Legumes
Soybean grown for 60 to 74 d accumulated a minimum of 2.8 Mg total biomass ha-1 and 106 kg N ha-1 in all locations and seasons in Taiwan and the Philippines (Table 3) . A maximum of 5.9 Mg biomass ha-1 and 140 kg N ha-1 was achieved in the wet season at AVRDC. Indigofera yields were more variable and always inferior to those of soybean. Mungbean biomass and N accumulation at BRCI were comparable to average indigofera yields of 0.9 Mg biomass ha-1 and 25 kg N ha-1. Soybean biomass in the wet season at AVRDC was nearly double that produced in the dry season, while indigofera biomass production was less affected by the season. At AVRDC, indigofera produced greater biomass yields on raised beds, while soybean biomass production was not influenced by the bed system. More than 90% of the legume N in 60- to 74-d soybean and indigofera was found in the shoot.
Soybean grown at MMSU derived 84.4% of the N accumulated in its plant biomass from biological N
2 fixation, compared with indigofera at 71.8%.
Tomato Yield, Nitrogen Uptake
Tomato yield in response to GM management and N fertilizer rates differed depending on season and location (Table 4) . Greatest tomato yields of 60 to 70 Mg ha-1 were achieved with 150 kg N ha-1 at MMSU and 120 kg N ha-1 at AVRDC in the DS. High tomato yields (40 Mg ha-1) were obtained in the control in the DS at AVRDC and at BRCI although no N was applied. The addition of 38 and 30 kg N ha-1 doubled tomato yields compared with the control at MMSU and in the WS at AVRDC, respectively, while 30 kg N ha-1 increased yields only by 13% in the DS at AVRDC, and had no effect at BRCI. A strong seasonal effect occurred at AVRDC, as DS yields were 10 times greater than those of the WS. Tomato yields responded lineraly to fertilizer N applications of 38 to 150 kg N ha-1 at MMSU (r2 = 0.81**), and 30 to 120 kg N ha-1 in the DS at AVRDC (low beds: r2 = 0.69**; raised beds: r2 = 0.54*). Simple regression coefficients between tomato yields and applied fertilizer N were significant neither in the WS at AVRDC nor at BRCI. With the exception of MMSU (r2 = 0.66**) tomato yields did not correlate with GM N amendments.
Tomato crops grown under adverse and suboptimal conditions in the hot tropical WS were able to use N released from GM (soybean mulch and indigofera incorporation) to produce yields comparable to those reached with 120 kg N ha
-1 applied as fertilizer. The effect of GM management (incorporation vs. mulch) on tomato yields differed with season and locations: tomato yields were greater with incorporated GM in the DS at AVRDC and in the WS at AVRDC with mulched GM; no differences of GM management on tomato yields were found at MMSU and BRCI.
Tomato N was doubled with soybean GM compared with the control in raised beds in the wet season at AVRDC and at MMSU, comparing favorably with that with 120 kg N ha-1 at AVRDC and to that with 38 kg N ha-1 at MMSU (Table 5) . In the DS at AVRDC and BRCI, tomato N was not increased by green manuring compared with the control. Tomato N was correlated with fertilizer N applied at MMSU (r2 = 0.90**), in the DS (low beds: r2 = 0.71**; raised beds: r2 = 0.55**) and in the WS (low beds: r2 = 0.56**) at AVRDC. Tomato N accumulation in controls differed greatly between experiments, with 20 kg N ha-1 in the WS at AVRDC and at MMSU, compared with 70 to 90 kg N ha-1 in the DS at AVRDC and at BRCI (Table 5).
The yields due to early transplanting of tomato plants increased by 10 Mg ha
-1 for tomato fruit and 10 kg ha
-1 for N, compared with the late transplanting at MMSU (data not shown).
At BRCI, tomato yields and N uptake did not respond to any of the treatments (Tables 4 and 5). However greatest concentration of nitrate in tomato petiole sap (1000–1472 mL NO3–N L-1 plant sap) was found in 30, 60, and 120 kg N ha-1 fertilizer treatments in early stages (7 wk after transplanting), while an average of 600 mL NO3–N L-1 plant sap was measured in control and GM treatments (data not shown). Thereafter, nitrate sap contents decreased gradually in all treatments and reached an average of 200 mL NO3–N L-1 at 9 wk after transplanting. From 10 wk after transplanting until final tomato harvest, nitrate sap dropped further in all treatments ranging between 9 and 100 mL NO3–N L-1 plant sap.
Residual Effect on Maize
All four GM treatments in raised beds and soybean GM in low beds increased maize biomass and N compared with control in the WS at AVRCD (Table 6) . In the DS, maize biomass and N were markedly enhanced by soybean GM in raised beds and by soybean mulch in low beds. The residual effect of soybean GM applied to tomato on the following maize was similar to that of 120 kg N ha-1. In both seasons, greater maize biomass was found on raised beds than on low beds.
Plant Nitrogen-15 BalanceThirty percent of the
15N applied for legume
15N enrichment at MMSU was recovered in soybean and 0.8% in indigofera. Most of the
15N applied was recovered in the shoot.
Recovery of GM 15N in tomato at MMSU was comparable among soybean and indigofera (Table 7) , indicating that 8.5 to 15% of legume N was taken up by the tomato crop. Slightly higher GM 15N was recovered in early-transplanted tomato. Of the 15N taken up by tomato, 59 to 70% accumulated in the fruits.
Answers to all your Biology Questions
