Potential for enhancement of root growth and nodulation of soybean co-inoculated with Azospirillum and Bradyrhizobium in laboratory systems

Soil Biology & Biochemistry 33 (2001) 457±463

www.elsevier.com/locate/soilbio

Potential for enhancement of root growth and nodulation of soybean coinoculated with Azospirillum and Bradyrhizobium in laboratory systems A.H. Molla a,*, Z.H. Shamsuddin b, M.S. Halimi b, M. Morziah c, A.B. Puteh d a

Department of Crop Botany, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Gazipur 1703, Bangladesh b Department of Land Management, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor DE, Malaysia c Department of Biochemistry and Microbiology, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor DE, Malaysia d Department of Crop Science, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor DE, Malaysia Received 6 December 1999; received in revised form 11 April 2000; accepted 20 July 2000

Abstract The potential enhancement of root growth and nodulation in vegetable soybean (AGS190) was studied with application of Azospirillum brasilense (Sp7) and A. lipoferum (CCM3863) co-inoculated with two Bradyrhizobium japonicum strains (TAL102 and UPMR48). Signi®cant root growth stimulation and nodulation were observed in Azospirillum as well as during its co-inoculation with Bradyrhizobium. Nodule formation is linked with the initiation of new roots; nodules were almost absent even in Bradyrhizobium inoculated plant due to the absence of new roots development in clipped rooted seedlings. Total root length, root number, speci®c root length, root dry matter, root hair development and shoot dry matter were signi®cantly increased by Azospirillum alone and its co-inoculum. Co-inoculated plants signi®cantly in¯uenced the number of nodules and its fresh weight. A. brasilense seemed to perform better in root growth and nodule development compared to A. lipoferum. q 2001 Elsevier Science Ltd. All rights reserved. Keywords: Azospirillum; Bradyrhizobium; Co-inoculation; Root growth; Nodulation

1. Introduction Stimulation and enhancement of root hairs and root growth is one of the several methods of plant growth promotion by a group of bacteria commonly known as PGPR (Plant Growth Promoting Rhizobacteria). The PGPR has a close association with plant roots and can enhance the growth of many plants. The bene®ts of using Azospirillum, a common PGPR used as a biofertilizer for cereals and vegetable crops, have been well documented (Klucas et al., 1979; Okon and Kapulnik, 1986; Sarig et al., 1986; Burdman et al., 1997). Improved plant growth is attributed to Azospirillum through subsequent increase of lateral root number and root hair formation (Tien et al., 1979; Jain and Patriquin, 1985), water and mineral uptake (Okon and Kapulnik, 1986) and N2 ®xation (Bashan and Holguin, 1997). The legume±Rhizobium symbiosis is known as the * Corresponding author. Address for correspondence: Biochemical Engineering Laboratory, Department of Chemical and Environmental Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor DE, Malaysia. Tel.: 160-3-89413951; fax: 160-3-89488939. E-mail address: [email protected] (A.H. Molla).

most ef®cient system for biological nitrogen ®xation (BNF) through nodulation in legume roots, which help to reduce the application of inorganic N. Enhanced root system of legumes and infectivity of Bradyrhizobium have an important role in expediting this process. The young and appropriate new roots are one of the key factor for suf®cient infectivity by Bradyrhizobium in most of the legumes, because it becomes attached to new roots and root hairs, producing root hair curling followed by infection thread development for nodulation (Bellone et al., 1997). In co-inoculation, Azospirillum promoted epidermal cell differentiation in root hairs that increased the number of potential sites for Bradyrhizobial infection (Yahalom et al., 1990), as a result more nodules were developed (Andreeva et al., 1993). It may be due to Azospirillum ensuring the availability of appropriate type of roots for effective infection when co-inoculated with Bradyrhizobium in legumes. Several encouraging, but inconsistent results on nodulation, and N2 ®xation have been reported in different legumes in co-inoculation studies of Azospirillum and Bradyrhizobium (Iruthayathas et al., 1983; Sarig et al., 1986; Del Gallo and Fabbri, 1991; Burdman et al., 1997). Besides inhibition of nodulation, decreased infection thread developments were also

0038-0717/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S 0038-071 7(00)00186-3

458

A.H. Molla et al. / Soil Biology & Biochemistry 33 (2001) 457±463

reported in white clover (Planzinski and Rolfe, 1985) and Medicago (Itzigsohn et al., 1993) when co-inoculated with Azospirillum. In some cases nodulation in legumes by Azospirillum co-inoculated with Bradyrhizobium can be replaced by application of phytohormones (Volpin and Kapulnik, 1994), because phytohormones enhanced root growth and Azospirillum can synthesize these (Tien et al. 1979; Baca et al., 1994; Patten and Glick, 1996). This implies that Bradyrhizobium can attach themselves to the new roots to infect and nodulate, irrespective of the roots being initiated by Azospirillum or IAA or any other means. In spite of this, Bothe et al. (1992) reported that IAA has no effect on lateral root formation, but nitrite added either directly or excreted by Azospirillum in nitrate respiration cause a drastic change in lateral root formation in wheat. Therefore, the issues of increased root growth and enhanced nodulation in co-inoculation by Azospirillum still appear inconclusive. However, there is a need to conduct a series of research in a controlled and homogeneous environment under close observation to evaluate root growth promotion and nodulation by Azospirillum co-inoculated with Bradyrhizobium. This research was carried out to evaluate the role of Azospirillum in root growth and enhancement of nodulation when co-inoculated with Bradyrhizobium. 2. Materials and methods Two experiments were conducted, one in test tubes containing nutrient solution for evaluating the stimulation of root growth and nodulation, and the other in small pots of sand culture for SEM (Scanning Electron Micrograph) studies to observe root hair stimulation and development. N-free nutrient solution was used for the experiments in test tubes (15 cm £ 2.2 cm) with nine treatments: uninoculated (control), Sp7, CCM3863, TAL102, UPMR48, Sp7 & TAL102, Sp7 & UPMR48, CCM3863 & TAL102 and CCM3863 & UPMR48. Sand culture for SEM studies were conducted in disposable foam glasses (9 cm £ 7.5 cm £ 4.5 cm) with six treatments: control, Sp7, CCM3863, TAL102, Sp7 & TAL102 and CCM3863 & TAL102. Experiments were carried out in the glasshouse and the Soil Microbiology Laboratory of Universiti Putra Malaysia (UPM), Malaysia. During the period of experiments, the available average sunshine at the glasshouse was 7.5 h, the average maximum and minimum temperatures were 35 and 228C, respectively, and the relative humidity was 76%. 2.1. Bacteria and inocula preparation In the test tube experiment, two strains of Azospirillum (Sp7, Azospirillum brasilense and CCM3863, A. lipoferum) and two strains of Bradyrhizobium japonicum (TAL102 and UPMR48) were used. Azospirillum Sp7 & CCM3863, and

Bradyrhizobium TAL102 were used in the sand culture experiments. The strains Sp7 and CCM3863 were used from the laboratory stocks obtained from Dr Johanna DoÈbereiner, Embrapa National Centre for Agrobiol. Res., Seropedica 23851-970, Brazil, and Dr I. Sedlacek, Czechoslovak Collection of Microorganisms, Tvrdeho 14, Brno, CS 60200, Czechoslovakia, respectively. Bradyrhizobium strain TAL102 was obtained from the University of Hawaii, College of Tropical Agriculture & Human Resources, Department of Agronomy & Soil Science (NifTAL), Paia, Hawaii 96779, USA. UPMR48 was locally isolated from soybean nodules. Azospirillum and Bradyrhizobium were cultured for 24 and 72 h in Okon (Okon et al., 1977) and YM (Somasegaran and Hoben, 1985) broth, respectively, in a 250 ml Erlenmeyer ¯ask containing 100 ml nutrient media in a rotary shaker at 125 rpm and at 30 ^ 18C. The inocula used have an optical density of 0.5±0.6 for Azospirillum and 0.3±0.4 for Bradyrhizobium at 600 nm. 2.2. Growing seedlings One hundred ®fty healthy equal size seeds of vegetable soybean (variety AGS190 of AVRDC) were surface sterilized by washing for 1 min in 95% ethanol, followed by 0.1% HgCl2 and ®nally washed six times with sterile distilled water. The seedlings were sown in distilled water-moist sterilized sand in a plastic tray. A single 6day old seedling was sown in the test tube. The shoot was kept above the mouth of the test tube using a suitable sponge. Similar size seedlings were used after clipping of all lateral roots by sharp sterilized scissors to make it more homogeneous. Treatments were applied by soaking roots (after clipping) in inoculum for 30 min for single inoculation; for co-inoculation the roots were soaked in Azospirillum suspension for 30 min followed by soaking for another 30 min in Bradyrhizobium. Finally, 1 ml of inoculum was applied in each test tube before planting, in co-inoculated tubes 2 ml (one from each bacterial inoculum) were applied. 2.3. Nutrient solution N-free nutrient solution was used according to Shamsuddin (1987). The solution contained (l 21) 1.00 ml of 1 M KH2PO4, 2.66 ml of 0.5 M K2SO4 and 1.00 ml of 1 M CaCl2´H2O; 1.66 ml of nutrient elements contained [(g l 21) MgSO4´7H2O, 13.4; MnSO4´4H2O, 8.845; ZnSO4´7H2O, 1.41; CuSO4´5H2O, 0.785; H3BO3, 0.475; CoSO4´7H2O, 0.24; Na2MoO4´2H2O, 0.12] and 1.66 ml of 3 mM of H Fe EDTA. The pH of the solution was adjusted to 6.5 with 1 M NaOH. A 50-ml sterile solution was placed in each test tube and was changed every 72 h by sterilized syringes for maintaining a favourable environment for bacterial growth and nutrient uptake by roots, otherwise the pH might be changed to acidic resulting in an ionic imbalance.

A.H. Molla et al. / Soil Biology & Biochemistry 33 (2001) 457±463

2.4. Root length measurement The plants were harvested at 42 days after sowing (DAS) and the root length measured by grid lines intersections method proposed by Newman (1966) and Tennant (1975), the proposed equation was L ˆ pND=4; where L is the root length, N the number of intersects and D the grid size. 2.5. Root growth study under scanning electron microscope (SEM) Plants were grown in sterilized sand (275 g) media in disposable foam glasses. Surface sterilized sprouted seeds were sown after applying proper treatments. Two-thirds of the main root (subtracting the basal part) containing root hair and elongation zones were sampled at 7 DAS and washed carefully with distilled water. The roots were immersed in 4% glutaraldehyde for 24 h at 48C followed by washing three times with 0.1 M cacodylate buffer (10 min for each washing). In the next step, the roots were immersed in osmium-buffer solution (50% each of 2% osmium and 0.2 M cacodylate buffer, v/v) for 2 h at 48C. The buffer solution was drained and was again immersed in 0.1 M cacodylate buffer and kept overnight at 48C. The samples were then dehydrated serially with acetone in different concentrations [35, 50, 75, 95% at each step for 10 min and ®nally with 100% (thrice) for 15 min] and dried in critical point of drying (CPD). Then suitable sized samples were mounted on aluminium stubs with carboncoated tape followed by coating with gold palladium in a Polaroid sputter coater and ®nally viewed by SEM and the photographs taken. 2.6. Experimental design and statistical analysis Both experiments were conducted using a completely randomized design (CRD). Test tube experiment and sand culture experiment in pots were replicated 10 and eight times, respectively. An uninoculated control was included in each replication. Analyses of variance and comparison means were done separately by the Statistical Analysis System (SAS, 1987) package. Data were tested by Duncan's Multiple Range Test (DMRT). In each group of data different letters showed a statistically signi®cant difference at the P # 0:05 level. Increments in root dry matter (RDM), root volume (RV), speci®c root length (SRL), lateral root number (LRN) and total root length (TRL) were expressed as percentage compared to control. 3. Results In the test tube experiment, A. brasilense (Sp7), A. lipoferum (CCM3863) and B. japonicum (TAL102 and UPMR48) were used as single and mixed inoculants. The results showed that Sp7 alone or in combination with UPMR48 gave signi®cantly higher …P # 0:05† RDM, RV

459

and shoot dry matter (SDM) than other treatments. Speci®c root length was observed to be signi®cantly higher …P # 0:05† in all Azospirillum and its co-inoculated treatments than other treatments, and the highest value (20.9) was recorded in CCM3863 but not signi®cantly different from other treatments of Azospirillum and its co-inoculated plants (Table 1). Sp7 used as a single inoculant showed 63.4% increase in RDM, 121.3% increase in RV and 575.0% increase in SRL than the control plants. In combination with UPMR48, the enhanced results were 45.0, 89.4 and 670.2% for RDM, RV and SRL, respectively (Table 1). The highest total root length (TRL) (124.0 cm/plant) was recorded in Sp7 alone as well as during its co-inoculation with UPMR48 followed by CCM3863 and its co-inoculated plants. The highest lateral roots number (LRN) was in Sp7 (58.0 roots/plant) followed by CCM3863 (50.0 roots/plant). CCM3863 with TAL102 produced 45.0 roots/plant and in Sp7 with TAL102 it was 43.0 roots/plant, while the control treatment showed TRL and LRN of 11.4 cm/plant and 4.8 roots/plant, respectively (Table 1). Enhancement of root development led to a better nodulation when UPMR48 was co-inoculated with Sp7, where the nodules number (NN) and fresh nodule weight (FNW) were observed to be 9.5 nodules/plant and 64.0 mg nodules/plant, respectively. This was followed by co-inoculation of UPMR48 and CCM3863 in which the results were 9.0 nodules/plant for NN and 54 mg nodules/plant for FNW. But the highest nodules number/plant (11.3) was recorded in TAL102 with Sp7 and its fresh weight was 44.3 mg. Nodules were almost absent in Bradyrhizobium-inoculated plants, UPMR48 treated plants produced only 1.3 nodules/ plant (Fig. 1). In the SEM study, intensive root hair stimulation was observed in Sp7 followed by its co-inoculation with TAL102 compared to control and TAL102 treated plants. A similar trend of root hair development was also observed in CCM3863 and its co-inoculation with TAL102, but higher results were attributed in treatments where inoculating was with Sp7 than CCM3863 (Fig. 3). Clipping roots of seedlings did not initiate new roots in control treatment but a signi®cant number of roots were regenerated in Azospirillum treated plants (Fig. 2). 4. Discussion Root growth such as lateral root number, total root length and root dry matter showed the highest number in Sp7 inoculated and Sp7 co-inoculated with UPMR48 treatments (Table 1). The values were lower to some extent in CCM3863 probably due to the different bacterial species. In co-inoculation, values decreased slightly compared to single inoculation of Azospirillum due to competition problems on the root surface for existence (Del Gallo and Fabbri, 1991). Enhanced effects on the growth of wheat roots were reported by Azospirillum inoculation even in

000.0 988.0 950.0 20.4 172.0 895.0 988.0 736.7 654.0 0000.0 1108.4 941.7 40.6 431.2 795.8 790.6 837.5 582.3 000.0 575.0 829.0 18.2 189.8 710.2 670.2 628.4 602.7 00.0 121.3 61.7 00.0 23.4 55.3 89.4 48.9 44.7 00.0 63.4 23.3 06.9 03.9 24.7 45.0 16.3 07.9 44.3abc 57.8a 52.1ab 34.9c 39.0bc 39.7bc 49.1abc 48.1abc 36.2bc 29.5ab 33.4a 27.7abc 27.5abc 29.5ab 21.5c 24.1bc 22.2bc 25.5bc 275.0b 377.5a 305.0b 257.5b 260.0b 302.5b 325.0ab 295.0b 295.0b 02.25b 15.19a 20.90a 02.66b 06.52b 18.23a 17.33a 16.39a 15.81a 0.47d 1.04a 0.76bc 0.47d 0.58dc 0.73bc 0.89ab 0.70c 0.68c 50.50c 82.50a 62.25b 54.00bc 52.50c 63.00b 73.25a 58.75bc 54.50bc 11.40c 124.00a 119.71ab 13.73c 31.01c 113.43ab 124.00a 95.38ab 85.96b Uninoculated Sp7 CCM3863 TAL102 UPMR48 Sp7 & TAL120 Sp7 & UPMR48 CCM3863 & TAL102 CCM3863 & UPMR48

4.80e 58.00a 50.00ab 6.75e 25.50d 43.00bc 42.75bc 45.00abc 32.75cd

TRL LRN SRL RV RDM

Total root length (cm) Treatments

Lateral root number

Root dry matter (mg)

Root volume (cm 3)

Speci®c root length (mm/m)

Shoot dry matter (mg)

Plant height (cm)

Leaf area (cm 2)

Increment in percent compared to control

A.H. Molla et al. / Soil Biology & Biochemistry 33 (2001) 457±463 Table 1 Root and shoot characters of vegetable soybean (per plant) at 42 DAS, inoculated with Azospirillum and Bradyrhizobium as single and co-inoculation. Different letters in column indicate signi®cant differences at P # 0:05 by DMRT (RDM ˆ root dry matter, RV ˆ root volume, SRL ˆ speci®c root length, LRN ˆ lateral root number and TRL ˆ total root length)

460

hydrophonic culture (Kapulnik et al., 1985a,b). In test tubes culture, increased root numbers in wheat seedlings by Azospirillum were also reported by Zimmer and Bothe (1989). High regeneration of roots in clipped root of Azospirillum treated plants might be the effect of growth promoting biochemicals like IAA, synthesized by this bacteria (Jain and Patriquin, 1985). It was also reported that Azospirillum can produce phytohormones IAA (mostly), gibberellins, cytokinins (Tien et al., 1979; Zimmer and Bothe, 1989; Bashan and Levanony, 1990; Rademacher, 1994; Patten and Glick, 1996) in in-vitro culture and have an important role in enhancing root growth. In addition, Bashan and Holguin (1997) cited that A. brasilense Cd synthesize IAA rapidly with the beginning of the stationary phase in liquid culture. Since higher performance by Sp7 (A. brasilense) would be the result of its potential to induce more biochemicals than others. Nodules were observed only in co-inoculated plants (Fig. 2). Bradyrhizobium alone did not produce nodules indicating that those plants have no roots or root hairs for infection, i.e. Bradyrhizobium treated plants did not regenerate any appropriate roots and root hairs (Figs. 2 and 3). It was also observed that the root hairs are positively correlated with nodulation by Bradyrhizobium (Bellone et al, 1997). Azospirillum application with Bradyrhizobium altered the situation and induced root growth leading to nodulation. These results implied that synergistic action of Azospirillum and Bradyrhizobium would enhance root development, which was favourable to better nodulation of vegetable soybean. Azospirillum in¯uenced nodulation in chickpea by anticipating nodule formation, increasing nodule number, size and dry weight (Del Gallo and Fabbri, 1991), higher nodules number were also reported at low N level in common bean in co-inoculation treatment rather than Rhizobium alone (Burdman et al., 1997). Infectivity by Bradyrhizobium might have a positive relationship with initiation of new roots. In the present study new roots development as well as nodulation were almost absent in Bradyrhizobium inoculated plants, but co-inoculated plants possessed signi®cant number of nodules with its increased fresh weight and roots number. Crown nodules might be the result of infection by Bradyrhizobium at an early stage of root initiation which nodulated without extending root length. In the SEM studies, intensive root hair development was only shown for Azospirillum inoculated plants rather than when co-inoculated with Bradyrhizobium (Fig. 3). It also appeared that the higher root length was only observed in Azospirillum treated plants. Apparently denser and longer root hairs were recorded in wheat and sorghum by Azospirillum inoculation than dead cells plants (Okon and Kapulnik, 1986). Similar results have also been reported in SEM studies for clover roots after Azospirillum coinoculation with Rhizobium (Planzinski and Rolfe, 1985). Speci®c root length implies the root length in unit dry weight of root that explain root diameter. However, the

A.H. Molla et al. / Soil Biology & Biochemistry 33 (2001) 457±463

461

Fig. 1. Effect of co-inoculation of Azospirillum and Bradyrhizobium on nodule number (NN) and fresh nodule weight (FNW, mg) per plant of vegetable soybean.

Fig. 2. Stimulation of root growth and nodulation of vegetable soybean in test tubes by co-inoculation with Azospirillum and Bradyrhizobium: (a) at 14 DAS, root stimulation (from left to right test tubes were uninoculated (-Inoc), Sp7, CCM3863 & TAL102), and (b) at 42 DAS nodulation [from left to right test tubes were, in single inoculation (1Bradyrhizobium), ®rst & second tubes Ð TAL102, third Ð UPMR48 and in co-inoculation (1Azos. 1 Bradyrhizo.), ®rst Ð Sp7 & TAL102, second Ð Sp7 & UPMR48 and third Ð CCM3863 & UPMR48].

462

A.H. Molla et al. / Soil Biology & Biochemistry 33 (2001) 457±463

Fig. 3. Scanning electron micrograph of root hair development of vegetable soybean (AGS190): (a) uninoculated; (b) inoculated with Bradyrhizobium (TAL102; (c) inoculated with Azospirillum (Sp7); (d) inoculated with Azospirillum (CCM3863); (e) co-inoculated (Sp7 & TAL102); and (f) co-inoculated (CCM3863 & TAL102).

signi®cant results of this parameter were obtained in CCM3863 and its co-inoculated plants rather than Bradyrhizobium and control treatments. Between the two Azospirillum species, the performance of A. brasilense (Sp7) in enhancing root growth of vegetable soybean was more encouraging than A. lipoferum (CCM3863). In co-inoculation, decreased shoot growth was reported

by Del Gallo and Fabbri (1991) and was explained as the result of competition of bacteria for food and survival. In our results, plant top parameters were not in¯uenced by Azospirillum inoculation, since it decreased to some extent when co-inoculated compared with control except for shoot dry matter, which supported the results of Del Gallo and Fabbri (1991) and Burdman et al. (1997).

A.H. Molla et al. / Soil Biology & Biochemistry 33 (2001) 457±463

5. Conclusions This laboratory study has shown that Azospirillum has the potential to signi®cantly stimulate root growth even in clipped rooted plants, implying that it has a positive in¯uence on root growth and development. But in co-inoculation with Bradyrhizobium, the decline in root growth stimulation might be the result of competition for survival. Azospirillum not only in¯uences root growth, but can also improve nodule initiation and development in vegetable soybean by co-inoculation with Bradyrhizobium. This study also implies that the strain Sp7 may play an encouraging role than CCM3863 in root growth and root hair development and also nodule initiation in co-inoculation with Bradyrhizobium in legumes. Acknowledgements The authors express their gratitude to the Ministry of Science, Environment and Technology, Malaysia, for providing funds for this study. The authors also acknowledge the help of the Institute of Bioscience, UPM, for providing facilities and cooperation for SEM work. References Andreeva, L.P., Red'kina, T.V., Ismailov, S.F., 1993. The involvement of indoleacetic acid in the stimulation of Rhizobium-legume symbiosis by Azospirillum brasilense. Russian Journal of Plant Physiology 40, 901± 906. Baca, B.E., Soto-Urzua, L., Xochihua-Corona, Y.G., Cuervo-Garcia, A., 1994. Characterization of two aromatic amino acid aminotransferases and production of indoleacetic acid in Azospirillum stains. Soil Biology & Biochemistry 26, 63±67. Bashan, Y., Holguin, G., 1997. Azospirillum-plant relationships: environmental and physiological advances (1990±1996). Canadian Journal of Microbiology 43, 103±121. Bashan, Y., Levanony, H., 1990. Current status of Azospirillum inoculation technology: Azospirillum as a challenge for agriculture. Canadian Journal of Microbiology 36, 591±608. Bellone, C.H., De Bellone, S.D.V.C., Pedraza, R.O., Monzon, M.A., 1997. Cell colonization and infection thread formation in sugar cane roots by Acetobacter diazotrophicus. Soil Biology & Biochemistry 29, 965± 967. Bothe, H., Korsgen, H., Lehmacher, T., Hundeshagen, B., 1992. Differential effects of Azospirillum, auxin and combined nitrogen on the growth of the roots of wheat. Symbiosis 13, 167±179. Burdman, S., Kigel, J., Okon, Y., 1997. Effects of Azospirillum brasilense on nodulation and growth of common bean (Phaseolus vulgaris L.). Soil Biology & Biochemistry 29, 923±929. Del Gallo, M., Fabbri, P., 1991. Effect of soil organic matter on chickpea inoculated with Azospirillum brasilense and Rhizobium leguminosarum bv. ciceri. Plant and Soil 137, 171±175. Iruthayathas, E.E., Gunasekaran, S., Vlassak, K., 1983. Effect of combined inoculation of Azospirillum and Rhizobium on nodulation and N2-

463

®xation of winged bean and soybean. Scientica Horticulturae 20, 231±240. Itzigsohn, R., Kapulnik, Y., Okon, Y., Dovrat, A., 1993. Physiological and morphological aspects of interactions between Rhizobium meliloti and alfalfa (Medicago sativa) in association with Azospirillum brasilense. Canadian Journal of Microbiology 39, 610±615. Jain, D.K., Patriquin, D.G., 1985. Characterization of a substance produces by Azospirillum, which causes branching of wheat root hairs. Canadian Journal of Microbiology 31, 206±210. Kapulnik, Y., Gagny, R., Okon, Y., 1985a. Effects of Azospirillum spp. inoculation on root development and NO32 uptake in wheat (Triticum aestivum cv. Miriam) in hydrophonic systems. Canadian Journal of Botany 63, 627±631. Kapulnik, Y., Okon, Y., Henis, Y., 1985b. Changes in root morphology of wheat caused by Azospirillum inoculation. Canadian Journal of Microbiology 31, 881±887. Klucas, R.V., Pederson, W., Shearman, R.C., Wood, U., 1979. Nitrogen ®xation associated with winter wheat, sorghum and Kentucky bluegrass. In: Vose, P.B., Ruschel, A.P. (Eds.). Associative N2-Fixation, vol. 1. CRC Press, Boca Raton, FL, pp. 119±128 (Proceedings of the International Workshop on Associative N2 Fixation, University of Sao Paulo). Newman, E.I., 1966. A method for estimating the total length of root in a sample. Journal of Applied Ecology 3, 139±145. Okon, Y., Albrecht, S.L., Burris, R.H., 1977. Methods for growing Spirillum lipoferum and for counting it in pure culture and in association with plants. Applied and Environmental Microbiology 33, 85±88. Okon, Y., Kapulnik, Y., 1986. Development and function of Azospirilluminoculated roots. Plant and Soil 90, 3±16. Patten, C.L., Glick, B.R., 1996. Bacterial biosynthesis of indole-3-acetic acid. Canadian Journal of Microbiology 42, 207±220. Planzinski, J., Rolfe, B.G., 1985. Interaction of Azospirillum and Rhizobium strains leading to inhibition of nodulation. Applied and Environmental Microbiology 49, 990±993. Rademacher, W., 1994. Gibberellin formation in microorganisms. Plant Growth Regulator 15, 303±314. Sarig, S., Kapulnik, Y., Okon, Y., 1986. Effect of Azospirillum inoculation on nitrogen ®xation and growth of several winter legumes. Plant and Soil 90, 335±342. SAS, 1987. SAS/STAT Guide to Personal Computers, Version 6. SAS Institute Inc., Cary, North Carolina, USA. Shamsuddin, Z.H., 1987. Growth, infectivity and nodulating abilities of some winged bean rhizobia in acid conditions. PhD thesis, Murdoch University, Perth, Australia. Somasegaran, P., Hoben, H.J., 1985. Methods in Legume-Rhizobium Technology. University of Hawaii (NifTAL), p. 273. Tennant, D., 1975. A test of a modi®ed line intersect method of estimating root length. Journal of Ecology 63, 995±1001. Tien, T.M., Gaskins, M.H., Hubell, D.H., 1979. Plant growth substances produced by Azospirillum brasilense and their effects on the growth of pearl millet (Pennisetum americanum L.). Applied and Environmental Microbiology 37, 1016±1024. Volpin, H., Kapulnik, Y., 1994. Interaction of Azospirillum with bene®cial soil microorganisms. In: Okon, Y. (Ed.). Azospirillum/Plant Associations, CRC Press, Boca Raton, FL, pp. 111±118. Yahalom, E., Okon, Y., Dovrat, A., 1990. Possible mode of action of Azospirillum brasilense strain Cd on the root morphology and nodule formation in burr medic (Medicago polymorpha). Canadian Journal of Microbiology 36, 10±14. Zimmer, W., Bothe, H., 1989. The phytohormonal interactions between Azospirillum and wheat. In: Nitrogen Fixation with Non-legumes. Kluwer, Dordrecht, pp. 137±145.