The objective of this study was to assess the feasibility of meeting …

• Abstract
• Introduction
• Materials and methods
• Results
• Discussion
• Conclusions
• References
• Tables

Biology Articles » Agriculture » Plant Production » Tomato Crop Response to Short-Duration Legume Green Manures in Tropical Vegetable Systems » Materials and methods

Materials and methods

- Tomato Crop Response to Short-Duration Legume Green Manures in Tropical Vegetable Systems

 

Field Trials
Six field experiments in major vegetable growing areas of Taiwan and the Philippines were conducted between 1993 and 1995: four experiments on the experiment farm of the Asian Vegetable Research and Development Center (AVRDC) in southern Taiwan and one each in the north and south of the Philippines. At AVRDC, two experiments were conducted during the wet season of 1993, with average air temperatures of 27.6°C and a total rainfall (6-mo experiment period) of 1348 mm. Two experiments were conducted in the dry season of 1993–1994, with average air temperatures of 21.6°C and a total rainfall (6 mo) of 52.9 mm. The cropping history of the field used for the 20 yr prior to our experiments was a rotation of vegetables grown in the DS and flooded rice grown in the WS. The soil is of the Take series (loamy, mixed, hyperthermic Fluvaquentic Entochrepts), with pH (H₂O) of 8.2, total Kjeldahl N of 0.7 g kg^-1 (Bremner, 1965), and total C 6.4 g kg^-1 (Walkley–Black method). Prior to the first run of our experiments, the field area was cropped with maize for about 1 mo, to obtain a homogeneous soil mineral N distribution. Maize stubble was removed before the trial started.

In the Philippines, the first experiment was conducted on the experiment farm of the Mariano Marcos State University (MMSU) in Batac, Ilocos Norte (IRRI rainfed lowland consortium site). The rainfed lowlands of the province Ilocos Norte are characterized by intensive cropping systems, although soil fertility and rainfall distribution appear unfavorable (Tripathi, 1995). Rice is grown during the wet season and upland crops (legumes, maize, and vegetables) are grown in the dry season. Average air temperature during the DS experiment period of 6 mo was 27.3°C (max. 33°C, min. 20°C). After strong rainfall events in October (175 mm), no more rainfall occurred within the 6-mo experiment period. The soil is a Fluvaquentic Ustropept (fine-silty, mixed isohyperthermic), with pH (H₂O) of 8.1, total Kjeldahl N of 0.7 g kg^-1 (Bremner, 1965), and total C of 5.9 g kg^-1 (Walkley–Black method). To obtain a homogeneous soil mineral N distribution, this soil has been previously cropped to one rice crop without N fertilizer application. Rice straw was removed from the field before the trail started.

The second experiment in the Philippines was conducted in collaboration with the tomato processing company Bukidnon Resources Company, Inc. (BRCI), at their experiment farm in San Juan, Bukidnon, Mindanao. An extensive subsistence farming is practiced in this area. Soils in Bukidnon are rich in organic matter and of volcanic origin. Average air temperatures during the 6 mo were 24.1°C (max. 28.3°C, min. 19.8°C). Total rainfall during the experiment period was 1471 mm, with an average daily rainfall of 25 mm. The soil is a clayey, kaolinitic, hyperthermic Ultisol, with pH (H₂O) of 5.7 (after liming with 5 Mg ha^-1 CaCO₃), total Kjeldahl N of 2.1 g kg^-1 (Bremner, 1965), and total C of 19.5 g kg^-1 (Walkley–Black method).

Experiment Design
Asian Vegetable Research and Development Center Experiments
Experiments were run simultaneously on two fields, each with different bed systems; raised or low beds. The raised beds were 45 cm high and 2 m wide, with 2-m furrows between the beds. The furrows were sown with rice (O. sativa L.) and permanently flooded. The low beds were 20 cm high and 2 m wide, with 50-cm-wide irrigation furrows between beds. Both experiments (raised and low beds) were adjacent, such that the soil type, the cropping history, and meteorological conditions were the same. The experiment design for each experiment was a randomized complete block. Treatment plots measured 2 by 6 m, with four replicates. The eight treatments were (Table 1) two legume species, two green manure management (mulch and incorporation), and four controls having weed-free fallow (instead of legumes) and receiving 0, 30, 60, or 120 kg N ha^-1 applied to the tomato crop. In the DS treatment, plots were split into three equal-sized subplots prior to planting the vegetable crop to study N release in planted (tomato; cabbage, Braissica oleracea var. capitata L.) vs. unplanted plots (Thönnissen Michel, 1996).

 
Mariano Marcus State University Experiment
Experimental design was a randomized complete block with split plots and four replicates. Main treatment plots were 6 by 6 m; subtreatment plots were 2 by 6 m. The eight main treatments were similar to those at AVRDC (Table 1). In contrast to AVRDC experiments, N rates (0, 38, 75, or 150 kg N ha^-1 applied to the tomato crop) in fertilizer N treatments were adjusted to local recommendations. Tomato planting time was varied in two of the subtreatments, early and late transplanting: transplanting tomato plants immediately after vs. 2 wk after GM application. The third subplot was kept unplanted for inorganic N measurements (Thönnissen Michel, 1996).

Bukidnon Resources Company, Inc., Experiment
Experimental design was a randomized complete block with split plots and three replicates. Main treatment plots were 4.8 by 4.6 m; subtreatment plots grown to tomato were 4.8 by 3 m and subtreatments kept unplanted were 4.8 by 1.5 m. The eight treatments were comparable to those at AVRDC and MMSU, with the difference that mungbean, which is locally grown, replaced indigofera at BRCI (Table 1).

Green Manure and Tomato Crop
Legumes and tomato crops were grown in a 6-mo experimental cropping pattern (Table 2) . Legumes were inoculated with a Rhizobium strain mixture, specific for each legume. Bacterial strains were provided by the Soil Science Department of the Chung Hsing University in Taichung, Taiwan for AVRDC trials, the Soil Microbiology Unit at IRRI for the MMSU trial, and the Department of Agriculture for the BRCI trial. Legumes were hand-sown at 80 seeds m^-2 for soybean and 1.32 g m^-2 for indigofera at AVRDC and MMSU. Soybean and mungbean were hand-sown at 55 seeds m^-2 at BRCI. Phosphorus at 35 kg P ha^-1 as superphosphate and K at 83 kg K ha^-1 as KCl were broadcast in all beds before legume sowing at all three locations. At BRCI, soil was limed at a rate of 5 Mg ha^-1 CaCO₃. From 60 to 74 d after sowing, legumes were cut at ground level, chopped into 5- to 10-cm pieces, and either incorporated by rototilling to the 15-cm depth or left as mulch on the soil surface, as required for each treatment (Table 2). At BRCI, legumes for the incorporation treatment were left on the soil surface for 1 wk before incorporation.

 
Tomato (AVRDC accession no 5915-93-1-0-3, a short-duration, determinate bushy type at AVRDC, Northern Food Corporation (NFC-line) at MMSU and BRCI variety 1403, selection 1584 at BRCI) seedlings (34 d old at AVRDC, 24 d old at MMSU, and 14 d old at BRCI) were transplanted in two rows per bed (40 cm plant to plant and 100 to 150 cm between rows). Nitrogen fertilizer was applied and split in basal at tomato transplanting and two side dressings, 2 wk and 4 to 5 wk after transplanting. At AVRDC, 30 kg N ha^-1 was applied to Ck120, Ck60, and Ck30 as basal N application, 30 kg N ha^-1 to Ck120, Ck60 for the first side dressing, and 60 kg N ha^-1 to Ck120 for the second side dressing. At MMSU, N basal fertilizer application were 48, 24, and 12 kg N ha^-1 for Ck120 (normal), Ck60 (one-half), and Ck30 (one-quarter), respectively. Nitrogen at 60, 30, and 15.5 kg ha^-1 was applied to Ck120, Ck60, and Ck30, respectively, for the first side dressing and at 42, 21, and 10.5 kg N ha^-1 to Ck120, Ck60, and Ck30, respectively, for the second side dressing. At BRCI, 2000 kg ha^-1 of poultry manure (Gallus sp.) containing 2% N, 1.3% P, and 2.1% K w/w was applied immediately before transplanting tomato plants in all treatments (Table 2). For N fertilizer treatments, 30 kg N ha^-1 was applied to Ck30 and Ck60 and 60 kg N ha^-1 to Ck120 for the first side dressing, and 30 and 60 kg N ha^-1 for Ck60 and Ck120, respectively, for the second side dressing. The tomato crop was harvested, depending on season and variety, at 80 to 120 d after transplanting (Table 2). At AVRDC, maize was sown 2 d after final tomato harvest in six rows per bed (30 seeds m^-2) and sampled at 30 d (Table 1).

Plant Analysis
Legumes were sampled at 60 to 74 d (Table 2). Plants from 0.5 m² at AVRDC and BRCI and from a microplot of 0.64 m² at MMSU (see ¹⁵N experiment) of each treatment replicate, which was afterwards excluded from further sampling, were carefully dug out to a depth of 15 to 20 cm and the soil was separated from the roots. Shoots, roots, and nodules were dried at 60°C for 72 h and weighed for biomass determination. Plant samples were ground in a Wiley Laboratory Mill Model 4 (Thomas Scientific, Philadelphia) to pass through a 1-mm sieve, subsampled, and ground again in a vibrating sample mill (Heiko T1-100, Heiko Seisakusho Ltd., Tokyo, Japan). Nitrogen content in shoots and roots including nodules were determined by the Kjeldahl distillation method (Bremner, 1965).

At tomato harvest, marketable fruit fresh weight, and fresh and dry weights and N content of tomato fruits and plants were determined. Maize plants (30 d; including roots) were pulled out from the soil and biomass and total N were determined as a relative indicator of the inorganic N available in the soil after tomato harvest.

Plant Petiole Sap Nitrate Analysis
Tomato was sampled weekly for plant petiole sap nitrate content (sap N) at BRCI between 0600 and 1000 h, 2 d after the second weekly irrigation. Sampling took place from 42 d after GM application (9 Aug. 1995) to 91 d after GM application (26 Sept. 1995). The fifth leaf (counted from the top) of five randomly selected plants per treatment was collected, in order to sample the most recently matured leaf (Drews and Fischer, 1989). Petioles were chopped into 1-cm pieces and squeezed with a garlic press. Petiole sap was diluted by 50 times with distilled water, and mixed thoroughly for 1 min. One drop of this solution was poured onto two reaction zones of Reflectoquant nitrate test strips, and sap N was determined by refractometric reading on the RQ-flex instrument (RQflex, Merck, Darmstadt, Germany).

Nitrogen Fixation and Nitrogen-15 Experiment
Biological Nitrogen Fixation
The amount of N acquired through biological N₂ fixation by legumes was estimated using the N difference method (Talbott et al., 1985). Legumes and reference plants were grown in a small experiment conducted in parallel in a field adjacent to the main field experiment. Seeding rates and harvest dates were the same as those for legumes in the main field experiment. A nonnodulating soybean line (provided by the NifTAL Project, Hawaii) was used as a non-N₂–fixing reference plant for soybean and an upland rice variety (IF 600 80-46A) as a reference for indigofera. Plants were grown on 8-m² plots, with two replicates.

Enriched Nitrogen-15 Balance in Tomato
Production of Nitrogen-15 Labeled Legume Plant Material
Legumes were enriched with ¹⁵N by foliar application (Zebarth et al., 1991) of 1 0.5% urea (30 atom % ¹⁵N) solution at a total rate of 10 kg N ha^-1. The ¹⁵N fertilizer was split for progressive foliar applications at 21, 28, 35, 42, and 48 d.

Application of Nitrogen-15 Labeled Legume Material as Green Manure
One day before legume harvest, metal frames measuring 0.8 by 0.8 by 0.3 m (length by width by height) (microplots) were pushed into the soil to a depth of 25 cm in soybean and indigorera incorporation and mulch treatments of the main field experiment. Legumes within the metal frames were removed, including roots. On the same day, ¹⁵N-labeled legumes were carefully dug out to a depth of 20 cm, the soil was separated from the roots, and the legumes were chopped into 5- to 10-cm pieces and applied (incorporated or as mulch) to microplots of the corresponding treatments in the main field experiment.

Legume Nitrogen-15 Recovery in Vegetables
Two tomato seedlings were transplanted into each microplot. Tomato fruit yield and plant biomass, N and ¹⁵N content of tomato fruits and plant were determined to calculate ¹⁵N recovery in the tomato plant. Percent N in tomato derived from soybean (S¹⁵N) was calculated using

S15N = a/b x 100

(1)

where a is the ¹⁵N atom % excess of tomato and b is the ¹⁵N atom % excess of soybean. Recovery of soybean ¹⁵N by tomato was calculated using

Recovery of soybean - 15N by tomato =  (S15N x TN)/AN

(2)

where TN is the total N uptake by tomato and AN is the amount of soybean N applied (Harris and Hesterman, 1990).

Statistical Analysis and Presentation of Yield Data
Data were analyzed by analysis of variance (ANOVA) procedure using JMP Version 2 (SAS Inst., 1989) and SAS version 6.03 (SAS Inst., 1991). Yields and N accumulations of legumes, tomato crops, and maize of the raised beds are presented on the basis of planted area, without allowance for space occupied by rice.