with a Phosphorus-Solubilizing Penicillium bilaji Strain and with Vesicular-Arbuscular Mycorrhizal Fungi

APPLIED AND ENVIRONMENTAL MICROBIOLOGY. Dec. 1987, p. 2699-2703 0099-2240/87/122699-05$02.00/0 Vol. 53. No. 12 Increased Phosphorus Uptake by Wheat and Field Beans Inoculated with a Phosphorus-Solubilizing Penicillium bilaji Strain and with Vesicular-Arbuscular Mycorrhizal Fungi R. M. N. KUCEY Agricuiltuire Canada Research Staition, Lethbridge, Alberta, Received 9 April 1987/Accepted 16 June 1987 TIJ 4BI, Canada Greenhouse and field experiments were conducted to test the effect of a P-solubilizing isolate of Penicillium bilaji on the availability of Idaho rock phosphate (RP) in a calcareous soil. Under controlled greenhouse conditions, inoculation of soils with P. bilaji along with RP at 45 jig of P per g of soil resulted in plant dry matter production and P uptake by wheat (Triticum aestivum) and beans (Phaseolus vulgaris) that were not significantly different from the increases in dry matter production and P uptake caused by the addition of 15,ug of P per g of soil as triple superphosphate. Addition of RP alone- had no effect on plant growth. Addition of vesicular-arbuscular mycorrhizal fungi was necessary for maximum effect in the sterilized soil in the greenhouse experiment. Under field conditions, a treatment consisting of RP (20 kg of P per ha of soil) plus P. bilaji plus straw resulted in wheat yields and P uptake equivalent to increases due to the addition of monoammonium phosphate added at an equivalent rate of P. RP added alone had no effect on wheat growth or P uptake. The results indicate that a biological system of RP solubilization can be used to increase the availability of RP added to calcareous soils. The use of unprocessed rock phosphate (RP) as a phosphorus (P) fertilizer has merit for areas without easy access to P fertilizer production plants. The direct use of RP is, however, restricted to acidic soils since the benefits from the addition of RP to neutral calcareous soils are low in comparison to benefits obtained with more acidic soils (10). Vesicular-arbuscular (VA) mycorrhizal fungi exist naturally in most agricultural soils and can affect plant P uptake. Uptake of P from RP and P fertilizers has been shown to be increased by the addition of mycorrhizal fungi to soils if the numbers of mycorrhizae in the soil are initially low (5, 13, 16). A group of soil microorganisms affecting soil P availability are the P-solubilizing bacteria (15), fungi (8), and actinomycetes (14). Such organisms have the ability to solubilize P-containing minerals by the secretion of organic acids and have been reported to increase the P uptake of rice from inoculated soils (3). P-solubilizing organisms are present naturally in prairie soils and have been found to comprise 0.1 to 0.5% of the total soil microbial population (9). Gaur et al. (6) found increased P uptake from RP by wheat and greengram when a P-solubilizing culture of Pseiudotnonas striata and Aspergilllus awamori was added along with corn stover. Khalafallah et al. (7) found increased P uptake from superphosphate by Vicea faba in a desert calcareous soil after inoculation of the soil with Bac illlus megatheriiim var. phosphaticum. Azcon et al. (2) observed P absorption from RP by lavender plants when they were inoculated with both mycorrhizal fungi and a P-solubilizing bacterium, but no absorption in the absence of the microorganism. In this paper we report the results of greenhouse experiments comparing the P uptake by wheat and field beans from Idaho RP and triple superphosphate in soils inoculated with mycorrhizal fungi and a P-solubilizing strain of Penicillium bilaji. We also report the results of a field experiment conducted on a Chernozemic calcareous soil in southern Alberta comparing Idaho RP and monoammonium phosphate with and without P-solubilizing fungal inoculation. 2699 MATERIALS AND METHODS Greenhouse experiment. A Brown Chernozemic soil (loamy sand, Cavendish series) was collected from a cultivated field near Iron Springs, Alberta. The soil was air dried, sieved (2 mm), and mixed 1:1 (vol/vol) with sand. The final mixture had a ph of 7.2 (1:1 in H,O) and contained 2 jig of available P per g of soil (NaHCO3-extractable) (12). One kilogram of the soil mixture was placed in 20-cm ceramic pots, autoclaved for 8 h (115 C, 12 lb/in2) to kill the native mycorrhizal fungi, and allowed to reequilibrate for 6 weeks. The greenhouse treatments used in this experiment were as follows: RP only, P. bilaji only, P. bilaji-rp, ground wheat straw only, straw-rp, straw-p. bilaiji, straw-rp-p. biklji, preincubated straw alone, preincubated straw-rp, preincubated straw-p. bilaji, preincubated straw-rp-p. bilaiji, triple superphosphate, and an untreated control. Field treatments were RP only, RP-P. bilaji, straw only, straw-p. bilk{ji, straw-rp, straw-rp-p. bilaji, monoammonium phosphate only, monoammonium phosphate-p. bilaji, control (no P added), and P. bilaji only. Idaho RP (10.3% P) was sieved (60 mesh), and the fine material was added to provide 45 jig of P per g of soil (437 mg of RP per pot). Triple superphosphate was added at a rate of 15 jg of P per g of soil (76.5 mg of triple superphosphate per pot). Wheat straw was ground (2 mm) and added at a rate of 1.0 g per pot. RP was thoroughly mixed with the straw when both were used. P. bilaiji, a P-solubilizing isolate, was inoculated on autoclaved ground (2 mm) wheat straw amended with 1.0% glucose for 2 weeks at room temperature to produce hyphae and spores. The inoculum of P. bilaiji consisted of 0.1 g of moist colonized straw particles per pot. The inoculum was placed on top of the straw when both were used or placed on top of the RP when straw was not included. Preincubated straw treatments consisted of the same materials as the soil-inoculated treatments but combined and incubated under laboratory conditions for 6 weeks before

2700 KUCEY adding them to the soils. Twenty-five grams of ground wheat straw was placed in plastic jars. RP (436 mg x 25 = 10.9 g of RP per jar) or P. bilaji (2.5 g of P. bilaji-colonized wheat straw) or both were added if used. The jar was covered with a perforated lid to allow air transfer, and the whole unit was weighed. Five milliliters of distilled water was added. The contents of the jar were mixed by shaking every 3 days and incubated on the bench top at room temperature. At weekly intervals, a sample of the straw was removed, the moisture content was measured, and water was added to maintain the moisture content at 20%. After 6 weeks of incubation, fungal hyphae were visible on the straw particles. Straw subsamples equal to 1.0 g of fresh straw, corrected for weight loss, were measured out for pot inoculation. VA mycorrhizal inoculum consisted of a mixed culture of VA mycorrhizal fungi isolated from southern Alberta soils and propagated on strawberry roots in the greenhouse. Dried colonized roots and adhering soil (5.0 g per pot) were used as the inoculum. All amendments to the soil, except for the mycorrhizal fungi, were placed in the bottom of a 5-cm-deep hole in the center of each pot and covered with soil. Mycorrhizal inoculum was placed in the same hole as the seeds and was evenly divided among seeds. Five wheat seeds (Triticum aestivum "Chester") or four bean seeds (Phaseolus vulgaris "GN1140") were planted in each pot and thinned to two plants per pot after emergence. Each of the treatments listed above was used for both test crops. NH4NO3 (30 pg of N per g of soil) was added to each pot at the beginning of the experiment. Pots containing wheat received a further 30,ug of N per g of soil at the beginning of week 4. Bean seeds were inoculated with a commercial Rhizobium phaseoli inoculant (Nitragin Co., Milwaukee, Wis.). Five replicates of each treatment for each test crop (5 replicates x 13 treatments x 2 mycorrhizae x 2 crops) were randomly arranged on greenhouse tables. The arrangement of the pots was altered every 2 weeks for the duration of the experiment. Supplemental lighting was used to maintain a 16-h-8-h day-night cycle. N- and P-free Long-Ashton nutrient solution (20 ml) was added to each pot on weeks 2, 3, and 8. Plants were harvested at maturity, dried, weighed, and ground. Subsamples of the plant material were acid-digested (4) and analyzed for P content (17). The presence or absence of VA mycorrhizae in the inoculated and uninoculated pots was verified by microscopically viewing roots from randomly selected pots after clearing and staining the roots with acid fuchsin (0.01% in lactoglycerol) (1). Statistical analysis of the data was performed using the General Linear Model program of the Statistical Analysis Systems (SAS Inc., Cary, N.C.) package. Single degree of freedom comparisons were made between individual treatments to determine whether the differences observed were significant. Statistical analysis was performed on the complete data set for each crop and on a data subset of each crop containing only treatments with mycorrhizae. This data subset was used to evaluate treatments under conditions of mycorrhizal root colonization such as would be encountered in nonsterilized field soils. Field study. A plot was selected on the Vauxhall Substation of the Lethbridge Research Station adjacent to a plot that had previously shown responses of wheat to P fertilizer application. The soil was an Orthic Brown Chernozem with a clay-loam texture and contained 2 p.g of NaHCO3- extractable P per g of soil. The treatments used in the field experiment are listed above. Methods used for preparation of the RP, P. bilaji, and APPL. ENVIRON. MICROBIOL. straw were the same as used for the greenhouse experiment. Monoammonium phosphate was used instead of triple superphosphate. Each treatment plot measured 2.0 m long and five rows (spaced 23 cm apart) wide. Both RP and monoammonium phosphate were added to the soil at a rate equal to 20 kg of P per ha, the recommended rate of P addition for wheat in this soil (3.83 g of monoammonium phosphate per row, 8.93 g of RP per row). Straw was added at a rate of 4.0 g per row. Two grams (fresh weight) of P. bilaji inoculum was added per row. Nitrogen as NH4Cl was added to all rows not receiving monoammonium phosphate to equalize N additions. All of the amendments added to a row were thoroughly mixed before addition to the soil. Furrows were hand dug to a 5-cm depth, and preweighed subsamples of the amendments were hand spread evenly along the length of the furrow. Wheat seeds (T. aestivum "Chester") were spread along the row 2 cm apart (4.1 g of seed per row). The furrows were covered over by hand, and N as NH4NO3 was spread over the plot surface at a rate equal to 30 kg of N per ha. Five replicates of the 10 treatments were arranged in a randomized block design. The plots were irrigated as needed to maintain soil moisture tension below 45 KPa. At maturity, the central 1.3 m of the central three rows of each plot were harvested. Straw and seed dry matter weights were determined before the grain and straw from each plot was ground, combined, and thoroughly mixed. The P content of the combined sample was measured after acid digestion by the same procedures used for the greenhouse study. Statistical analysis of the data was performed using the General Linear Model program of the Statistical Analysis Systems package. Single degree of freedom comparisons were made between individual treatments to determine the significance of differences observed. RESULTS Greenhouse study. Overall, the addition of VA mycorrhizal fungi had a significant positive effect on the dry matter production of both wheat and beans and resulted in a significant increase in the P content of both crops (Tables 1 and 2). Visual examination of roots showed mycorrhizae to be present in inoculated treatments and absent from those pots not receiving mycorrhizal inoculum. Taken individually, the effect of mycorrhizae on each treatment was generally to increase the plant weights and P content of wheat, even though the effect was not significant in all comparisons. For beans, the effect of mycorrhizal fungi was to increase plant weights for individual treatments, but the effect on P content was not always positive. Since mycorrhizae would be present under natural conditions and had a significant effect, further comparisons of the data were made only for those treatments receiving mycorrhizal fungi. Single degree of freedom comparisons of the data for the treatments receiving mycorrhizal fungi show that for both plant dry weight and P content, the complete treatment (straw plus RP plus P. bilaji) resulted in levels significantly higher than those of the unfertilized control and also higher than the treatment receiving only straw plus RP (Table 2). The addition of RP alone or in combination with straw in the absence of P. bilaji did not have a significant effect on plant weights or P content of either crop plant. The addition of P. bilaji alone had a positive effect on wheat dry matter production and P content but did not affect these parameters for the beans. The addition of P. bilaji plus straw increased wheat and bean plant weights but did not affect the P content of either crop.

VOL. 53, 1987 P-SOLUBILIZING FUNGUS INCREASES PLANT P UPTAKE 2701 TABLE 1. Dry matter production and P uptake by wheat and beans in pots amended with RP and inoculated with VA mycorrhizal fungi (VAM) and a P-solubilizing P. bilaiji Dry matter (g per pot) P content (mg per pot) Treatment' Wheat Beans Wheat Beans -VAM +VAM -VAM +VAM -VAM +VAM -VAM +VAM Triple superphosphate 2.6 2.7 5.8 6.1 3.9 4.8 12.1 10.1 Straw-RP-P. biltji 2.3 3.2 4.9 5.6 3.1 5.1 9.3 8.6 PIS-RP-P. bilaji 2.2 2.7 4.3 4.5 4.1 4.4 6.7 7.6 Straw-P. bilaji 2.1 2.8 4.6 5.2 3.2 4.3 8.5 8.0 PIS-P. bilaji 1.9 2.5 4.7 4.6 3.2 5.0 6.5 7.1 Straw-RP 1.9 2.3 4.6 4.3 3.4 3.5 9.2 6.9 PIS-RP 1.9 2.4 3.9 4.1 3.6 4.5 5.9 6.7 RP-P. bilaji 2.1 2.5 4.3 4.9 2.8 4.0 7.8 7.5 P. bilaji 2.0 2.5 4.6 4.5 3.1 4.3 8.8 7.3 RP 2.1 2.4 3.9 4.6 3.1 3.6 7.8 7.2 Control 1.9 2.3 3.0 3.6 3.0 3.9 6.7 6.6 ' PIS, Preincubated straw. The addition of straw in combination with the other amendments had a significant positive influence on the effectiveness of RP and P. bilaji (Table 2). In the case of wheat plant weights and P content, preincubation of the straw before addition neither benefited nor deterred the effectiveness of the treatments. For bean plant weights and P content, however, the preincubated straw-rp-p. bilaji treatment was significantly less effective than the straw-rp-p. bilaji treatment. When compared with the effectiveness of triple superphosphate, the straw-rp-p. bilaji treatment was equivalent for bean dry weights or P content by wheat or beans. The complete treatment was significantly more effective than triple superphosphate alone in the case of wheat dry weights (Table 1). Field study. Grain and straw dry matter production was 1.4- and 1.5-fold higher, respectively, than that in the untreated controls as a result of the addition of straw-rp-p. bilaji to the soil. The increase was equivalent to that observed from the addition of monoammonium phosphate at equal rates of P (Table 3) and greater than that observed from the addition of straw-rp. The addition of P. bilaji alone or in combination with straw or RP also had a significant effect on grain and straw dry weights (Table 4). Inclusion of straw as a carbon source for fungal growth was shown to TABLE 2. Effects and statistical analysis (sums of squares) of data from greenhouse trials on the use of rock phosphate and VA mycorrhizal fungi (VAM) and P-solubilizing fungi on wheat and bean dry matter production (DMP) and P uptake Data and source df Sums of squares Wheat DMP Bean DMP Wheat P Bean P All data included Total 129 19.7 114.4 128.7 377.1 Replicates 4 0.2 2.4 2.5 2.0 Treatment 25 12.6" 54.9" 55.5" 219.6" Error 100 6.9 57.1 70.8 155.4 Contrast with versus without VAM 1 6.2" 4.2" 24.6" 6.3"b VAM treatments only Total 64 8.9 50.4 78.8 137.6 Replicates 4 0.4 0.3 6.6 0.6 Treatment 12 4.2" 24.9" 31.8b 53.7" Error 48 4.3 25.2 50.4 83.3 Contrast" S-RP-P. biltaji versus TSP 1 0.4"7 0.8 0.2 5.8 S-RP-P. bilaiji versus control 1 2.0" 9.5" 6.2" 9.8"b S-RP-P. biltiji versus S-RP 1 2.0"l 3.9" 5.9b 7.5b S versus PIS' 1 0.2 3.5 b 1.4 2.6 S versus no S' 1 0.4" 3.3" 1.2 7.3" S-P. bilaji versus control 1 0.5" 5.4" 1.5 2.8 P. bilaji versus control 1 0.8" 1.6 7.2"b 0.1 S-RP versus control 1 0.1 1.2 0.4 0.1 RP versus control 1 0.1 0.1 0.9 3.0 " Significant at P < 0.01. b Effect significant at P < 0.05. S, Straw; PIS. preincubated straw; TSP, triple superphosphate. All treatments with straw versus equivalent treatments with preincubated straw. All treatments with straw versus equivalent treatments without straw.

2702 KUCEY TABLE 3. Grain and straw yields and P uptake by wheat after application of RP and P-solubilizing P. bilaji fungus to a field soil Treatment Grain Straw P uptake Treatment ~(g/o.9 in2) (gio.9 in2) (gio.9 in2) Straw-RP-P. bilaji 232 303 1.10 Straw-P. bilaji 188 232 0.87 Straw-RP 185 224 0.84 RP-P. bilaji 198 223 0.82 P. bilaji 207 237 0.95 RP 167 204 0.73 Straw 151 192 0.69 MAPa 238 296 1.12 MAP-P. bilaji 231 285 1.09 Control 163 202 0.71 - MAP, Monoammonium phosphate. significantly increase the effectiveness of the RP-P. bilaji treatment. The addition of P. bilaji to monoammonium phosphate did not affect grain or straw yields above the level of monoammonium phosphate alone. RP addition alone or with straw did not have a significant effect. Wheat P content was increased by the addition of the complete treatment (1.5-fold) or monoammonium phosphate (1.6-fold) in comparison with the control (Table 3). Phosphorus content with these two treatments was not significantly different (Table 4). The addition of P. bilaji to the straw-rp treatment increased the effectiveness of that combination. The addition of P. bilaji alone had a significant effect on P content but when added in combination with RP did not have an effect. RP added alone or with straw had no effect on the wheat P content. DISCUSSION Increased P uptake by plants colonized by mycorrhizal fungi has been previously reported (5, 13, 16). The beneficial effect of the mycorrhizae on a P-solubilizing system such as the one used in this study is also clear. The presence of mycorrhizal fungi had negligible effects on the growth and P uptake of wheat and beans when added along with triple superphosphate, which is not surprising since triple superphosphate is generally quite available for plant uptake. The mycorrhizal fungi did have a greater effect on the P. bilaji- RP system, increasing both plant production and P uptake. In the presence of mycorrhizae, the P-solubilizing system was as effective as triple superphosphate. Under prairie field conditions, soils lacking mycorrhizal fungi are almost never encountered (11), so the problem of reduced effectiveness of the P-solubilizing system in the absence of mycorrhizae would not be serious. This work confirms the synergistic effect of mycorrhizal fungi and P-solubilizing microorganisms reported by Azcon et al. (2). Their study used a P-solubilizing bacterium and rates of RP addition much higher than those in this study and did not compare the resulting P uptake with that from a processed P fertilizer. The present study shows that the mycorrhizae are a necessary part of a P-solubilizing system and, if present, can make the system significantly more effective by increasing the ability of the plant root system to absorb P. Previous work in similar soils has shown that various RP sources added at a rate of 109 mg/kg of soil to a moderately acidic (ph 5.3) Chernozemic soil varied in effectiveness from negligible to 75% of the effect of adding 10.9 mg per kg of soil as triple superphosphate, as measured by increases in plant APPL. ENVIRON. MICROBIOL. P uptake (10). Idaho RP in that study proved to be 18% as effective as triple superphosphate when added at 10 times the rate of P. In the present study, Idaho RP added alone in the pot study was 17% as effective as triple superphosphate when added at 3 times the P rate and 5% as effective as monoammonium phosphate in the field study when added at equivalent P rates. Considering that RP availability usually decreases with increasing soil ph, it is safe to assume that the estimates of Idaho RP availability are similar to those in previous studies and that this availability is low in comparison with processed P fertilizers. The addition of the P- solubilizing Penicillium fungus along with a carbon source (straw) and RP resulted in a real increase in P uptake by beans in the greenhouse study and by wheat in both the greenhouse and the field. Previous work (10) showed plant P uptake to be well correlated with rock P availability, so the P uptake measurements of this study can be considered to be reflections of the P availability of the various treatments. Tracers were not used in this study, so it is not possible to determine whether the increased P uptake was due only to solubilization of the added RP or whether solubilization of some soil P forms occurred as well. Indeed, P. bilaji addition alone or in combination with a carbon source without RP did result in significant increases in plant production in the greenhouse and in increases in P uptake in the field, showing that some soil P was likely solubilized. The effect of the combined treatment was greater than the effect of P. bilajistraw or RP-P. bilaji or straw-rp, showing that all three components are necessary for maximum effects in this study. In the pot experiment, straw was added fresh or was preincubated with the RP or P. bilaji or both. This was an attempt to see whether a preincubation period would increase the effectiveness of the complete P-solubilizing system and to see whether other microorganisms which occur as natural contaminants under nonsterile conditions would have any P-solubilizing effect on the RP. The preincubation TABLE 4. Effects and statistical analysis of dry matter production (DMP) and P uptake data from wheat after application of RP and P-solubilizing fungus to a field soil Sums of squares Data source df Grain Straw DMP DMP P uptake Total 49 62,543 139,730 1.79 Replicates 4 7,111 17,808 0.17 Treatment 9 42,226 73,870 1.24 Error 36 13,206 48,051 0.39 Contrasta S-RP-P. bilaji versus 1 11,764b 25,401b 0.38b control S-RP-P. bilaji versus 1 90 130 0.01 MAP S-RP-P. bilaji versus 1 5,429b 15,840b 0.16b S-RP S-RP-P. bilaji versus 1 2,856b 15,920b 0.19b RP-P. bilaji RP-P. bilaji versus 1 3,028b 1,103 0.03 control P. bilaji versus control 1 3,808b 3,478 0.14b RP versus control 1 40 10 0.01 MAP-P. bilaji versus 1 130 314 0.01 MAP a S, Straw; MAP, monoammonium phosphate. b Effect significant at P < 0.01.

VOL. 53, 1987 P-SOLUBILIZING FUNGUS INCREASES PLANT P UPTAKE 2703 of straw did not increase the effectiveness of the treatments with P. bilaji inoculation and, in fact, resulted in a yield reduction in the case of bean growth. Biological P solubilization of RP has been shown to be as effective as processed P fertilizers in calcareous soils under greenhouse and field conditions. Use of an RP-solubilizing system alleviates the need for increasing the rate of P addition when directly applying RPs. Further work is necessary to test the effectiveness of this system on a wider range of soils with a wider range of RPs and to elucidate the mechanisms involved. LITERATURE CITED 1. Ambler, J. R., and J. L. Young. 1977. Techniques for determining root length infected by vesicular-arbuscular mycorrhizae. Soil Sci. Soc. Am. J. 41:551-556. 2. Azcon, R., J. M. Barea, and D. S. Hayman. 1976. Utilization of rock phosphate in alkaline soils by plants inoculated with mycorrhizal fungi and phosphate-solubilizing bacteria. Soil Biol. Biochem. 8:135-138. 3. Banik, S., and B. K. Dey. 1982. Available phosphate content of an alluvial soil as influenced by inoculation of some isolated phosphate-solubilizing micro-organisms. Plant Soil 69:353-364. 4. Bray, R. H., and L. T. Kurtz. 1945. Determination of total, organic and available forms of phosphorus in soils. Soil Sci. 59:39-45. 5. Cabala-Rosand, P., and A. Wild. 1982. Direct use of low grade phosphate rock from Brazil as fertilizer. II. Effects of mycorrhizal inoculation and nitrogen source. Plant Soil 65:363-373. 6. Gaur, A. C., R. S. Mathur, and K. V. Sadasivam. 1980. Effect of organic materials and phosphate-dissolving culture on the yield of wheat and greengram. Indian J. Agron. 25:501-503. 7. Khalafallah, M. A., M. S. M. Saber, and H. K. Abd-EI-Maksoud. 1982. Influence of phosphate dissolving bacteria on the efficiency of superphosphate in a calcareous soil cultivated with Vicia faba. Z. Pflanzenernaehr. Bodenkd. 145:455-459. 8. Khan, J. A., and R. M. Bhatnagar. 1977. Studies on solubilization of insoluble phosphates by microorganisms. I. Solubilization of Indian phosphate rocks by Aspergillus niger and Penicillium sp. Fert. Technol. 14:329-333. 9. Kucey, R. M. N. 1983. Phosphate-solubilizing bacteria and fungi in various cultivated and virgin Alberta soils. Can. J. Soil Sci. 63:671-678. 10. Kucey, R. M. N., and J. B. Bole. 1984. Availability of phosphorus from 17 rock phosphates in moderately and weakly acidic soils as determined by P-32 dilution, A value and total P uptake methods. Soil Sci. 138:180-188. 11. Kucey, R. M. N., and E. A. Paul. 1983. Vesicular-arbuscular mycorrhizal spore populations in various Saskatchewan soils and the effect of inoculation with Glomus mosseae on faba bean growth in greenhouse and field trials. Can. J. Soil Sci. 63:87-95. 12. Olsen, S. R., C. V. Cole, F. S. Watanabe, and L. A. Dean. 1954. Estimation of available phosphorus in soils by extraction with sodium bicarbonate. U.S. Department of Agriculture Circular no. 939. Government Printing Office, Washington, D.C. 13. Powell, C. L. 1979. Effect of mycorrhizal fungi on recovery of phosphate fertilizer from soil by ryegrass plants. New Phytol. 83:681-694. 14. Rao, A. V., B. Venkateswarlu, and P. Kaul. 1982. Isolation of a phosphate dissolving soil actinomycete. Curr. Sci. 51:1117-1118. 15. Taha, S. M., S. A. Z. Mahmoud, A. Halim El-Damaty, and A. M. Abd El-Hafez. 1969. Activity of phosphate-dissolving bacteria in Egyptian soils. Plant Soil 31:149-160. 16. Waidyanatha, U. P., N. Yogaratnam, and W. A. Ariyaratne. 1979. Mycorrhizal infection on growth and nitrogen fixation of Pueraria and Stylosanthes and uptake of phosphorus from two rock phosphates. New Phytol. 82:147-152. 17. Ward, G. M., and F. B. Johnston (ed.). 1953. Chemical methods of plant analysis. Contribution no. 238. Chemical Division, Canada Department of Agriculture, Ottawa. Downloaded from http://aem.asm.org/ on September 6, 2018 by guest