Soil and fertilizer phosphorus: Effects on plant P supply and mycorrhizal development

Transcription

1 Soil and fertilizer phosphorus: Effects on plant P supply and mycorrhizal development Cynthia Grant 1, Shabtai Bittman 2, Marcia Montreal 1, Christian Plenchette 3, and Christian Morel 4 Can. J. Plant Sci. Downloaded from by on 05/09/18 1 Agriculture and Agri-Food Canada, Brandon Research Centre, Box 10000A, R.R. #3, Brandon, Manitoba, Canada R7A 5Y3 ( cgrant@agr.gc.ca); 2 Agriculture and Agri-Food Canada, 6947 No. 7 Highway, P. O. Box 1000, Agassiz, British Columbia, Canada V0M 1A0; 3 INRA, 17 rue Sully, Dijon, Cedex France; 4 UMR TCEM Transfert Sol-Plante et Cycle des éléments minéraux dans les Ecosystèmes Cultivés, INRA-ENITA, BP Villenave d Ornon, Cedex France. Received 7 October 2003, accepted 18 May Grant, C., Bittman, S., Montreal, M., Plenchette, C. and Morel, C Soil and fertilizer phosphorus: Effects on plant P supply and mycorrhizal development. Can. J. Plant Sci. 85: Plants require adequate P from the very early stages of growth for optimum crop production. Phosphorus supply to the crop is affected by soil P, P fertilizer management and by soil and environmental conditions influencing P phytoavailability and root growth. Phosphorus uptake in many crops is improved by associations with arbuscular mycorrhizal fungi. Cropping system and long-term input of P through fertilizers and manures can influence the amount and phytoavailability of P in the system and the development of mycorrhizal associations. Optimum yield potential requires an adequate P supply to the crop from the soil or from P additions. Where early-season P supply is low, P fertilization may improve P nutrition and crop yield potential. Alternately, under low-p conditions, encouragement of arbuscular mycorrhizal associations may enhance P uptake by crops early in the growing season, improving crop yield potential and replacing starter fertilizer P applications. Soil P supply that exceeds P requirements of the crop may preclude mycorrhizal development. To encourage arbuscular mycorrhizal association, threshold levels of soil solution P that restrict mycorrhizal development must not be exceeded. Sustainable P management practices must be applied both in conventional and in alternative biologically based agricultural systems. Key words: Microbiology, fertility, colonization Grant, C., Bittman, S., Montreal, M., Plenchette, C. et Morel, C Le phosphore dans le sol et les engrais : incidence sur l absorption du P par les plantes et sur le développement des mycorhizes. Can. J. Plant Sci. 85: Les plantes ont besoin d une quantité suffisante de P dès le début de leur croissance si l on veut que leur culture donne un rendement optimal. Les apports de P dépendent de la concentration de cet élément dans le sol, de l application d engrais phosphatés et des conditions environnementales qui affectent la quantité de P disponible pour la plante et la croissance des racines. Chez maintes cultures, l association avec des mycorhizes à arbuscules améliore l absorption du phosphore. Les pratiques agricoles et l apport prolongé de P résultant de l application d engrais et de fumier peuvent modifier la quantité de P dont les plantes disposent dans le milieu et le développement d une symbiose avec les mycorhizes. Pour atteindre son meilleur rendement, la culture a besoin de tirer une quantité suffisante de P du sol ou des amendements. Quand la concentration de P ne suffit pas au début de la période végétative, la fertilisation facilite parfois l assimilation de cet élément et accroît le rendement potentiel de la culture. Dans les mêmes conditions, favoriser l association avec les mycorhizes à arbuscules peut aider la plante à mieux absorber le P au début de la saison de croissance, ce qui en accroîtra le rendement éventuel et remplacera l épandage initial d engrais phosphatés. Lorsqu il contient plus de P que la plante n en a besoin, en revanche, le sol peut empêcher le développement des mycorhizes. Pour favoriser l association avec les mycorhizes à arbuscules, il convient de ne pas dépasser la concentration de P soluble qui entravera le développement de ces cryptogames dans le sol. On doit recourir à des pratiques durables de gestion du P dans les systèmes agricoles traditionnels comme dans ceux qui font appel aux méthodes de culture biologiques. Mots clés: Microbiologie, fertilité, colonisation Plants require adequate P from the very early stages of growth for optimum crop production (Grant et al. 2001). Restricted early-season P supply frequently limits crop production, and P fertilizer is commonly applied to ensure that sufficient P is available to optimize crop yield and maturity. Total soil P usually ranges from 100 to 2000 mg P kg 1 soil representing approximately 350 to 7000 kg P ha 1 in the surface 25 cm of the soil, although only a small portion of this P is immediately available for crop uptake (Morel 2002). Crop removal may range from 3 to 30 kg P ha 1, therefore 3 crop production will gradually deplete available soil P in the absence of P application. Conversely, excess P supply in the soil is a major environmental concern. Accumulation of P in the soil from applications of animal manures, biosolids or chemical fertilizer in excess of that taken up by the crop can increase the risk of P movement to surface and groundwaters. Excess P in water degrades the quality of aquatic Abbreviations: AM, arbuscular mycorrhizal; FYM, farmyard manure

2 4 CANADIAN JOURNAL OF PLANT SCIENCE ecosystems by encouraging eutrophication (Schindler 1977). Therefore it is important that P management balances the goal of providing sufficient P to the crop to optimize crop yield with the goal of avoiding excess P and environmental risk. Where plant-available P in the soil is low, efficient applications of fertilizer P or manure and/or improved mycorrhizal association may improve crop P levels. The reserves of P in the world are finite and are gradually being depleted (Tiessen 1995). Thus there is a need to develop agricultural systems based on meeting minimum P requirements for crops. Management of the cropping system to improve the availability of P to the crop early in the growing season may improve P nutrition while reducing the potential for excess accumulation of P in soils and risk of movement of P into water systems. This will require a detailed understanding of the processes governing soil P cycling and availability in which mycorrhizal symbiosis may play a significant role. This paper discusses P dynamics in agricultural systems and outlines the potential for improving P nutrition of crops by enhancing mycorrhizal associations and improving P fertilizer use efficiency for sustainable crop production. Crops Require P during Early Growth The importance of adequate tissue P concentrations during early-season growth has been reported in many different crop species (Grant et al. 2001). Studies in Ontario have shown that corn grain yield was strongly affected by P supply and tissue P concentration in the L4 to L5 stage, rather than by P concentration later in growth (Barry and Miller 1989; Lauzon and Miller 1997). Gavito and Miller (1998a) reported that enhanced early-season P nutrition in corn increased the dry matter partitioning to the grain at later development stages. Similarly, in wheat (Gericke 1924, 1925; Boatwright and Viets 1966) and barley (Brenchley 1929), P supply prior to 6 wk of growth had a much greater effect on final grain yield than P supply in later growth. Intermediate wheatgrass (Boatwright and Viets 1966), broadleaved willow (Atkinson and Davidson 1971), radish and lettuce (Avnimelech and Scherzer 1971) and a variety of other crops (Crafts-Brandner 1992; Elliott et al. 1997) also showed persistent reductions in growth after early-season P deficiencies. In contrast, studies by Plénet et al. (2000) reported the maximum difference in biomass production of corn under P deficiency in field conditions at 400 to 600 growing degree days ( C) after sowing. The aboveground biomass accumulation was severely reduced ( 60%) during early stages of corn growth although only slight differences were observed on biomass accumulation at harvest and grain yield. The spectacular effect of P deprivation on early reduction in shoot growth is explained by a slight although rapid stimulation of root growth, which has often been reported (Mollier and Pellerin 1999). The ultimate effect of initial reductions in growth related to P deficiency on the final crop yield will be influenced by other growth-limiting factors experienced by the crop through the remainder of the growing season. Phosphorus Supply to Crops The early supply of P to the crop is influenced by soil P and P application as well as by soil and environmental conditions that affect P phytoavailability and root growth. Roots absorb P ions from the soil solution. The ability of the plant to absorb P will depend on the concentration of P ions in the soil solution at the root surface and the area of absorbing surface in contact with the solution. Mass flow and diffusion govern the movement of P ions in soil, with diffusion being of primary importance (Barber et al. 1963; Barber 1984). Therefore, the rate of P uptake is related to the rate of water uptake and P concentration in soil solution. The P ions near the root hairs are absorbed quickly, resulting in a depletion zone with a decreasing P concentration gradient near the root surface (Walker and Barber 1962; Bagshaw et al. 1972). Diffusion occurs in the depletion zone down the concentration gradient (Barber 1984). In highly P fertilized soils, the P concentration in soil solution is high (>1 ppm) and the depletion zone is readily replenished, but the replenishment is slow when soil solution P is low especially for soil solid phase with a low buffer capacity (Morel 2002). While P uptake by the root is a function of the concentration of P at the root surface, the quantity of P ions in soil solution at any given time generally represents less than 1% of P annually taken up by crops, with approximately 99% of P taken up by plants being bound to soil constituents before uptake. In soils cultivated for decades, about 75% of the total P in soils is in inorganic forms, more than 20% is organic P and a few percent is in soil microbial biomass (Morel 2002). The inorganic P in the soil solution is present as orthophosphate P ions, usually H 2 PO 4 and HPO 4 2 depending on soil ph. Replenishment of the ions in soil solution relies on mobilization of P from soil constituents by physico-chemical and biological mechanisms (Hinsinger 1998). The depletion of the P ion concentration of the solution at the root surface by absorption controls the release of P ions at the soil-solution interface by diffusion and their transport by diffusion in solution. The release of P ions from soil solid phase to solution varies with time and the gradient of P ion concentration and can be quantified by sorptiondesorption and isotopic dilution methods, which gave the same results if both variables are taken into account (Schneider and Morel 2000; Morel 2002). Phosphorus supply to a crop will be influenced by the ability of the soil to replenish the depletion zone at the root surface. In the depleted zone, the transfer of P ions appears to become very low after a few days and does not appear to be a very efficient strategy of P uptake even if the zone remained active over long periods to gain access to slowly released P (Table 1). The release of depleted P ions drastically decreased after a few days. The rate of release per day after 7 d of depletion represented only a few percent of that released during the first day (Morel 2002). To adequately characterize plant-available soil P it is therefore necessary to determine the dynamic of the amount (quantity factor, Q) of P released from soil to solution as a function of the concentration in soil solution P (intensity factor, I) and time (Morel et al. 2000). From Q-I relationships it is then possible to deduce phosphate buffering and

3 GRANT ET AL. PHOSPHORUS AND MYCORRHIZAE 5 Table 1. The daily change over 7 d in the gross amount (Pr) of P ions transferred between soil and solution for 1 wk in three soils. Calculations were made every day over a 7-d period using v, w and p parameters of the kinetic Freundlich (Pr = v Cp w t p ) published in Morel (2002), the P concentration in solution of 0.2 mg P L 1 Soil texture Pr(1d) Pr(2d-1d) Pr(3d-2d) Pr(4d-3d) Pr(5d-4d) Pr(6d-5d) Pr(7d-6d) mg P kg 1 d 1 soil Sandy soil Loamy soil Loamy clayed soil Can. J. Plant Sci. Downloaded from by on 05/09/18 sorbing capacity by soils (Morel 2002). Generally, soil testing procedures predict P availability using chemical extractants such as dilute concentration of strong acids, alone or in combination with complexing ions, dilute concentration of weak acids and buffered alkaline solution. Chemical extraction removes P from the solid soil phase by dissolution, anion replacement, complexing and hydrolysis of cations (Table 2). Although chemical extractions are still useful for the purpose of making fertilizer P recommendations, there are clear limitations to their ability to assess plant-available soil P (Fardeau et al. 1988). Soil solution P can be high even in a soil with a low chemically extracted P. For example, very light-textured silicate sands may have very low P content, but not react with phosphates. Application of a small amount of P fertilizer will increase the P concentration in solution because the sand phase does not react with P but the chemically extracted P will remain quite low. Therefore, reliable assessment of P supply to crops would require analysis of both solubility and mobility of P in soils, as well as an estimate of plant and mycorrhizal influences on P mobilization. Potential Benefits from Mycorrhiza Effective arbuscular mycorrhizal (AM) associations may help to improve early-season P nutrition in crops. The external hyphae of arbuscular mycorrhizae extend from the root surface to the soil beyond the P depletion zone and so access a greater volume of undepleted soil than the root alone (Hayman 1983; Jakobsen 1986; Plenchette and Fardeau 1988). Some hyphae may extend more than 10 cm from root surfaces (Jakobsen et al. 1992) which is a hundred times further than most root hairs. Also, the small diameter of hyphae (20 50 µm) allows access to soil pores that cannot be explored by roots. Therefore, a root system that has formed a mycorrhizal network will have a greater effective surface area to absorb nutrients and explore a greater volume of soil than nonmycorrhizal roots. In one study, the volume was calculated to be at least 100 times greater with mycorrhizal association than in its absence (Sieverding 1991). Moreover, mycorrhizal colonization may induce formation of lateral roots or increase root branching (Citernesi et al. 1998; Schellenbaum et al. 1991) further increasing the volume of soil explored. Mycorrhizal plants can absorb more P at lower concentration in the soil solution than nonmycorrhizal plants, as shown by for soybeans by Plenchette and Morel (1996) (Table 3). One possible explanation is that mycorrhizal hyphae have a higher affinity (lower Km) for P than roots (Howlever et al. 1981). But this phenomenon is not necessary to explain better P uptake by mycorrhizal roots. Barber (1984) explained that there is a very limited concentration gradient around hyphae (i.e., minimal depletion zone) since the radius of hyphae is much smaller than that of roots + root hairs (0.005 mm versus 0.15 mm). Hence, P concentration in soil solution around hyphae is always higher than in the P depletion zone around roots, and hyphae may absorb more P in low P soil even without having a higher affinity for P. Mycorrhizae also have biochemical and physiological characteristics which differ from those of roots which can enhance P availability. They can acidify the rhizosphere through increased proton efflux or pco 2 enhancement (Rigou and Mignard 1994), which can mobilize P (Bago and Azcon-Aguilar 1997), particularly in neutral or calcareous soils. In acidic soils, where phosphorus is mainly bound with Fe or Al, excretion of chelating agents (citric acid and siderophores) by mycorrhizae can enhance bioavailable P supply of the soil (Cress et al. 1986; Haselwandter 1995). Moreover, mycorrhizae also produce phosphatases, which can mobilize P from organic sources (Tarafdar and Marschner 1994a, b). These studies suggest that mycorrhizae can access, by various methods, pools of soil P that are not available to nonmycorrhizal plants. However, this was tested in several studies reviewed by Bolan (1991) where 32 P-labeled phosphate ions were mixed thoroughly in the soil to determine if the specific activity ( 32 P/ 31 P) of the P absorbed differed in mycorrhizal and non-mycorrhizal plants. Plants which access additional pools of P would have a lower specific activity (Barrow 1980). No difference was found in the specific activity of soybeans (Fig. 1) despite higher P uptake in the AM plants (Fig. 2) (Morel and Plenchette 1994). Specific activity was also similar in a range of other crop plants, AM-inoculated or not, despite large differences in growth and P uptake (Bolan 1991). More recently, a review of 51 published values in plant and in soil solution in different combination of soil types, plant species, and P fertilizers determined that specific activities of P for mycorrhizal and non-mycorrhizal plants did not differ (Morel 2002). Plant-available P, often called L value, did not differ after mycorrhizal inoculation. Also, as a corollary result, the relative contribution of P applied as a water-soluble form did not significantly differ between mycorrhizal and non-mycorrhizal plant. Other experimental approaches based on the use of low-solubility P sources applied to soils also generally concluded that mycorrhizal association does not allow the use of highly insoluble mineral sources (Tinker 1980). Therefore it appears that the primary benefit

4 6 CANADIAN JOURNAL OF PLANT SCIENCE Table 2. Examples of chemical extractants used to characterize plant-available soil P in few studies on AM association Authors Soil type Method Reagent Country Liu et al. (2000) Coarse-silty (humaquept) Mehlich-3 NH 4 F + NH 4 NO 3 + HOAc Canada + HNO 3 + EDTA Morton et al. (1990) Loam (hapudult) Olsen 0.5 M NaHCO 3, ph = 8.5 USA Joner (2000) Moraine soil (dystric cambisol) Olsen 0.5 M NaHCO 3, ph = 8.5 France Siqueira and Saggin-Junior (2001) Clay dark-reddish latosol (oxisol) Mehlich-1 HCl + H 2 SO 4 Brazil Plenchette et al. (1983) Podzol humo-ferric orthic Bray II 0.1 NH 4 F NHCl Canada Arihara and Karsawa (2000) Haplic wet andosol Truog N H 2 SO 4, ph = 3 Japan Wilson and Hartnett (1997) Silty clay loam (aquic arguidoll) Bray I 0.03 N NH 4 F N HCl USA Can. J. Plant Sci. Downloaded from by on 05/09/18 Table 3. Effect of mycorrhizal inoculation and phosphorus concentration in soil solution on dry matter yield and phosphorus content of soybean (Plenchette and Morel 1996) Dry matter yield (g plant 1 ) P content (mg kg 1 ) P fertilizer P in equilibrium supplied (mg kg 1 ) Shoot Grain Shoot Grain solution (µg ml 1 ) NM z M NM M NM M NM M h 3.42g 0.70j 3.01ef 0.37e 0.46de 2.71h 3.36f h 3.68g 1.25ij 3.39de 0.52de 0.51de 2.83gh 3.99e h 3.62g 1.53hi 3.47de 0.54de 0.57d 2.89fgh 4.20ed g 4.83f 2.12gh 3.45de 0.53de 0.50de 2.77h 4.32ed g 5.47ef 2.64fg 4.03cd 0.44de 0.53de 2.58h 3.96e g 5.30ef 2.94cf 3.91cd 0.47de 0.54de 2.90fgh 4.86bc g 5.86de 3.71cd 4.42bc 0.42de 0.50de 2.84gh 4.92bc bc 6.58cd 4.40bc 5.18a 0.43de 0.60d 3.34fg 4.58cd abc 6.59cd 4.38ab 4.85ab 0.61d 0.78c 4.23cd 5.17b ab 7.66a 5.04ab 5.47a 1.36b 1.60a 6.30a 6.35a z M = mycorrhizal, NM = nonmycorrhizal. a h For each tissue parameter, means followed by the same letter within a column or a row are not significantly different (Duncan s multiple range test P < 0.05). of mycorrhizae is in the extension of the zone of P uptake. The hydrolysis of organic P by exocellular phosphatases secreted by mycorrhizae, excretion of protons, hydroxyls and organic anions, and modifications of the redox potential around mycelium and roots of the mycorrrhizal association might also hasten the release of P ions from soil to solution (Hinsinger 2001). The roles of these different mechanisms have been shown especially in response to phosphorus deficiency (Lambers et al. 2003). However, it is not yet clear what the quantitative contribution to plant nutrition will be in field conditions for agricultural soils that are not highly P- deficient. Since P status during the early stages of crop development is important in determining the potential crop yield, mycorrhizal associations must be established early in crop growth to be of most benefit. Growing mycorrhizal crops after other mycorrhizal crops and using no-till management can encourage more rapid establishment of mycorrhizal associations and potentially enhance early-season P status of the crop (Miller 2000). Winter cover crops and even weeds may support the mycorrhizal network. In this way, the mycorrhizal association can enhance the P inflow and P concentration of crop during early development (Lu and Miller 1989). Miller (2000) suggested that an effective mycorrhizal system would function in a similar fashion to P fertilizer placed with the seed. In studies on corn in Ontario, rapid development of a crop-mycorrhizal association led to an increase in shoot-p concentration comparable to that obtained by applying 7 kg P ha 1 with the corn seed (Gavito and Miller 1998a). Mycorrhizal associations in mycorrhizal-dependant soybeans resulted in a decrease in critical soil P requirement (Table 3) (Plenchette and Morel 1996). Mycorrhizal associations increased flax yield at soil P levels below 40 kg P ha 1 (Thingstrup et al. 1998), with the mycorrhizal effect on flax in an unfertilized low-p soil corresponding to a P dressing of 90 kg P ha 1 (Kahiluoto et al. 2001). The benefit of mycorrhizal association was greater in flax than clover, with barley receiving no benefit, illustrating that the response to mycorrhizal association will vary considerably with crop species (Kahiluoto et al. 2001). It is important to note that, while mycorrhizal associations may be beneficial, they do not necessarily enhance P uptake sufficiently to maximize crop yields. Ryan and Ash (1999, 2000) reported that in spite of enhanced mycorrhizal association in biodynamic pastures, the level of P in the forage was below that of conventionally fertilized pastures (Table 4) and that P deficiency may have been restricting yields on the biodynamic system. In studies in SE Australia, neither field pea nor autumn-sown wheat showed a benefit in yield or P uptake from enhanced mycorrhizal colonization, even under low-p conditions, possibly due to growing season temperature or moisture regime which affect crop growth rate and P availability or to the type of AM fungal community present (Ryan and Angus 2003). Effect of Plant Phosphorus Status on Mycorrhizal Association A major impediment to exploiting mycorrhizal association

5 GRANT ET AL. PHOSPHORUS AND MYCORRHIZAE 7 Can. J. Plant Sci. Downloaded from by on 05/09/18 Fig. 1. Specific activity of P taken up by soybean, AM-inoculated or not, and specific activity of phosphate ions in solution for different concentration of P ions in solution (after Morel and Plenchette 1994). Soybean was inoculated with Glomus mosseae, cultivated for 95 d on sterilized soil in pot experiment. The range of soil solution P was obtained after applying increasing rates of water-soluble P fertilizer. The kinetics of P ions specific activity in soil solution was determined in soil suspensions (1 g:10 ml) for 35 d. The calculated change of the P specific activity after 35 d was negligible. in agricultural systems is that mycorrhizal association tends to decline as P concentration in the plant increases (Menge et al. 1978; Lu et al. 1994; Valentine et al. 2001). Higher tissue P in the plant reduces the production of spores (De Miranda and Harris 1994) and of secondary external hyphae (Bruce et al. 1994). Exudation from host plant roots of signal molecules that encourage hyphal branching is enhanced by P limitation in host roots (Nagahashi et al. 1996; Nagahashi and Douds 2000). Therefore, increasing P status of the root may reduce the secretion of these signal molecules, thus reducing hyphal branching and mycorrhizal association. Phosphorus status of the root may affect membrane phospholipids, thus influencing membrane permeability and the release from the roots of carbohydrates that nourish the fungi (Graham et al. 1981; Schwab et al. 1991). Therefore, where P concentration in the plant is low, carbohydrate exudation will encourage mycorrhizal association, which then enhances the uptake of P from the soil. Muthukumar and Udaiyan (2000) reported that concentration of soluble carbon in cowpea root increased with decreasing tissue P levels. As root carbohydrate concentration increased, mycorrhizal association was enhanced, although cause and effect was not necessarily proven. In this study, the percentage root length with arbuscules and vesicles and sporulation was more closely associated with carbohydrate concentration than was the total percentage mycorrhizal colonization, indicating that carbohydrates may influence the nature of the association. Olsson et al. (1997) also suggested, based on analysis of fatty acid signatures, that less carbohydrate is allocated to the root when P levels are higher and fewer spores are produced when there is less Fig. 2. Dry matter yield and P taken up by AM-inoculated soybean (M) and not AM-inoculated soybean (NM) as a function of soil solution P in a French luvisoil (after Morel and Plenchette 1994). Soybean was inoculated with Glomus mosseae, cultivated for 3 mo on sterilized soil in pot experiment with four replicates. The range of soil solution P was obtained after applying increasing rates of water-soluble P fertilizer. Symbols are for individual results. Lines connect the mean (+ or ) two standards deviations for each Cp value.figure 1. Specific activity of P taken up by soybean, AMinoculated or not, and specific activity of phosphate ions in solution for different concentration of P ions in solution (after Morel and Plenchette, 1994). Soybean was inoculated by Glomus mossae, cultivated for 95 days on sterilized soil in pot experiment. The range of soil solution P was obtained after applying increasing rates of water-soluble P fertilizer. The kinetics of P ions specific activity in soil solution was determined in soil suspensions (1g:10 ml) for 35 days. The calculated change of the P specific activity after 35 d was negligible. carbohydrate available for the fungus to form storage structures. However, increasing P levels had a limited effect on the extraradical/intraradical biomass ratio, except that at the highest level of P applied the extraradical hyphae tended to contribute a lower proportion of the biomass. In subsequent studies, Olsson et al. (1999) observed that increasing P concentrations reduced extraradical mycelium less than colonization.

6 8 CANADIAN JOURNAL OF PLANT SCIENCE Table 4. Concentrations of P and N in shoots of clover and ryegrass grown, with no nutrient additions, in soil from the conventional and biodynamic farms (Ryan and Ash 1999) Crop System Phosphorus Nitrogen mg g 1 Clover Conventional (0.003) 0.31 (0.01) Biodynamic (0.002) 0.29 (0.01) Ryegrass Conventional (0.001) 0.21 (0.01) Biodynamic (0.002) 0.21 (0.01) Mean and standard error of the mean in parentheses. As available P increases, increasing mycorrhizal association may depress plant growth, since there is a carbon cost associated with supporting the association (Kahiluoto et al. 2000). The relative benefit to the plant of the mycorrhizal symbiosis depends on the P supply and on other alternative strategies, such as root proliferation (Strong and Soper 1974a,b) that the plant can utilize to access sufficient P. The phenomenon was clearly demonstrated in hydroponic studies, where cucumber plants inoculated with mycorrhizal fungi produced 19% lower biomass yield than uninoculated plants when provided with a full-strength nutrient solution (Valentine et al. 2001). In contrast, if the concentration of P in the nutrient solution was reduced the biomass yield of the inoculated plants was 66% higher than the uninoculated plants. Even at very low levels of available P there may be a transient depression in plant growth associated with mycorrhizal association because the fungal partner may compete with the plant for the limited available P (Kahiluoto et al. 2000). Plant growth may be reduced if the AM fungus utilizes plant carbon for fungal growth and metabolism, but tissue P concentration and photosynthesis do not increase. Effect of Plant Characteristics on Mycorrhizal Association Plant species differ both in their mycorrhizal dependency and in the degree of association at varying levels of available P (Kahiluoto et al. 2000). For a given available P level, mycorrhizal dependency is related in part to the morphology of plant root system, with plants having extensive fibrous roots often less dependent on mycorrhizae than plants with less extensive root systems (Plenchette et al. 1983). For example flax, which has a small, fibrous root system, is more dependent and responsive to mycorrhizae than barley, a plant with an extensive, fibrous root system (Kahiluoto et al. 2000). In this study, relative mycorrhizal effectiveness for both species also varied with soil type, even at the same level of extractable P. The range of soil P concentration where a positive response to mycorrhizal association occurred was broader in flax, the crop with the higher mycorrhizal dependency and lower in barley, the crop with the lower dependency. There is some evidence that cultivars may differ in their relation with mycorrhizal fungi, their mycorrhizal response and the response of AM association to soil P level. In studies with cowpea, Rajapakse et al. (1989) observed that the extent to which natural colonization was reduced by addition of P did not differ between cultivars, but cultivars varied in response to inoculation at different P levels. Impact of Soil P Levels on Mycorrhizal Association Mycorrhizal association tends to decrease with increasing tissue P concentrations; therefore, if plant-available P increases and consequently tissue P concentration increases, mycorrhizal association can be affected. Thus, mycorrhizal association tends to decrease with increasing background levels of soil P. For example, Al-Karaki and Clark (1999) reported that both mycorrhizal association and the benefits of association on seed yield and tissue P concentration in durum wheat was reduced on soils with high as compared to low soil P content. Similarly, Kahiluoto et al. (2001) reported higher mycorrhizal colonization and greater benefit of mycorrhizal association in flax grown on soils with a low rather than intermediate initial P content. Where soil P was high, mycorrhizal association could reduce plant growth, presumably due to the carbon cost of the symbiosis (Kahiluoto et al. 2001). The P supply to the crop is largely influenced by the concentration of available P in the soil solution and the speed of replenishment of the solution. Therefore, factors which reduce solution P concentration can influence plant P status and mycorrhizal associations. In studies by Plenchette and Fardeau (1988), increases in dry matter yield and P uptake with mycorrhizal association were higher in soils with higher P fixing capacity than in soils with a lower P fixing capacity. In fact, the P fixing capacity had a greater influence on plant-available P and on the mycorrhizal effect than the content of available P as determined by the chemical extraction method. The effect of P level in the soil on mycorrhizal association appears to be indirect, through its influence on plant tissue P concentration rather than directly on the soil fungi. This was elegantly demonstrated in split-root studies where colonization of roots by mycorrhizal fungi occurred, even in the presence of high concentrations of soil P, as long as the concentration of P in the general root system was low (Menge et al. 1978). This suggests that soil P content may not affect colonization of juvenile plants before they begin to absorb significant amounts of soil P; in corn, for example, this would be near the three-leaf stage. Impact of phosphorus application and other nutrients Phosphorus fertilizers are frequently applied to improve the P nutrition of crops. Since P fertilizers can increase the P concentration in plant tissue, they can depress mycorrhizal association (Menge et al. 1978; Lu and Miller 1989; Kahiluoto et al. 2000, 2001; Liu et al. 2000). Phosphorus fertilization can reduce both mycorrhizal colonization and length of extraradical hyphae (Liu et al. 2000). Improving the P status of the plant can also reduce the potential benefit for the plant of the mycorrhizal association. In studies by Kahiluoto et al. (2000), freshly applied P fertilizer suppressed mycorrhizal colonization and function more than would be expected from its effect on P concentration in the soil solution. The short-term effect of P fertilization was not explained by the increase in soil P availability, but rather by an increase in plant P concentration, which affected mycorrhization (Menge et al. 1978). If the addition of fresh P was

7 GRANT ET AL. PHOSPHORUS AND MYCORRHIZAE 9 omitted, mycorrhizal formation was increased, but not necessarily crop growth because of the C cost of the additional mycorrhizae. Addition of the isoflavanoid formononetin to the soil partially overcame the reduction in AM colonization brought about by application of P (Fries et al. 1998). Formonenetin significantly reduced peroxidase activity, which could reduce the rigidity of cell walls, making the cells more susceptible to fungal colonization (Fries et al 1996). Phosphorus fertilization does not always reduce mycorrhizal association. In studies reviewed by Miller et al. (1995), extensive mycorrhizal colonization was observed even at very high rates of fertilization (Fairchild and Miller 1988). Localized concentration of P in fertilized systems could also influence mycorrhizal activity. Lu et al. (1994) suggested that colonization was controlled locally by the P concentration in the portion of the root system being colonized. Therefore, colonization of roots in a P band may be restricted by high P concentration in the local root tissue while the colonization of roots growing further from the band may not be influenced as much, until the overall plant P concentration is substantially increased. Paradoxically, if the available P in the soil is very low, mycorrhizal colonization and spore production may be restricted and mycorrhizal associations may be increased by P application. For example, sunflower grown in near P-free sand medium showed only poor mycorrhizal development, with infection increasing as P was added (Koide and Li 1990). In pot studies, both total weight of infected roots and maximum percentage of root length infected increased with increasing P to levels, which produced between 6 and 66% of the maximum shoot growth (Bolan et al. 1984). Xavier and Germida (1997) also reported an increase in the percentage of AM colonized lentil roots with moderate levels of P fertilization. Similarly, mycorrhizal inoculation increased the grain yield of Neepawa wheat by 20% when 20 kg P ha 1 was applied (Xavier and Germida 1997). The beneficial effect of P fertilization in these studies may relate to enhancement of growth of the mycorrhizal fungus or the host plant when P in the soil is very low. With higher available soil P, application of fertilizer P may depress mycorrhizal association. Lower association may benefit crop growth by reducing the carbon drain to the fungi (Kahiluoto et al. 2001). However, with moderate to high P levels in the soil, the incremental influence of fertilizer P on mycorrhizal association may be negligible, as mycorrhizal association would be suppressed by the residual soil P, whether P fertilizer was applied or not (Gavito and Miller 1998b). Source of P fertilization may also influence mycorrhizal development, with the effects being primarily related to the solubility and availability of the P source. While readily soluble and phytoavailable forms of P will rapidly increase P supply to the plant and decrease the mycorrhizal association, less-soluble forms of P such as rock phosphate have less effect on P supply to the plant and mycorrhizal association. Hence, mycorrhizal association may be especially important when less-soluble P forms are used for crop production. The slow release of P may prevent the tissue P concentration from reaching levels that tend to inhibit mycorrhizal associations. This effect has been shown for various slowly available P sources such as rock phosphate (Toro et al. 1997), plant residues (Joner and Jakobsen 1995) and manure (Joner 2000). For example, Steffens (1992) found that after many years of application of rock phosphate on a calcareous soil, severe P deficiency was observed in non-mycorrhizal sugar beets, but not in well-colonized sunflowers, while wheat was intermediate in colonization and P deficiency. Similarly, in pot studies by Joner (2000) on soils with a long-term (74-yr) history of farmyard manure (FYM) or chemical fertilization, accumulation of P from past manure applications had a smaller effect in decreasing mycorrhizal associations than accumulation of P from past fertilizer applications. Recent additions of NK fertilizer and FYM had no effect on mycorrhizal formation, while additions of NPK led to reduced colonization. Therefore, it appeared that moderate applications of FYM would have less of an adverse effect on AM than application of the same amount of nutrients in the form of chemical NPK fertilizers. The difference between the sources was possibly due to temporal difference in P availability, with the FYM gradually releasing P in balance with plant demand. Organic amendments may also benefit root growth and mycorrhizae by improving soil physical condition through addition of organic matter (Campbell et al. 1986). The effect of P fertilization may also vary depending on the balance of other nutrients present. Mycorrhizal association tends to be highest when low P is combined with an ample supple of other nutrients. Valentine et al. (2001) observed that mycorrhizal association in cucumber was fourfold higher when the solution culture contained 10% of the full level of P and the full level of other nutrients than when all nutrients including P were at the full level and threefold higher than if all nutrients including P were apply at 10% of the full level. This demonstrates that although P availability is a major driving factor in mycorrhizal association, an adequate supply of other nutrients is required for high levels of association. Plenchette and Corpron (1987) reported that application of P or K alone to fescue decreased propagule densities, but application of P and K together reduced the negative impact. Similarly, Saif (1986) reported that application of P alone reduced mycorrhizal colonization in tropical forages under field conditions while fertilization with a balanced NPK blend did not. In growth chamber studies with maize, mycorrhizal association was highest with low levels of P and intermediate levels of N (Liu et al. 2000). The increased mycorrhizal association and extraradical hyphal growth observed with intermediate N fertilization was associated with an increase in plant shoot weight. The authors suggested that increased growth may lead to dilution of the P in the root and enhanced carbohydrate exudation, leading to stimulation of AM fungi and enhanced colonization (Liu et al. 2000). Colonization would then be lower under conditions where N does not stimulate growth, such as very high or very low N levels. Interactions among N, P and K fertilization were noted in corn by Guttay and Dandurand (1989) who observed an increase in mycorrhizal association with N and K fertilization at low P levels and a decrease at high P levels. The N and K fertilization

8 10 CANADIAN JOURNAL OF PLANT SCIENCE also increased leaf P concentration, which may have influenced mycorrhizal association. Hepper (1983) also noted that mycorrhizal colonization at low root P concentrations was regulated by N concentration in the tissue, with high N concentrations having higher colonization. The nutrient concentration and their ratios in roots can influence arbuscule formation, with formation being promoted by low tissue nutrient content or nutrient imbalances (Muthukumar and Udaiyan 2000). Impact of Long-term P Fertilization As well as having immediate effects on plant growth and mycorrhizal colonization, P fertilization may have longterm effects on the P supply in the soil and on mycorrhizal fungi populations. Mycorrhizal infectivity of soils has been reported to decline with cumulative P fertilization in a number of studies (Anderson et al. 1987; Clapperton et al. 1997; Thingstrup et al. 1998; Kahiluoto et al. 2000, 2001). Longterm P fertilization will generally increase the background level of P in the soil, which will increase the P status of the plants. Nevertheless, producers observe that crops may still respond to P fertilizer placed near the seed, so many high P fields continue to receive additional P each year, further increasing accumulation of P in the soil (Jokela 1992). The increased P status of the plants can influence colonization. For example, Jensen and Jakobsen (1980) reported that long-term fertilization led to high total P in the soil and lower AM colonization. Similarly, Thomson et al. (1992) determined that colonization of roots by Scutellospora calospora was inversely related to increasing residual soil P from different fertilization rates, even after 6 yr without P. Anderson et al. (1987) reported that mycorrhizal association declined after 7 yr of application of 180 kg P ha 1 under conventional tillage, but was not significantly decreased by P accumulation in the surface layer under reduced tillage. Reduced tillage is known to enhance colonization for reasons unrelated to P stratification (Miller 2000). Low-input crop production systems may enhance mycorrhizal activity, especially if P supply to the crop is restricted. Mäder et al. (2000) reported that root colonization under field conditions was up to 60% higher in low- rather than high-input systems, with the majority of the effect being attributable to soil chemical properties, particularly soluble soil P. Ryan and Ash (1999) evaluated mycorrhizal association in greenhouse studies using soil from three longterm conventional and three biodynamically managed dairy cattle pastures. The biodynamic soils had no external inputs of organic or inorganic fertilizers for an average of 17 yr, while the conventional soils had received regular inputs of N and P fertilizers. Colonization of the pasture species was lower on the conventional than on the biodynamic soils, presumably because the background soil P level was two- to threefold higher in the conventional than the biodynamic soil. Similarly, colonization of clover was reduced under conventional as compared to biodynamic pastures and the decrease was correlated with the increased level of soil P resulting from long-term P addition in the conventional system (Ryan and Ash 2000). However, under both systems, colonization decreased in a similar fashion with increasing shoot P concentration, indicating that the mycorrhizal fungi in conventional soils were not more or less tolerant of P. In fact, application of P over many years may decrease the infectivity and affect the functional properties of the mycorrhizal fungi communities (Kahiluoto et al. 2000). In a longterm experiment (>70 yr), repeated P fertilizer application decreased the mycorrhizal soil infectivity whether mineral or organic P was applied (Plenchette 1989). Long-term P fertilization did not affect the number of species of mycorrhizal fungi, but decreased spore densities and root colonization (Kahiluoto et al. 2001). Kahiluoto et al. (2000) suggested that the difference in infectivity was probably due to a difference in size of the community, but could also indicate a difference in its functional structure. Strategies to Optimize P Nutrition of Crops Phosphorus nutrition must be at an adequate level in the early stages of plant growth to optimize the yield potential of the crop. Plant species differ in their P requirements, in the methods used to access available P and in their response to P fertilizer applications (Kalra and Soper 1968; Strong and Soper 1974a,b). Phosphorus management practices must be designed to address the nutrient requirements of the individual crops and the nutrient management goals for the crop production systems. In annual, fertilizer-based crop production systems, where plant-available P is low, efficient fertilizer P applications can be used to increase P status of the crop, provided it is economically feasible. Plant-available P sources placed in bands or near the seed-row can improve P use efficiency by allowing the crop roots to access the P early in crop growth and by slowing the reaction of the P with Ca, Mg or with Fe or Al oxides (Mitchell 1957; Sample et al. 1980). Improving P use efficiency can encourage decreased P application rates, which will reduce accumulation of P in the soil and thus decrease the risk of negative environmental impacts through. If P has accumulated in the soils to high levels through excessive applications of manure or P fertilizer, there can be an environmental risk of enhanced movement of P to surface or ground water. Producers may still wish to apply P fertilizer in a belief that application of P will enhance early crop growth and optimize yield potential. Improved strategies for P soil-testing, which include assessment of factors other than extractable P may help to more clearly identify fields where P responses are likely to occur. Improved prediction would require consideration of crop type and soil P fixing ability as well as factors which affect mycorrhizal infectivity, such as cropping sequence and tillage system. Much better quantitative information is needed on the benefits and costs of mycorrhizal associations. Mycorrhizal associations may be most useful in situations where application of fertilizer P is undesirable. These may include organic or low-input farming systems or soils with moderate to high P content. Where the risk of P movement to water is high, it may be important to maintain the level of P near the soil surface at very low levels. Mycorrhizal associations could be of great benefit in enhancing the ability of the crop to extract P from the soil and improve nutritional status of the crop. Strategies could be used to enhance early

9 establishment of mycorrhizal association; fortunately very early colonization may avoid the depressing effect of P application on colonization. Such strategies would include use of reduced tillage management, elimination of fallow and sequencing of mycorrhizal-dependent crops following one another (Miller 2000). Genetic improvement of both crops and mycorrhizae may enhance both the ability of crops to access P from the soil and their ability to form useful mycorrhizal associations. While the economic benefit of using mycorrhizal inoculants in typical farming systems has not been confirmed (Miller 2000), there may be situations where native mycorrhizal populations are low and inoculants may be beneficial. CONCLUSIONS Effective P management is important to optimize crop yield potential, reduce production costs and decrease the risk of environmental damage. Plants require adequate P from the very early stages of growth for optimum crop production. Phosphorus supply to the crop is affected by soil P and by soil and environmental conditions influencing P phytoavailability and root growth. Cropping system and long-term input of P through fertilizers and manures can influence the amount and phytoavailability of P in the system and the development of mycorrhizal associations. Phosphorus uptake in many crops is improved by associations with arbuscular mycorrhizal fungi, particularly in low P soils. In many agricultural soils of western Europe, the United States of America and Canada, accumulation of P beyond crop requirement has a negative impact on mycorrhizal association. The long term use of commercial fertilizers has increased the plant-available soil P of many agricultural soils to excessive levels. Also, in areas of intensive livestock production, manure P, once considered a resource, is increasingly seen as a source of pollution. Effective soil testing is needed to predict the early-season P supply. Where early-season P supply is low, P fertilization may improve P nutrition and crop yield potential. Alternately, encouragement of AM associations may enhance P uptake by crops early in the growing season, improving crop yield potential and replacing fertilizer P applications. Quantitative analysis of the mechanisms employed by mycorrhizal associations in phosphorus nutrition of crop plants is still required to understand the contribution of these associations in the functioning of the soil-plant system. To be relevant for agronomic purposes, this analysis should be developed under field conditions representative of commercial production systems. Use of effective P management practices, whether through efficient fertilizer use or encouragement of mycorrhizal associations can optimize the economics of crop production while avoiding negative effects of P on environmental quality. Al-Karaki, G. N. and Clark, R. B Mycorrhizal influence on protein and lipid of durum wheat grown at different soil phosphorus levels. Mycorrhiza 9: Anderson, E. L., Millner, P. D. and Kunishi, H. M Maize root length density and mycorrhial infection as influenced by tillage and soil phosphorus. J. Plant Nutr. 10: GRANT ET AL. PHOSPHORUS AND MYCORRHIZAE 11 Arihara, J. and Karasawa, T Effect of crops on arbuscular mycorrhizal formation and growth of succeeding maize. Soil Sci. Plant Nutr. 46: Atkinson, D. and Davidson, A. W The effects of phosphorus deficiency on the growth of Epilobium montanum L. New Phytol. 70: Avnimelech, Y. and Scherzer, S The effect on yield of phosphorus uptake by young plants. Pages in R. M. Samish, ed. Recent advances in plant nutrition. Proceedings of the 6th International Colloquium on Plant Analysis and Fertilizer Problems. Tel Aviv. March Gordon and Breach Science Publishers, London, UK. Vol. 2. Bago, B. and Azcon-Aguilar, C Changes in the rhizospheric ph induced by arbuscular mycorrhiza formation in onion (Allium cepa L). Z. Pflanz. Bodenkunde. 160: Bagshaw, R., Vaidyanathan, L. V. and Nye, P. H The supply of nutrient ions by diffusion to plants roots in soil. V. Direct determination of labile phosphate concentration gradients in a sandy soil induced by plant uptake. Plant Soil 37: Barber, S. A Soil nutrient bioavailability: A mechanistic approach. John Wiley & Sons Inc., New York, NY. 398 pp. Barber, S. A., Walker, J. M. and Vasey, E. H Mechanisms for the movements of plant nutrients from the soil and fertilizer to the plant root. J. Agric. Food Chem. 11: Barrow N. J Evaluation and utilization of residual phosphorus in soils. Pages in F. E. Khasawneh, E. C. Sample, and E. J. Sample, eds. The role of phosphorus in agriculture. ASA, CSSA, SSSA Publication, Madison, WI. Barry, D. A. J. and Miller, M. H Phosphorus nutritional requirement of maize seedlings for maximum yield. Agron. J. 81: Boatwright, G. O. and Viets, F. G., Jr Phosphorus absorption during various growth stages of spring wheat and intermediate wheatgrass. Agron. J. 58: Bolan N. S Critical review on the role of mycorrhizal fungi in the uptake of phosphorus by plants. Plant Soil 141: Bolan, N. S., Robson, A. D. and Barrow, N. J Increasing phosphorus supply can increase the infection of plant roots by vesicular-arbuscular mycorrhizal fungi. Soil Biol. Biochem. 16: Brenchley, W. E The phosphate requirement of barley at different periods of growth. Ann. Bot. 43: Bruce, A., Smith, W. E. and Tester, M The development of mycorrhizal infection in cucumber: effects of P supply on root growth, formation of entry points and growth of infection units. New Phytol. 1276: Campbell, C. A., Schnitzer, M., Stewart, J. W. B., Biederbeck, V. O., Selles, F Effect of manure and P fertilizer on properties of a black chernozem in southern Saskatchewan. Can. J. Soil Sci. 66: Citernesi, A. S., Vitagliano, C. and Giovannetti, M Plant growth and root system morphology of Olea europa L. rooted cuttings as influenced by arbuscular mycorrhizae. J. Hortic. Sci. Biotechnol. 73: Clapperton, M. J., Janzen, H. H. and Johnston, A. M Suppression of VAM fungi and micronutrient uptake by low-level P fertilization in long-term wheat rotations. Am. J. Altern. Agric. 12: Crafts-Brandner, S. J Significance of leaf phosphorus remobilization in yield potential in soybean. Crop Sci. 32: Cress, W. A., Johnson, G. V. and Barton, L. L The role of endomycorrhizal fungi in iron uptake by Hilaria jamesii. J. Plant Nutr. 9: