African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 7, Num. 4, 1999, pp. 375-382

African Crop Science Journal, Vol. 7. No. 4,  pp. 375-382, 1999     

Residual nitrogen benefits of promiscuous soybeans to maize under field conditions

  P.  Kasasa, S. Mpepereki, K. Musiyiwa, F. Makonese and  K.E. Giller
Department of Soil Science and Agricultural Engineering, University of Zimbabwe,
P. O. Box MP 167, Mt Pleasant, Harare,  Zimbabwe

Code Number CS99028

ABSTRACT

Residual N benefits of incorporated stover of two promiscuous and two specific soybean varieties to a subsequent, unfertilised maize crop were determined at three  sites. Soybean stover was either incorporated onto field plots previously under four soybean varieties or removed soon after harvesting. Maize yield increases were higher when promiscuous soybean stover was incorporated compared to where specific soybean stover was incorporated. Yields were higher with stover incorporation compared with stover removal. Plots where soybean stover was removed gave significantly higher maize grain yield (1 - 2 t ha^-1) on-farm and 3 - 5 t ha^-1 on-station compared to the control (maize after maize) plots which gave 0.4 t ha^-1 on-farm and 2 t ha^-1 on station. Our results indicate that soybean has significant residual fertility effects for maize and maize-based cropping systems.

Key Words: Glycine max, residual nitrogen benefits, maize yields

RÉSUMÉ

Les effets résiduels bénéfiques de l’azote des fanes de deux variétés de soja non spécifiques et de deux variétés de soja spécifiques incorporés dans une culture ultérieure de maïs sans apport d’engrais, ont été déterminés dans trois sites d’essai en milieu réel.  Les rendements du maïïs ont augmenté quand les fanes de variétés non spécifiques de soja étaient incorporées comparés aux résidux de soja spécifiques.  Les rendements étaint plus élévés avec l’incorporation de fanes en comparaison de l’enlèvement des fanes.  Les parcelles o les fanes de soja étaient enlevées ont  donné  de rendements  en grains significativement  plus  élévés en milieu réel (1 - 2 t ha^-1) et station (3 - 5 t ha^-1) plus que les parcelles témoins (maïs sur maïs) qui ont donné  0.4 t ha^-1 en milieu réel et 2 t ha^-1 en station. Nos résultats montrent  que le soja a des effets résiduels de fertilité significatifs sur le maïs et les systèmes de cultures à base du maïs.

Mots Clés: Glycine max, bénéfice de l’azote résiduel, rendements de maïs

Introduction

Nitrogen (N) is the key nutrient limiting crop yields on Zimbabwe’s sandveld soils (Tanner  and  Mugwira, 1984). Maize is the staple and therefore the most important crop in Zimbabwe’s smallholder cropping system in terms of area cropped. However the crop removes large quantities of nutrients from the soil and requires substantial N applications for optimum yields. The national average maize yields in the smallholder agricultural sector of Zimbabwe  range  from 1.0 to 1.5 t ha^-1 (Mashingaidze, 1994) compared with yields of over 10 t ha^-1 in the commercial farming sector. Boosting maize production per unit area in the low fertility sandy soils that are characteristic of most smallholder farming areas is not possible without the use of mineral fertilisers ((Mashiringwani, 1983). However, smallholder farmers do not normally apply the recommended rates of fertiliser partly because of its prohibitive costs and the highly variable yield responses under dryland conditions (Blackie, 1994).

Soil fertility improvements through incorporation of plant residues from N fixing legumes have been reported (Giller and Wilson, 1991; Brown et al., 1993; McDonagh et al., 1993; Toomsan et al,. 1995). Farmers have traditionally incorporated N2-fixing legume cover crops as green manures or legume residues (stover) to increase soil organic matter and available N to subsequent crops. The soil N enrichment has been attributed to the residual effects of legumes on soil fertility (Senaratne and Hardarson, 1988) or the N sparing effects of N fixing legumes (Herridge et al., 1995).

However, most grain legumes have been found to remove more N from the soil than they leave behind because of their high N harvest indices (NHI) (Eaglesham et al., 1982; Giller and Wilson, 1991; Giller et al., 1994; Toomsan et al., 1995). Grain legumes have been reported to contribute N to cropping systems when the percent N derived from fixation is greater than or equal to the nitrogen harvest index of the legume and when the stover is incorporated (Giller and Wilson, 1991; Giller et al., 1994; Toomsan et al., 1995).

Grain legumes with low NHI leave more N in their stover. Promiscuous soybean (Glycine max (Merryl) varieties such as Magoye from Zambia and ‘Local’ which is thought to be an unimproved Hernon land race from northern Zimbabwe (Tattersfield, pers. comm.) are leafy varieties.  It has also been suggested that they fix N during late growth phases, such as podfilling stage. Incorporation of their stover could improve soil fertility and reduce the quantities of inorganic N fertilisers smallholder farmers require to produce an economic maize yield.

Although residual benefits of local legumes to subsequent maize  are often assumed, no research has been carried out to quantify the residual N benefits to maize of incorporating stover of promiscuous soybean varieties (which do not require rhizobial inoculation) in the smallholder farming sector of Zimbabwe. This research was therefore carried out to determine the potential soil fertility benefits of growing promiscuous soybeans in rotation with maize.

Materials and methods

Site selection and soil sampling. Experiments to determine the residual benefits of promiscuous and specific soybean varieties to a subsequent maize crop were carried out at three sites, Hotera and Tapera in the smallholder sector (on-farm) and the University of Zimbabwe (UZ) farm (on-station) in the 1997/98 season. The three sites were selected on the basis of their high potential for soybean production. The on- farm sites were located in Hurungwe communal area, northern Zimbabwe (Natural region IIb, mean annual rainfall 850 - 1000 mm), which was chosen because the soils in this area are typical of many smallholder soils in Zimbabwe. Most farmers in these areas are resource-poor and require an affordable alternative source of N to boost their maize yields. The soils had also been reported to harbour indigenous soybean rhizobia (Mpepereki and Makonese,1995). A third site, the UZ farm near Harare (NR IIa, mean annual rainfall 960 - 1200 mm) was included as an on- station control site to determine yield potentials under high management. However,  maize had been grown continuously at the UZ site for four seasons without any mineral fertiliser applications to revamp the fertility of the soils.  Prior to planting at each site composite soil samples were collected from the top 30 cm by digging with a hoe. Soils were air-dried and sieved to pass through a 2 mm sieve before analysing for physical and chemical properties (Table 1).

Table 1.   Soil characteristics at study sites

The stover plus fallen leaves of the four soybean varieties, Magoye, Local, Roan and Nyala grown in the 1996/97 season with and without inoculation were either incorporated or removed. The experimental design was a split-split plot with soybean stover incorporation/removal being superimposed on the split-plot design which had inoculation as the main plot and the four soybean varieties as the sub-plots. Main plot sizes measured 5 m x 6 m and  subplots 1.5 m x  3 m. Removal of stover was to mimic farmer practices where stover is removed for livestock feed or is grazed by animals during the dry season.

No conventional tillage operations were carried out on any of the experimental plots. Land preparations were done by hand on all experi-mental plots as ploughing could have moved incorporated stover control plots. Two maize varieties, one short season (SC401) and another medium season (SC625) were planted at the three sites, with SC 401 planted at Tapera and Hotera in Hurungwe and SC 625 at UZ farm. Weather forecasts had earlier predicted that the 1997/98 rainy season would be short because of the El Nino effect.

Two seeds were planted per station at 0.9 m inter-row and 0.30 m intra-row spacing to get a plant population of 37,000 plants per hectare at each of the 3 sites. Plants were thinned to 1 plant per station two weeks after germination. No fertiliser, basal or top dressing was applied in any of the experimental plots.

The experimental plots were hand-weeded three times during the season and Dipterex (2.5%) was applied twice to control maize stalk borer (Busseola fusca). At physiological maturity, plants in each sub-plot were counted and harvested leaving out one hedge row on each side of a plot and one plant at the end of each row.

A total of 18 plants from three rows were harvested from each sub-plot. The harvested plants were air-dried and weighed to get total dry matter. Maize cobs were removed from the maize stalks and shelled. Sub-samples of the shelled maize and maize stover were weighed and dried at 700C for 24 hr, before re-weighing them to adjust total dry matter and grain yields for moisture content.

Seed and stover sub-samples were ground to pass through a 1 mm sieve using a laboratory hammer mill. Samples were analysed for N using a Roboprep automatic C/N analyser (Europa Scientific, Crewe) coupled to a 20 - 20 mass spectrometer at Wye College, University of London, United Kingdom.

Results

Maize grain yields. Incorporation of soybean stover from both promiscuous and specific varieties increased maize grain yields at the three sites (Tables 2, 3 and 4). At all the study sites maize grain yields following soybeans with or without stover were higher than those from the control (maize after maize) plots which averaged  0.37 t ha^-1 at the smallholder sites and 1.81 t ha^-1 at the UZ site, respectively (Tables 2 to 4).Total N contents and quantities of the incorporated stover for the soybean varieties tested ranged from 9 to 54 kg N  ha^-1 and 1.0 to 3.9 t ha^-1 at the smallholder sites respectively (Tables 2 and 3).  Maize grain yields without stover incorporation ranged from 0.74 to 1.87 t ha^-1  in plots across varieties at the Tapera and Hotera  sites  compared with 0.89  to 2.01 t ha^-1 on plots where the stover was incorporated (Tables 2 and 3). At Hotera and Tapera, incorporating stover from promiscuous soybean varieties Magoye and Local gave higher maize yields compared with the specific Nyala and Roan (Tables 2 and 3).

Table 2.  Maize dry matter and grain yields in (+) or (-) soybean stover plots at Tapera (1997/98 season)

s.e.d = standard error of the difference; (-) inoc = not inoculated with rhizobia; (+)inoculated with B. japonicum strain MAR 1491, (-) stover = soybean stover removed from plots; (+) stover = soybean stover incorporated.  DM = dry matter (grain + stover)

Table 3. Maize dry matter and grain yields in (+) or (-) soybean stover plots at Hotera (1997/98 season)

(-) inoc = not inoculated; (+)inoculated with B. japonicum strain MAR 1491; (-) stover = soybean stover removed from plots;  (+) stover = soybean stover incorporated; DM = dry matter (grain + stover).

At the UZ site total N contents and quantities of incorporated stover ranged from 30 to 170 kg N ha^-1 and 2.24 to 10.44 t ha^-1, respectively (Table 4). Maize grain yields at the smallholder sites were lower than those obtained  at  the UZ site,  which  ranged  from 2.56 to 4.59  t  ha^-1 on the plots where  stover  was  removed  and 3.58  to  5.26 t ha^-1 on plots where stover was incorporated (Table 4).  There were no yield differences between plots where promiscuous and specific soybean stover was incorporated at the UZ site (Table 4).

Table 4. Maize grain yields in (+) or (-) soybean stover at the University of Zimbabwe site (1997/98 season)

(-) inoc = not inoculated; (+) = inoculated with B. japonicum strain MAR 1491; (-) stover = soybean stover removed;  (+) stover = soybean stover incorporated; DM = dry matter (stover + grain)

Maize dry matter yields.  Maize dry matter  yields were higher on plots where soybean stover had been incorporated compared with where stover had been removed.  At all the study sites maize DM yields following soybeans with or without stover were higher than those from the control (maize after maize) plots which averaged 1.1 and 1.4 t  ha^-1 at  Tapera  and  Hotera sites  and  5.3 t  ha^-1 at the University of Zimbabwe site, respectively (Tables 2 to 4). Maize DM yields at all the study sites were not affected by whether the incorporated stover came from inoculated or uninoculated soybean plots. Dry matter yields at the  smallholder  sites  ranged from 1.87 to  4.1 t ha^-1 and 1.19 to 3.24 t ha^-1 on plots with and without soybean stover incorporation, respectively (Tables 2 and 3).  At all the study sites maize DM yields following soybeans with or without stover were higher than those from the control (maize after maize) plots which  averaged 1.1 and 1.4 t  ha^-1 at Tapera and Hotera sites and 5.3 t ha^-1 at the University of Zimbabwe site, respectively (Tables 2 to 4). Maize DM yields at all the study sites were not affected by whether the incorporated stover came from inoculated or uninoculated soybean plots.  In contrast, at University of Zimbabwe site,  maize  dry  matter yields  ranged from 8.4  to 11.6  t  ha^-1 and 7.4 to 9.9 t ha^-1 on plots with and without soybean stover, respectively (Table 4).

Plots where promiscuous soybean stover was either incorporated or removed had higher maize DM yields compared with those where specific soybean stover was dug in or removed at the smallholder sites (Tables 2 and 3). At this site maize DM yields did not differ on plots where soybean stover had been incorporated or removed (Table 4).

Maize seed and stover total N contents. Maize stover N contents were increased by soybean stover incorporation. Whether the stover came from promiscuous or specific soyabean varieties did not affect total N accumulation in the maize stover (Table 5). Total maize stover N contents at Hotera and Tapera, smallholder sites ranged from 3 - 16 kg N ha^-1 and 6 - 22 kg N ha^-1 on plots where soybean stover had been removed or incorporated,  respectively (Table 5). At the University of Zimbabwe (UZ) site, total maize stover N contents were higher; 15 - 21 kg N ha^-1 and  20 - 39 kg N ha^-1 on plots where soybean stover had been removed and incorporated, respectively. Whether soybean stover had been removed or incorporated, total maize stover N was consistently higher than that of the maize after maize control plots e.g.,  2.6 kg N ha^-1 at Tapera  (Table 5) and 3.2 kg N ha^-1 at the UZ site (data not shown).

TABLE 5. Maize seed and stover N accumulated in (+) and (-) soybean stover plots at Tapera site (1997/98)

(-) inoc = not inoculated; (+)inoculated with B. japonicum strain MAR 1491; (-) stover = soybean stover removed;
(+) stover = soybean stover incorporated

Maize total seed N content was strongly affected by stover incorporation x variety x inoculation interaction at Hotera, and stover incorporation only at Tapera and the UZ sites. Incorporation of promiscuous soybean stover from inoculated plots resulted in subsequent maize accumulating more N in the  grain than stover removal at Hotera (data not shown) . At Hotera and Tapera, total maize seed N  contents ranged from 6 - 20  kg N ha^-1 on plots where soybean stover had been removed compared with 10 - 25 kg N ha^-1 on plots where the soybean stover had been incorporated (see Table 5 for Tapera site). At the UZ site total maize seed N ranged from 26 - 50 kg N ha^-1 and 42 - 70 kg N ha^-1 on plots where soybean stover had been removed and incorporated, respectively (data not shown).

Discussion

Soybean stover was incorporated soon after harvesting (end of April). This was to mimic a practice in the smallholder agricultural sector where farmers plough-in plant residues before they are grazed by livestock which in most cases belong to other farmers. At this time of the year the soils will still be moist which aids decomposition of the plant residues before the onset of the hot dry months of August to mid-November.

Incorporation of soybean stover at the end of the 1996/97 season benefited the maize crop that was grown in the 1997/98 season as shown by the higher maize DM and grain yields and maize stover and seed N contents compared with those from plots where stover was removed. The incorporated soybean stover must have decomposed and released some of its N which become available for uptake by the maize to cause differences in maize DM, grain yields, stover and seed N contents between the (+) and (-) stover plots.

Generally promiscuous soybean residual effects  resulted in higher maize DM, grain, stover and seed N contents compared to specific soybean regardless of stover incorporation or removal. This was probably due to the low NHI of the promiscuous varieties coupled with their higher stover yields.  Smallholder farmers, the majority of whom cultivate on low fertility sandy soils,  should be encouraged to grow promiscuous and leafy soybean varieties such as Magoye and Local in rotation with maize to benefit from the residual fertility of these varieties. Overall,  soybean has residual soil fertility benefits regardless of variety or type.

Although total stover N ranged from 9 - 170 kg ha^-1, not all the N was released and taken up by the maize crop. Small quantities of legume residue N are taken up by cereals grown after the legume (Ladd et al., 1983; Toomsan et al., 1995). Some of the mineralised N is lost from the soil through leaching, erosion, volatilisation and denitrification while some becomes part of the recalcitrant organic matter pool (Giller and Wilson, 1991).

Removal of the soybean stover gave lower maize DM and grain yields than the plots where stover was incorporated. Legume stover removal from the field has been reported to result in net N removal from the soil amounting to the N content of the removed stover (Toomsan et al., 1995). Maize grown in the (-) stover plots only benefited from decomposing roots, nodules and leaves that fell before harvesting and could not be picked. Below-ground plant materials of most grain legumes have been reported to contribute less than 10 % of the total plant N (Kipe-Nolt  and Giller, 1993). However, these small contributions from fallen leaves and below-ground plant parts of the four soybean varieties gave significantly higher maize DM and grain yields compared with the control (maize after maize) plots. The sparing of soil N by the N2-fixing soybeans in the preceeding season could also have contributed to the higher maize DM and grain yields compared with the control plots (Giller et al., 1994; Herridge et al., 1995). It is possible  that below ground N contributions of legume plant parts  are probably underestimated.

Maize DM and grain yields at the smallholder sites were higher on plots where stover of promiscuous soybean varieties had either been incorporated or removed compared with the specific varieties Roan and Nyala. This could be due to higher quantities of DM and total N contents of the incorporated promiscuous soybean stover compared with those of the specific varieties.

 Most grain legumes have been found to remove more N from the soil than they leave behind because of their high N harvest indices of up to 0.90 (Eaglesham et al., 1982; Giller and  Wilson, 1991; Giller et al., 1994; Green and Blackmer, 1995; Toomsan et al., 1995). In this study promiscuous soybean varieties Magoye and Local had lower NHI of 0.6 to 0.7 at the smallholder sites and 0.4 to 0.5 at the UZ farm compared with the specific varieties Roan and Nyala (NHI = 0.7 to 0.9 at the smallholder and UZ sites). Incorporating or removing promiscuous soybean stover gave higher maize DM and grain yields compared with the specific varieties.  The  significantly higher residual N benefits to the subsequent maize crops were partly due to their low NHI compared with the specific varieties.

Smallholder farmers practising a promiscuous soybean-maize rotation and incorporating crop residues could therefore boost their maize grain yields  above the  national average  of 1.0 - 1.5 t  ha^-1 (Mashingaidze, 1994). Our results show that even smallholder farmers removing the soybean stover and feeding it to livestock, could still improve maize yields by rotating it with any of the four soybean varieties.

We conclude that smallholder cropping enterprises could be improved if farmers rotate maize with promiscuous soybean varieties and provide modest inputs of other nutrients. While the benefits in this study are firstly attributable to the residual N benefits, improvements in the soil physical and chemical properties such as water holding capacities and improved cation exchange capacities due to increased organic matter contents must also play a part.

Further research is still needed to determine the amounts of basal and nitrogenous fertilisers to apply to a maize crop grown in rotation with soybeans for optimal and sustainable maize production. There is also need to look at the synchrony between the time of soybean stover  incorporation, peak N release from the stover and peak N demand by the maize crop.

Acknowledgements

This work was funded by the Rockefeller Foundation’s Forum on Agricultural Resource Husbandry.

References

Blackie , M. J.  1994.  Maize  productivity for the 21 st century: the African challenge . Outlook on Agriculture  23:189-195.

Brown, R. E., Varvel , G. E. and Shapiro, C. A. 1993. Residual effects of inter-seeded hairy vetch  on soil nitrate -nitrogen levels. Soil Science Society of America Journal  57: 121-123.

Giller, K. E. and Wilson, K. J. 1991. Nitrogen Fixation in Tropical Cropping Systems. CAB International, Wallingford.

Giller, K. E., McDonagh, J. F. and Cadisch, G.  1994. Can biological nitrogen fixation sustain agriculture in the tropics?  In: Soil Science and sustainable Land Management in the Tropics. Syers, J. K.  and Rimmer, D. L. ( Eds.), pp. 173-191.  CAB International, Wallingford.

Eaglesham, A. R. J., Anayaba, A., Ranga Rao, V. and  Eskew, D. L. 1982. Mineral N effects on cowpea and soyabean crops in a Nigerian soil. Development, nodulation, acetylene reduction and grain yield. Plant and Soil 68:171-181.

Herridge, D. F.,  Marcellos, H., Felton, W. L., Turner, G. L. and Peoples, M. B. 1995. Chickpea increases soil-N fertility in cereal systems through nitrate sparing and N-fixation. Soil Biology and Biochemistry 27:545-551.

Kipe-Nolt, J. A. and Giller, K. E. 1993. A field evaluation using 15 N isotope dilution method of lines of Phaseolus vulgaris L. bred for increased nitrogen fixation. Plant and Soil 152:107-114.

Ladd, J. N., Amato, M., Jackson, R. B. and Butler, J. H. A.  1983. Utilization by wheat crops of nitrogen from legume residues decomposing in soils in the field. Soil Biology and Biochemistry 15:231-238

Mashingaidze, K. 1994. Maize research and development. In: Zimbabwe’s Agricultural revolution. Rukuni, M. and Eicher, C. K. (Eds.), pp. 208-218. University of Zimbabwe Publications, Harare. Zimbabwe

Mashiringwani, N. A. 1983. The present nutrient status of the soils in communal areas of Zimbabwe. Zimbabwe Agricultural Journal 80:73-75.

McDonagh, J. F., Toomsan, B., Limpinuntana, V. and Giller, K. E. 1993. Estimates of the residual nitrogen benefit of groundnut to maize in Northeast Thailand. Plant and Soil 154: 267-277.

Mpepereki, S. and Makonese, F. 1995. Prevalence of cowpea and soyabean rhizobia in field soils of Zimbabwe. Zimbabwe Journal of Agricultural Research 33:191- 205.

Senaratne, R. and  Hardarson, G. 1988. Estimation of residual N effect of faba bean and pea on two succeeding cereals using the 15N methodology. Plant and Soil 110:81-89.

Tanner, P. D. and Mugwira, L. M. 1985. Effectiveness of communal area manures as sources of nutrients for young maize plants. Zimbabwe Agricultural Journal  81:31-35.

Toomsan, B., McDonagh, J. F., Limpinuntana, V. J. H. A. and Giller, K. E. 1995. Nitrogen fixation by groundnut and soyabean and residual nitrogen benefits to rice in farmers’ fields in Northeast Thailand. Plant and Soil  175:45-56.

©1999, African Crop Science Society