Roots, foraging and the exploitation of soil nutrient patches: the role of mycorrhizal symbiosis

There is little doubt that plant roots respond to patches of elevated fertility (Drew 1975; Crick & Grime 1987; Fitter 1994) and that the foraging of clonal plants (coupled with phenotypic morphological plasticity) may allow an integrated exploitation of such spatial heterogeneity (Slade & Hutchings 1987a,b; Birch & Hutchings 1994; Caldwell 1994; Hutchings & de Kroon 1994; Piglliucci, Diiorio & Schlichting 1997).

Available evidence indicates that, while plant roots can colonize fertile sites within soil, sufficient time (weeks rather than days) and relatively large patches of nutrient (c. 5 cm3) are often required for the roots to respond (Eissenstat & Caldwell 1988; Hodge et al. 2000; Jackson & Caldwell 1989, 1991; van Vuuren, Robinson & Griffiths 1996). Exceptionally, Jackson & Caldwell (1989) observed root proliferation only 1 day after the addition of aqueous fertilizer and went on to demonstrate that uptake kinetics of roots may also respond within this time-scale (Jackson, Manwaring & Caldwell 1990).

In natural ecosystems, nutrient patches are likely to be associated with decomposing organic residues, therefore experiments using organically derived nutrient patches are more realistic than those involving additions of mineral salts. On a large scale (> 5 cm3) residues might include leaf litter, large fruits, and the faeces and bodies of vertebrates. Smaller-scale patches may derive from dead seeds, pollen, smaller fruit, dead bacteria and fungi, and the faeces and bodies of invertebrates. Variations in microtopography, microclimate and parent material may also influence soil heterogeneity (see Stark 1994). van Vuuren et al. (1996) found that the roots of wheat plants proliferated within large patches enriched with 15N labelled organic matter (plant litter) and observed that the maximum rate of uptake of N occurred prior to the ‘spectacular proliferation of roots’. This indicates that root proliferation (greatest at 22–34 days) might have been too slow to take full advantage of the period of maximal mineralization and potential nutrient acquisition (10–22 days). Hodge et al. (1998) reached a similar conclusion with 15N/13C labelled plant litter patches and found roots were poor competitors for N in the presence of soil microbial populations. This finding may be fundamental in understanding the competitive partitioning of nutrients derived from readily decomposable patches. Based on such evidence, we may come to perceive the role of roots, in terms of acquiring nutrients from organic patches, as relatively inert structures from which symbiotic organs, that can better compete with the wider microbiota (i.e. mycorrhizal hyphae) are dispatched.

Studies on the effect of soil nutrient patches on the growth of clonal plants [that are well adapted to exploit nutrient heterogeneity (Hutchings & de Kroon 1994)] assessed the effects of plant foraging for, and scale changes in, soil nutrient patches (e.g. Humphrey & Pyke 1997; Wijesinghe & Hutchings 1997). Humphrey & Pyke (1997) showed a morphological response to nutrient patches and found that root invasion into high nutrient zones was the primary mechanism of nutrient capture in caespitose Elymus lanceolatus ssp. wawawaiensis and could be as important as above-ground colonization in rhizomatous E. lanceolatus spp. lanceolatus. Using Glechoma hederacea, Wijesinghe & Hutchings (1997) demonstrated the inability of this herb to respond to small nutrient patches, whereas large nutrient patches (comparable to those used by Humphrey & Pyke 1997) were well exploited. In neither case was the importance of mycorrhizal fungi considered.

Both Elymus spp. and G. hederacea are known to be facultative mycotrophs with arbuscular mycorrhizal fungi (AMF) (Harley & Harley 1987). Generally, the association is thought to be mutualistic with a compatible exchange of carbon and nutrients between the two partners (Harley 1989; Allen 1991; Johnson, Graham & Smith 1997). AMF tend to form mycorrhizas with facultative mycotrophs under nutrient-poor conditions (Allen 1991). Both Humphrey & Pyke (1997) and Wijesinghe & Hutchings (1997) used substrates of extremely low nutrient status for their low fertility patches, such that, under natural conditions, mycorrhizas might be expected to form.

Where roots (and foraging shoots) do not have the morphological plasticity required to respond to small-scale nutrient patches, mycorrhizal fungi may be important. The structure and function of vegetative mycelium allows comprehensive exploration and rapid colonization of patches of nutrient in soils (Boddy 1993; Bolton & Boddy 1993; Prosser 1993; Hughes & Boddy 1996; Wells, Donnelly & Boddy 1997; Boddy & Abdalla 1998). These attributes, combined with the physiological plasticity of mycelium (e.g. Cairney & Burke 1996) make fungi ideal for soil exploration (foraging) and capture of discrete resources (nutrient patches). The association of plant roots with mycorrhizal fungi may provide the autotroph with a more favourably distributed and extensive system of absorbing organs (Smith & Read 1997), and with a greater chance of encountering fertile microsites not available to the root system.

Ericoid and ectomycorrhizal fungi may be particularly adapted to exploit patches of high quality organic matter by production of appropriate exo-enzymes. For example, when small (0·4 g) pockets of fermenton horizon organic matter (FHOM) were placed within an otherwise uniform peat substrate they were rapidly colonized by ectomycorrhizal mycelium (in symbiosis with Pinus sylvestris) (Bending & Read 1995a,b). Within the colonized patches of FHOM an increase in levels of nutrient mobilizing enzymes and a loss of macro nutrients could be detected. Similar findings have also been obtained from other studies using organically derived enrichment and ectomycorrhizal (Read 1991; Tibbett & Sanders 1998) and ericoid mycorrhizal plants (Kerley & Read 1998). When inorganic salts exclusively were added (rather than FHOM) there was no mycelial colonization of the fertilized patches (Read 1991). It is conceivable a system operates where mycorrhizal mycelia proliferate into fresh organic patches and that roots respond subsequently to mineral (or mineralized) patches.

AMF are likely to exhibit similar morphological plasticity to ericoid and ectomycorrhizal counterparts but generally secrete lower levels of enzymes with nutrient-mobilizing potential. Nonetheless, there is some evidence to suggest that the hyphae of AMF proliferate in, and intimately associate with, patches of organic matter (Facelli 1998; St John, Coleman & Reid 1983a,b). This might be important in the acquisition of patchy nutrients, especially for poorly mobile PO4 ions (see Cui & Caldwell 1996a,b; Farley & Fitter 1999). If this were so (and this remains unclear given the current paucity of corroborating data), it would be important in our understanding of both plant growth response to, and the ecological function of, mycorrhizal symbiosis in typically heterogeneous soils. For example, in the experimental work of Wijesinghe & Hutchings (1997) (discussed above), acquisition of nutrients by AMF from the smaller-scale nutrient patches (and transfer to the host) might have significantly influenced their results, had their plants been mycorrhizal. The negative effect of small patch size on biomass production may have been annulled as the external mycelium would have served to integrate resource heterogeneity (see Read, Francis & Finlay 1985). One might even speculate that the smallest biomass might arise from the largest (50:50) patch, as a result of competition for light. The results of Humphrey & Pyke (1997) may also have been influenced by the presence of mycorrhizas. The exploitation of nutrient patches by AMF could possibly increase the importance of below-ground nutrient capture, compared to aerial foraging. Also significant in this respect is the recent evidence that glomalean colonization can cause changes in the growth traits of clonal plants (Prunella spp.) (Streitwolf-Engel et al. 1997) and in the root morphology of grasses (Miller, Hetrick & Wilson 1997). AMF infection has also been shown to alter the competitive balance between species (e.g. West 1996; van der Heijden et al. 1998), although interpretations can vary (Watkinson & Freckleton 1997).

The significance ascribed to root system response and foraging behaviour in nutrient patch exploitation may be easily over-estimated in accounting for plant access to transient nutrient patches in nature. The importance of mycobionts are ignored in these systems at peril. Mycorrhizal mycelia may be more important than root proliferation (and shoot foraging of clonal plants) in nutrient patch exploitation. One must consider their morphological and physiological plasticity, rapid response to substrates, and exo-enzyme capabilities of some fungi, in combination with the inherent competitive disadvantage of roots compared with the soil microbiota in exploiting organic nutrient patches (see Hodge et al. 1998). A likely role of root proliferation may be to acquire free ions resulting from decomposition of organic residues, but by this time the mineralized nutrients would have already been cycled through the microbial biomass. Therefore, the intimate, rapid and, in some cases, direct access to nutrient patches supplied by mycorrhizal mycelia is likely to be the primary source of patch-derived nutrients in mycotrophic plants. Current and historical emphasis on plant roots may have over estimated their significance in this regard. There is a need, therefore, for more data to establish clearly the relative importance in nutrient-patch exploitation of clonal shoots, roots and their mycorrhizal partners in order to determine the contribution of each to plant nutrition and ecology.


My thanks to F. E. Sanders and A. H. Fitter for their comments on the manuscript.