Abstract

Key Words

Ectomycorrhizas, oxalate, phosphatase activity, phosphate rock

Introduction

Ectomycorrhizal (ECM) fungi inoculation of Pinus radiata D. Don seedlings has been considered to improve phosphorus (P) uptake and growth of the seedlings in low fertility soils (Chu-Chou and Grace 1985). Our previous studies have shown that ECM hyphae, like roots, induce soil acidification and promote production of phosphatase enzyme and oxalate anions, causing changes in soil P availability (Liu et al. 2004a). There may be differences in the effectiveness of ECM fungal species or isolates in inducing changes in rhizosphere properties. However, no detailed study on this aspect has been reported previously on P. radiata, although relevant studies have reported on some other conifers, e.g. Pinus sylvestris (Wallander 2000).

The aims of this study were: (i) to compare the efficiency of Rhizopogon rubescens Tul. and Suillus luteus (L. ex. Fr.) S. F. Gray with that of ECM in natural soil in promoting growth and P uptake of P. radiata seedlings treated with and without a phosphate rock (PR) fertiliser; and (ii) to determine the effect of these inoculations on the rhizosphere properties, particularly the P chemistry.

Materials and methods

Experimental design

A factorial experiment with 2 levels of PR fertiliser treatments and 5 soil treatments was conducted on P. radiata seedlings in pots under glasshouse conditions. The two PR fertilisation levels were: P-unfertilised soil and P-fertilised soil (100 µg P g^-1 soil applied as Ben Guire phosphate rock (BGPR); particle size 150 – 250 µm; total P 13.1%; and citric acid extractable P 2.75%). The soil treatments were: (1) autoclaved soil; (2) soil treated with fungicide; (3) natural forest soil; (4) S. luteus inoculated soil (soil inoculated with S. luteus spores); and (5) R. rubescens inoculated soil (soil inoculated with R. rubescens spores). Each treatment was replicated five times, giving a total of 50 experimental pots.

Soil preparation

The soil material used in this study was from the 0 - 15 cm horizon of an Orthic Allophanic Soil (Andosol) under a second rotation P. radiata plantation. The soil had total P 205µg g^-1 soil, Olsen P 0.9 µg g^-1 soil, P retention 90%, pH (1:2.5 soil:H₂O) 5.8 and CEC (mol_ckg^-1 soil) 0.14. The autoclaved soil was prepared by autoclaving the air-dried soil for 30 min at 120ºC temperature and 15 lb in² pressure using a steam autoclave. The fungicidal treatment consisted of soils leached with a dilute solution of commercial Terrazole 35WP at the rate of 0.05 ml fungicide per pot containing 0.5 kg soil (fungicide applied after soils were potted). The soils for the P-fertilised treatments were prepared by thoroughly mixing BGPR with the soil at a rate of 100 µg P g^-1 soil. The nitrogen (N) and potassium (K) were applied to all treatments as solutions of NH₄NO₃ and KCl at a rate of 100 g N or K kg^-1 soil. After 5 months of growth the seedlings received another addition of N and K at the same rate.

Seedling inoculation

Two-week-old seedlings of P. radiata , pre-germinated in perlite medium, were transplanted into pots containing 0.5 kg air-dried soils packed to a bulk density of 0.85 g cm^-3, at the rate of 5 seedlings per pot. The surface of the soil was covered with two filter papers (Newman No.31) to minimise cross infection by air-borne inocula.

Sporocarps of R. rubescens. and S. luteus were collected from P. radiata forests and spore suspensions were prepared in sterile water in the laboratory. The spore concentration in the suspension was diluted to 3 - 4 × 10⁷ spores ml^-1. Each seedling was inoculated with 5 ml of the diluted spore suspension by pipetting the suspension into holes made to 40 mm depth in the soil near the seedling stem base. Seedlings were inoculated again 4 weeks after transplanting to ensure that inoculation was successful. Uninoculated seedlings, namely the seedlings in the autoclaved, fungicide and natural soil treatments, also received the same amount of autoclaved inoculum suspensions.

Trial management and harvesting of plants

Treatments were arranged in a randomised complete design inside a glasshouse maintained at 16ºC during nights and 28ºC during days. The pots were watered using distilled water to “pot field capacity” at 2-day intervals by weighing the pots. The direction of the bench on which the pots were arranged was changed weekly so that the seedlings in all treatments received equal amounts of sunlight.

Seedlings were harvested after 10 months growth. To enable sampling of soil solutions at approximately constant soil moisture content, forty-eight hours before harvesting of the seedlings the soil moisture in all pots was adjusted to “pot field capacity”. The seedling height was recorded as the mean of the heights of five seedlings in each pot. The total root length and ECM root tip numbers were calculated from the root length and ECM tips measured in a sub-sample of 5 randomly selected lateral roots per pot (1 lateral root per seedling). The root length was measured using a gridline technique described by Newman (1966). The number of ECM root tips (forked fine roots) was counted manually.

Soil and soil solution sampling

At plant harvest, the top 10 mm root-free soil was discarded to minimise the influence of the atmosphere on the soils. The rhizosphere soil was collected as described by Wang and Zabowski (1998). The whole seedling was removed from each pot with minimum injury to its roots, by shaking the roots until the soil not tightly adhering to the roots was removed and then collecting the soil closely adhering to the root system (rhizosphere soil) by vigorously shaking the roots. The “bulk” soil (soil not influenced by roots) was collected from areas in the pots where there were no roots.

Soil solutions were sampled from two of the five replicates of each treatment as the subsequent analyses of these solutions were expensive. Solutions were removed from a portion of moist soil within 6 h of sample collection, using a double-centrifuge-tubes technique (centrifuged for 0.5 h at 10000 rpm and 4ºC; filtered immediately through a 0.45 µm millipore filter) (Wang and Zabowski 1998). The solutions were stored frozen until they were analysed for oxalate and phosphate concentrations.

Mycorrhiza identification and ECM hyphal length measurement

The ECM fungal species which infected the roots were identified by visibly examining the fine root characteristics as described by Chu-Chou and Grace (1983). The ECM hyphal length density in the soils was measured using a simplified agar film technique described by Nicholas and Parkinson (1967).

Plant, soil and solution analyses

The herbage shoot total N and P contents (Kjeldahl digestion; Blackmore et al. 1987), soil pH (1:2.5 of soil:H₂O; Blackmore et al. 1987), acid (pH 6.5) and alkaline (pH 11) phosphatase enzyme activities (Tabatabai 1994) and soil P fractions (Hedley et al. 1994) were determined as described in the relevant methods. The oxalate concentrations in the soil solutions were determined using a Waters ion chromatographic system at Landcare Ltd. Palmerston North, New Zealand.

Statistical analyses of data

Significant differences in seedling parameters between the treatments were tested using a two-way ANOVA procedure. Differences in soil properties between treatments were analysed according to two factorial design (2 soils - bulk and rhizosphere soils × 5 soil treatments) and three factorial design (2 P fertilisation levels × 2 soils × 5 soil treatments). Significant differences between means were established by calculating LSDs at a significance level of P < 0.05.

Results and discussion

ECM infection and its effects on P uptake by seedlings

ECM infection is associated with swollen short root tips. The majority of mycorrhizae formed in the treatments inoculated with R. rubescens and S. luteus were characteristic of these fungi as described by Chu-Chou and Grace (1983). The seeding roots in natural and fungicide treated soils had a mixture of ECM fungi. The main species present were R. rubescens and Laccaria laccata. The two ECM inoculated treatments and the natural soil had higher ECM tip density than fungicide treated soil and the latter had higher ECM tip density than autoclaved soil (Table 1). There was no significant difference in hyphal density between inoculation treatments.

The BGPR-P application resulted in significant (P < 0.05) increase in ECM tip and hyphal densities (Table 1). The ECM tip densities in P-fertilised treatments were 3 to 5-fold higher than those in P-unfertilised treatments. This indicates that in extremely P deficient soils, such as the one used in this study, an addition of BGPR at the P rate of 100 µg g^-1 soil stimulated the infection of ECM fungi in newly germinated P. radiata seedlings. These results are consistent with the findings of Lamb and Richards (1974). However, we have found that higher rate of P application may also sometimes decrease the ECM tip density (Liu et al. 2004b).

In the P-fertilised soils the P uptake and shoot dry matter weight were significantly higher for the natural soil and the ECM inoculated treatments compared to the autoclaved and fungicide treatments, but there was no difference between the natural and the two ECM inoculated treatments. The P concentrations in shoots and roots of the seedlings were significantly correlated with ECM tip density in P-fertilised soil (Figure 1), but no such relationship was observed in unfertilised soil probably due to the lower degree of ECM infection in P-unfertilised treatments.

Soil pH

In both P-unfertilised and P-fertilised soils, soil pH in autoclaved soils was significantly (P < 0.05) higher than in the other treatments (Table 2), probably due to excretion of protons by the mycorrhizal roots of P. radiata in the non-autoclaved soils. However, there was no significant difference in pH between the soils inoculated with the two ECM species.

Figure 1. Relationship between ECM root tip density and P concentration in roots (A) and shoots (B) of P. radiata seedlings.

Table 1. Effects of P fertilisation and soil treatment (autoclaved, fungicide treated, natural untreated, and ECM inoculated) on ECM infection of roots and P uptake by P. radiata (numbers associated with same letters are not significantly different at P < 0.05; upper case letters after numbers in columns– difference between BGPR-P fertilised and unfertilised soils; lower case letters after numbers in rows – difference between soil treatments)

Seedling parameters		Autoclaved	Fungicide treated	Natural untreated	S.luteus Inoculated	R. rubescens inoculated	Diff. at P <0.05
ECM Tip density (tip m-1 root)	P unfertilised	15 ± 1.7 bB	57 ± 6 bB	141 ± 20 aB	109 ± 20 aB	126 ± 20 aB	P < 0.0001 LSD(soil) =29
	P fertilised	99 ± 29 dA	256 ± 20 cA	350 ± 36 bA	349 ± 35 bA	422 ± 16 aA
	P < 0.0001; LSD(inoc.) = 46
Hyphal Density in rhizosphere soil (m g-1 soil)	P unfertilised	22±3 B	27±10 B	35±10 B	20±12 B	34±17 B	P < 0.0001 LSD(soil) =6
	P fertilised	34±3 A	70±14 A	36±14 A	64±14 A	42±14 A
	NS (inoc.)
P conc in shoots (mg g-1 DM)	P unfertilised	0.24 ± 0.01 bB	0.34 ± 0.06 aB	0.36 ± 0.04 aB	0.29 ± 0.03 bB	0.29 ± 0.04 bB	P < 0.0001 LSD(soil) =0.06
	P fertilised	0.45 ± 0.09 bA	0.70 ± 0.03 aA	0.70 ± 0.07 aA	0.78 ± 0.04 aA	0.71 ± 0.04 aA
	P < 0.001; LSD(inoc.) = 0.1
P conc in roots (mg g-1 DM)	P unfertilised	0.36 ± 0.03 bB	0.50 ± 0.03 aB	0.51 ± 0.06 aB	0.41 ± 0.03 aB	0.43 ± 0.05 aB	P < 0.0001 LSD(soil) =0.07
	P fertilised	0.61 ± 0.05 bA	0.91 ± 0.04 aA	0.86 ± 0.04 aA	0.90 ± 0.10 aA	0.80 ± 0.05 aA
	P < 0.001; LSD(inoc.) = 0.1
P uptake in shoots (mg pot-1)	P unfertilised	0.5 ± 0.04 aB	0.6 ± 0.10 aB	0.8 ± 0.14 aB	0.6 ± 0.13 aB	0.6 ± 0.06 aB	P < 0.0001 LSD(soil) =0.55
	P fertilised	2.4 ± 0.89 cA	5.8 ± 0.43 bA	6.5 ± 0.82 aA	6.9 ± 0.37 aA	6.8 ± 0.38 aA
	P < 0.0001; LSD(inoc.) = 0.88
Shoots DM (g)	P unfertilised	1.9 ± 0.02 aB	1.9 ± 0.02 aB	2.3± 0.04 aB	2.1 ± 0.05 aB	2.2 ± 0.04 aA	P < 0.0001 LSD(soil) = 0.5
	P fertilised	4.9 ± 0.18 cA	8.2 ± 0.07 bA	9.5 ± 0.08 aA	8.9 ± 0.11 bA	9.5 ± 0.04 aB
	P < 0.0001; LSD(inoc.) = 0.7

Soil pH in P-fertilised rhizosphere soils was significantly (P < 0.05) lower than that in the corresponding bulk soils for all inoculation treatments (Table 2B). However, in P-unfertilised soils which had significantly lower ECM infection, pH was significantly (P < 0.05) higher in rhizosphere soil than in bulk soil, particularly for ECM inoculation treatments (Table 2A). This contrasting observation in pH between P-fertilised and P-unfertilised soils may be associated with the form of N taken up by P. radiata seedlings under varying degrees of ECM infection. Olykan and Adams (1995) reported that ECM infected P. radiata seedlings take up N predominantly in the cationic NH₄⁺ form than in the anionic NO₃^- form from soil. Skinner (1978) found that non-mycorrhizal P. radiata seedlings grown in solution culture preferred to take up NO₃^- -N rather than NH₄⁺-N. This difference in N uptake form between mycorrhizal and non-mycorrhizal seedlings suggests that the degree of mycorrhizal infection on pine roots may have a strong influence on changes in rhizosphere pH (Figure 2).

Table 2. Effects of P fertilisation and soil treatments (autoclaved, fungicide treated, natural untreated and ECM inoculated) on soil properties. Numbers associated with same letters are not significantly different at P < 0.05; upper case letters after numbers in columns – difference between bulk and rhizosphere soils; lower case letters after numbers in rows – differences between soil treatments

A. BGPR-P unfertilised treatments
Soil properties	Treatment	Autoclaved	Fungicide treated	Natural untreated	S.luteus inoculated	R. rubescens inoculated
Soil pH (1:2.5 H2O)	Bulk soil	5.27±0.01aA	4.85±0.03 cB	4.99±0.02 bB	4.93±0.05 bB	4.95±0.03 bB	P < 0.0001 LSD(soil) =0.07
	Rhizosphere	5.31±0.08 aA	5.07±0.04 bA	5.09±0.05 bA	5.17±0.01 bA	5.12±0.07 bA
	P < 0.0001; LSD(inoc.) = 0.11
Acid P-tase (µg g-1 h-1)	Bulk soil	498±39 d	701±46 c	887±38 a	788±33 b	862±19 ab	NS (soil)
	Rhizosphere	561±51 c	766±33 b	855±66 a	850±43 a	931±15 a
	P < 0.0001; LSD(inoc.) = 82
Alkaline P-tase (µg g-1 h-1)	Bulk soil	115±26 b	253±20 a	284±38 a	292±44 a	241±20 a	NS (soil)
	Rhizosphere	147±15 b	226±27 a	273±46 a	277±42 a	278±34 a
	P < 0.001; LSD(inoc.) = 68
Oxalate conc. (mg L-1)	Bulk soil	0.40±0.00 c	0.85±0.25 b	0.30±0.00 c	0.51±0.00 c	1.35±0.35 a	NS (soil)
	Rhizosphere	0.75±0.05 b	0.45±0.05 c	0.35±0.05 c	0.45±0.05 c	1.65±0.25a
	P < 0.001; LSD(inoc.) =0.36
Solution P (mg L-1)	Bulk soil	0.01	0.02	0.02	0.01	0.01	NS (soil)
	Rhizosphere	0.02	0.01	0.01	0.01	0.01
	NS (inoc.)
Resin P (µg g-1)	Bulk soil	0.44±0.1	0.32±0.1	0.76±0.1	0.59±0.2	0.60±0.1	NS (soil)
	Rhizosphere	0.44±0.1	0.58±0.1	0.62±0.1	0.55±0.1	0.68±0.3
	NS (inoc.)
B. BGPR-P fertilised treatments
Soil properties	Treatment	Autoclaved	Fungicide treated	Natural untreated	S.luteus inoculated	R. rubescens inoculated
Soil pH (1:2.5 H2O)	Bulk soil	5.57±0.02 aA	5.56±0.03 aA	5.50±0.02 bA	5.52±0.02 bA	5.54±0.01 aA	P < 0.0001 LSD(soil) =0.03
	Rhizosphere	5.45±0.00 aB	5.36±0.03 bB	5.33±0.02 bB	5.41±0.02 aB	5.36±0.02 bB
	P < 0.01; LSD(inoc.) = 0.05
Acid P-tase (µg g-1 h-1)	Bulk soil	245±68cA	551±57bA	698±80aA	618±40abB	587±61bB	P < 0.05 LSD(soil) =83
	Rhizosphere	321±97cA	629±65bA	763±52aA	748±78aA	724±51aA
	P < 0.0001; LSD(inoc.) = 132
Alkaline P-tase (µg g-1 h-1)	Bulk soil	95±12bA	166±15abB	189±29aA	185±20aA	135±18bB	P < 0.05 LSD(soil) =30
	Rhizosphere	100±12bA	234±12aA	206±33aA	207±25aA	216±33aA
	P < 0.001; LSD(inoc.) = 48
Oxalate conc. (mg L-1)	Bulk soil	0.40±0.10a A	0.25±0.05 B	0.45±0.15 A	0.15±0.05 B	0.35±0.25 B	P < 0.05 LSD(soil) =0.28
	Rhizosphere	0.40±0.00 A	0.80±0.50 A	0.25±0.05 A	1.00±0.20 A	1.10±0.10 A
	NS (inoc.)
Solution P (mg L-1)	Bulk soil	0.02 A	0.01 B	0.02 A	0.02 B	0.02 B	P < 0.01 LSD(soil) =0.005
	Rhizosphere	0.02 A	0.02 A	0.02 A	0.04 A	0.03 A
	NS (inoc.)
Resin P (µg g-1)	Bulk soil	0.60±0.1 A	1.20±0.3 B	1.03±0.3 B	0.79±0.1 B	1.15±0.1 A	P < 0.05 LSD(soil) =0.5
	Rhizosphere	1.09±0.4 A	2.08±0.7 A	1.59±0.3 A	1.69±0.4 A	1.24±0.3 A
	NS (inoc.)

Figure 2. Relationship between ECM root tip density and the difference in soil pH between the rhizosphere and bulk soils. P-fertilised and P-unfertilised treatments not distinguished in the figure.

Acid and alkaline phosphatase activities

The autoclaved soils had significantly lower phosphatase activities than natural or ECM inoculated soils, probably due to the higher ECM activity in these latter treatments (Table 2).

In P-unfertilised treatments root processes did not result in any significant difference in acid and alkaline phosphatase activities between the bulk and rhizosphere soils (Table 2A). This is probably due to low ECM infection of the roots. However, in the P-fertilised treatments higher acid and alkaline phosphatase activities were generally observed in the rhizosphere soils than in the bulk soils for the ECM inoculated treatments (Table 2B). The higher phosphatase activities in the rhizosphere soil were associated with higher ECM tips and hyphal length densities in soils (Table 1), suggesting that ECM fungi may play an important role in increasing phosphatase production. This suggestion is also supported by much higher acid and alkaline phosphatase activities in non-autoclaved soils in both P-unfertilised and P-fertilised treatments (Table 2A and 2B). However, it should be noted that the increases in ECM tip density and hyphal density in P-fertilised treatments (Table 1) did not result in higher phosphatase activities in these treatments compared to P-unfertilised treatments. Conversely, P application resulted in significant (P < 0.05) decreases in both acid and alkaline phosphatase activities for overall data (Table 2A and 2B). The decreases in phosphatase activities with P application are consistent with the suggestion that synthesis of phosphatases can be induced when plants and mycorrhizal fungi are deficient in P (Hedley et al., 1982).

Oxalate exudation by roots

The concentrations of oxalate in soil solution were highly variable in the different treatments. This is partly because some of the oxalate excreted by the roots were decomposed in the soil or was adsorbed to the soil. Nevertheless, the overall data (bulk and rhizosphere soil solution) of the two replicates analysed showed that a significantly (P < 0.05) higher oxalate concentration was measured in P-unfertilised soils compared to P-fertilised soils (Figure 3). Soil P stress in P-unfertilised soils may have triggered oxalate exudation from pine roots or ECM fungi as in the case of phosphatase enzyme release. Lipton et al. (1987) found that the rate of citrate release from VAM infected alfalfa roots increased by 182% when P concentration in the nutrient growth solution decreased from 0.1 mM to 0.01 mM.

Amongst the soil treatments, in general, the oxalate concentrations were found to be higher in the R. rubescens inoculated treatment than in the others (Figure 3), although significant difference was only found in P-unfertilised rhizosphere soils (Table 2). In P deficient soils, the higher oxalate production is expected to increase release of P from P-fixing soil minerals, thereby increasing plant P availability (Fox and Comerford, 1992).

Figure 3. Concentration of oxalate in soil solution (line bars show 1SE)

Significantly higher oxalate concentrations were measured in rhizosphere soil solutions from P-fertilised fungicide and S. luteus and R. rubescens inoculated treatments compared to the bulk soils (Table 2B). However, in P-unfertilised soils there was no difference in oxalate concentration between rhizosphere and bulk soil solutions (Table 2A), probably due to the relatively large variation within treatments and lower ECM activity.

BGPR dissolution and soil P fractions

Application of BGPR significantly (P < 0.05) increased P concentration in soil solution (1.3 fold) and all P fractions (except for residual-P) in the soils (data not shown). This is due to high BGPR dissolution in soil (dissolution of 57 – 69%).

Soil treatments did not cause any change in the concentration of soil P fractions except for significant (P < 0.05) decreases in H₂SO₄-P concentration in non-autoclaved treatment compared to the autoclaved treatment in P-fertilised soils (data not shown). The significantly lower H₂SO₄-P concentrations in non-autoclaved treatments were associated with lower soil pHs and this resulted in significant increase in BGPR dissolution.

Possibly due to the smaller root biomass production in the P-unfertilised treatments (data not shown), root processes did not cause any statistically significant differences in concentration of P fractions between the bulk and rhizosphere soils. In the P-fertilised treatments, however, significantly higher solution P and resin-P concentrations were found for overall data in the rhizosphere soils than in the bulk soils (Table 2B). The increased plant-available P concentrations in the rhizosphere soils were generally associated with the higher hyphal length and root tip densities, acid and alkaline phosphatase activities and oxalate concentration in the rhizosphere soils than in the bulk soils (Table 2B). These root-induced changes in the rhizosphere soils may have produced favorable conditions for organic P (P_o) conversion to inorganic P (P_i) and also remediated P fixation by allophane and hydrous oxides of Fe and Al compounds in soil (Fox and Comerford, 1992). Nevertheless, it should be noted that the techniques employed in this study for sampling rhizosphere soil (collecting rhizosphere soil by shaking the roots) may have resulted in overestimation of the concentration of plant-available P, because the P in broken ECM tips and mycelium is likely to be released into rhizosphere soil/soil solution.

Conclusions

In the P-fertilised soils the P uptake and shoot dry matter weight were significantly higher for the natural soil and the ECM inoculated treatments compared to the autoclaved and fungicide treatments, but there was no difference between the natural and the two ECM inoculated treatments. Mycorrhizal infection of seedlings grown in natural or ECM inoculated soils, compared with seedlings grown in autoclaved soils, significantly increased phosphatase activities and oxalate concentration, but decreased soil pH. In comparison with the various ECM fungal types, seedlings inoculated with R. rubescens produced relatively higher oxalate concentrations in soil solution (particularly in P-unfertilised soils) and ECM root tip density. No significant differences in concentration of soil P fractions between rhizosphere and bulk soils were found. In P-fertilised treatments, however, significantly higher concentrations of solution P and resin-P were observed in the rhizosphere soils than in the bulk soils. The increased plant-available P concentrations in the rhizosphere soils were associated with higher hyphal length density, acid and alkaline phosphatase activities and oxalate concentration in the rhizosphere than bulk soils.

Acknowledgments

We wish to thank New Zealand Forest Research (Institute Ltd.), Rotorua for providing part financial support through the Centre for Sustainable Forest Management for this study, and Landcare Research New Zealand Ltd. Palmerston North for the analysis of the oxalate concentration of the samples.

References

Blakemore LC, Searle PL, Day BK (1987) ‘Methods for chemical analysis of soils.’ New Zealand Soil Bureau Scientific Report 80. (Wellington, New Zealand)

Chu-Chou M, LJ Grace (1985) Comparative efficiency of the mycorrhizal fungi Laccaria laccata, Hebeloma crustuliniforme and Rhizospogen species on growth of radiata pine seedlings. New Zealand Journal of Botany 23, 417-424.

Chu-Chou M, LJ Grace (1983) Characterization and identification of mycorrhizas of radiata pine in New Zealand. Australia Forestry Research 13, 121–132.

Fox TR, Comerford NB (1992) Influence of oxalate loading on phosphorus and aluminum solubility in Spodsols. Soil Science Society of America Journal 56, 290–294.

Hedley MJ, Steward JWB, Chauhan BS (1982) Changes in organic soil phosphorus fractions induced by cultivation practices and lab incubations. Soil Science Society of America Journal 46, 970–976.

Hedley MJ, Kirk GJD, Santos MB (1994) Phosphorus efficiency and the forms of soil phosphorus utilized by upland rice cultivars. Plant and Soil 158, 53–62.

Lamb RJ, Richards BN (1974) Inoculation of pines with mycorrhizal fungi in natural soils: I. Effects of density and time of application of inoculum and phosphorus amendment on mycorrhizal infection. Soil Biology and Biochemistry 6, 167–171.

Lipton D, Blancher B, Blevins B (1987) Citrate, malate and succinate concentration in exudates from P-sufficient and P-stressed Medicago sativa L. seedlings. Plant Physiology 85, 315–317.

Liu Q, Loganathan P, Hedley MJ (2004a). Influence of ectomycorrhizal hyphae on phosphate fractions and dissolution of a phosphate rock in the rhizosphere soils of Pinus radiata. Journal of Plant Nutrition (accepted).

Liu Q, Loganathan P, Hedley MJ, Skinner MF (2004b) The mobilisation and fate of soil and rock phosphate in the rhizosphere of ectomycorrhizal Pinus radiata seedlings in an Allophanic Soil. Plant and Soil (in press).

Newman EI (1966) A method of estimating total length of root in a sample. Journal of Applied Ecology 3, 139–145.

Nicholas DP, Parkinson D (1967) A comparison of methods for assessing the amount of fungal mycelium in soil samples. Pedobiologia 7, 23-41.

Olykan ST, Adams JA (1995) Pinus radiata seedling growth and micronutrient uptake in a sand culture experiment, as affected by the form of nitrogen. New Zealand Journal of Forestry Science 25, 49–60.

Tabatabai MA (1994) Soil enzymes. In ‘Methods of soil analysis. Part 2. Microbiological and biochemical properties’. (SSSA Book Series No. 5: Madison, WI).

Wallander H (2000) Uptake of phosphorus from apatite by Pinus sylvestris seedlings colonized by different ectomycorrhizal fungi. Plant and Soil 218, 249-256.

Wang X, Zabowski D (1998) Nutrient composition of Douglas-fir rhizosphere and bulk soil solutions. Plant and Soil 200, 13–20.