiosis and especially the issue of whether different fungus}host interfaces have specialized roles in transfer of inorganic nutrients and organic carbon between the partners. If there is no such specialization between hyphal coils and arbuscules, then the latter might not be necessary for the function of Paris-types. This would account for reports of the infrequency or absence of arbuscules in this class. The control exerted on structures by the genomes of host and fungus, and possible reasons (anatomical and physiological) for the existence of the VA mycorrhizal structures, are discussed. The presence or absence of extensive intercellular spaces and differences in the wall structure of cortical cells might be particularly important in determining which type of VA mycorrhiza is formed. Key words : VA mycorrhizas, root anatomy, plant taxonomy, nutrient transport.

* To whom correspondence should be addressed. E-mail : asmith!botany.adelaide.edu.au

383 383 384 384 385 386 386

374

F. A. Smith and S. E. Smith

.            : 

A R U M-    



P A R I S-

Gallaud (1905) described two major structural classes of (vesicular)–arbuscular (VA) mycorrhizas, which he named Arum-types and Paris-types, after the plants in which they were first described. In both classes, the initial epidermal penetration results in intracellular hyphae, often forming coils (‘ pelotons ’) in the hypodermis, where present, and outer cortex. The Arum-type mycorrhiza was defined by the existence of extensive intercellular hyphae underlying this initial phase of colonization after which, following penetration into cortical cells, arbuscules are formed, normally as terminal structures on socalled trunk hyphae. These structures are summarized in Figure 1. Figure 2 shows some of the many illustrations by Gallaud (1905) and Figure 3 shows Arum-type structures in Allium as seen by confocal laser scanning microscopy. Vesicles can be intercellular or intracellular, and are not formed by all VA mycorrhizal fungi. Accordingly, the term ‘ arbuscular mycorrhiza ’ is coming into use. Nevertheless, as explained previously (Smith, 1995 ; Smith & Smith, 1996 ; Smith & Read, 1997), we will retain this abbreviation in the present review for overall simplicity and for consistency with most of the literature. Also we wish to avoid undue emphasis on arbuscules alone, since not all VA mycorrhizas appear to form arbuscules. Thus we use ‘ VA ’ generically for mycorrhizas formed by fungi of the order Glomales sensu Morton & Benny (1990) : the issue is further discussed below. Gallaud (1905) defined the Paris-type VA mycorrhiza by the absence of intercellular hyphae. Instead, there are extensive coils of intracellular hyphae within cortical cells, from which arbuscules are normally formed as intercalary structures (Figs 1, 2). Figure 3 shows hyphal coils in the cortical cells of Panax as seen by confocal laser scanning microscopy. One of the cells illustrated contained no arbuscules, but small and indistinct arbuscules were present in adjacent cells. Vesicles, where present in Paristypes, are intracellular. On the basis of his survey, Gallaud believed this class to be the less common. Much subsequent experimental work has been done with temperate crop plants that have Arum-type mycorrhizas (e.g. Allium, Trifolium, Vicia etc), and this class has – almost by default – come to be regarded as the norm. In an earlier and classic study (Janse, 1897), the two classes are also clearly distinguishable, although the fine detail of arbuscules was not recognized. In that case lack of taxonomic knowledge of the fungi led to some confusion, with emphasis on similarities with the intracellular hyphal structures – especially the hyphal coils – in other endomycorrhizas that were studied, i.e. in what are now known to be ericoid and orchid mycorrhizas. The situation was further

confused by Peyronel (1923, 1924), who studied both types of structures and concluded that the fungi that produced hyphal coils in Paris-types were secondary colonists in roots that were already colonized by the true VA mycorrhizal fungi. Peyronel believed that these secondary fungi did not produce arbuscules or vesicles and that the hyphae were septate to a degree that does not occur in the fungi that form VA mycorrhizas. Harley (1969) gave a summary of Peyronel’s work ; see also Harley (1991) for a more detailed account and biography. Peyronel’s view was certainly incorrect for Paristypes in general because, as shown by Gallaud (1905), some do produce both arbuscules and vesicles, whereas the intraradical hyphae in Arumtypes can and do produce septa depending on the host}fungus combination (e.g. Hildebrand & Koch (1936) and many papers subsequently). Also, both intra- and extra-radical hyphae can become septate under various conditions which include damage (Gerdemann, 1955), age (Kinden & Brown, 1976) and pre-penetration stages of colonization (Giovannetti et al., 1993). The confusion over whether Paris-types are ‘ genuine ’ VA mycorrhizas should have been laid to rest by experimental investigations which showed conclusively that VA mycorrhizal fungi isolated from Arum-type hosts produce Paris-type structures in other hosts and vice versa. The first such report that we have found was a conference abstract by Barrett (1958), stating that an isolate of Rhizophagus – the generic name then used for Glomalean fungi – ‘ that developed the Arum type (sic) in Zea mays produced the Paris type in Solanum tuberosum, and a peloton type in Iris. A Paris type from Heterocallis developed the Arum type in Zea mays. ’ (We have not identified ‘ Heterocallis ’ – it might be the liliaceous plant Hemerocallis.) Barrett continued : ‘ These results indicate that the host, not the endophyte, determines the host–fungus relationships ’. This immensely important point has been confirmed by others (see below). However, Barrett (1958) also said that ‘ the tests reported here do not confirm Gallaud’s findings ’, which seems odd, seeing that Gallaud’s classification did not directly address the issue of control by the host or fungus. Also, Barrett’s distinction between the Paris-type and peloton type did not help, seeing that pelotons (coils) are a definitive feature of Paris-types. Work by others suggests that the Solanaceae includes representatives of both classes : see below. A similar investigation was by Gerdemann (1965), who showed that the same Glomalean isolate produced a Paris-type VA mycorrhiza in Liriodendron (tulip-tree) and an Arumtype in Zea. Jacquelinet-Jeanmougin & GianinazziPearson (1983) likewise showed that the development of Paris-type structures in Gentiana was produced by Glomalean fungi that formed Arumtypes in Allium. The Gentianaceae, which includes

Structural diversity in VA mycorrhiza (a) Arum-type (a) Paris-type

Figure 1. Summary of cortical structures in the two classes of VA mycorrhizas described by Gallaud (1905). The size of arbuscules, especially in Paris-types, can vary greatly. From Smith & Smith (1996), with permission.

achlorophyllous (mycoheterotrophic sensu Leake, 1994) genera is a classic Paris-type family, as first shown by Janse (1897) ; see also Gay, Grubb & Hudson (1982), McGee (1985) and other references given later. Control of many major mycorrhizal structures by the genome of the host was also emphasized by Bonfante-Fasolo & Fontana (1985), Daniels-Hetrick, Bloom & Feyerherm (1985) and Smith (1995). Nevertheless, despite the impressive experimental evidence, it is still possible that the fungal genome might also exert some control over the structures. We discuss this point later. There have been numerous studies of the development of Arum-type structures, including those by Mosse & Hepper (1975) with Trifolium, BonfanteFasolo & Scannerini (1977) with Ornithogalum, Holley & Peterson (1979) with Phaseolus, Morton (1985) with Zea, Alexander et al. (1989) with Allium, Phaseolus and Zea, and Weber, Klahr & MarronHeimbuch (1995) with plants in the Apocynaceae. The review by Bonfante-Fasolo (1984) gives other references. Detailed work on development of Paristypes (or near-Paris-types) includes that by Kinden & Brown (1975, 1976) with Liriodendron, BonfanteFasolo & Fontana (1985) with Ginkgo, Yawney & Schultz (1990) and Cooke, Widden & O’Halloran (1992, 1993) with Acer, Weber et al. (1995) with plants in the Apocynaceae, and Whitbread, McGonigle & Peterson (1996) with Panax. Bonfante-Fasolo’s (1984) review covers Paris-types, although she does not use this term or distinguish them explicitly. Surveys of the comparative anatomy of mycorrhizas by Brundrett & Kendrick (1988, 1990 a, b), Brundrett, Murase & Kendrick (1990) and Cooke et al. (1992) have shown the widespread occurrence of Paris-types in wild plants – both herbs and woody plants. Widden (1996) has described structures in members of the Liliaceae (sensu Cronquist, 1981) that are variations within the Paristypes and which were also noted by Gallaud (1905) in Colchicum (also Liliaceae sensu Cronquist), as shown in Figure 2. The elegant illustrations in the recent papers (light and scanning electron micrographs) show the presence of arbuscules associated with intracellular hyphal coils. Brundrett & Kendrick (1990 b) suggested that it is not yet clear which class is the most common. This comment has

375 led to our surveying the literature in an attempt to assess the extent of Paris-types in relation to Arumtypes. In addition, we have made some observations of our own with tropical plants including Durio zibethinus (for which there have been conflicting comments over its mycorrhizal status), Artocarpus and Nephelium spp. All of these turned out to have Paris-type mycorrhizas with small arbuscules attached to very extensive intracellular hyphal coils (Smith et al., 1997). There are many reports in the literature of absence of arbuscules, or their presence in only limited numbers, in Paris-type VA mycorrhizas (e.g. Boullard, 1953 a, b ; Alexander, 1988 ; see also Smith & Smith, 1996). This obviously renders the definition of such mycorrhizas rather hazardous in both structural and functional terms if it is assumed that arbuscules are the sole basis of the mutualistic symbiosis. Apparent absence of arbuscules might be a result of seasonal effects or environmental stress (Brundrett & Kendrick, 1990 a ; Yawney & Schultz, 1990 ; Mullen & Schmidt, 1993). The formation of arbuscules in American ginseng (Panax quinquefolius) has been shown to depend on the season but also differs greatly among combinations of farms and plant age-classes, apparently reflecting the characteristics of individual seed-beds (Whitbread et al., 1996). .    Our own interest in diversity in VA mycorrhizas arose from considerations of differences in ‘ mycorrhizal efficiency ’ – i.e. the variation in nutritional benefits and costs to the plant that are associated with differences in amounts of inorganic nutrients such as P or Zn transferred to the plant in return for organic carbon (e.g. sugar) transferred to the fungus. We suggested (Smith & Smith, 1996) that the physiological basis of the symbiosis might be related to the structural features summarized above. Previously, it has been widely assumed that bidirectional transfer of inorganic nutrients and organic C occurs solely across arbuscules. Indeed, in their taxonomic revision of the Glomales, Morton & Benny (1990) took this assumedly unique role of the arbuscules as a fundamental feature of the Order. As we pointed out (Smith & Smith, 1996), there is no firm evidence to rule out other fungus–host interfaces as sites of transfer in VA mycorrhizas. In other mycorrhizas which lack arbuscules (both ecto- and endomycorrhizas) other such interfaces must obviously be the sites of transfer. The dynamics and life-span of Arum-type arbuscules have been measured in several studies (e.g. Cox & Tinker, 1976 ; Toth & Miller, 1984 ; Alexander et al., 1989 and references therein) but estimates of the surface areas of both the arbuscules and intraradical fungal hyphae in VA mycorrhizas generally are notably lacking. This is

376

F. A. Smith and S. E. Smith

Figure 2. For legend see opposite.

Structural diversity in VA mycorrhiza due to the technical problems in measuring them, and especially in determining longevity and activity. In experiments with Allium porrum, the relative surface area of the intercellular interface with respect to that of the arbuscules increased with time and became the larger after about 8 wk of plant growth (Smith & Dickson, 1991 ; Smith et al., 1994). There have as yet been no attempts to make comparable measurements in Paris-types, an omission that we hope to rectify soon using the techniques developed previously (Smith & Dickson, 1991 ; Smith et al., 1994). It has been suggested that in Arum-types such as Allium, P and possibly other inorganic nutrients are transferred to the host across the arbuscular interface – as is generally accepted – and that the intercellular interface is a major route for transfer of sugar from host to fungus (Gianinazzi-Pearson et al., 1991 b). Spatial separation of transport would provide a structural basis for differences in mycorrhizal efficiency. In the absence of functional arbuscules, a VA mycorrhizal fungus could still survive in the intercellular phase, essentially as a parasite or ‘ cheater ’ (sensu Soberon & Martinez del Rio, 1985). ‘ Cheating ’ by the fungus – i.e. gaining organic carbon without supplying P to its co-symbiont – would either prevent a positive growth response by the plant or produce a negative response (Janos, 1985, 1987). Obviously, if transfer of solutes occurred solely across arbuscules, extensive growth of the fungus would not be possible when arbuscules are lacking or inactive. Evidence for the occurrence of cheating by VA mycorrhizal fungi is reviewed by Smith & Smith (1996) ; see also Johnson, Graham & Smith (1997). The experimental basis of our suggestion remains slim, though there is extensive evidence for decreases in numbers of arbuscules in Arum-types under various controlled environmental conditions including low light, high external P and sometimes low temperature (see Smith & Smith, 1996). More important, however, are the ‘ myc−# ’ mutants of the Arum-type plant Pisum, in which colonization occurs but functional arbuscules are not formed (GianinazziPearson et al., 1991 a ; Gianinazzi-Pearson et al., 1994). In these cases the intercellular interface must be involved in obtaining organic C to support fungal growth. Clearly, our suggestion is not valid in the same form in Paris-type mycorrhizas, because the extensive intercellular interface is absent. However,

377 despite the lack of quantitative measurements of surface areas, there is no doubt whatsoever that the intracellular hyphal coils can be extremely extensive compared with arbuscules, as shown by illustrations in many publications (e.g. Brundrett & Kendrick, 1990 b ; Whitbread et al., 1996 ; Widden, 1996), and as we have found with our own observations (Smith et al., 1997). If there is spatial separation of transfer, the coils could be the site of transfer of organic C, with transfer of P across the localized arbuscular interface. Cheating would again be possible if arbuscules were absent or were inactive. As yet there is no evidence for separation of transfer of P and organic C in Paris-type VA mycorrhizas. If there is no separation of transfer the hyphal coils might totally take over the role of arbuscules when the latter are absent or not functioning. This would mean that there is nothing physiologically special about arbuscules, i.e. that they are not essential for a functional Paris-type VA mycorrhiza. There is nothing novel about this suggestion. Possible involvement of hyphal coils in nutrient exchange has been mentioned by several workers who have studied Paris-type mycorrhizas in autotrophic plants (e.g. Bonfante-Fasolo & Scannerini, 1977 ; Reinsvold & Brent Reeves, 1986 ; Louis, 1990 ; Whitbread et al., 1996) but this role has been ignored by most VA mycorrhizologists. Demuth & Weber (1990) and Imhof & Weber (1997) suggested that Paris-type structures and absence of arbuscules in the mycoheterotrophic plant Voyria (Gentianaceae) reflects a ‘ structural incompatibility ’ in which the fungus obtains no organic C from the host. In this case, transfer of organic C and inorganic nutrients to the plant must involve the intracellular hyphal coils. Our purpose in the present review is not again to rehearse the detailed arguments for spatial separation of nutrient transfer or to assess techniques by which the issue might be resolved (see Smith & Smith, 1996). Here we consider the relative extent of the two classes of VA mycorrhizas, the validity and usefulness of this division, and its structural basis in terms of growth of the roots and mycorrhizal fungi. .          1. Bryophytes and Pteridophytes The absorbing organs of a wide range of lower land plants have long been known to be mycotrophic and

Figure 2. Assemblage of illustrations photocopied from Gallaud (1905), retaining the original numbers. 27 : Allium sphaerocephalum (Arum-type), longitudinal section. 28 : A. sphaerocephalum, longitudinal section, showing passage cell and point of penetration. 31 : Anemone nemorosa (Paris-type), longitudinal section. 21 : Colchicum automnale, transverse section, showing point of penetration and hyphal swellings. 41 : arbuscule of Arum maculatum. 42 : ‘ compound ’ arbuscule of Sequoia gigantea (Paris-type). Abbreviations : ap, epidermis ; as, hypodermis ; pa, root-hair ; c, passage cell ; end, endodermis ; l, air-space ; noy, nucleus ; r, hyphal swellings ; v, young vesicles (according to Gallaud).

378

F. A. Smith and S. E. Smith (a)

(b)

Figure 3. Extended focus images using laser scanning confocal microscopy. (a) Paris-type structures in the cortex of Panax quinquefolius ; Z-reconstruction of 17 slices (2 µm between slices). (b) Arum-type structures in the cortex of Allium porrum ; Z-reconstruction of 14 optical slices (1 µm between slices). Samples were embedded in LR White resin and stained with Phloxine B. S. Dickson (unpublished) with thanks to L. Peterson, M. Farquhar & Y. Uetake.

Structural diversity in VA mycorrhiza these groups were studied by Janse (1897) and Gallaud (1905), among many other early workers. Rayner (1927) and Harley (1969) gave good summaries of the early work, and Bonfante-Fasolo (1984) has reviewed more recent developmental aspects. Bryophytes are colonized by a range of fungal endophytes and previous uncertainties about the fungal taxa are now being addressed (see, for example, Duckett & Read, 1995). VA mycorrhizal (Glomalean) fungi occur as endophytes in many bryophytes and they are very widespread in pteridophytes. Where the cells of the absorbing organs have ready access to the soil environment, as in some thalloid structures lacking epidermal-type layers, there can be direct penetration of external hyphae between cells (i.e. formation of an intercellular phase) but in the intracellular phase extensive aseptate hyphal coils are common, and arbuscules (where present) are intercalary rather than terminal on hyphae. A good example is the gametophyte of Phaeoceros (Bryophyta : Anthocerophyta), which has been studied in detail by Ligrone (1988) ; see also Stahl (1949). The prothalli (gametophytes) of different Lycopodium spp. (pteridophytes) show variation in ‘ mycorrhizal ’ structures (the absorbing organs are not roots !), as first shown by Treub (1884–89, cited and discussed by Rayner, 1927). Whereas L. cernuum had intracellular hyphae in the peripheral region, there were intercellular hyphae in the central tissues. In other species described by Treub the hyphae are entirely intracellular. Interestingly, the prothalli of L. cernuum contain photosynthetic cells while those of the second group studied by Treub are achlorophyllous. A more recent investigation of Lycopodium by Duckett & Ligrone (1991) confirmed Treub’s findings and extended them to the ultrastructural level. These examples aside, and allowing for the simple anatomy of the absorbing organs, bryophytes and ‘ lower ’ pteridophytes (nonFilicales) mostly have VA structures that are Paristypes. However, absence of arbuscules seems quite common, as found for the thalloid hepatic Cryptothallus by Pocock & Duckett (1984), for Lycopodium by Duckett & Ligrone (1991) and for Psilotum (pteridophyte) by Peterson, Howarth & Whittier (1981) ; see also Janse (1897). Both Arum-type and Paris-type VA mycorrhizal structures have been recorded in roots of the Filicales (true ferns). Extensive Arum-type intercellular hyphae and arbuscules were first described in Angiopteris (Marattiaceae : Marattiales) by Janse (1987) and Gallaud (1905). This structure was later confirmed by Boullard (1958), who showed that in contrast Danaea (same family) forms the Paris-type. The position in the Ophioglossales is also not clearcut. Janse (1897) described intercellular hyphae in ‘ Ophioderma ’ (Ophioglossum) but Gallaud (1905) included the mycorrhiza in Ophioglossum as a Paris-

379 type, stating that coils are absent but the hyphae are always intracellular. Descriptions and illustrations by Boullard (1958) show Paris-type features including coils, vesicles and the remnants of arbuscules in Ophioglossum and the related genus Botrychium. As far as we are aware, representatives of all families in the leptosporangiate Filicales that have been examined appear to have Paris-type structures (e.g. Boullard, 1958 ; Cooper, 1976 ; Berch & Kendrick, 1982 ; see also Harley, 1969). Last, and not least, Remy et al. (1994) have now unequivocally demonstrated the presence of arbuscules in the early Devonian fossil Aglaophyton, which shows features of both bryophytes and vascular plants. It would be very interesting to know if this plant was a Paris-or Arum-type : it ought to be possible to detect the occurrence of intercellular hyphae or intracellular coils in the fossils. The structures described by Kidston & Lang (1921) from the same fossil flora (Rhynie chert) included, as well as vesicles, coiled intracellular hyphae, suggesting that they were Paris-types. 2. Gymnosperms There have been many studies of VA mycorrhizas in gymnosperms, and in all cases except possibly one they can be classified as Paris-types, irrespective of whether or not the investigators have used this term. Examples include Podocarpaceae : (Janse, 1897 ; Shibata, 1902 ; Gallaud, 1905 ; Baylis, McNabb & Morrison, 1963) ; Taxaceae : (Prat, 1926 ; Strullu et al., 1981) ; Taxodiaceae (Gallaud, 1905 ; Konoe$ , 1957). Bonfante-Fasolo (1984) gives a general survey. The possible exception is Ginkgo (Ginkgoaceae), in which Bonfante-Fasolo & Fontana (1985) found ‘ rare ’ intercellular hyphae and vesicles and abundant intracellular hyphal coils and intercalary arbuscules in the inner cortex. Accordingly, Ginkgo should perhaps strictly be included in a ‘ near-Paris ’ or ‘ Intermediate ’ class ; it is certainly not an Arumtype. We note here that the presence of rare intercellular hyphae in other examples classified as Paris-types might compromise the validity of the classification. More such examples will occur below. The methodology of the investigation might be a problem in that the presence of rare intercellular hyphae may be obscured by the dense intracellular coils and missed altogether when root squashes are examined, rather than sections. Also, standard staining methods are not uniformly successful. Early workers, such as Janse (1897) and Gallaud (1905), and others including Bonfante-Fasolo & Fontana (1985) did not fall into these traps. 3. Angiosperms Arum-type and Paris-type VA mycorrhizas are widely scattered through both monocotyledonous and dicotyledonous families in the angiosperms.

380

F. A. Smith and S. E. Smith

Table 1. Family groupings of angiosperms having distinct Arum-type and Paris-type VA mycorrhizas and those with both types and}or intermediate types Arum-types

Paris-types

Monocots

Agavaceae (1)# Araceae* (5)",#,#* ‘ Liliaceae ’ Alliaceae (1)","&,"',#& Asphodelaceae (1)# Anthericaceae (1)# Convallariaceae (4)#,#* Hyacinthaceae (4)# Hypoxidaceae (1)" Ruscaceae (1)# Zingiberaceae (1)"

Dicots

Alangiaceae (1)" Anacardiaceae (1)%# Ascepiadaceae* (4) #,"%,$% Balsaminaceae (1)" Begoniaceae (1)" Boraginaceae (2)# Campanulaceae (1)" Combretaceae (1)( Compositae (5)",#,( Cucurbitaceae (2)") Elaeocarpaceae (2)" Guttiferae (2)(,"% Malvaceae (5)( Oleaceae (1)"%,$! Proteaceae (1)" Rosaceae (2)",#,"%,$!,$( Staphyleaceae (1)" Symplocaceae (1)" Thymeleaceae (1)"% Turneraceae (1)( Urticaceae (2)"% Vitaceae (1)(

Burmanniaceae (1)" Cannaceae (1)( Dioscoraceae (2)#,( Heliconiaceae (1)( ‘ Liliaceae ’ Colchicaceae (2)#,#) Liliaceae s. s. (1)#* Trilliaceae (3)#,#*,$( Uvulariaceae (2)",$( Marantaceae (1)( Thismiaceae (1)",) Triuridaceae (1)" Aceraceae (1)","#,"%,#' Annonaceae (1)( Araliaceae (2)(,$) Aristolochiaceae (10)#* Bombacaceae (2)(,%! Caricaceae (1)( Casuarinaceae (1)" Cecropiaceae (1)( Cornaceae (1)"% Cunoniaceae (1)$" Gentianaceae (7) ","%,##,#%,#(,$* Grossulariaceae (1)" Hamamelidaceae (10)" Hippocastanaceae (10)%" Lecythidaceae (1)( Linaceae (1)"%,##,%$ Loganiaceae (1)$$ Magnoliaceae (2)","!,"( Malpighiaceae (1)( Melastomaceae (1)( Moraceae (2)",%! Myrsinaceae (1)" Myrtaceae (1)" Polygalaceae (1)"% Rubiaceae* (6)",(,"% Sapindaceae (1)%! Saxifragaceae (1)# Theaceae (1)" Ulmaceae (2)",$! Umbelliferae (6)#,"% Violaceae (1)#,"%

Both types (B) and}or intermediate types (I) Araceae* (1P)( Gramineae (" 10) (B,I)%,(,*,"!,"% Arecaceae (3) (B)",(,#" Pandanaceae (1) (I)"

Apocynaceae (14) (B,I)",$& Asclepiadaceae* (I)$' Burseraceae (1) (I)",( Caprifoliaceae (1) (I)","% Euphorbiaceae (6) (B,I)",(,"", Flacourtiaceae (1) (I)" Labiatae (7) (B)",#,(,"% Leguminosae Caesalpinoideae (1) (I)( Mimosoideae (3) (B)",&,( Papilionoideae (13) (B) #,(,"*,#!,#$ Meliaceae (1) (I)","$ Menyanthaceae (2) (I)$# Ranunculaceae (3) (B)#,"% Rubiaceae* (1A)( Rutaceae (6) (B,I)",$,( Scrophulariaceae (2) (B)( Solanaceae (4) (B)",( Sterculiaceae (2) (I)',(,#" Verbenaceae (5) (B)(,%%

The system of Cronquist (1981) – see also Mabberley (1989) – has been used except for families within ‘ Liliaceae ’ (sensu Cronquist), which are as given by Dahlgren et al. (1985). Key : (1, etc), number of genera recorded ; (B), both types recorded ; (I), intermediate characters (extensive coils and intercellular hyphae). Araceae* (etc) in two columns occurs where there is a single record that differs from the majority ; (1P) or (1A) indicates the minority record in the ‘ Both ’ category. Superscripts : ",# etc ¯ references, as follows. " Janse (1897) ; # Gallaud (1905) ; $ Mcluckie & Burges (1932) ; % Endrigkeit (1937) ; & Asai (1944) ; 'Laycock (1945) ; ( Johnston (1949) ; ) McLennan (1958) ; * Nicolson (1959) ; "! Gerdemann (1965) ; "" Wastie (1965) ; "# Kessler (1966) ; "$ Redhead (1968) ; "% Stelz (1968) ; "& Mosse (1973) ; "' Hayman (1974) ; "( Kinden & Brown (1975) ; ") Saif (1977) ; "* Abbott & Robson (1978) ; #! Holley & Peterson (1979) ; #" Nadarajah (1980) ; ## Gay et al. (1982) ; #$ Carling & Brown (1982) ; #% Jacquelinet-Jeanmougin & Gianinazzi-Pearson (1983) ; #& Brundrett, Piche! & Peterson (1985) ; #' Frankland & Harrison (1985) ; #( Ku$ hn & Weber (1986) ; #) McGee (1986) ; #* Brundrett & Kendrick (1990 b) ; $! Brundrett et al. (1990) ; $"McGee (1990) ; $#Weber & Kramer (1994) ; $$ Tiemann, Demuth & Weber (1994 a) ; $% Tiemann, Demuth & Weber (1994 b) ; $& Weber, Klahr & Marron-Heimbuch (1995) ; $' Untch & Wber (1996) ; $( Widden (1996) ; $) Whitbread et al. (1996) ; $* Imhof & Weber (1997) ; %! Smith et al. (1997) ; %" H. J. Hudson (pers. comm.) ; %# L. Haugen & S. E. Smith (unpublished) ; %$ B. Thomas & S. E. Smith (unpublished) ; %% P. O‘Connor & F. A. Smith (unpublished). Other references to genera that have been widely studied (e.g. Allium and Acer) are given in the text.

Structural diversity in VA mycorrhiza Table 1 lists families in each class, giving numbers of genera that have been studied in each family, and using the classification by Cronquist (1981). In some cases the names in the early literature had to be updated, using Mabberley (1989) for this purpose. While it is not established that all species in one genus are all within the same class, we have found only one case (Ranunculus) where different species within a single genus appear to form VA mycorrhizas of both classes, and another (Zea) where there may be major differences between cultivars. These are discussed below. Table 1 relies heavily on the surveys by Janse (1897), Gallaud (1905), Johnston (1949) and Stelz (1968), with examples from other papers as listed. Not all authors have used Gallaud’s classification, and we have placed families in the Arumtype and Paris-type classes on the basis of records of presence or absence of intercellular hyphae in the cortex ; and in the latter class, presence of hyphal coils. Generally, all examples included have been claimed to be VA (or arbuscular) mycorrhizas by the authors cited, except of course where the work was done before these terms were introduced. Fortunately, early workers usually included detailed illustrations which have helped confirm the basis of the classification. We have not been deterred by reports of total absence of arbuscules, since all such cases in Table 1 (though not identified there) included statements of presence of vesicles (sometimes ‘ rare ’) and hyphae that are mainly aseptate. Some potentially useful lists have been excluded because, although they include plants with extensive intracellular hyphae, they do not specify whether intercellular hyphae are absent. Examples are surveys by Thapar & Khan (1973) of forest trees (both gymnosperms and angiosperms) in India, and by Louis (1990) of coastal plants in Singapore. As already noted, Louis was conscious of the possible functional implication of absence of arbuscules. In many cases, Table 1 gives records for only one species or genus per family, which is a hazardous basis for generalizing, but where different workers have made examinations of the same species or genera they have almost always come to the same conclusion about structures. Examples to the contrary are discussed below. Thus, classic Arum-type families include Compositae and Malvaceae, whereas Paris-type families include Aceraceae and Gentianaceae. This list is not exhaustive, and it does not include a number of early surveys published in languages other than English that we have not seen. Nevertheless, it shows that Paris-types are indeed very common, as suggested by Brundrett & Kendrick (1990 b). The Liliaceae – as defined by Cronquist – seems at first sight a particularly ‘ difficult ’ family, with representatives of both classes (Table 1). It includes variations of Paris-types that have distinctive in-

381 tracellular hyphal swellings (‘ bobbits ’). These were described in detail by Widden (1996) in Clintonia and Medeola. Rather similar features in Colchicum were noted by Gallaud (1905), as shown in Figure 2. The swellings are a consistent feature of these plants. The classification of liliaceous plants is currently under review and the different structural classes can be distributed quite neatly (bearing in mind the limited numbers of genera surveyed) between the families listed by Dahlgren, Clifford & Yeo (1985). This has been done in Table 1. Table 2 gives more details, and shows the distinct distribution within the Superorder Liliiflorae sensu Dahlgren et al. (1985), which includes taxa from outside the Liliaceae sensu Cronquist (1981). Ornithogalum (Hyacinthaceae) was described as an Arum-type by Gallaud. A detailed study of O. umbellatum by Bonfante-Fasolo & Scannerini (1977) showed extensive intracellular hyphal coils in the outer cortex with intercellular hyphae in the inner cortex. This difference appears in other plants (see below). Extensive ramifications of intercellular hyphae between epidermis and cortex in Thysanotus (Anthericaceae) are another example of a variant in the Arum-type class (McGee, 1988). Structural variation within Paris-type dicots is illustrated by the presence of lobed intracellular hyphae in Linum catharticum (illustrated in Gay et al., 1982). Families with genera of both classes are also listed in Table 1 (last column : ‘ B ’). Examples include the Apocynaceae, Euphorbiaceae and Solanaceae (this last being a favourite for experimental studies). Some plants contain Paris-type features (especially extensive hyphal coils) but also many intercellular hyphae : see Table 1 : last column ‘ I ’. The Apocynaceae again includes genera in this category (Weber et al., 1995). Obviously, as with the gymnosperm Ginkgo, discussed above, there is a complication where rare intercellular hyphae have been recorded in roots that are otherwise distinctly Paristype. Theobroma (cacao) is one such example (Laycock, 1945 ; Nadarajah, 1980). Gallaud (1905) wrestled with these issues when he found that some Ranunculus spp. fell into the Paris class, whereas others had Arum-type features. This appears to be the sole example where species of one genus were put into both classes by a single investigator. The most complicated family is the Leguminosae, where the situation is not clarified when it is divided into the different sub-families. The Gramineae are also complicated (Table 1) and we have not attempted to sort out distribution of the VA mycorrhizal classes into the graminaceous sub-families. Nicolson’s (1959) survey of grasses in Britain suggested that Paris-types are very common. Greny (1973) made a detailed anatomical– morphological study of several important members of the Gramineae and concluded that Zea, Avena, Triticum, Hordeum, Lolium and Holcus all have

382

F. A. Smith and S. E. Smith

Table 2. Distribution of Arum-types and Paris-types within the Superorder Liliiflorae sensu Dahlgren et al. (1985) (for simplicity, only names of genera are listed) Arum-types O. Asparagales Convallariaceae*

Ruscaceae* Agavaceae Hypoxidaceae* Asphodelaceae* Anthericaceae* Hyacinthaceae*

Alliaceae* O. Burmanniales Thismiaceae

Paris-types

Convallaria Maianthemum Polygonatum Smilacina Ruscus Agave Curculigo Aloe Anthericum Hyacinthoides Muscari Ornithogalum

Gallaud (1905) Gallaud (1905) Gallaud (1905) Brundrett & Kendrick (1990 b) Gallaud (1905) Gallaud (1905) Janse (1897) Gallaud (1905) Gallaud (1905) Gallaud (1905) Gallaud (1905) Gallaud (1905), BonfanteFasolo &Scannerini (1977) Gallaud (1905) Janse (1897) ; et multi alii

Scilla Allium

O. Dioscorales Dioscoraceae Trilliaceae*

O. Liliales Liliaceae sensu stricto* Uvulariaceae*

References

Thismia

Janse (1897) McLennan (1958)

Tamus Dioscorea Paris Trillium Medeola† Colchicum† Wurmbea Erythronium Disporum† Clintonia† Uvularia†

Gallaud (1905) Gallaud (1905) Gallaud (1905) Brundrett & Kendrick (1990 b) Widden (1996) Gallaud (1905) McGee (1986) Brundrett & Kendrick (1990 b) Janse (1897) Widden (1996) Girard (1985)

* Included in the Liliaceae sensu Cronquist (1981). † Hyphal swellings : see text.

Arum-type mycorrhizas whereas Phleum, Alopecurus, Agrostis and Agropyrum have Paristypes. By contrast, Johnston (1949) described Zea as having Paris-type mycorrhizas. Although he did not refer to Gallaud’s classification, his survey of plants in the West Indies split them on the basis of (a) the presence of an extensive intercellular phase in the outer and inner cortex, or (b) the presence of extensive intracellular hyphal coils in the cortex. Apart from the distinction between outer and inner cortex, these are essentially the same as Gallaud’s definitions. Since the many other workers who have used Zea have shown it to contain extensive intercellular hyphae, this poses a problem, and it must be emphasized that a number of families are in the ‘ B ’ category in Table 1 as a result of Johnston’s survey of plants from the West Indies. Geographical factors aside, there may be no need to have doubts about Johnston’s powers of observation, because a number of investigators have noted that Zea has hyphal coils in the outer cortex (e.g. Winter, 1951 ; Meloh, 1963 ; Gerdemann, 1965 ; Greny, 1973 ; Kariya & Toth, 1981 ; Morton, 1985). Hence in this case there might be plasticity, which might be

exhibited in different maize cultivars. Triticum and Hordeum can also have extensive hyphal coils (Greny, 1973). Hyphal coils in the outer cortex occur in other genera, including tobacco (Hildebrand & Koch, 1936) and some legumes, including Medicago and Trifolium spp. (Abbott, 1982). They may reflect differences in cortical structure and function. In addition to formation in the outer cortex, hyphal coils are frequently formed during penetration of fungal hyphae through cells of hypodermes of both Arum-types and Paris-types, as shown by several early workers, including Gallaud (1905) ; see also Smith, Long & Smith (1989) – we were then in ignorance of the earlier work. Hypodermes, though frequently overlooked, are present in many plants (e.g. Peterson, 1988 ; Perumalla, Peterson & Enstone, 1990). Weber and his collaborators have published a valuable series of papers that survey development of VA mycorrhizas in the Apocynaceae, Asclepiadaceae, Gentianaceae and Loganiaceae (Table 1). These families are all members of the Order Gentianales. As already noted, the achlorophyllous members of the Gentianaceae are classic

Structural diversity in VA mycorrhiza Paris-types. However, the other families in the Order include Arum-types, both types or intermediates – the Apocynaceae contains all of these (Table 1 ; Weber et al., 1995). Weber and his colleagues have proposed that Paris-types are phylogenetically more advanced than Arum-types, reflecting weakened fungal vigour in some hosts and hence an evolutionary tendency towards successful mycoheterotrophy (e.g. Imhof & Weber, 1997). Given the frequency of Paris-types, and their dominance in lower plants, we are not convinced by this argument. Much of the information in Tables 1 and 2 has been obtained with field-grown plants and hence plants of uncertain age. Preliminary results by So$ derstro$ m et al. (1996) have suggested that up to 18 d Petroselinum crispum (parsley) and Daucus carota (carrot) – both in the Umbelliferae – display some Arum-type features, with Paris-type features predominating afterwards. These developmental effects deserve further study in relation to root anatomy and the ageing and turn-over of individual fungal colonization units.

.      ? We believe that the answer to this question lies in the context in which it is asked. Systematists who are interested in structures other than reproductive ones might question the usefulness of the classification if it is blurred by the occurrence of intermediates or (especially) of taxa at family or genus level with representatives of both classes. However, some problems can be cleared up by revision of plant taxa, as with the ‘ old ’ Liliaceae, and it is even possible that distribution of VA mycorrhizal classes might be helpful where taxonomic characters include nonreproductive features, as with liliaceous plants (Dahlgren et al., 1985). Developmental changes – as suggested by So$ derstro$ m et al. (1996) with parsley and carrot – are another complication in this context. If the interest lies in the development and physiology of the symbiosis and (especially) any differentiation in function between the different interfaces, then the distinction might still be important. This especially relates to whether structural differences can be correlated with differences in mycorrhizal responsiveness (growth increases compared with non-mycorrhizal controls) and the whole question of cheating, as summarized above. The issue is then not whether any given interface is totally absent, but whether limitations or changes in its surface area (or molecular activity) affect rates of solute transfer. The presence or absence of arbuscules, at least, is a feature in which most VA mycorrhizologists ought to be interested. Obviously, environmental considerations must to be taken into account in considering structural and functional

383 plasticity, given evidence for changes in colonization rates, intraradical hyphae and numbers of arbuscules under low light, high soil P, low temperature etc (see Smith & Smith, 1996, for discussion). However, there is no evidence that the structural basis of the classes is under environmental control.

.    The distinction between the two VA mycorrhizal classes and the occurrence of intermediates is certainly important if features of roots (morphological, physiological or both) are being sought that might promote the formation of one class or another. The simplest possibility is that Arum-types occur in roots with large intercellular spaces within the cortex. This possibility was considered over-simple by Gallaud (1905). For example, he pointed out that those Ophioglossaceae that have mycorrhizas falling within the Paris class do so despite the presence of intercellular spaces. Other Paris-types, such as Voyria (Imhof & Weber, 1997) and Colchicum (see Fig. 2) also have obvious intercellular spaces in their roots and there can be few roots in which intercellular spaces of one sort or another are completely absent, given the need for aeration. Nevertheless, Brundrett & Kendrick (1988, 1990 b) have proposed that Arumtypes are formed in roots which have continuous longitudinal air-spaces within their cortices (or inner cortices, where most intercellular hyphal growth often occurs), thus allowing a pathway of low physical resistance for hyphal growth, compared with the intracellular pathway. This intercellular route then gives the opportunity for rapid sequential penetration of cortical cells and growth of arbuscules, and intracellular vesicles, where present. The occurrence of limited or discontinuous intercellular spaces in roots, or differences between outer and inner cortices, might account for the intermediate VA mycorrhizal structures. Brundrett & Kendrick (1990 a) demonstrated that the intracellular spread of colonization in Paris-types was relatively slow when compared with spread in Arum-types. However, it is still not clear why, in Paris-types, hyphae do not grow extensively between appressed root cell walls and along tortuous intercellular pathways. Perhaps these pathways are physically and biochemically less favourable than sequential penetrations of cell walls. In other words, differences in cell wall structure and modifications produced during fungal colonization might be important (Bonfante-Fasolo & Fontana, 1985 ; Bonfante-Fasolo, 1988). The structural basis of the classes proposed by Brundrett & Kendrick is based on few examples and needs to be tested more rigorously with a wider range of plants. Justin & Armstrong (1987) surveyed the root structure and porosity of 91 plant species,

384

F. A. Smith and S. E. Smith

grown in non-mycorrhizal conditions, with the aim of assessing anatomical responses to flooding. Their results certainly showed quantitatively that even in drained soil there are large air-spaces in roots of some plants that form Arum-type mycorrhizas (e.g. Zea – particularly in the inner cortex) or that are from Arum-type families (e.g. Compositae : Table 1). However, the porosity of roots of Allium and other putative Arum-types (e.g. Vicia and Lotus spp.) is no greater than that of Paris-types such as Viola and members of the Umbelliferae (Justin & Armstrong, 1987). What is required to put Brundrett and Kendrick’s proposal on a firmer foundation is comparison of the (three-dimensional) intraradical spread of the fungal hyphae with the development and distribution of intercellular spaces, as determined by appropriate infiltration techniques, in a wider range of Arum-type, Paris-type and intermediate plants. Intercellular spaces in roots have received little direct attention from most VA mycorrhizologists and their distribution and extent can become obscured where root squashes or crosssections are used to identify the presence of VA mycorrhizas. If the formation of the individual classes of VA mycorrhiza depends primarily on the presence or absence of cortical intercellular spaces, or indeed on any other anatomical property of the plant root, then the taxonomic issue is transferred immediately to that of the value of root anatomy as a taxonomic character. Certainly, it is not surprising that there is much consistency of root anatomy in individual genera or families, especially where plants have not been bred for cultivation ; i.e. for rapid growth. It then seems a truism that the formation of the two classes or intermediates between them is primarily under the genetic control of the host. We note again here the seminal demonstrations of the formation of Arum-types or Paris-types when different hosts were colonized by the one species or isolate of VA mycorrhizal fungi (Barrett, 1958 ; Gerdemann, 1965 ; Jacquelinet-Jeanmougin & Gianinazzi-Pearson, 1983). Inevitably, however, there are complications and more questions to address.

.       Abbott (1982) compared the anatomy of ten Glomalean endophytes in the Arum-type Trifolium subterraneum. She showed that isolates differed with respect to production of vesicles, hyphal branching and diameter, etc, as found by others. More important in the present context, she found that isolates of Gigaspora and Glomus formed extensive ‘ loops ’ of intracellular hyphae in the outer cortex with intercellular hyphae plus (intracellular) arbuscules in the inner cortex. Acaulospora was different in that it produced ‘ predominantly ’ in-

tracellular hyphae (including coils) throughout the cortex of T. subterraneum. Differences between fungal species or strains in (primarily) Arum-types might result from differences in their ability to penetrate the various cell types within the root. However, it is still not clear why some fungi prefer a more tortuous intracellular pathway when extensive intercellular spaces are common. Demuth, Forstreuter & Weber (1991) found differences in the size of hyphal coils and in the size and shapes of arbuscules when species of Gentiana (Paris-type) were colonized by different isolates of Glomus and Acaulospora. We conclude that it is possible for the fungal genome to exert significant control on VA mycorrhizal structure at the Arum}Paris-type level. It seems unlikely that an individual fungal species or strains can greatly affect the extent of intercellular spaces within the root. The amount of plasticity in hyphal structures that can be superimposed by genomic variation in fungi as well as in hosts clearly requires more study with other host}fungus combinations, and especially the quantification of the relative sizes of intracellular and intercellular interfaces that can be produced by different fungi on the one plant species. More subtle variation has been found by B. So$ derstro$ m & S. Dickson (unpublished) in Petroselinum (parsley), in which Glomus sp. (WUM 16) formed extensive coils and few arbuscules. The reverse was the case with G. mosseae. The distinction was not apparent in Daucus (carrot), again pointing to interactions between genomes of fungus and host in determining structural features. .   The issue of whether the intercellular and intracellular environments are the more favourable for growth of the fungal endophyte is important to the plant physiologist interested in VA mycorrhizas. Intuitively, an intracellular interface (whether a coil or arbuscule) seems a more favourable environment for nutrient transfer than an intercellular one. Even though the intracellular interface is topologically still apoplastic, with modified cell walls between the fungal and plant plasma membranes, the latter are much more closely appressed than they are in an intercellular interface – especially one of comparatively large volume. Intuition aside, however, there is no reason to suspect that cortical intercellular spaces in roots are nutritionally unfavourable for growth of VA mycorrhizal fungi. Concentrations of 40–50 m hexose equivalents (20–25 m sucrose) have been measured in cortical apoplasts of nonmycorrhizal roots of barley (Farrar, 1985) and Ricinus (Chapleo & Hall, 1989). These concentrations were quite similar to the cytoplasmic concentrations. The condition that is necessary for sustained fungal uptake of organic C in the cortical apoplast is

Structural diversity in VA mycorrhiza that the rate of release from cortical cells must keep pace with (or be increased by) fungal demand. There are at present too many unknowns to approach this issue mathematically. In addition, the intercellular phase might be more favourable for maintenance of O levels for fungal respiration. As already noted, the # presence or absence of hypodermes – which increase the isolation of the cortical apoplast from the external (soil) environment – does not appear to be associated with the occurrence of the two classes. Possible interrelationships between mycorrhizal structures and function are well worth further investigation in mycoheterotrophic plants. Those that are VA mycorrhizal appear all to have Paristype mycorrhizas. Besides the Gentianaceae, examples include Thismia spp. (Thismiaceae : sometimes included in Burmanniaceae) : see Table 1. Since VA mycoheterotrophic plants receive organic C via their fungal endophyte from an adjacent donor plant (rather than directly from soil) there must be significant physiological differences from ‘ normal ’ VA mycorrhizal plants, at least at the level of membrane transport proteins and their operational control. There seem to be strong parallels in structure between the Paris-type VA mycorrhizas in heterotrophic gametophytes of Lycopodium, plants of the Burmanniaceae and Gentianaceae and the (non-VA) mycorrhizas in orchids. The cells of the last contain hyphal coils, as shown by Janse (1897) and many workers subsequently : representative illustrations are in Smith & Read (1997). Although heterotrophic growth of orchids mostly depends on a supply of organic C from soil via the fungal hyphae, there can be transfer from an autotrophic ‘ donor ’ plant via the hyphae. Control of mycorrhizal structures by the plant is nicely shown by the formation of hyphal coils in orchids by fungi from an ectomycorrhizal host (Zelmer & Currah, 1995). Since protocorms or roots of orchids seem to lack large intercellular spaces (as shown by illustrations in Smith & Read, 1997), absence of an intercellular phase is not surprising if the reasoning applied above to VA mycorrhizas is again applied. There can also be transfer of organic C between linked autotrophic VA mycorrhizal plants of the same or a different species : see Smith & Smith (1996). Despite previous uncertainties about the significance of amounts of organic C transferred and its form (Smith & Smith, 1996), the use of stable isotopes is now giving good evidence of significant net transfer of organic C between ectomycorrhizal plants at least (Simard et al., 1997). There is nothing yet to implicate the structural considerations in this review in transfer of solutes between linked VA mycorrhizal plants. Nevertheless, following the reasoning of Imhof & Weber (1997), it would be nice to know if Paris-type interfaces are better adapted than Arum-type interfaces as sites for transfer of organic C in autotrophic receiver plants

385 .  This survey of the relative extent of Arum- and Paris-type VA mycorrhizas has shown that the latter are not rare and anomalous cases, as has been sometimes thought. They occur in a vast number of wild plants, including woody ones in which it is not easy to distinguish details of structures – as we found when examining Durio, Nephelium and Artocarpus (Smith et al., 1997). In examples such as these, arbuscules can often be obscured by the dense hyphal coils and vesicles can be rare or lacking, depending on the fungal endophyte and environmental conditions. Accordingly, some Paris-types may well have not have been recognized as VA mycorrhizas. There is now considerable scope for DNA ‘ finger-printing ’, to identify unknown fungal endophytes, especially in lower plants and in other examples where normal features used for identification – whether vesicles, arbuscules or both – are absent. Increasing use of this technique to identify mycorrhizal fungi from plants collected from the field can be anticipated. Whether there are significant functional differences between the various interfaces in the two classes remains to be seen. It is particularly important to establish if the presence of extensive intracellular hyphal coils in Paris-types can render arbuscules unnecessary as sites for nutrient transfer. This would help explain the numerous reports of absence of arbuscules in this class. However, it is possible that in at least some cases absence of arbuscules only occurs in some phases of root growth, as shown with an Alpine Ranunculus sp. by Mullen & Schmidt (1993). This will be a problem with surveys of field-grown material where the age of the roots is uncertain. Control of formation of arbuscules by the host vis-a[ -vis the fungus is highly relevant to the vexed question of nomenclature (cf. Smith, 1995 ; Walker, 1995). If a so-called ‘ arbuscular ’ mycorrhizal fungus can form a functional mycorrhiza without arbuscules, owing to control solely exerted by the host (the emphasis is important) what is the best nomenclature to use – for fungus and symbiosis ? Perhaps we should start referring to ‘ Glomalean mycorrhizas ’ to avoid all structural and functional issues. In that case, it will certainly be unwise to define the Glomales by the presence of arbuscules, with an explicitly stated functional role for these structures in nutrient interchange, as was done by Morton & Benny (1990). It is for this reason that we have not adopted ‘ Glomalean mycorrhizas ’ throughout the present review. Another possibility – though difficult to pronounce at first sight – is ‘ zygomycetous ’ or – simpler – ‘ Z-mycorrhiza ’ (Trappe, 1987). This avoids all structural issues. Less provocatively, we hope that this review will prompt renewed interest in the distribution of the

386

F. A. Smith and S. E. Smith

two classes between and within families and will awaken interest in comparative studies of structure and function. It would be useful if check-lists could include structural details of the VA mycorrhizas. We suggest that classifications of Arum-types and Paristypes, based on the presence or total absence of intercellular hyphae in the cortex, and intermediate types (e.g. ‘ near-Paris ’), based on limited or rare intercellular hyphae but extensive intracellular coils in the inner cortex and (where present) intercalary arbuscules are useful in this regard. Presence or absence of arbuscules should certainly be recorded, along with presence or absence of vesicles. We recognize that this may increase the work-load in future studies of VA mycorrhizal diversity !                This review was commenced while F. A. S. was on Study Leave in the Department of Soil Science, University of Adelaide, and use of facilities there is gratefully acknowledged. It was inspired in large measure by a memorable visit by F. A. S. and S. E. S. to the laboratory of Larry Peterson, University of Guelph, Canada. We thank Larry and his research group for their hospitality and help, including the assistance that they gave Sandy Dickson in producing Figure 3. Carol Peterson deserves our thanks for very valuable discussions about the occurrence, structure and function of hypodermes, and Paul Widden kindly discussed with us his then unpublished findings. Peter Barlow helped point us towards publications about intercellular spaces, and Stephan Imhof alerted to us his recent work and to other papers by Professor H. C. Weber and associates that we nearly overlooked. We also thank Valentin Furlan for providing us with a copy of Janse (1897), David Read for a very prolonged ‘ loan ’ of his copy of Gallaud (1905), John Conran for helping to sort out the Liliaceae for us, and Bengt So$ derstro$ m for stimulating discussions during his sabbatical visit to S. E. S.’s laboratory. Last, and certainly not least, a referee made some most helpful suggestions that greatly improved the manuscript. Our experimental work is supported by the Australian Research Council.

 Abbott LK. 1982. Comparative anatomy of vesicular–arbuscular mycorrhizas formed on subterranean clover. Australian Journal of Botany 30 : 485–99. Abbott LK, Robson AD. 1978. Growth of subterranean clover in relation to the formation of endomycorrhizas by introduced and indigenous fungi in a field crop. New Phytologist 81 : 575–85. Alexander I. 1988. Mycorrhizas of indigenous tropical forest trees : some research priorities. In : Ng FSP, ed. Proceedings of the Asian Seminar. Kuala Lumpur, Malaysia : Forest Research Institute, 79–89. Alexander T, Toth R, Meier R, Weber HC. 1989. Dynamics of arbuscule development and degeneration in onion, bean and tomato with reference to vesicular–arbuscular mycorrhizae in grasses. Canadian Journal of Botany 67 : 2505–2513. Asai T. 1944. U$ ber die Mykorrhizenbildung der leguminosen Pflanzen. Japanese Journal of Botany 13 : 463–485. Barrett JT. 1958. Synthesis of mycorrhiza with pure cultures of Rhizophagus. Phytopathology 48 : 391. Baylis GTS, McNabb RFR, Morrison TM. 1963. The mycorrhizal nodules of Podocarps.Transactions of the British Mycological Society 46 : 378–384.

Berch SM, Kendrick B. 1982. Vesicular–arbuscular mycorrhizae of Southern Ontario. Mycologia 74 : 769–776. Bonfante-Fasolo P. 1984. Anatomy and morphology of VA mycorrhizae. In : Powell C Ll, Bagyaraj DJ, eds. VA Mycorrhiza. Boca Raton, FL, USA : CRC Press, 5–33. Bonfante-Fasolo P. 1988. The role of the cell wall as a signal in mycorrhizal associations. In : Scannerini S, Smith D, BonfanteFasolo P, Gianinazzi-Pearson V, eds. Cell to Cell Signals in Plant, Animal and Microbial Symbiosis. Berlin & Heidelberg : Springer-Verlag, 219–235. Bonfante-Fasolo P, Fontana, A. 1985. VAM fungi in Gingko biloba roots : their interactions at cellular level. Symbiosis 1 : 53–67. Bonfante-Fasolo P, Scannerini S. 1977. A cytological study of the vesicular–arbuscular mycorrhiza in ‘ Ornithogalum umbellatum ’ L. Allionia 22 : 6–20. Boullard B. 1953 a. Les champignons endophytes des Araliace! es. Bulletin de la SocieteU Botanique de France 100 : 76–77. Boullard B. 1953 b. Les champignons endophytes des Cornace! es. Bulletin de la SocieteU Botanique de France 100 : 150–152. Boullard B. 1958. La mycotrophie chez pteU ridophytes : sa freU quence, ses caracte[ res, sa signification. Bordeaux : E. Drouillard. Brundrett MC, Kendrick B. 1988. The mycorrhizal status, root anatomy and phenology of plants in a sugar maple forest. Canadian Journal of Botany 66 : 1153–1173. Brundrett MC, Kendrick B. 1990 a. The roots and mycorrhizas of herbaceous woodland plants. I. Quantitative aspects of morphology. New Phytologist 114 : 457–468. Brundrett MC, Kendrick B. 1990 b. The roots and mycorrhizas of herbaceous woodland plants. II. Structural aspects of morphology. New Phytologist 114 : 469–479. Brundrett MC, Murase G, Kendrick B. 1990. Comparative anatomy of roots and mycorrhizae of common Ontario trees. Canadian Journal of Botany 68 : 551–578. Brundrett MC, Piche! Y, Peterson RC. 1985. A developmental study of the early stages in vesicular–arbuscular mycorrhiza formation. Canadian Journal of Botany 63 : 184–194. Carling DE, Brown MF. 1982. Anatomy and physiology of vesicular–arbuscular and non-mycorrhizal roots. Phytopathology 72 : 1108–1114. Chapleo S, Hall JL. 1989. Sugar unloading in roots of Ricinus communis L. II. Characteristics of the extravascular apoplast. New Phytologist 111 : 381–90. Cooke MA, Widden P, O’Halloran I. 1992. Morphology, incidence and fertilization effects on the vesicular–arbuscular mycorrhizae of Acer saccharum in a Quebec hardwood forest. Mycologia 84 : 422–430. Cooke MA, Widden P, O’Halloran I. 1993. Development of vesicular–arbuscular mycorrhizae in sugar maple (Acer saccharum) and effects of base-cation amendment on vesicle and arbuscule formation. Canadian Journal of Botany 71 : 1421– 1426. Cooper KM. 1976. A field survey of mycorrhizas in New Zealand ferns. New Zealand Journalof Botany 14 : 169–181. Cox G, Tinker PB. 1976. Translocation and transfer of nutrients in vesicular–arbuscular mycorrhizas. I. The arbuscule and phosphorus transfer : a quantitative study. New Phytologist 77 : 371–378. Cronquist A. 1981. An integrated system of flowering plants. New York : Columbia University Press. Dahlgren RMT, Clifford HT & Yeo PF. 1985. The families of the monocotyledons : structure, evolution and taxonomy. Berlin : Springer-Verlag. Daniels-Hetrick BA, Bloom J, Feyerherm SM. 1985. Root colonization of Glomus epigaeum in nine host species. Mycologia 77 : 825–828. Demuth K, Forstreuter W, Weber HC. 1991. Morphological differences in vesicular–arbuscular mycorrhizae of Gentianaceae produced by different endophytes. Flora 185 : 127–132. Demuth K, Weber HC. 1990. Strukturelle inkompatibita$ t der vesikula$ r–arbuskula$ ren Mycorrhizapilze in Enzian (Gentianaceae). Angewandte Botanik 64 : 247–252. Duckett JG, Ligrone R. 1991. A light and electron microscopy study of the fungal endophytes in the sporophyte and gametophyte of Lycopodium cernuum with observations on the gametophyte–sporophyte junction. Canadian Journal of Botany 70 : 58–72. Duckett JG, Read DJ. 1995. Ericoid mycorrhizas and rhizoid–

Structural diversity in VA mycorrhiza ascomycete associations in liverworts share the same mycobiont : isolation of the partners and resynthesis of the associations in vitro. New Phyologist 129 : 439–447. Endrigkeit A. 1937. Beitra$ ge zum erna$ hrungsphysiologischen Problem der Mykorrhiza unter besonderer Beru$ cksichtigung des Baues und der Funktion der Wurzel- und Pilzmembranen. Botanisches Arkiv, Leipzig 39 : 1–87. Farrar JF. 1985. Fluxes of carbon in roots of barley plants. New Phytologist 99 : 57–69. Frankland JC, Harrison AF. 1985. Mycorrhizal infection of Betula pendula and Acer pseudoplatanus : relationships with seedling growth and soil factors. New Phytologist 101 :133–151. Gallaud I. 1905. E! tudes sur les mycorrhizes endotrophes. Revue GeU neU rale de Botanique 17 : 5–48, 66–83, 123–136, 223–239, 313–325, 425–433, 479–500. Gay PE, Grubb PJ, Hudson, HJ. 1982. Seasonal changes in the concentrations of nitrogen, phosphorus and potassium, and in the density of mycorrhiza, in biennial and matrix-forming perennial species of closed chalk-land turf. Journal of Ecology 70 : 571–593. Gerdemann JW. 1955. Wound-healing of hyphae in a phycomycetous mycorrhiza fungus.Mycologia. 47 : 916–918. Gerdemann JW. 1965. Vesicular–arbuscular mycorrhizas formed on maize and tulip tree by Endogone fasciculata. Mycologia, 57 : 562–575. Gianinazzi-Pearson V, Gianinazzi S, Guillemin, GP, Trouvelot A, Duc G. (1991 a). Genetic and cellular analysis of resistance to vesicular–arbuscular (VA) mycorrhizal fungi in pea mutants. Advances in Molecular Genetics of Plant–Microbe Interactions 1 : 336–42. Gianinazzi-Pearson V, Gollotte A, Dumas-Gaudot E, Giovannetti M, Lherminier J, Berta G, Gianinazzi S. 1994. Studies of cellular and molecular processes involved in the development of arbuscular mycorrhiza using plant mutants. Abstracts, 4th European Symposium on Mycorrhizas, Granada, Spain : 87. Gianinazzi-Pearson V, Smith SE, Gianinazzi S, Smith FA. 1991 b Enzymatic studies on the metabolism of vesicular– arbuscular mycorrhizas. V. Is H+-ATPase a component of the enzyme activities in plant–fungus interfaces ? New Phytologist 117 : 61–74. Giovannetti M, Sbrana C, Avio L, Citernesi AS, Logi C. 1993. Differential hyphal morphogenesis in arbuscular mycorrhizal fungi during pre-infection stages. New Phytologist 125 : 587–593. Greny A. 1973. E! tude anatomo-morphologique des endomycorrhizes constitue par le MaıX s, l’Avoine, le BleU , l’Orge et diverses GramineU es prairiales et adventices. Me! moire presente au Conservatoire des Arts et Me! tiers pour obtenir le Diplome d’Ingenieur C.N.A.M. Harley JL. 1969. The biology of mycorrhiza. 2nd edn. London : Leonard Hill. Harley JL. 1991. The history of research on mycorrhiza and the part played by Professor Beniamino Peyronel. In : Funghi, Piante e Suolo. Torino : Centro di Studio sulla Micologia del Terreno, C.N.R., 31–73. Hayman DS. 1974. Plant growth responses to vesicular– arbuscular mycorrhizas. VI. Effects of light and temperature. New Phytologist 73, 71–80. Hildebrand AA, Koch LW. 1936. A microscopical study of infection of the roots of strawberry and tobacco seedlings by micro-organisms of the soil. Canadian Journal of Research 14 : 11–25. Holley JE, Peterson RL. 1979. Development of a vesicular– arbuscular mycorrhiza in bean roots. Canadian Journal of Botany 57 : 1960–1978. Imhof S, Weber HC. 1997. Root anatomy and mycotrophy (AM) of the achlorophyllous Voyria truncata (Standley) Standley & Steyermark (Gentianaceae). Botanica Acta 110 : 127–134. Jacquelinet-Jeanmougin S, Gianinazzi-Pearson V. 1983. Endomycorrhizas in the Gentianaceae. I. The fungus associated with Gentiana lutea L. New Phytologist 95 : 663–666. Janos DP. 1985. Mycorrhizal fungi : agents or symptoms of tropical community succession ? In : Molina R, ed. Proceedings, 6th NACOM. Oregon State University, 98–106. Janos DP. 1987. VA mycorrhizas in humid tropical ecosystems. In : Safir G, ed. VA Mycorrhizae : an Ecophysiological Approach. Boca Raton, FL, USA : CRC Press, 107–134. Janse JM. 1897. Les endophytes radicaux de quelques plantes

387 Javanaises. Annales du Jardin Botanique de Buitenzorg 14 : 53–201. Johnson NC, Graham JH, Smith FA. 1997. Functioning of mycorrhizal associations along the mutualism–parasitism continuum. New Phytologist 135 : 575–585. Johnston A. 1949. Vesicular–arbuscular mycorrhiza in sea-island cotton and other tropical plants. Tropical Agriculture (Trinidad) 26 : 118–121. Justin SHFW, Armstrong W. 1987. The anatomical characteristics of roots and plant response to flooding. New Phytologist 106 : 465–495. Kariya N, Toth R. 1981. Ultrastructure of the mycorrhizal association formed between Zea diploperennis and Glomus fasciculatus. Mycologia 73 : 1027–1039. Kessler KJ. 1966. Growth and development of sugar maple (Acer saccharum Marsh.). Canadian Journal of Botany 44 : 1413–1425. Kidston R, Lang WH. 1921. On the old red sandstone plants showing structure from the Rhynie chert bed, Aberdeenshire. V. The thallophyta occurring in the peat bed, the succession of the plants through a vertical section of the bed, and the conditions of accumulation and preservation of the peat. Transactions of the Royal Society of Edinburgh 52 : 855–902. Kinden DA, Brown MF. 1975. Electron microscopy of vesicular– arbuscular mycorrhizae of yellow poplar. I. Characterization of endophytic structures by scanning electron stereoscopy. Canadian Journal of Microbiology 21 : 989–993. Kinden DA, Brown MF. 1976. Electron microscopy of vesicular– arbuscular mycorrhizae of yellow poplar. IV. Host-endophyte interactions during arbuscular deterioration. Canadian Journal of Microbiology 22 : 64–75. Kno$ bel M, Weber HC. 1988. Vergleichende Untersuchungen zur Mycotrophie bei Gentiana verna L. und Voyria truncata (Stand.) Stand. & Stey. (Gentianaceae). BeitraX ge zur Biologie der Pflanzen 63 : 463–477. Konoe$ R. 1957. U$ ber das vorkommen der Wurzelpilze bei Metasequoia und den na$ chst verwantden in Pflanzen. Journal of the Institute of Polytechnics, Osaka City University 8 : 179–184. Ku$ hn KD, Weber HC. 1986. Zur vesikula$ ren–arbuskula$ ren Mykorrhiza von Gentiana asclepiadea L. (Gentianaceae) an natu$ rlichen standorten. Angewandte Botanik 60 : 427–439. Laycock DH. 1945. Preliminary investigations into the function of the endotrophic mycorrhiza of Theobroma cacao L. Tropical Agriculture (Trinidad) 22 : 77–80. Leake JR. 1994. The biology of myco-heterotrophic (’saprophytic’) plants. New Phytologist 127 : 171–216. Ligrone R. 1988. Ultrastructure of a fungal endophyte in Phaeoceros laevis (L.) Prosh (Anthocerophyta). Botanical Gazette (Chicago) 149 : 92–100. Louis I. 1990. A mycorrhizal survey of plant species colonizing coastal reclaimed land in Singapore. Mycologia 82 : 772–778. Mabberley DJ. 1989. The plant-book. Cambridge : Cambridge University Press. McGee PA. 1985. Lack of spread of endomycorrhizas of Centaurium (Gentianaceae). New Phytologist 101 : 451–458. McGee PA. 1986. Mycorrhizal associations of plant species in a semiarid community. Australian Journal of Botany 34 : 585–593. McGee PA. 1988. Growth response to and morphology of mycorrhizas of Thysanotus (Anthericaceae : Monocotyledonae). New Phytologist 109 : 459–463. McGee PA. 1990. Survival and growth of seedlimgs of coachwood (Ceratopetalum apetalum) : effects of shade, mycorrhizas or a companion plant. Australian Journal of Botany 38 : 583–592. McLennan EI. 1958. Thismia rodwayi F Muell. and its endophyte. Australian Journal of Botany 6 : 25–37. McLuckie J, Burges A. 1932. Mycotropism in the Rutaceae. I. The mycorrhiza of Eriostemum crowei F. v. M. Proceedings of the Linnean Society of New South Wales 57 : 291–312. Meloh K-A. 1963. Untersuchungen zur Biologie der endotrophen Mykorrhiza bei Zea mays L. und Avena sativa L. Archiv fuX r Mikrobiologie 46 : 369–381. Morton JB. 1985. Variation in mycorrhizal and spore morphology of Glomus occultum and Glomus diaphanum as influenced by plant host and soil environment. Mycologia 77 : 192–204. Morton JB, Benny GL. 1990. Revised classification of arbuscular mycorrhizal fungi (Zygomycetes) : a new order, Glomales, two new suborders, Glomineae and Gigasporineae, and two new families, Acaulosporaceae and Gigasporaceae, with an emendation of Glomaceae. Mycotaxon, 37 : 477–491.

388

F. A. Smith and S. E. Smith

Mosse B. 1973. Plant growth responses to vesicular–arbuscular mycorrhiza. IV. In soil given additional phosphate. New Phytologist, 72 :127–136. Mosse B, Hepper CM 1975. Vesicular–arbuscular infections in root organ cultures. Physiological Plant Pathology 5 : 215–223. Mullen RB, Schmidt SK. 1993. Mycorrhizal infection, phosphorus uptake, and phenology in Ranunculus adoneus : implications for the functioning of mycorrhizae in alpine systems. Oecologia 94 : 229–234. Nadarajah P. 1980. Species of Endogonaceae and mycorrhizal association of Elaeis guineensis and Theobroma cacao.In : Mikola P, ed. Tropical Mycorrhiza. Oxford : Clarendon Press, 233–237. Nicolson TH. 1959. Mycorrhiza in the Gramineae. I. Vesicular– arbuscular endophytes, with special reference to the external phase. Transactions of the British Mycological Society 42 :421–438. Perumalla CJ, Peterson CA, Enstone DE. 1990. A survey of angiosperm species to detect hypodermal Casparian bands. I. Roots with a uniseriate hypodermis and epidermis. Botanical Journal of the Linnean Society of London 193 : 93–112. Peterson CA. 1988. Exodermal Casparian bands : their significance for ion uptake by roots. Physiologia Plantarum 72 : 204–208. Peterson RL, Howarth MJ, Whittier DP. 1981. Interactions between a fungal endophyte and gametophyte cells in Psilotum nudum. Canadian Journal of Botany 59 : 711–720. Peyronel B. 1923. Prime ricerche sulle micorize endotrofiche e sulla micoflora radicola normale delle fanerogame. Rivista di Biologia 5 : 463–85. Peyronel B. 1924. Prime ricerche sulle micorize endotrofiche e sulla micoflora radicola normale delle fanerogame. Rivista di Biologia 6 : 17–53. Pocock K, Duckett JG. 1984. A comparative ultrastructural analysis of the fungal endophytes in Cryptothallus mirabilis Malm. and other British thalloid hepatics. Journal of Bryology 13 : 227–233. Prat H. 1926. E! tude des Mycorrhizes du Taxus baccata. Annales des Sciences Naturelles, Botanique et Biologie VeU geU tale, Paris 8 : 141–162. Rayner MC. 1927. Mycorrhiza. An account of non-pathogenic infection by fungi in vascular plants and bryophytes. New Phytologist Reprint No 15. London : Wheldon and Wesley. Redhead JF. 1968. Mycorrhizal associations in some Nigerian forest trees. Transactions of the British Mycological Society 51 : 377–387. Reinsvold RJ, Brent Reeves F. 1986. The mycorrhizae of Juniperus osteosperma : identity of the vesicular–arbuscular symbiont, and resynthesis of VA mycorrhizae. Mycologia 78 : 108–113. Remy W, Taylor TN, Hass H, Kerp H. 1994. Four-hundredmillion-year-old vesicular arbuscular mycorrhizae. Proceedings of the National Academy of Sciences, USA : 91 : 11841–11843. Saif SR. 1977. The influence of stage of host development on vesicular–arbuscular mycorrhizae and endogenous spore population in field-grown vegetable crops. I. Summer-grown crops. New Phytologist 79 : 341–348. Shibata K. 1902. Cytologische Studien u$ ber die endotrophen Mykorrhizen. Jahrbucher fuX r die wissenschaftliche Botanik 37 : 643–684. Simard SW, Perry DA, Jones MD, Myrold DD, Durall DM, Molina R. 1997. Net transfer of carbon between ectomycorrhizal species in the field. Nature 388 : 579–582. Smith FA, Smith SE. 1996. Mutualism and parasitism : diversity in function and structure in the ‘ arbuscular ’ (VA) mycorrhizal symbiosis. Advances in Botanical Research 22 : 1–43. Smith HF, O’Connor PJ, Smith SE, Smith FA. 1997. (Vesicular)–arbuscular mycorrhizas of durian and other plants of forest gardens in W. Kalimantan, Indonesia. In : Schulte A, Ruhiyat D, eds. Forest Soils in the Humid Tropics : Characteristics, Ecology and Management. Berlin, Heidelberg, New York : Springer. (In press.) Smith SE. 1995. Discoveries, discussions and directions in mycorrhizal research. In : Varma A, Hock, B eds. Mycorrhiza : Structure, Function, Molecular Biology and Biotechnology. Berlin : Springer-Verlag, Berlin, 3–24.

Smith SE, Dickson S. 1991. Quantification of active vesicular– arbuscular mycorrhizal infection using image analysis and other techniques. Australian Journal of Plant Physiology 18 : 637–648. Smith SE, Dickson S, Morris C, Smith FA. 1994. Transfer of phosphate from fungus to plant in VA mycorrhizas : calculation of the area of symbiotic interface and of fluxes of P from two different fungi to Allium porrum L. New Phytologist 127 : 93–99. Smith SE, Long CM, Smith FA. 1989. Infection of roots with a dimorphic hypodermis : possible effects on solute uptake. Agriculture, Ecosystems and Environment 29 : 403–407. Smith SE, Read DJ. 1997. Mycorrhizal smbiosis, 2nd edn. London : Academic Press. Soberon MJ, Martinez del Rio C. 1985. Cheating and taking advantage in mutualistic symbioses. In Boucher DH, ed. The Biology of Mutualism. London : Croom Helm, 192–216. So$ derstro$ m B, Dickson S, Smith FA, Smith SE. 1996. Arumand Paris-type of endomycorrhiza in some plant–AM fungus species combinations. In : Szaro TM, Bruns TD, eds. Program and Abstracts, First International Congress on Mycorrhizae. Berkeley, California : University of California, 111. Stahl M. 1949. Die Mykorrhizen der Lebermoose mit besonderer Beru$ cksichtigung der Thallosen formen. Planta 37 : 103–148. Stelz T. 1968. Mycorrhizes et vegetation des pelouses calcaires. The' se pour l’obtention du Doctorat, Faculte! des Sciences de l’Universite! de Rouen, France. Strullu DG, Gourret JP, Garrec JP, Fourcy A. 1981. Ultrastructure and electron- probe microanalysis of the metachromatic vacuolar granules occurring in Taxus mycorrhizas. New Phytologist 87 : 537–545. Thapar HS, Khan, SN. 1973. Studies on endomycorrhiza in some forest species. Proceedings, Indian National Science Academy B39 : 687–694. Tiemann C, Demuth K, Weber HC. 1994 a. Zur VAMycorrhiza von Gelsemium rankinii und G. sempervirens (Loganiaceae). BeitraX ge zur Biologie der Pflanzen 68 : 311–321. Tiemann C, Demuth K, Weber HC. 1994 b. Zur symbiose von Cynanchum vincetoxicum (L.) PERS., Asclepias curassavica L. und Ceropegia woodii SCHL. (Asclepiadaceae) mit mycorrhizapilsen (VAM). Flora 189 : 1–6. Toth R, Miller RM. 1984. Dynamics of arbuscule development and degeneration in a Zea mays mycorrhiza. American Journal of Botany 71 : 449–460. Trappe JM. 1987. Phylogenetic and ecologic aspects of mycotrophy in the angiosperms from an evolutionary viewpoint. In : Safir GR ed. Ecophysiology of VA Mycorrhizal Plants. Boca Raton, FL, USA : CRC Press, 5–25. Untch, Weber HC. 1995. Strukturen der Mycorrhiza (AM) bei Ceropegia dichotoma Haw., Ceropegia fusca Bolle und Periploca laevigata Ait. (Asclepiadaceae). BeitraX ge zur Biologie der Pflanzen 69 : 129–140. Walker C. 1995. AM or VAM : What’s in a word ? In : Varma A, Hock B, eds. Mycorrhiza : Structure, Function, Molecular Biology and Biotechnology. Berlin : Springer-Verlag, 25–26. Wastie RL. 1965. The occurrence of an Endogone type of endotrophic mycorrhiza in Hevea brasiliensis. Transactions of the British Mycological Society 48 : 167–178. Weber HC, Klahr A, Marron-Heimbuch M. 1995. Anatomical structures of the VA mycorrhiza in the Apocynaceae (Gentianales). Botanica Acta 108 : 525–534. Weber HC, Kra$ mer M. 1994. VA–Mycorrhiza bei Menyanthaceae. BeitraX ge zur Biologie der Pflanzen 68 : 351–362. Whitbread F, McGonigle TP, Peterson, RL. 1996. Vesicular– arbuscular mycorrhizal associations of American ginseng (Panax quinquefolius) in commercial production. Canadian Journal of Botany 74 : 1104–1112. Widden P. 1996. The morphology of vesicular–arbuscular mycorrhizae in Clintonia borealis and Medeola virginiana. Canadian Journal of Botany 74 : 679–685. Winter AG. 1951. Untersuchungen u$ ber die Verbreitung und Bedeutung der Mykorrhizen bei kultivierten Gramineen und einigen anderen landwirtschaftlichen Nutzpflanzen. Phytopathologische Zeitschrift 17 : 421–432. Yawney WJ, Schultz RC. 1990. Anatomy of a vesicular– arbuscular endomycorrhizal symbiosis between sugar maple (Acer saccharum Marsh) and Glomus etunicatum Becker and Gerdemann. New Phytologist 114 : 47–57. Zelmer CD, Currah RS. 1995. Evidence for a fungal liaison between Corallorhiza trifida (Orchidaceae) and Pinus contorta (Pinaceae). Canadian Journal of Botany 73 : 862–866.