Summary 1The arbuscular mycorrhizal (AM) fungi colonizing plants at a woodland site in North Yorkshire (UK) have been characterized from the roots of five plant species ( Rubus fruticosus agg. L., Epilobium angustifolium L., Acer pseudoplatanus L., Ajuga reptans L. and Glechoma hederacea L.), and identified using small-subunit rRNA (SSUrRNA) gene amplification and sequencing. 2 Interactions between five plant species from the site and four co-occurring glomalean fungi were investigated in artificial one-to-one AM symbioses. Three of the fungi were isolated from the site; the fourth was a culture genetically similar to a taxon found at the site. Phosphorus uptake and growth responses were compared with non-mycorrhizal controls. 3 Individual fungi colonized each plant with different spatial distribution and intensity. Some did not colonize at all, indicating incompatibility under the conditions used in the experiments. 4 Glomus hoi consistently occupied a large proportion of root systems and outperformed the other fungi, improving P uptake and enhancing the growth of four out of the five plant species. Only G. hoi colonized and increased P uptake in Acer pseudoplatanus , the host plant with which it associates almost exclusively under field conditions. Colonization of all plant species by Scutellospora dipurpurescens was sparse, and beneficial to only one of the host plants ( Teucrium scorodonia ). Archaeospora trappei and Glomus sp. UY1225 had variable effects on the host plants, conferring a range of P uptake and growth benefits on Lysimachia nummularia and T. scorodonia , increasing P uptake whilst not affecting biomass in Ajuga reptans and Glechoma hederacea , and failing to form mycorrhizas with A. pseudoplatanus . 5 These experimental mycorrhizas show that root colonization, symbiont compatibility and plant performance vary with each fungus-plant combination, even when the plants and fungi naturally co-exist. 6 We provide evidence of physical and functional selectivity in AM. The small number of described AM fungal species (154) has been ascribed to their supposed lack of host specificity, but if the selectivity we have observed is the general rule, then we may predict that many more, probably hard-to-culture glomalean species await discovery, or that members of species as currently perceived may be physiologically or functionally distinct.

S U M M A R YThe arbuscular mycorrhizal (AM) association between fungi in the order Glomales and the roots of a very wide range of vascular plants is of global ecological significance hut has proved particularly intractable to study in the field. We have developed a reliable technique to identify the fungal symbionts in roots taken directly from natural communities. Selective Enrichment of Amplified DNA combines the use of recently-developed specific DNA primers with a novel method based on the principle of subtractive hybridization to remove interfering plantderived DNA after amplification with the polymerase chain reaction. Using this technique we have shown that cndomycorrhizas of bluebells {Hyaciiithoides non-scripta) sampled directly from a woodland habitat are multispecies communities of varying composition which contain at least three genera of mycorrhizal fungi. The technique works well on a range of plant species and should have wide application to the identification of other symbionts, including pathogens. A spore survey has indicated that two particular AM types are associated with bluebells and this observation corroborates the molecular data. The presence of a Glomus species in bluebell roots was not expected from the spore data.

The arbuscular mycorrhizas of bluebell (Hyacinthoides non-scripta (L.) Chouard ex Rothm.) involve several symbiotic fungi of the order Glomales. We have previously simplified the system by ignoring the taxonomic diversity of the fungi, but it is unlikely that all fungal species contribute in the same manner or to the same extent to the functioning of the symbiosis. To discover how many and which fungi take part in the bluebell mycorrhiza we sought to identify the range of arbuscular mycorrhizal (AM) fungi found in bluebell roots sampled during a complete growing season, September to June.Although the taxonomy of the Glomales by their spores is not yet fully understood, identification is, to a large extent, possible. Arbuscular mycorrhizal communities are usually characterized by their spores but, since spores can rarely be directly associated with individual plants or plant species, a more satisfactory approach would be to identify fungal symbionts where they interact with the host plant, in the roots. Unfortunately, the intraradical mycelia of the fungi are less easily distinguished than the spores and, as yet, classification is possible only to the family level.We have developed a method whereby different AM fungal taxa in the roots of bluebell can be distinguished by objective assessment. A large suite of morphological characteristics of the fungi in roots was recorded in samples taken at monthly intervals. Hierarchical cluster analysis of the resulting data separated six distinct AM fungal morphotypes (Scutellospora type, Acaulospora type, three Glomus types and fine endophytes) and a classification system created by which identification by eye was possible.We compared the fungi in roots with glomalean spores in soil from the root zone of the same bluebell plants. Two species occurred in most samples, Scutellospora dipurpurescens Morton & Koske (emend. Walker, 1993) and Acaulospora koskei Blaskowski. A further six occurred sporadically, five Acaulospora spp. and Glomus rubiforme Gerdemann & Trappe) Almeida & Schenck ( l Sclerocystis rubiformis). The presence of a single species of Scutellospora was consistent with a Scutellospora root morphotype which varied little. By contrast, the diversity of Acaulospora in the spore assemblage was reflected by variation within the Acaulospora morphotype. Glomus spores were very rarely found in field collections, yet Glomus morphotypes were found to be an important component of the bluebell mycorrhiza.Because some important species are not represented in spore assemblages in the field and those that are found can only be associated with vegetational groups, not individual plants or single species, glomalean spore populations provide only a partial account of the fungi which contribute to arbuscular mycorrhizas. Although it is still not possible to identify AM fungi in roots with the same precision as their spores, the method reported here permits assessment of diversity in the roots of individual plant species, which may be applied to the investigation of mycorrhi...

Summary 1The composition of communities of arbuscular mycorrhizal (AM) fungi can have a large effect on the performance of their plant hosts. The role of individual fungal species in shaping this response is as yet unresolved. 2 We have used the fungicide benomyl to alter the community of AM fungi in undisturbed monoliths of soil in a natural community. Changes in the community were characterized by root colonization (%RLC), cloning, sequencing and tRFLP of a partial SSUrDNA fragment. Eleven plant species were sufficiently abundant in the monoliths to be examined. 3 In the highly mycorrhiza-dependent perennial herb Ajuga reptans , phosphate concentration was significantly reduced after benomyl treatment over a full growing season. The other plant species showed low colonization and no significant difference in phosphate concentration after benomyl treatment. 4 Although colonization in A. reptans was reduced, many mycorrhizal fungi survived in the roots. Some became more abundant following fungicide treatment, suggesting competitive release. Fungi that increased were generalists that have been identified in field samples from published studies colonizing a wide range of plant species. Those that declined were specialists with a narrow host range; five types had not been recorded previously in field samples. 5 AM fungi in this study differed greatly in their response to perturbation, independent of the identity of the host plant. If such functional diversity is widespread, then elucidating the part played by AM fungal diversity in regulating plant community structure will be key to our understanding and management of ecosystems.

Roots of bluebell (Hyacinthoides non-scripta (L.) Chouard ex Rothm.) are colonized by a range of fungal symbionts from several genera of the order Glomales. Using the identification scheme described in Merryweather & Fitter (1998), arbuscular mycorrhizal (AM) fungi in bluebell roots were quantified throughout the single growing season of 1994-5 and compared with populations of spores found in the soil around the roots.In the early part of the growing season, when its activity is entirely subterranean (autumn and winter), bluebell habitually associates with a Scutellospora morphotype which is almost certainly S. dipurpurescens Morton & Koske (emend. Walker, 1993) whose spores occur in the root zone. This is the time of maximum phosphorus inflow and bulb-stored carbohydrate utilization by this mycorrhiza. A diverse flora of other AM fungal morphotypes (Acaulospora and Glomus), which might also form mycorrhizas with other plants in the vicinity of bluebells, invade the roots later in the season (spring), when P inflow is reduced and carbohydrate is available as fresh photosynthate. Their contribution to the mycorrhiza might be less than that of Scutellospora, particularly in terms of P assimilation.Both AM fungi in roots and glomalean spores recovered from soil around bluebell roots showed a significant degree of correlation with the vegetation within which the test plants grew. In the case of AM fungi in roots, Scutellospora showed no special preference for either, but Glomus correlated with a canopy of sycamore (Acer pseudoplatanus L.) and Acaulospora with oak (Quercus petraea (Mattuschka) Liebl.). Spores which most closely resembled S. dipurpurescens and Acaulospora gerdemannii Schenck & Nicolson were significantly more numerous under sycamore, but a spore like Acaulospora koskei Blaskowski, the most numerous and frequently encountered glomalean spore in the system, showed no preference for areas dominated by either tree. There was no significant relationship between AM fungal populations in bluebell roots and glomalean spores recovered from associated soil.The two spore taxa most frequently found in the vicinity of bluebell roots (A. koskei and S. dipurpurescens) were also found in lower numbers in soil from a region of the field site in which bluebell was absent, indicating that the main bluebell AM fungi do not exclusively associate with that host.