This manuscript has been created by the University of Chicago as operator of Argonne National Laboratory under Contract No. W-31-109-ENG-38 with the US Department of Energy. The US government retains for itself, and others acting on its behalf, a paid-up, nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the government.
Mycorrhizal mediated feedbacks influence net carbon gain and nutrient uptake in Andropogon gerardii
Article first published online: 18 JUN 2002
Volume 155, Issue 1, pages 149–162, July 2002
How to Cite
Miller, R. M., Miller, S. P., Jastrow, J. D. and Rivetta, C. B. (2002), Mycorrhizal mediated feedbacks influence net carbon gain and nutrient uptake in Andropogon gerardii. New Phytologist, 155: 149–162. doi: 10.1046/j.1469-8137.2002.00429.x
- Issue published online: 18 JUN 2002
- Article first published online: 18 JUN 2002
- Received: 5 December 2001 Accepted: 26 February 2002
- arbuscular mycorrhizal fungi;
- carbon gain;
- Andropogon gerardii;
- carbon allocation;
- P : N ratio
- • The carbon sink strength of arbuscular mycorrhizal fungi (AMF) was investigated by comparing the growth dynamics of mycorrhizal and nonmycorrhizal Andropogon gerardii plants over a wide range of equivalent tissue phosphorus : nitrogen (P : N) ratios.
- • Host growth, apparent photosynthesis (Anet), net C gain (Cn) and P and N uptake were evaluated in sequential harvests of mycorrhizal and nonmycorrhizal A. gerardii plants. Response curves were used to assess the effect of assimilate supply on the mycorrhizal symbiosis in relation to the association of C with N and P.
- • Mycorrhizal plants had higher Cn than nonmycorrhizal plants at equivalent shoot P : N ratios even though colonization did not affect plant dry mass. The higher Cn in mycorrhizal plants was related to both an increase in specific leaf area and enhanced photosynthesis. The additional carbon gain associated with the mycorrhizal condition was not allocated to root biomass. The Cn in the mycorrhizal plants was positively related to the proportion of active colonization in the roots.
- • The calculated difference between Cn values in mycorrhizal and nonmycorrhizal plants, Cdiff, appeared to correspond to the sink strength of the AMF and was not an indirect result of enhanced nutrition in mycorrhizal plants.
The outcome of a mycorrhizal relationship depends on the balance between fungal demands for energy and the plant’s need for nutrients. In general, a plant growing under conditions of nutrient limitation increases allocation of carbohydrate assimilates from leaves to roots (Marschner et al., 1996). The transport of carbohydrate assimilates from leaves to arbuscular mycorrhizal fungi (AMF) in the root has been studied in less detail, although it is clear that the fungus receives all of its carbohydrate from the host plant (Jakobsen & Rosendahl, 1990; Eissenstat et al., 1993; Peng et al., 1993; Smith & Read, 1997). The colonization of a plant’s roots by AMF creates a ‘sink’ demand for carbohydrates, which could create an extra 4–26% drain of C from the host (Pang & Paul, 1980; Kucey & Paul, 1982; Snellgrove et al., 1982; Koch & Johnson, 1984; Douds et al., 1988; Jakobsen & Rosendahl, 1990; Black et al., 2000).
Some investigators have suggested that the growth of the mycorrhizal fungus is a C-limited process, with the host having ultimate control over fungal activity through regulation of the transfer of carbohydrates to the roots (Schwab et al., 1991; Reinhard et al., 1994; Smith & Smith, 1996). However, many experiments conducted at elevated CO2 concentrations show no effect on AMF due to increasing the C supply, suggesting a give–take balance between fungus and plant (Staddon & Fitter, 1998). Other investigators have hypothesized that the increased demand for C by the growing fungus increases the rate of C assimilation by the host plant (Johnson, 1984; Koch & Johnson, 1984; Brown & Bethlenfalvay, 1988; Fitter, 1991; Wright et al., 1998a,b). However, it has proven difficult to differentiate whether this increased C gain is a direct effect of fungal colonization or an indirect effect of the enhanced leaf P status that accompanies fungal colonization (Graham, 2000). Explanations for increased sink strength caused by mycorrhizal fungal colonization of roots include rapid utilization of C and its rapid conversion into fungal-specific compounds (Bevage et al., 1975; Losel & Cooper, 1979), increased respiration by the mycorrhizal root system (Pang & Paul, 1980; Snellgrove et al., 1982; Harris et al., 1985), and increased activity of sucrolytic enzymes (Wright et al., 1998b; Müller et al., 1999).
Increasing soil P supply to a host has been demonstrated to inhibit mycorrhizal colonization and hyphal growth (Abbott et al., 1984; Thomson et al., 1986). A host appears to regulate the symbiotic cost of colonization when AMF provide little or no P benefit (i.e. with high soil P supply). This regulation results in more available C for processes that are likely to provide a more immediate return on investment, such as leaf production for photosynthesis (Graham & Eissenstat, 1994). However, evidence suggests that the balance between P and N supply to the host may be more important than the effects of P alone for regulating colonization by AMF (Smith et al., 1986; Sylvia & Neal, 1990). For example, suppression of fungal colonization did not accompany increased P when plants were deficient in N, but P additions reduced colonization when N was sufficient (Sylvia & Neal, 1990). Hepper (1983) reported that the P : N ratio of the nutrient addition was the most important factor governing fungal colonization when P supply was low to adequate. It is evident that when either P or N is deficient, as is the case for most natural systems, a delicate balance exists between the need of the host for C assimilates and nutrients and the allocation of these assimilates for growth of AMF.
The P : N ratio of the host tissue may delimit the conditions favorable for growth of AMF by defining the boundaries for host C partitioning to sinks available to mycorrhizal fungi. Thus, the mycorrhizal symbiosis may be largely controlled by the fluxes of host-supplied C and fungus-supplied nutrients in a feedback system. In the most simplistic view, partitioning of C assimilate to the root increases in response to plant nutrient deficiencies and/or imbalances, allowing the growth and activity of AMF to increase and resulting in greater nutrient inflow to the plant. When plant nutrient demands are ameliorated, allocation of C to the root declines until nutrients once again become limiting. In reality, this feedback system can be moderated and perturbed by a myriad of abiotic and biotic factors affecting host and fungus physiology. For example, flowering is one phenological event that may compete strongly with AMF for assimilated C, causing plant C allocation patterns to undergo dramatic shifts at this time (Johnson et al., 1982).
The mycorrhizal symbiosis represents a series of complex feedbacks between host and fungus that is governed by their physiology and nutrition. Most experiments on C dynamics have added P to matched mycorrhizal and nonmycorrhizal plants in order to achieve similar biomass and P contents (Syvertsen & Graham, 1990). However, comparing host response curves at various levels of host tissue P and N may be a better way to evaluate the amount of C assimilates available for allocation to below-ground sinks (Rousseau & Reid, 1991; Eissenstat et al., 1993). Response curves allow for simultaneous evaluation of the effects of inoculation on both nutrients and C assimilation.
The objective of this study was to assess the sink strength of AMF by comparing mycorrhizal and nonmycorrhizal plants over a wide range of equivalent P : N ratios. We investigated these relationships in an Illinois population of the prairie grass Andropogon gerardii by measuring the dynamics of host growth, apparent photosynthesis, net C gain, and P and N uptake in sequential harvests of mycorrhizal and nonmycorrhizal plants. A second objective was to use response curves to assess the effect of assimilate supply on the mycorrhizal symbiosis in relation to the association of C with N and P. The amount and activity of the AMF in relation to host assimilate and nutrient dynamics were also evaluated.
Materials and Methods
The response of Andropogon gerardii Vitman to inoculation with AMF at various P and N levels in the soil was evaluated. Seeds of A. gerardii were collected from the restored tallgrass prairie of the Fermi National Environmental Research Park (Fermilab) near Batavia, Illinois, USA (Schultz et al., 2001), located in a relatively fertile area known as the Prairie Peninsula. Plants were grown in a full factorial design with two levels of AMF (inoculated and uninoculated control), two levels of soil P and three levels of soil N, with each treatment replicated six times. Host and endophyte responses to the treatments were quantified at five equally spaced sequential harvests (30, 45, 60, 75, and 90 d after planting, DAP) comprising 360 plants.
The growth medium consisted of pasteurized soil mixed with calcinated clay (Terragreen Soil Conditioner, Oil-Dri Corporation of America, Chicago, IL, USA) and red flint sand (mixture of equal parts by volume #20 and #32 mesh) in a 3 : 1 : 1 ratio by volume. The soil was a Mundelein silt loam collected from the Fermilab site at various locations, including the restored prairies and pasture grass areas (Miller et al., 1995). Soils were passed through a 6.35-mm sieve to remove roots and rhizomes and to mix the soil. The soil was then autoclaved for 1.5 h and allowed to cool, mixed, and autoclaved again.
The two soil P treatments were created by spreading equal portions of the growth medium on a polyethylene sheet to a depth of 2 cm and spraying with a solution of either KCl or KH2PO4 in distilled water. The P0 (no P added) treatment had a Bray-1 extractable soil P (Bray & Kurtz, 1945) of 18.6 µg cm−3 of soil; no additional P was added. For the P enhancement treatment (P1 treatment), Phosphate (PO4)-P at approximately 40 µg cm−3 of soil was added, resulting in a Bray-1 extractable soil P of 38 µg cm−3. The application of KCl to the P0 treatment provided an equivalent amount of K as did the KH2PO4 addition. The N treatments consisted of no addition (N0), a total N addition of 12.6 µg cm−3 (N1), and a total addition of 126 µg cm−3 (N2). The N additions were applied as an NH4NO3 solution at weekly increments during watering (total of 12 wk), with N0 treatment receiving an equivalent volume of deionized water. The initial growth medium (N0 treatment) had ammonium (NH4)-N at 13.6 µg cm−3 and nitrate (NO3)-N at 1.7 µg cm−3, for a total mineral N content of 15.3 µg cm−3 (based on a 2 m KCl extract of the soil).
The AMF used in the experiment were a mixture of primarily Glomus mosseae (Nicol. & Gerd.) Gerdemann & Trappe; Paraglomus occultum (Walker) Morton & Redecker; Glomus microaggegatum Koske, Gemma & Olexia; and Glomus geosporum (Nicol. & Gerd.) Walker isolated from soils at the Konza Prairie Long-Term Ecological Research site near Manhattan, Kansas (Schultz et al., 2001). The inoculum was developed and increased by cultivating AMF on roots of A. gerardii and Sorghum vulgare L. in pasteurized Fermilab soil diluted with red flint sand (2 : 1 ratio). After 4 months of growth in a controlled-environment room, the soil with roots containing inoculum was air dried, homogenized by mixing the soil and cutting roots into pieces < 1 cm in length, and refrigerated at 4°C for 1–2 months before use.
Seedling establishment and growth conditions
Before planting, 6 × 15 cm cylindrical pots were filled with 500 cm3 of the growth medium. For the mycorrhizal treatments, 25 cm3 of soil inoculum was layered on top of the pasteurized growth medium, and 25-cm3 of pasteurized soil was placed above the inoculum. For nonmycorrhizal treatments, 50 cm3 of pasteurized growth medium was added to the pots. Soil microflora and microfauna, minus the mycorrhizal component, were added by washing nonpasteurized soil (collected from around mycorrhizal A. gerardii roots) with deionized water, passing the soil wash through a 38-µm sieve, and adding 50 ml of the sievate to the growth medium. A single 2-wk-old seedling of A. gerardii (pregerminated in vermiculite and roots trimmed to uniform size) was planted in each pot. Plant growth occurred in a controlled-environmental room on a 16 h : 8 h day–night cycle. Day and night temperatures were set at 28°C and 22°C, respectively. Photosynthetically active radiation (PAR) measured daily at the leaf surface ranged from 550 µmole photons m−2 s−1 at the beginning of the experiment to 460 µmole photons m−2 s−1 at the end of the experiment. Relative humidity was maintained at approximately 50% for the duration of the experiment. The plants were watered every other day with 25 ml of deionized water and were watered biweekly to field capacity. Although PAR was about one-third of saturation for photosynthesis in a C4 grass, the levels used allowed maximization of the range of responses imposed by the mycorrhizal and nutrient addition treatments. Photoassimilate was most likely limiting to plant growth only at the very highest P and N treatments.
The day before harvesting, plants were irrigated to field capacity. Photosynthesis measurements were conducted midmorning in the growth chamber under conditions described above to allow for control of air temperature, relative humidity and irradiance. Apparent photosynthesis (Anet) (CO2 assimilation in µmol m−2 s−1) was measured for each plant by using a LI-6200 gas exchange system (Li-Cor, Lincoln, NE, USA) on the most recently fully expanded leaf held perpendicular to the light source. These leaves were chosen to avoid measuring different maturation states within a harvest date. Plant height, total numbers of living and dead leaves, and length (b1) and width (b2) of the most recently fully expanded leaf were measured before plants were removed from pots. Leaf area (La) of the most recently expanded leaf was calculated from the equation La = (b1 × b2 × 0.68), as derived from 30 single leaf measurements. Estimated shoot area (SAc) was calculated from the equation SAc = La × Ln, where Ln is the number of living leaves. The relationship between actual shoot area and the estimated shoot area, SAc, was determined in a separate experiment by measuring total leaf area (SA) with a LI-3000 portable area meter (Li-Cor) and using this measured area to correct the estimated value (SA = 2.06 + (0.47 × SAc)). This adjusted value for shoot area is used in this study.
Measurements of growth, allocation patterns, and carbon gain
Shoot, root, and total plant mass were measured after drying to constant weight at 65°C at each harvest. Because the inflorescence was projecting at 90 DAP, shoot biomass was separated into vegetative and reproductive parts before dry weight determination at this harvest date. Mean relative growth rate (RGR) for the treatments was calculated from the product of the slopes of the least squares regressions for RGR = ΔlogeW/Δt, where W and t are plant dry weight and DAP. Leaf area ratio (LAR) was calculated from the equation LAR = SLA × LWR, where specific leaf area (SLA) is leaf area (cm2) per leaf mass (g), and leaf weight ratio (LWR) is the fraction of the total biomass allocated to the leaves (mass of leaf (g)/mass of plant (g)).
Net carbon gain (Cn) was calculated from the equation Cn = A × SA, where A is the measured CO2 assimilation rate. Also calculated was the net symbiotic C gain difference (Cdiff) between mycorrhizal and nonmycorrhizal plants on the basis of equivalent P : N ratios by using predicted response curves for shoot P : N ratio. For this calculation, nonmycorrhizal net C gain (Cn−amf) was subtracted from mycorrhizal net C gain (Cn+amf) according to the equation Cdiff = Cn+amf − Cn−amf, at equivalent shoot P : N ratio. Both linear and nonlinear regressions were performed to determine the best fit for the relationship between Cn and shoot P : N ratio for both mycorrhizal and nonmycorrhizal hosts at each harvest. The Cdiff for each measured shoot P : N was then calculated as the difference in Cn for mycorrhizal vs nonmycorrhizal plants. The empirical relationship between Cn and shoot P : N ratio for mycorrhizal and nonmycorrhizal A. gerardii at 75 and 90 DAP was then determined by using response curves to predict Cn and to calculate Cdiff. The parameter Cdiff was defined as the predicted Cn for the mycorrhizal host minus the predicted Cn for the nonmycorrhizal host at an equivalent shoot P : N ratio.
Root length and mycorrhizal colonization
Roots were separated into coarse (> 1.0 mm diameter) and fine (< 1.0 mm diameter) categories. Fine roots were cut into 2-cm segments and mixed thoroughly, and a 0.25 g (fresh weight) subsample was removed for determination of root length and colonization. The dry weight : fresh weight ratio of the remaining roots was used to calculate the dry weight of the subsample. Root length of the 0.25 g subsample was estimated by using the Ag-Vision (Decagon Devices, Pullman, WA, USA) digital-line-intercept method (Harris & Campbell, 1989). Total fine root length was calculated as the ratio of subsample length to subsample dry mass, times the total fine root dry mass.
The root subsample was subjected to nitro blue tetrazolium chloride (NBT)/acid fuchsin ‘vital’ staining for quantification of AMF. The protocol, which is outlined in Schaffer & Peterson (1993), involved staining the freshly harvested roots with a buffered NBT-succinate solution and counterstaining with acid fuchsin. This procedure allowed succinate dehydrogenase (SDH)-active AMF structures in the roots to be distinguished. Stained roots were arranged lengthwise on microscope slides and mounted in polyvinyl alcohol-lacto-glycerol. The proportion of root length colonized by AMF was quantified by the magnified-intersects method at ×160 (McGonigle et al., 1990). At each intersection, the type and activity of the AM fungal structure was noted. In 150 intersections per slide, the presence or absence of intracellular hyphae, coils, arbuscules, or vesicles was recorded. Activity of the fungal structure, as determined by the vital staining procedure was also quantified as the ratio of SDH-active to total fungal structures.
Tissue nutrient and nonstructural carbohydrate concentrations
Shoot and root tissue N and P concentrations were determined after wet digestion with sulfuric acid and hydrogen peroxide of dried plant material ground through a 20-µm mesh. Phosphorus concentrations were determined on the digest by using the molybdate-blue method (Murphy & Riley, 1962), and N concentrations were determined by a Kjeldahl method with a rapid-flow autoanalyser.
Total nonstructural carbohydrate concentrations were determined for both shoot and root samples by summing free glucose, ketone sugars, sucrose, soluble starch and insoluble starch pools (Roe et al., 1949; Ashwell, 1957; Van Handel, 1968; Smith, 1981). The basic procedure involved hot-water extraction (12 min at 135°C) of dried root tissue, followed by centrifugation at 2250 g for 7 min to separate soluble and insoluble materials. The supernatant was analysed for free glucose by using glucose oxidase and for ketone sugars (free fructose) by resorcinol reagent. Sucrose was determined by destruction of reducing sugars with hot concentrated KOH, followed by anthrone determination of the fructose moiety of sucrose. Insoluble and soluble starch was determined by digesting the pellet and a subsample of supernatant with amyloglucosidase. Starch quantities were determined as glucose by the glucose oxidase method (Sigma Chemical Company, St Louis, MO, USA) and soluble starch glucose was corrected by subtraction for free glucose in the supernatant.
Extractable soil C
After removal of roots at harvest, the soil mixture was placed in a zipper-closure bag and stored at −20°C. Before extraction, the soil was thawed overnight in a refrigerator and passed through a 4-mm sieve to remove any remaining root material. Moisture content was determined gravimetrically. Subsamples equivalent to 10 g of dry soil were wetted to 60% water holding capacity, incubated at room temperature for 27 h, and shaken for 60 min with 50 ml of 0.5 m K2SO4 on a wrist-action shaker. The extract was centrifuged, filtered, and analysed for dissolved organic C with a Dohrmann DC-180 C Analyzer (Tekmar-Dohrmann, Cincinnati, OH, USA).
Responses of A. gerardii to inoculation and fertilization were evaluated by using anova, Pearson product-moment correlations, and a statistical comparison of response curves (SAS Version 8.0; SAS Institute, Cary, NC, USA). Four-way anova (time, inoculation, N and P) was used to evaluate main effects and interactions; variables were natural log-transformed where necessary to meet model assumptions of homogeneity of variance. Response curves were compared statistically by using methods described in Potvin et al. (1990). Briefly, for each comparison, regression equations were estimated for each treatment separately and for all treatments combined. The curvilinearity of the data required the use of a second-order polynomial equation. All curves were fitted by using the nonlinear regression procedure (PROC NLIN) in SAS. For each treatment, the fit of the single line to all of the data points was compared with the fit for two separate lines representing the combined mycorrhizal and nonmycorrhizal populations. The F-values were computed by using SAS Type III residual sums of squares according to published formulae (Potvin et al., 1990).
Effects of AMF and nutrients on growth
The experimental manipulation of nutrients and AMF inoculation produced shoot P : N ratios in A. gerardii tissue ranging from 0.05 to 0.35; the ratios were generally at the lower end of this range at 45 DAP and spread widely across the range at 90 DAP. The P : N ratios were especially wide for mycorrhizal plants at 75 DAP and 90 DAP. All measures of A. gerardii growth were influenced by harvest date, and many were influenced by the nutrient treatments (Table 1). However, mycorrhizal plants were not larger overall than nonmycorrhizal plants (Fig. 1). This was further supported by no inoculation effect on total shoot (leaf) area (P < 0.43). In terms of second- and third-order interactions, inoculation × P produced the greatest number of significant effects (plant weight, shoot weight, root weight, SA and LWR), while all other interactions had significant effects on a subset of these variables (Table 1).
|Main effects and interactions||Plant wt||Shoot wt||Root wt||SA||LWR||LAR|
|Time × inoculation||0.0276|
|Time × N||0.0002||0.0001||0.0003|
|Inoculation × N||0.0660|
|Time × inoculation × N||0.0856||0.0516|
|Time × P||0.0986|
|Inoculation × P||0.0010||0.0531||0.0265||0.0083||0.0714|
|Time × inoculation × P||0.0897||0.0035||0.0001|
Mycorrhizal and nonmycorrhizal plants did not differ in biomass production or root length (Fig. 1; Table 1). However, there was a significant time × inoculation interaction for shoot dry weight, suggesting a mycorrhizal effect for mean RGR of the shoots. An inoculation effect on shoot dry weight was also found for third-order interactions time × inoculation × N and time × inoculation × P. For LWR (the proportion of plant mass allocated to shoot), the mycorrhizal effect was associated with the second- and third-order interactions of inoculation × P and time × inoculation × P. These interactions indicate that both soil P fertility and time modify the magnitude of the mycorrhizal effect on LWR. Under low P fertility, mycorrhizal plants allocated a greater proportion of their growth to shoots, whereas at higher P levels, the nonmycorrhizal plants allocated more to shoot mass.
Above- and below-ground allocation
The only significant main effect for the inoculation treatment on a growth parameter was on SLA (cm−2 g−1 leaf; anova main effect for inoculation F = 3.23, df = 1, P = 0.0735). Mycorrhizal plants had slightly greater SLA than nonmycorrhizal plants (110 cm2 g−1 compared with 104 cm2 g−1), suggesting that mycorrhizal plants produced longer, thinner leaves than nonmycorrhizal plants. The significant inoculation effect for SLA was eliminated when leaf nonstructural carbohydrate content was subtracted from leaf dry mass before calculating SLA (main effect for inoculation treatment: F = 1.02, df = 1, P < 0.3149).
Mycorrhizal and nonmycorrhizal plants did not differ in shoot total nonstructural carbohydrate (TNC) concentration (F = 0.16, df = 1, P < 0.6884). However, there was a significant second-order inoculation × time and inoculation × P interactions for shoot TNC concentration suggesting AMF reduced the accumulation of TNC in leaves (Fig. 2a,c). By contrast, we found a significant main effect with inoculation for the total amount of shoot TNC (F = 3.80, df = 1, P < 0.0534), indicating that mycorrhizal plants allocated greater amounts of nonstructural carbohydrates to below-ground sinks. Like shoot TNC concentration, the total amount of accumulated TNC also demonstrated significant second-order inoculation × time and inoculation × P interactions (Fig. 2b,d). These significant interaction effects suggest the relationship of shoot TNC with AMF are influenced not only by growth stage, but by also soil P fertility. At a higher soil P supply the mycorrhizal treatment shoots have less accumulated TNC in them than nonmycorrhizal shoots. Allocation of TNC to below-ground structures did not result in mycorrhizal plants having significantly higher root TNC concentration (F = 0.62, df = 1, P < 0.4318); there was no significant inoculation effect for second- or third-order interactions.
Although the above-ground biomass of mycorrhizal and nonmycorrhizal plants was similar (Fig. 1a), mycorrhizal plants allocated more biomass to the inflorescence and less to the vegetative growth, whereas nonmycorrhizal plants did the opposite (Fig. 3, P < 0.059 on the basis of a t-test on the ratio of vegetative dry mass to total shoot dry mass for mycorrhizal vs nonmycorrhizal plants). Furthermore, although not shown, the proportion of each of the measured nonstructural carbohydrates in reproductive tissue at 90 DAP was consistently higher in the mycorrhizal plants.
Both time and inoculation with AMF had significant effects on nutrient concentrations in the tissue of A. gerardii (Table 2). Mycorrhizal shoots had greater P concentrations than nonmycorrhizal shoots at all harvest dates (Fig. 4a), with mycorrhizal plants having an overall mean shoot P concentration of 1.62 mg g−1, compared with 1.27 mg g−1 for nonmycorrhizal plants (F = 133.01, df = 1, P < 0.0001). By contrast, shoot tissue N concentrations were greater in nonmycorrhizal plants than in mycorrhizal plants (13.18 mg/g compared with 12.28 mg g−1; F = 17.15, df = 1, P < 0.0001). Harvest time differences for shoot N were evident in mycorrhizal and nonmycorrhizal plants at 45 DAP and 60 DAP (Fig. 4b). Furthermore, shoot P and N concentrations appeared to be interrelated (Fig. 4c). When nutrition of the host is expressed by P : N ratio, the greatest proportion of the sums of squares is accounted for by the inoculum treatment (Table 3). Furthermore, the disparate trends for N and P concentration within shoot tissue resulted in a tendency for mycorrhizal shoots (mean P : N ratio = 0.16) to have a wider ratio than nonmycorrhizal shoots (mean P : N ratio = 0.11). The trends for root tissue nutrient content were similar to those reported above for shoots (Table 2).
|Main effects and interactions||Shoot N||Shoot P||Root N||Root P|
|Time × inoculation||0.0015||0.0001|
|Time × N||0.0218||0.0525|
|Inoculation × N||0.0729||0.0088|
|Time × inoculation × N|
|Time × P||0.0267|
|Inoculation × P||0.0002||0.0010|
|Time × inoculation × P||0.0033||0.0054|
|Inoculation × N × P||0.0005||0.0379|
|Main effects and interactions||df||F-value||P > F||R2|
|Time × inoculation||3||20.43||0.0001|
|Time × N||6||24.67||0.0001|
|Inoculation × N||2||6.73||0.0014|
|Time × inoculation × N||6||4.61||0.0002|
|Time × P||3||2.41||0.0675|
|Inoculation × P||1||3.47||0.0638|
|Time × inoculation × P||3||0.08||0.9705|
|N × P||2||0.76||0.4675|
|Time × N × P||6||1.56||0.1606|
|Inoculation × N × P||2||7.06||0.0011|
|Time × inoculation × N × P||6||0.66||0.6820|
Apparent photosynthesis and net C gain
The effects of shoot nutrient concentration and inoculation on Anet were evaluated with a comparison of nonlinear regression response curves (Fig. 5). For plants at equivalent tissue N, the trend was for mycorrhizal plants to have higher Anet than nonmycorrhizal plants, although the difference in these response curves was only marginally significant (Fig. 5a). The reverse was true for plants at equivalent tissue P concentrations; Anet was lower in mycorrhizal plants except at high tissue P concentrations, where the values were not different (Fig. 5b). When shoot N and P were expressed as P : N ratio, Anet was the highest at low P : N ratios (Fig. 5c). At equivalent P : N ratios, mycorrhizal plants had higher Anet than nonmycorrhizal plants. Furthermore, as shoot P : N ratio increased (N was more limiting to plant growth than P) mycorrhizal plants exhibited the largest difference in Anet when compared with their nonmycorrhizal counterparts.
The response of net C gain (Cn) to shoot P : N status in mycorrhizal and nonmycorrhizal plants was evaluated for each harvest separately, because Cn integrates Anet with plant size (Fig. 6a–d). At 45 DAP, variations in shoot P : N ratio were small, but the inoculated and nonmycorrhizal populations were beginning to separate (Fig. 6a). At 60 DAP, the positioning of the data points suggests that the mycorrhizal effect was developing further (Fig. 6b). By 75 DAP, the mycorrhizal and nonmycorrhizal plants showed significantly different Cn patterns at equivalent P : N ratios (Fig. 4c). At low P : N ratios (e.g. 0.10), mycorrhizal plants had higher Cn rates; at higher P : N ratios (e.g. 0.20), Cn rates were approximately equal. By 90 DAP, when the plants had initiated flowering, mycorrhizal and nonmycorrhizal plants had similar rates at equivalent P : N ratios, although mycorrhizal plants were able to sustain Cn because more P and N is diverted to reproductive growth sooner than in nonmycorrhizal plants (Fig. 6d).
Assessment of sink strength
The Cdiff is the difference in net C gain for mycorrhizal vs nonmycorrhizal A. gerardii at equivalent P : N ratios; positive values indicate that mycorrhizal plants have higher rates of C gain than nonmycorrhizal plants. The Cdiff varied dramatically across a range of shoot P : N ratios at 75 DAP, whereas at 90 DAP the variation in Cdiff was greatly reduced (Fig. 7). At 75 DAP, the mycorrhizal plants gained approximately 0.4 mmol h−1 more C than nonmycorrhizal plants at the lowest P : N ratio of 0.10. Cdiff values approached zero as the shoot P : N ratio reached 0.20–0.25. Thus, at low P : N ratios (so that N is not limiting relative to P), mycorrhizal plants had higher rates of C assimilation than nonmycorrhizal plants, but this response was less pronounced as N became more limiting.
The Cdiff was positively associated with total and SDH active colonization. Higher percentages of fungal colonization were related to higher Cdiff values for both colonization measures (Fig. 8); the trend was similar for total and SDH-active colonized root length (data not shown). The positive relationship of Cdiff with colonization, particularly SDH active colonization, suggests that the additional Cn in mycorrhizal plants was allocated to the mycorrhizal fungi. Because mycorrhizal and nonmycorrhizal plants were the same size at the end of the experiment, this result is indirectly supported by a positive relationship between total colonized root length and total nonstructural carbohydrates in root tissue (Pearson-product moment correlation coefficient r = 0.57, P < 0.001), which indicates that AMF can respond to increased photosynthate in below-ground structures.
Mycorrhizal colonization and extractable C
Root length increased for mycorrhizal and nonmycorrhizal plants throughout the experiment (Fig. 1b). Total colonized root length leveled off at 75 DAP and 90 DAP (Fig. 9d). The amount of SDH-active colonized root length peaked at 75 DAP and declined at 90 DAP (Fig. 9d). The peaks in metabolic activity at 75 DAP and their subsequent decline was especially evident for SDH-active arbuscules and coils (Fig. 9a,c). In addition, total and SDH-active colonization were negatively associated with tissue P : N ratio, both in terms of per cent colonization (Table 4 and Fig. 10a,b) and colonized root length (Table 4). Thus, as shoot tissue N became more limiting in relation to P concentration, the amount of AMF associated with the root system decreased. Although the relationship between total colonization and shoot P : N ratio was virtually identical at 75 DAP and 90 DAP (Fig. 10a), the relationship between SDH active colonization and shoot P : N ratio was not (Fig. 10b). At 90 DAP, SDH-active colonization was significantly lower than it had been 75 DAP across a wide range of P : N ratios, even though total colonization did not vary between these two harvest dates. This suppression of SDH-active colonization may be related to the fact that the reproductive inflorescence, which was only being initiated at 75 DAP, was fully projecting by 90 DAP; since flowering is probably a strong competing sink for C, a significant decrease in SDH-active colonization at 90 DAP is not surprising.
|Source||Pearson-product moment correlations (r)|
|45 DAP||60 DAP||75 DAP||90 DAP|
|Arbuscule colonized length||NS||−0.36**||NS||NS|
|Arbuscule colonization (%)||NS||−0.49***||−0.33*||NS|
|SDH-active arbuscule colonized length||NM||−0.35**||NS||NS|
|SDH-active arbuscule colonization (%)||NM||−0.44**||NS||NS|
|Coil colonized length||NS||NS||−0.34**||−0.30*|
|Coil colonization (%)||NS||−0.32*||−0.37**||NS|
|SDH-active coil colonized length||NM||NS||−0.34**||−0.30*|
|SDH-active coil colonization (%)||NM||−0.32*||−0.38**||NS|
|Total colonized length||NS||NS||−0.32*||−0.38**|
|Total colonization (%)||NS||−0.37**||−0.42**||−0.40**|
|SDH-active colonized length||NM||NS||−0.37**||−0.29*|
|SDH-active colonization (%)||NM||−0.41**||−0.45***||−0.29*|
The amount of extractable soil C (for this study defined as exudate C) in the rhizosphere of the plants at 60, 75 and 90 DAP decreased with time (P < 0.0001) and was consistently lower in mycorrhizal plants (P < 0.0001; Table 5). A significant interaction effect on exudation quantity was observed with P fertility, where the lower the soil P content of the soil the higher was the extractable organic C (P < 0.0005). Furthermore, although an increase in extractable C was found with nonmycorrhizal plants, mycorrhizal plants’ active and total colonization responded in a similar positive manner to the amount of extractable C in the soil (Pearson product moment r = 0.32, P < 0013, and r = 0.25, P = 0.0114, respectively).
|Source||Extractable soil C (mg g−1 root)|
|60 DAP||75 DAP||90 DAP|
|Nonmycorrhizal||83.7 ± 1.2 a||73.3 ± 1.3 a||73.3 ± 1.2 a|
|Mycorrhizal||75.1 ± 1.3 b||71.1 ± 1.3 a||66.5 ± 1.3 b|
The factorial soil treatments of this study were applied to create a wide range in plant tissue P and N concentrations. Because C, N and P nutrition are closely coupled, we hypothesized that the P : N ratio of host tissue should be an indicator of the physiological and nutritional state of the host. We also believed that this ratio approach would be useful for elucidating mycorrhizal influences on growth, photosynthesis, and nutrition. Furthermore, although the experimental manipulations were applied in a complete factorial manner, we were interested primarily in the response of A. gerardii to the mycorrhizal inoculation treatments. Associated interaction effects of inoculation with P or N fertilization were of only secondary interest with respect to host response. Moreover, host and fungus have their own growth and developmental rates that are coupled to feedbacks in the plant–fungus interaction. Thus, an important focus of this study is the use of sequential harvests to capture and quantify the dynamics of plant and fungal responses.
In general, the carbohydrate available for plant dry matter production is the difference between carbohydrate acquired via gross photosynthesis and carbohydrate that leaves the plant through respiration, exudation, senescence, herbivory and growth of symbiotic organisms (Koide & Elliot, 1989). If the shoots of two plants are of similar size and leaf area, but one plant allocates a larger fraction of its photosynthate to below-ground structures (e.g. roots and mycorrhizal fungi) than the other, the former would be expected to grow more slowly (Tinker et al., 1994). Plants can continue to grow at the same rate and maintain similar shoot size only if the plant that transfers more C to roots, has a higher fixation rate per unit leaf area or requires less C to produce or support a unit of photosynthetic area. Therefore, if mycorrhizal and nonmycorrhizal plants have similar dry matter contents, as they did in this study, the mycorrhizal A. gerardii must have had an enhanced photo-assimilation rate, a larger photosynthetic surface area, or both, to support the energy requirements of the AMF. Our study demonstrates the importance of comparing mycorrhizal and nonmycorrhizal plants at equivalent P : N ratios in attributing differences in Anet or SA directly to colonization by the AMF.
At equivalent P : N ratios, the mycorrhizal A. gerardii showed an overall higher C assimilation rate than the nonmycorrhizal population. The enhancement of Cn in mycorrhizal A. gerardii was not related to enhanced nutritional status of the plant, but rather it appeared to be a direct result of fungal colonization, a response similar to that described by Wright et al. (1998b). Evidence for this conclusion includes (1) the lack of difference in Anet between mycorrhizal and nonmycorrhizal plants at equivalent tissue N and (2) the occurrence of generally higher Anet in nonmycorrhizal plants at equivalent tissue P. The latter finding is most likely due to the higher P concentrations found in mycorrhizal plant leaves than in nonmycorrhizal plants, resulting in wider shoot P : N ratios for mycorrhizal plants. This analysis illustrates the importance of considering Cn in terms of the balance of both P and N in the plant tissue.
The elimination of the mycorrhizal response to SLA by correcting for differences in shoot TNC suggest that the observed inoculation effect on SLA may be in response to an increase in sink strength brought on by a growing mycorrhizal fungus. Direct measures of the total amount of accumulated shoot TNC support this observation. Hence, by association with AMF, mycorrhizal A. gerardii had a greater amount of assimilate available for allocation to below-ground processes than did its nonmycorrhizal counterpart. These results are in agreement with other studies, which have found significantly higher rates of Cn in mycorrhizal plants that were attributable to allocation of photosynthate to the mycorrhizal fungi (Wright et al., 1998b). However, this effect is not universal, as other studies have found that the increase in assimilation rate results from improved P status of the mycorrhizal plants (Black et al., 2000).
The rate of Cn for mycorrhizal and nonmycorrhizal plants at 75 DAP and 90 DAP showed a trend towards decreasing at higher shoot P : N ratios, similar to the trend for Anet. This observation is not surprising and suggests that overall Cn was limited by N. An interesting aspect of the relationship among Cn, shoot P : N ratio, and inoculation is that even though the associations are not well developed for the early harvests, the positioning of data points suggest that mycorrhizal and nonmycorrhizal plants were acting as different populations; on average, inoculated plants had greater Cn values than noninoculated plants. The direct relationship identified between colonization and Cn further indicates that the fungus represents a considerable sink in A. gerardii. Generally, sink strength was strongest at low P : N ratios, with little difference in Cn between the mycorrhizal and nonmycorrhizal plants at the highest P : N ratios. The relationship between AMF colonization, Cn and nutrients changes over time with host developmental stage. Hence, this study demonstrates the need for caution in interpreting the outcome of single-harvest experiments.
When the difference between Cn values for mycorrhizal and nonmycorrhizal plants is expressed as Cdiff at 75 DAP and 90 DAP, it is apparent that the sink strength of the fungi differs across the shoot P : N gradient. For example, at 75 DAP, Cdiff (as C) dropped from a little over 0.4 mol h−1 at a shoot P : N ratio of 0.10 to about 0.05 mmol h−1 at a shoot P : N ratio of 0.30. At the 90 DAP harvest, the range in Cdiff was comparatively quite narrow, ranging from essentially zero at a shoot P : N ratio of 0.10 to a response equivalent to the Cdiff at 75 DAP for a P : N ratio of 0.30. Because the amounts of shoot and root dry matter were similar for mycorrhizal and nonmycorrhizal plants, Cdiff most likely represents the additional photosynthate produced in response to the mycorrhizal condition. The reason for the apparent reduction in sink strength of AMF at 90 DAP (both in Cn and Cdiff) is unknown. However, the inflorescence, which had started to project by 75 DAP and was fully projected by 90 DAP, probably acted as a competing sink for C and reduced the sink strength of the AMF. This hypothesis is supported by the trends in total and SDH-active colonization across the P : N gradient (Figs 9 and 10). Although total colonization levels remain similar at the two harvest dates, the drop in SDH-active colonization levels between 75 and 90 DAP supports the hypothesis that the sink strength of the AMF is lessened during flowering because of the presence of a stronger competing sink. Alternatively, the reduction in SDH-active colonization may be related to the physiological aging of the hyphae (Smith & Dickson, 1991; Saito et al., 1993).
Our data suggest the observed enhancement of C gain in mycorrhizal A. gerardii is related directly to colonization by AMF. Evidence for this conclusion is the strong positive association between Cdiff and mycorrhizal colonization, especially the association with SDH-active colonization, which represents the metabolically active portion of the fungus and hence a stronger sink. Wright et al. (1998b) found that the activity of cell wall and cytoplasmic invertases and sucrose synthase in plant roots were stimulated solely in response to mycorrhizal colonization, supporting the hypothesis that AMF can directly affect C assimilation by increasing sink strength. Mycorrhizal colonization often leads to increased C allocation to the root system, and this C may be incorporated into increased root biomass, root respiration, hyphal biomass, and hyphal respiration, or it may be lost in the form of exudation from the roots (Finlay & Söderström, 1992). Although we did not quantify C allocation to all possible below-ground compartments, we did rule out an increase in root biomass with mycorrhizal colonization.
In this study, both mycorrhizal colonization and high soil P were related to reduced levels of extractable soil C. This finding suggests that nonmycorrhizal plants may compensate for a lack of colonization by enhancing the release of organic exudates into the rhizosphere as a means for acquiring P (Graham et al., 1981; Díaz et al., 1993). An alternative explanation is that the C sink provided by the fungi precluded copious exudation of C into the rhizosphere of mycorrhizal plants. Although the quantity of extractable soil C was less for mycorrhizal plants, total colonized root lengths were positively associated with total nonstructural carbohydrates in the root (Graham et al., 1981, 1996, 1997). This observation is consistent with the alternative hypothesis above, that because AMF are a strong sink for C allocated to the root, their presence reduces C exudation to the rhizosphere.
In summary, we found that mycorrhizal plants had a higher rate of assimilation and higher net C gain than nonmycorrhizal plants. We further determined that this effect was not an indirect result of enhanced plant nutrition, but it appears to be a direct result of mycorrhizal colonization in A. gerardii roots. Although our lines of evidence that the allocation of this extra C gain goes directly to the fungi are indirect, we believe that, taken together, they are compelling. First, mycorrhizal plants had higher C gain at equivalent P : N ratios than did nonmycorrhizal plants, even though plant dry mass was not affected by the presence or absence of AMF. Second, we ruled out the possibility that this C was allocated to greater root biomass, as the root systems of mycorrhizal plants were not significantly larger. Third, the proportion of SDH-active fungi in the roots was positively related to Cn for the mycorrhizal plants, and the proportion of SDH-active fungi in roots declined when the inflorescence of the plants began to project, suggesting the presence of a competing C sink. Finally, although extractable soil C was positively related to colonized root length for mycorrhizal plants, it was lower overall in the rhizosphere of mycorrhizal plants than in nonmycorrhizal plants, suggesting that some portion of the C allocated to roots is being retained by fungal tissue.
This investigation demonstrates that the mycorrhizal symbiosis represents a series of complex feedbacks between host and symbiont, based on physiology and nutrition. Rather than viewing the mycorrhizal symbiosis in terms of P nutrition, our data demonstrate the need to view P in relation to its association with N and C. Experiments on C dynamics usually compare matched mycorrhizal and nonmycorrhizal plants by adding P in order to achieve a plant with similar biomass and P content (e.g. Syvertsen & Graham, 1990). For this study, we examined response curves against shoot P : N ratios as a means of determining the amount of assimilate available for allocation to below-ground sinks. We believe that, at least under the conditions of this study, the use of response curves allowed for evaluation of the effects of the AMF on host nutrient and C assimilation. To further support the hypothesis that plants respond directly to the increased sink strength of mycorrhizal fungi, it would be desirable to measure directly the fluxes of C from leaves of mycorrhizal and nonmycorrhizal plants at equivalent tissue P : N ratios.
We would like to thank Gail Wilson for supplying the inoculum and K. Barrios, E. Cates, G. Eckert, D. Gardner and J. Schultz for their contributions during the course of the study. We also would like to thank the two anonymous reviewers for thoughtful comments that greatly improved the manuscript. This research was supported by the Program for Ecosystem Research of the US Department of Energy, Office of Science, Office of Biological and Environmental Research, Global Change Program under contract W-31-109-Eng-38.
- 1984. The effects of phosphorus on the formation of hyphae in soil by the vesicular–arbuscular mycorrhizal fungus, Glomus fasciculatum. New Phytologist 97: 437–446. , , .
- 1957. Colorimetric analysis of sugars. Methods in Enzymology 3: 73–105. .
- 1975. Comparative carbohydrate physiology of ecto- and endomycorrhizas. In: SandersFE, MosseB, TinkerPB, eds. Endomycorrhizas. London: Academic Press, 149–175. , , .
- 2000. Effect of mycorrhizal-enhanced leaf phosphate status on carbon partitioning, translocation and photosynthesis in cucumber. Plant, Cell & Environment 23: 797–809. , , .
- 1945. Determination of total, organic and available forms of phosphorous in soil. Soil Science 59: 39–45. , .
- 1988. The Glycine–Glomus–Rhizobium symbiosis. VII. Photosynthetic nutrient-use efficiency in nodulated, mycorrhizal soybeans. Plant Physiology 86: 1292–1297. , .
- 1993. Evidence of a feedback mechanism limiting plant response to elevated carbon dioxide. Nature 364: 616–617. , , , .
- 1988. Carbon cost of the fungal symbiont relative to net leaf P accumulation in a split-root VA mycorrhizal symbiosis. Plant Physiology 86: 491–496. , , .
- 1993. Carbon economy of sour orange in relation to mycorrhizal colonization and phosphorus status. Annals of Botany 71: 1–10. , , , .
- 1992. Mycorrhiza and carbon flow to the soil. In: AllenMF, ed. Mycorrhizal functioning: an integrative plant–fungal process. New York, NY, USA: Chapman & Hall, 134–160. , .
- 1991. Costs and benefits of mycorrhizas: implications for functioning under natural conditions. Experientia 47: 350–355. .
- 2000. Assessing the cost of arbuscular mycorrhizal symbiosis in agroecosystems. In: PodilaGK, DoudsDD, eds. Current advances in mycorrhizal research. St Paul, MN, USA: The American Phytopathological Society, 127–140. .
- 1994. Host genotype and the formation and function of VA Mycorrhizae. Plant and Soil 159: 179–185. , .
- 1981. Membrane-mediated decrease in root exudation responsible for phosphorus inhibition of vesicular–arbuscular mycorrhiza formation. Plant Physiology 68: 548–552. , , .
- 1996. Carbon economy of sour orange in response to different Glomus spp. Tree Physiology 16: 1023–1029. , , .
- 1997. Carbohydrate allocation patterns in citrus genotypes as affected by phosphorus nutrition, mycorrhizal colonization, and mycorrhizal dependency. New Phytologist 1356: 335–343. , , .
- 1968. Direct micro-determination of sucrose. Analytical Chemistry 22: 280–283. .
- 1985. Carbon economy of soybean–Rhizobium–Glomus associations. New Phytologist 101: 427–440. , , .
- 1989. Automated quantification of roots using a simple image analysis analyzer. Agronomy Journal 81: 935–938. , .
- 1983. The effect of nitrate and phosphate on the vesicular–arbuscular mycorrhizal infection of lettuce. New Phytologist 92: 389–399. .
- 1990. Carbon flow into soil and external hyphae from roots of mycorrhizal cucumber plants. New Phytologist 115: 77–83. , .
- 1984. Phosphorus nutrition on mycorrhizal colonization, photosynthesis, growth and nutrient composition of Citrus auriantum. Plant and Soil 80: 35–42. .
- 1982. Effect of flower bud development in chrysanthemum on vesicular-arbuscular mycorrhiza formation. New Phytologist 90: 671–675. , , , .
- 1984. Photosynthate partitioning in split-root citrus seedlings with mycorrhizal and nonmycorrhizal root systems. Plant Physiology 75: 26–30. , .
- 1989. Cost, benefit and efficiency of the vesicular–arbuscular mycorrhizal symbiosis. Functional Ecology 3: 252–255. , .
- 1982. Carbon flow photosynthesis and N2 fixation in mycorrhizal and nodulated fava beans (Vicia fabia L.). Soil Biology and Biochemistry 14: 407–412. , .
- 1979. Incorporation of 14C-labelled substrates by uninfected and VA mycorrhizal roots of onion. New Phytologist 83: 415–431. , .
- 1996. Effects of mineral nutritional status on shoot–root partitioning of photoassimilates and cycling of mineral nutrients. Journal of Experimental Botany 47: 1255–1263. , , .
- 1990. A new method which gives an objective measure of colonization of roots by vesicular–arbuscular mycorrhizal fungi. New Phytologist 115: 495–501. , , , , .
- 1995. Growth of extraradical hyphae of vesicular–arbuscular mycorrhizal fungi in two temperate grassland communities. Oecologia 103: 17–23. , , .
- 1999. Pool sizes of fructans in roots and leaves of mycorrhizal and nonmycorrhizal barley. New Phytologist 142: 551–559. , , , , , .
- 1962. A modified single solution method of the determination of phosphate in natural waters. Analytica Chimica Acta 27: 31–36. , .
- 1980. Effects of vesicular–arbuscular mycorrhiza on 14C and 15N distribution in nodulated faba beans. Canadian Journal of Soil Science 60: 241–250. , .
- 1993. Growth depression in mycorrhizal citrus at high-phosphorus supply: analysis of carbon costs. Plant Physiology 101: 1063–1071. , , , , .
- 1990. The statistical analysis of ecophysiological response curves obtained from measurements involving repeated measures. Ecology 71: 1398–1400. , , .
- 1994. Influence of phosphorus supply and light intensity on mycorrhizal response in Pisum–Rhizobium–Glomus symbiosis. Experientia 50: 890–896. , , , .
- 1949. A photometric method for the determination of insulin in plasma and urine. Journal of Biological Chemistry 178: 839–845. , , .
- 1991. The effects of phosphorus fertilization and mycorrhizal development on phosphorus nutrition and carbon balance of loblolly pine. New Phytologist 117: 319–326. , .
- 1993. Succinate dehydrogenase activity of external and internal hyphae of a vesicular–arbuscular mycorrhizal fungus, Glomus mosseae (Nicol. & Gerd.) Gerdmann and Trappe, during mycorrhizal colonization of roots of leek (Allium porrum L.), as revealed by in situ histochemical staining. Mycorrhiza 4: 59–62. , , .
- 1993. Modifications to clearing methods used in combination with vital staining of roots colonized with vesicular–arbuscular mycorrhizal fungi. Mycorrhiza 4: 29–35. , .
- 2001. Evidence of a mycorrhizal mechanism for the adaptation of Andropogon gerardii to high and low-nutrient prairies. American Journal of Botany 88: 1650–1656. , , , , .
- 1991. Regulation of nutrient transfer between host and fungus in vesicular–arbuscular mycorrhizas. New Phytologist 117: 387–398. , , .
- 1981. Removing and analyzing total nonstructural carbohydrates from plant tissue. Publication R-2107. Madison, WI, USA: University of Wisconsin. College of Agriculture and Life Science. .
- 1996. Mutualism and parasitism: diversity in function and structure in the ‘arbuscular’ (VA) mycorrhizal symbiosis. Advances in Botanical Research 22: 1–43. , .
- 1991. Quantification of active vesicular–arbuscular mycorrhizal infection using image analysis and other techniques. Australian Journal of Plant Physiology 18: 637–648. , .
- 1997. Mycorrhizal symbiosis. New York, New York, USA: Academic Press. , .
- 1986. Effects of mycorrhizal infection on plant growth, nitrogen and phosphorus nutrition in glasshouse-grown Allium cepa L. New Phytologist 103: 359–373. , , , .
- 1982. The distribution of carbon and the demand of the fungal symbiont in leek plants with vesicular–arbuscular mycorrhizas. New Phytologist 92: 75–87. , , , .
- 1998. Does elevated atmospheric carbon dioxide affect arbuscular mycorrhizas? Trends in Ecology and Evolution 13: 455–458. , .
- 1990. Nitrogen affects the phosphorus response of VA mycorrhiza. New Phytologist 115: 303–310. , .
- 1990. Influence of vesicular arbuscular mycorrhizae and leaf age on net gas exchange of Citrus leaves. Plant Physiology 94: 1424–1428. , .
- 1986. Effects of phosphorus on the formation of mycorrhizas by Gigaspora calospora and Glomus fasciculatum in relation to root carbohydrates. New Phytologist 103: 751–765. , , .
- 1994. Carbon use efficiency in mycorrhizas: theory and sample calculations. New Phytologist 128: 115–122. , , .
- 1998b. Mycorrhizal sink strength influences whole plant carbon balance of Trifolium repens L. Plant, Cell & Environment 21: 881–891. , , .
- 1998a. Effects of VA mycorrhizal colonization on photosynthesis and biomass production of Trifolium repens L. Plant, Cell & Environment 21: 209–216. , , .