phosphorus uptake and morphological characteristics of vesicular-arbuscular mycorrhizal colonization of Trifolium suhterraneum

Transcription

1 Phytol. (1997), 135, Effects of soil compaction on plant growth, phosphorus uptake and morphological characteristics of vesicular-arbuscular mycorrhizal colonization of Trifolium suhterraneum BY H. NADIAN^*, S. E. SITH, A.. ALSTON AND R. S. URRAY Department of Soil Science, Waite Agricultural Institute, The University of Adelaide, Glen Osmond, SA 5064, Australia (Received 27 arch 1996; accepted 12 September 1996) SUARY We investigated the effect of soil compaction and phosphorus (P) application on morphological characteristics of mycorrhizal colonization and growth responses, to determine the reasons for reduced responses observed in our previous work with compacted soil. Growth, phosphorus (P) uptake and intensity of vesicular-arbuscular (VA) mycorrhizal colonization were studied in clover plants {Trifolium subterraneuvi L.) witli and without VA mycorrhizal colonization at two P applications and three levels of soil compaction. Phosphorus was supplied either at constant mass concentration (mg P kg"' soil) or at constant volume concentration (mg P dm"'' soil). Increasing bulk density of the soil from l-l to 1'6 g m~^ significantly decreased root length and shoot d. wt, but increased the diameter of both main axes and first order lateral roots regardless of P application. Total P uptake and shoot d. wt of clover plants colonized by Glomus intraradices (Schenck & Smith) were significantly greater than those of non-mycorrhizal plants at all levels of soil compaction and both P applications. However, soil compaction to a bulk density of 1-6 gm"^ (penetrometer resistance = 3-5 Pa at a matric potential of 33kPa) significantly decreased mycorrhizal growth response. There was no evidence that the increased volume concentration of P at high bulk densities was responsible for the reduced responses. Soil compaction had no significant effect on the fraction of root length containing arbuscules and vesicles, but total root length colonized by arbuscules, vesicles or by any combination of arbuscules, vesicles and intra-radical hyphae significantly decreased as soil compaction was increased. The air-filled porosity of highly compacted soil, which varied from 0-07 to 0-11 over the range of matric potentials encountered ( 33 and lookpa), had no significant eflect on the intensity of internal colonization. Key words: Arbuscule, compaction, mycorrhizai growth response, P uptake, vesicle. INTRODUCTION compaction, thereby reducing both the volume of soil explored by the roots and the uptake of less Agricultural machinery often causes soil compaction, mobile nutrients like P. Shierlaw & Alston (1984) which increases bulk density (and penetrometer found a significant negative correlation between resistance) and alters soil pore size distribution, shoot P concentration and difi'erent levels of soil movement of air, water and nutrients (Grable & compaction. Siemer, 1968; Hoffmann &Jungk, 1995). These soil The growth and development of ve.sicularproperties influence plant growth either via mech- arbuscular (VA) mycorrhizal fungi are afiected by anical resistance (Bengough & uuins, 1991) or by many environmental factors. The best documented poor aeration (Agnew & Carrow, 1985). A root of these is soil P supply, which is negatively cannot penetrate a soil pore smaller than the diameter correlated with mycorrhizal growth response and of the root cap if the soil is strong (Wiersum, 1957). with the extent of fungal colonization (Abbott, Root growth will, therefore, be impeded by the Robson & De Boer, 1984; Bruce, Smith & Tester, decrease in average pore size associated with soil 1994; Smith & Read, 1996). Soil aeration is another * To whom correspondence should he addressed. environmental factor which has been shown to aflfect hnadian@,\vaite.adelaide.edu.au growth and development of VA mycorrhizal fungi.

2 304 H. Nadian and others Saif (1983) showed that the influence of soil Og on the percentage of root length colonized was very small compared with that on the intensity of development of arbuscules, vesicles or entry points. However, the efficiency of Glomus macrocarpus in enhancing the growth of its hosts was more affected by soil COo than was its development in roots (Saif, 1984). There is no information on how soil compaction, and hence decrease in average pore size, affects the growth of extraradical fungal hyphae. In our previous work using Trifolium subterraneum, we found a significant difference between P uptake and shoot d. wt of mycorrhizal and nonmycorrhizal plants in both slightly and highly compacted soils, but mycorrhizal growth response decreased with increasing soil compaction (Nadian et al, 1996). One possible reason for this decrease was the increase in mass of P applied per unit volume of soil, resulting from compaction. Therefore, any changes in mycorrhizal colonization in slightly and highly compacted soil might have been due to either the physical effect of compaction or to differences in the concentration of P. One of the aims of the present study was to compare the effect on mycorrhizal colonization of holding constant either the mass concentration of P (mg P kg"^ soil) or the volume concentration of P (mg P dm"^ soil) in the immediate vicinity of the roots over the range of compaction levels studied. There have been no studies of how changes in root growth in compacted soil affect mycorrhizal colonization, particularly the development of arbuscules and vesicles. Increased root diameter in mechanically impeded roots, decreased Oj and increased COj contents of the soil atmosphere resulting from compaction of the soil might affect the frequency of arbuscules, vesicles and intra-radical hyphae. Changes in these frequencies, particularly arbuscules, with increasing soil compaction might also contribute to the relatively lower mycorrhizal growth response and P uptake observed. Thus, one of the other aims of the present study was to investigate the effect of soil compaction on the intensity of mycorrhizal colonization and the development of arbuscules, vesicles and external hyphae. ATERIALS AND ETHODS The experiment was a factorial in randomized complete block design with four blocks and 12 treatments (see Table 3). Choice of soil An easily compacted soil of low P content was created by mixing 20 % of a solonized brown soil and 80 % of a red brown earth, collected from the 0-^20 cm layers at Avon and Adelaide in South Australia. Some properties of this soil mix are given in Table 1. Soil compaction The soil was passed through a 2 mm mesh sieve autoclaved at 121 C for 1 h on two consecutive The soil was thoroughly mixed with a basal fertilize solution as described by Smith & Smith (1981). The desired amounts of P fertilizer (as NaHaPO.^) anti sufficient distilled water to bring the water content o; the soil to 0-2 kg kg~^ soil were mixed throughout ths soil. The soil was double wrapped in plastic and held for 2 d at room temperature before packing into PVC pots, 90 mm in diameter and mm in length (depending on the degree of soil compaction). The pots were filled with soil in successive layers of 3 cm, and each layer was compacted to the desired bulk density with a hydraulic ram. Between each compaction, the soil surface was roughened to obtain a homogenous soil structure. The soil was compacted to one of three bulk densities: 1-1, 1-4 and 1-6 g m"^, giving penetrometer resistances at a matric potential of -33 kpa of 0-8, 2-3 and 3-5 Pa respectively, measured with a 2 mm diameter cone (30 semi-angle, penetration rate of 5 mm 5 min"^). The penetrometer resistance (Qp) was calculated as: Qp = AF/nd^, where d was the diameter of the cone and F was the force required to penetrate the soil. P application Our previous work (Nadian et al, 1996) indicated that, in a slightly compacted soil, clover plants (T. subterraneum) colonized by Glomus intraradices had maximum mycorrhizal growth response when the soil was supplied with 15 mg P kg""^ soil; this level of P was therefore used in the present study. We compared the effect on mycorrhizal colonization of holding either the mass concentration of P constant at 15 mg P kg"^ soil or the volume concentration of P constant at 15 mg P dm"^ soil, over the range of compaction levels studied. In a treatment with 15 mg P kg"^ soil, the pots contained 620 g oven-dry soil and received 9-3 mg P. Phosphorus concentration (mg P kg~^ soil) was therefore constant, whereas the mass of P per unit volume of soil increased with increasing soil compaction. In a treatment with 15 mg P dm"^ soil, the pots (all containing 620 g oven-dry soil) with bulk densities of 1-1, 1-4 and 1-6 g m"^ soil received 840, 6-64 and 5-81 mg P respectively. Thus, the mass of P per unit volume of soil was the same for all levels of soil compaction with this P treatment, but concentration per unit soil mass varied. Plant rnaierial and growth conditiotis Seeds of clover (T. subterraneum) were sterilized with NaOCl solution (5 g dm"'*) and germinated on moist filter paper at 23 C. Inoculum was obtained from pot cultures of T. subterraneum grown for two

3 ycorrhizal growth matrix in compacted soil 305 Table 1. Some chemical a7id physical characteristics of the soil tnix used in experiments to assess the effects of soil OJYipaction on mycorrhizal colonization of Trifolium subterraneum EC* ph (ds m (1:5, soil:water) Extractable Pf (mg kg~^ soil) Organic C (g kg"' soil) echanical composition (% w/w) Sand Silt Clav * Electrical conductivity. t NaHCOj extraction (Colwell, 1963). months in a soil:sand mix (1:9) containing 10% of inoculum from pot cultures of G. intraradices. Roots were washed with deionized water, cut into segments c. 1 cm long, thoroughly mixed and used as fresh inoculum. For the mycorrhizal treatments, each seedling was inoculated by placing 0-25 g fresh inoculum in each planting hole. Seedlings for the non-mycorrhizal treatments received 0-25 g of uncolonized clover roots. Each seedling received 0-5 cm'* of a dense suspension of Rhizobium leguminosarum bv. trifolii. Soil matric potential in all pots was maintained between 33 and 100 kpa (according to measured soil water characteristic curves). Three clover seedlings were transplanted in each pot and grown in a growth room where the photoperiod was 16 h and the irradiance was 380 fimol TTT^ s'^. The day and night air temperatures were 20 and 16 C respectively. The surface of the soil was covered with a 10-mm-deep layer of white polythene beads to minimize evaporation. easurements Shoots were harvested after 7 wk and were weighed after drying (70 C, 2 d). The roots were carefully washed from soil with deionized water, cut into 1 cm segments and thoroughly mixed. Sub-samples of roots were taken for determination of root length, mycorrhizal colonization and d. wt. Root lengths were determined by the line-intersect method (Tennant, 1975). The diameters of the root axes and of the first order lateral roots were measured at 3 mm behind the root apex with a binocular microscope fitted with an eyepiece micrometer. The values of root diameters for the main axes and the first lateral roots reported in this paper are themeans of 12 and 24 measurements respectively. Dried ground shoots and roots were digested in HNO3/HCIO4 and P concentrations were determined by the phosphovanado-molybdaternethod (Hanson, 1950). Assessment of colonization by VA mycorrhizal fungi followed cgonigle et al. (1990). This method determines the fraction of the root length containing arbuscules (FAC), vesicles (FYC) and either arbuscules, vesicles or hyphae (FHC). The latter is equivalent to percentage root length colonized. A random subsample of roots was mounted on glass slides. At least 150 intersects were assessed from each root sample. Total root lengths colonized by arbuscules (ACL), by vesicles (VCL) and by any combination of arbuscules, vesicles or intra-radical hyphae (HCL) were calculated as the product of total root length and FAC, FVC and FHC respectively. Two days before harvest, samples of soil air were collected with syringes from the centre of the pots. Carbon dioxide was measured by injecting samples into an air stream passing through an infra-red gas analyser (Atkins & Pate, 1977). Oxygen was measured by means of a gas chromatograph fitted with a thermal conductivity detector and a coaxial column containing molecular sieve 5A. The carrier gas was helium at a flow rate of 60 cm" min"'. The length of total external hyphae and of living hyphae was determined after staining with acid fuchsin and FDA (fiuorescein diacetate) respectively. The method for extraction of hyphae from the soil was modified from Abbott et al., (1984). Four soil cores (13 mm in diameter) were taken from each pot just before the plants were harvested. The soil was mixed, and a 2 g sub-sample was added to 300 cm'' of sodium hexametaphosphate (35-7 g dm'-' in distilled water) and stirred with a magnetic stirrer for 10 min. A 5 cm'' aliquot was removed and filtered through a 8//.m illipore* filter. To measure the length of living external hyphae, the retained material was stained in 0-5 cm'' of a freshly prepared solution of FDA in 60 m sodium phosphate buffer ph 7-4 at a final concentration of 0-01 g dm"'' FDA (Schubert et al., 1987). All procedures, from filtering the soil suspension and addition of FDA, were completed within 30 min to overcome problems of loss of enzyme activity and fading of fluorescence. A similar procedure was carried out to measure the length of total external hyphae, but the retained material from filtering of a 5 cm^ aliquot was stained with 0-5 cm'' of 0-1 g dm"^ acid fuchsin solution for 30 min. The length of total external hyphae and living hyphae were determined by counting the number of intersections between hyphae and a grid eyepiece micrometer at X 160 magnification on a Zeiss Standard Lab 16 equipped for epifiuorescence microscopy. The excitation filter was BP and the barrier filter was LP 520.

4 306 H. Nadian and others The pore size distributions of the soil were determined from soil water characteristic curves for all levels of soil compaction, using pressure plates (for low potentials) or sintered glass funnels (for high potentials) on cores 90 mm in diameter and 20 mm long. ycorrhizal growth response was calculated from the following equation: Q. m 0) o 100 (a) (b) / Shoot d. v^'t of \mycorrhizal plants /Shoot d. wt of non- -1 \ mycorrhizal plants Shoot d. wt of non-mycorrhizal plants xloo. N 1 t 50 o For all analyses, the F-test was used to test the significance of the main factors or analysis of variance using Genstat (Genstat 5 Committee, 1987). RESULTS The effect of soil compaction on growth of mycorrhizal and non-mycorrhizal plants Increasing soil compaction from a bulk density (BD) of l'l to l-galgm"'' increased linearly the penetrometer resistance (PR) of the soil; a linear regression gave the relationship PR = 5-2 BD 4-97, j-2 = Increased bulk density also decreased shoot d. wt, whereas mycorrhizal colonization increased shoot d. wt at all levels of soil compaction and P applications (Fig. 1). However, with both P treatments, mycorrhizal growth response decreased with increasing soil compaction. No significant differences were observed in mycorrhizal growth response between 15 mg P kg"^^ soil and 15 mg P dm"^^ soil (Fig. 2 and Table 3). Soil compaction decreased root length (Fig. 3), but increased the diameter of the main axes and first order lateral roots, particularly in the case of the main axes (Fig o 1.00 Q, 0-75 x: E 0.50 T3 O Bulk density (g Figure 1. The effect of soil compaction on shoot d. wt of mycorrhizal (, D) and non-mycorrhizal (, ^) Trifolium subterraneum with 15 mg P kg"^ soil (a) and 15 mg P drn"''soil {b). Vertical bars represent SE of the means, n = 4. (b) Bulk density (g Figure 2. The effect of soil compaction on mycorrhizal growth response of Trifolium subterrajiemn at two levels of P application as described in Figure 1. Vertical bars represent SE of the lneans, n = 4. 4). No significant difference was observed between root diameters of mycorrhizal and non-mycorrhizal roots. The effect of soil compaction on P uptake by mycorrhizal and non-mycorrhizal plants Shoot and root P concentrations and consequently total P uptake of mycorrhizal plants were greater than those of non-mycorrhizal plants at all levels of soil compaction (Table 2). Soil compaction had no effect on the P concentrations in shoots and roots of either mycorrhizal or non-mycorrhizal plants. Although, total P uptake decreased with increasing soil compaction (Table 2), total P uptake per unit length of root increased as soil compaction was increased (Fig. 5) and mycorrhizal fungi made an increasing contribution to the P uptake per unit length of root. No significant difference was observed in either tissue P concentration or total P uptake between two methods of P application (see Table 3). The effect of soil co?npaction on mycorrhizal colonization The fraction of the root length colonized by arbuscules (FAC), by vesicles (FVC) and by either arbuscules, vesicles or hyphae (FHC) in slightly compacted soil were 0-27, 0-11 and 0-45 respectively. Compacting the soil to a bulk density of 1-6 g m~^ had no effect on the FAC, FVC and FHC, but the total root length colonized by arbuscules (ACL), by vesicles (VCL) or by either arbuscules, vesicles or hyphae decreased with increasing soil compaction (Fig. 6). For the FAC, FVC and FHC there was no difference between 15 mg P kg~^ soil and 15 mg P dm~^ soil. The air-filled porosity' of highly compacted soil, which varied from 0-07O'l 1 over the

5 ycorrhizal growth matrix in compacted soil r growth was at its maximum. The concentration of CO2 in the soil air increased from 0-05 to 0-10 m"' m"'' over the same range of soil compaction. Soil compaction led to the collapse of most large pores (Fig. 8). In slightly compacted soil, 18-5 "o of the total porosity resided in pores of //m diameter, whereas compacting the soil to a bulk density of l-6gm~^ (3-5 Pa) decreased this percentage to only 5-4. In highly compacted soil, 28 % of the total pore space comprised pores larger than 3 //m diameter, whereas the value was 63 % in slightly compacted soil '6 Bulk density (g m"^) Figure 3. The effect of soil compaction on root length of mycorrhizal (, D) and non-mycorrhizal (, ^) Trifolium. subterraneum. The soil was supplied with 15 mg P kg"^ soil. Vertical bars represent SE of the means Soil bulk density (g m"^) Figure 4. Diameters of mycorrhizal Trifolium subterraneum roots (axes, D and first-order laterals, H) as affected by soil compaction. The soil was supplied with 15 mg P kg"^ soil. Vertical bars represent SE of the means. range matric potentials encountered ( 33 and lookpa), had no significant effect on the FAC, FVC and FHC. In both P fertilization regimes, no significant difference was observed in the length of total external hyphae and of living hyphae between inoculated and non-inoculated soils. However, soil compaction decreased the length of living external hyphae (Fig. 7), but not the length of total external hyphae. Soil compaction decreased the O, content of the soil atmosphere from 0-18 m^m"'* in slightly compacted soil to 0-10 m^ m~^ in highly compacted soil. The concentrations of Og in the soil atmosphere are probably close to the minimum concentrations that occurred, since the samples were collected during the day, when temperature and presumably respiration were high and before harvesting when plant DISCUSSION ycorrhizal colonization increased shoot d. wt at all levels of soil compaction and both P treatments, but this increase was smaller as soil compaction increased. A similar trend was observed for total P uptake. Decreased total P uptake by mycorrhizal plants with increasing soil compaction might be attributed to the considerable reduction in root length, and length of root colonized by arbuscles (Fig. 6) leading to a significant reduction in surface area available for P uptake and transfer to the host plant. Despite the reduction in P uptake per plant in highly con-ipacted soil, P uptake per unit length of root increased with increasing soil compaction, particularly in the case ol mycorrhizal plants. This increase indicates that the efficiency of mycorrhizal roots in absorbing P was greater in highly compacted soil than in slightly compacted soil (Fig. 4) and also that the factor limiting growth of mycorrhizal plants was not the efficiency of the roots per se. There have been no previous reports of how the effects of changes in root growth in compacted soil interact with the development of characteristic mycorrhizal structures. Our data show that compaction had no effect on the fraction of root colonized by arbuscules (FAC), vesicles (FVC), or by any combination of arbuscles, vesicles and intra-radical hyphae (FHC). The simplest explanation is that colonization is not influenced by compaction or consequent changes in root growth. However, two conflicting effects could have had the same outcome: (1) increase in root diameter and consequently in root cortex (Atwell, 1988) increases surface area per unit length of root available for colonization, and (2) decrease in concentration of O2 in the atmosphere of compacted soil might decrease the intensity of internal colonization. This suggestion is supported by the results of Saif (1981), who found that a decrease in O«concentration of soil air decreased the number of arbuscules. oreover, the air-filled porosity of highly compacted soil varied from 0-07 to 0-11 over the range of matric potentials encountered ( 33 and kpa). The latter value is close to the air-filled porosity of 0-12 which has been suggested as inadequate to provide sufficiently rapid diffusion

6 308 H. Nadian and others Table 2. The effects of soil compaction on total P uptake and shoot and root P concentration of mycorrhizal ( and non-mycorrhizal clover () Bulk density (g m-'') P treatment* ycorrhiza Total P uptake (mg per pot) P concentration (jig P mg d. wt) Shoot Root PI P2 PI P2 PI P2 l-56(±0-ll)t 0-46 ( ) 1-86 (±0-15) 0-54 ( ) 0-81 (-t-0-08) 0-24 ( ) O ( ) 0-37 ( ) 0-54 ( ) 0-16 ( ) 0-70 ( ) 0-23 (±0-02) 1-25 ( ) 0-85 ( ) 1-26 ( ) 0-92 ( ) 1-23 (±0-03) 0-94 ( ) 1-31 ( ) 1-02 ( ) 1-25 ( ) 0-81 ( ) 1-31 ( ) 0-85 ( ) 1-20 ( ) 0-95 ( ) 1-21 ( ) 1-02 ( ) 1-20 ( ) 0-98 ( ) 1-31 ( ) 1-05 ( ) 1-24 ( ) 0-82 ( ) 1-27 ( ) 1-09 ( ) * PI = 15 mg P dm-3 soil; P2 = 15 mg P t SE of mean. soil Bulk density (g Figure 5. The effect of soil compaction on total P uptake per unit length of roots by mycorrhizal (, Q) and nonmycorrhizal (, ) Trifolium subterraneum at two levels of P application as described in Figure 1. Vertical bars represent SE of the means, n = 4. of oxygen in soil for plant growth (Grable & Siemer, 1968). However, Saif (1984) pointed out that the effect of poor soil aeration on the mycorrhizal endophyte was slight and the rate of colonization was less sensitive than was mycorrhizal efficiency. It seems the established colonization in the root can be buffered by the cortex, and it becomes less sensitive to poor aeration than the external hyphae which are in direct contact with the concentrations of Og and COg in the soil atmosphere. Total root length colonized by arbuscules (ACL), by vesicles (VCL) or by any combinations of arbuscles, vesicles or hyphae (HCL) decreased greatly with increasing soil compaction. This decrease can be attributed to the considerable reduction in the total length of roots in compacted soil. There seems, therefore, to be no 1-6 evidence that the decreased mycorrhizal growth response in compacted soil was due to reduction in arbuscule formation per unit root length colonized. In this experiment, no significant difference was observed in the FVC, FAC and FHC between 15 mg P kg"^ soil and 15 mg P dm"^ soil, although a decrease in intensity of internal colonization with increasing soil P concentration has been reported (Amijee, Tinker & Stribley, 1989; Smith & Gianinazzi-Pearson, 1990; Bruce et al., 1994). This might be related to the small difference in concentration of P between 15 mg P kg"^ soil and 15 mg P dm"^ soil. We have already demonstrated that mycorrhizal growth response is not very sensitive to P concentration in this range. Thus, decrease in mycorrhizal growth response with increasing soil compaction observed in this and in our earlier experiments (Nadian et al., 1996) is not likely to be a consequence of the increase in mass of P applied per unit volume of the soil, resulting from compacting the soil. We could not compare the length of external hyphae in inoculated and non-inoculated soil because of difficulties in differentiating between hyphae of VA mycorrhizal fungi and those of non-va mycorrhizal fungi (saprophytic Zygomycetes) when the hyphae were fragmented (Abbott & Robson, 1985; Jakobsen, Joner & Larsen, 1994). However, with both P treatments, soil compaction decreased the length of living external hyphae. It is difficult to know to what extent soil compaction might affect the length of external hyphae of VA mycorrhizal fungi compared with those of non-va mycorrhizal fungi. Soil compaction might directly or indirectly affect the length of external hyphae. Compacting the soil to a bulk density of 1-6 g m~^ soil significantly decreased the percentage of pores larger than 3 fim (Fig. 8) which can be occupied by the fine branches

7 ycorrhizal growth matrix in compacted soil 309 Table 3. Probability of F for some measurements (for abbreviations, see text) Source of variance df Compaction (C) 2 ycorrhiza () 1 Phosphorus (P) 1 Cx 2 CxP 2 xp 1 Cx xp 2 Shoot d.wt Shoot P concentration RootP concentration Total P uptake Total P uptake per unit length of root FHC FAC ACL VCL Root length Root diameter (axes) Root diameter (first lateral) Soil air O2 ycorrhizal growth response (%) #*# *** *#* *** *** ##* * *** *** ** t *, **, ***. significance at the 0-05, 0-01, probability level, respectively. J, not significant at P = ycorrhiza was not considered as source of variance in calculating the ANOVA of colonization. 20 I ft15 CD ^«10 c ^ O '" o Bulk density (g m"^ Figure 6. The effect of soil compaction on total root length colonized by arbuscules (ACL, Q) and by vesicles (VCL, ) of Glomus intraradices. The soil was supplied with 15 mg P kg^'^ soil. Vertical bars represent SE of tbe means. 1.6 t.^ Is Bulk density (g Figure 7. The length of living external hyphae in inoculated (, D) and uninoculated (, ) soils as afiected by soil compaction. Vertical bars represent SE of tbe means, w = of mycorrhizal mycelium. Thus development of larger runner hyphae (diameter 5-20/tm) might have been restricted in highly compacted soil. The change in pore size distribution of the soil also led to a reduction in air-filled porosity and consequently in oxygen concentration from 0-18 m^m"^ in slightly compacted soil to 0-10 m* m'* in highly compacted soil. These physical alterations in compacted soil might have directly affected the development of external hyphae. A possible indirect effect of compaction on the lengths of external hyphae may be attributed to the considerable reduction in total colonized root length from 27-3 m per pot in slightly compacted soil to 5*5 m per pot in highly compacted soil. If the mean length of external hyphae is assumed to be 200 m m~^ colonized root (based on the results of Jakobsen, Abbott & Robson, 1992 and Pearson &

8 310 H. Nadian and others a. ter( E ID oa. > Contribution to total porosity (%) Figure 8. Pore size distributions in slightly (bulk density = 1-1 g m"'', D) and highly (bulk density = 1-6 g m"^, ) compacted soils. Each mean is average of four replicates. Jakobsen, 1993), then the length of external hyphae in slightly and highly compacted soil are estimated to be 9-7 and 2-8 m cm~^ soil respectively. Such reduction in the length of external hyphae in highly compacted soil would be expected to follow from a decrease in the total length of root colonized in these soils, but as yet we have no evidence to confirm; this assumption. In conclusion, mycorrhizal colonization in compacted soil increased total P uptake and plant growth and compensated, in part, for the effect of soil compaction. However, mycorrhizal growth response decreased with increasing soil compaction as a result of decreased total root length colonized by arbuscules and intra-radical hyphae and consequently a considerable reduction in surface area of interface for P transfer to the host plant. Our observations lead us to conclude that the decline in benefit of mycorrhizal colonization in highly compacted soil might be due, at least in part, to inhibition of root growth and consequently to a considerable reduction in the surface area available for colonization. ACKNOWLEDGEENTS The technical assistance of r Colin Rivers and s Sandy Dickson is gratefully acknowledged. We thank s Rita iddelberg for providing statistical advice and s Kerri uuer for helping in measurement of O^ and COj in the soil atmosphere with gas chromatography. REFERENCES Abbott LK, Robsoa AD, De Boer G The effect of phosphorus on the formation of the hyphae in soil by the vesicular-arbuscular mycorrhizal fungus, Glomus faciculatum. New Phytotogist 97: Abbott LK, Robson AD Formation of external hyphae in soil by four species of vesicular-arbuscular mycorrhizal fungi. New Phytologist 99: Agnew L, Carro-w RN Soil compaction and moisture stress preconditioning in Kentucky bluegrass. I. Soil aeration, water use, and root responses. Agronomy Journal 77: Amijee F, Tinker PB, Stribley DP The development of endomycorrhizal systems VII. A detailed study of the effects of soil phosphorus on colonization. New Phytologist 111: 435^146. Atkins CA, Pate JS An IRGA technique to measure CO^ content of small volumes of gas from internal atmospheres of plant organs. Photosynthetica 11: Atwell BJ Physiological responses of lupin roots to soil compaction. Plant and Soil 111: Bengough AG, ullins CE Penetrometer resistance, root penetration resistance and root elongation rate in two sandy loam soils. Plant and Soil 131: Bruce A, Smith SE, Tester The development of mycorrhizal infection in cucumber: effects of P supply on root growth, formation of entry points and grovnh of infection units. New Phytologist ITl: Colwell JD The estimation of the phosphorus fertilizer requirements of wheat in southern New South Wales. Australian Journal of Experimental Agricultural and Animal Husbandry 3: Genstat 5 Committee GENSTAT 5: Reference anual Oxford: Clarendon Press. Grable AR, Siemer EG Effects of bulk density, aggregate size and soil water suction on oxygen diffusion, redox potentials and elongation of corn roots. Soil Science Society of America Proceedings 32: Hoffmann C, Jungk A Growth and phosphorus supply of sugar beet as affected by soil compaction and water tension. Plant and Soil 176: Hanson WC The photometric determination of phosphorus in fertilizers using the phosphovanado-molybdate complex. Journal of the Science of Food and Agriculture 1: Jakobsen I, Abbott LK, Robson AD External hyphae of vesicular-arbuscular mycorrhizal fungi associated with Trifolium subterraneum L. 1 Spread of hyphae and phosphorus inflow into roots. New Phytologist 120: Jakobsen I, Joner EJ, Larsen J Hyphal phosphorus transport, a keystone to mycorrhizal enhancement of plant growth. In: Gianninazzi S, Schuepp H, eds. Impact of Arbuscular ycorrhizas on Sustainable Agriculture and Natural Ecosystems. Basel/Switzerland: Birkhauser Verlag, cgonigle TP, iller H, Evans DAG, Fairchild GL, Swan JA A new method which gives an objective measure of colonization of roots by vesiculararbuscular mycorrhizal fungi. New Phytologist 115: Nadian H, Smith SE, Alston A, urray RS The effect of soil compaction on growth and P uptake by Trifolium subterraneum: interactions with mycorrhizal colonization. Plant and Soil (in press). Pearson JN, Jakobsen I The relative contribution of hyphae and roots to phosphorus uptake by vesicular-arbuscular mycorrhizal plants, measured by dual labelling with '^P and ^=P. New Phytologist 124: 489^94. Saif SR The influence of soil aeration on the efficiency of the vesicular-arbuscular mycorrhizas. I. Effects of soil oxygen on growth and mineral uptake of Eupatorium odoratum L. inoculated with Glomus macrocarpus. New Phytologist 88: Saif SR The influence of soil aeration on the efficiency of the vesicular-arbuscular mycorrhizas. II. Effects of soil oxygen on growth and mineral uptake in Eupatorium odoratum L., Sorghum bicolor (L.) oench and Guizotia abyssinica (L.f.) Cass. inoculated with vesicular-arbuscular mycorrhizal fungi. New Phytologist 95: Saif SR The influence of soil aeration on the efficiency of the vesicular-arbuscular mycorrhizas. III. Soil carbon dioxide and growth and mineral uptake in mycorrhizal and nonmycorrhizal plants of Eupattorium odoratum L., Guizotia abyssinica (L.f.) Cass. and Sorghum bicolor (L.) oench. New Phytologist 96: Schubert A, arzachi C, azzitelli, Cravero C, Bonfante-Fasolo P Development of total and viable extraradical mycelium in the vesicular-arbuscular mycorrhizal

9 ycorrhizal growth matrix in compacted soil 311 fungus Glomus clarum Nicol & Scenck. New Phytologist 107: Shierlaw J, Alston A Effect of soil compaction on root growth and uptake of phosphorus. Plant and Soil 77: Smith FA, Smith SE ycorrhizal infection and growth of Trifolium subterraneum: use of sterilized soil as a control treatment. New Phytologist 88: Smith SE, Gianinazzi-Pearson V Phosphate uptake and arbuscular activity in mycorrhizal Allium cepa L.: effects of photon irradiance and phosphate nutrition. Australian Journal of Plant Physiology 17: Smith SE, Read DJ ycorrhizal symbiosis, 2nd edition. London: Academic Press. Tennant D A test of a modified line intersect method of estimating root length. The Journal of Ecology 63: Wiersum LK The relationship of the size an structural rigidity of pores to their penetration by roots. Plant and Soil 9:

10