Nutrient Cycling in Agroecosystems 68: 59–72, 2004. © 2004 Kluwer Academic Publishers. Printed in the Netherlands.

59

The potential of Velvet bean (Mucuna pruriens) and N fertilizers in maize production on contrasting soils and agro-ecological zones of East Uganda Crammer K. Kaizzi1,2, Henry Ssali2 and Paul L.G. Vlek1,* 1

Center for Development Research (ZEF), University of Bonn, Walter-Flex Str. 3, D-53113 Bonn, Germany; Agricultural Research Institute (KARI), National Agricultural Research Organization (NARO), Box 7065, Kampala, Uganda; *Author for correspondence (e-mail: [email protected]; phone: +49-228-73-1866; fax: +49-228-73-1889) 2Kawanda

Received 11 December 2002; accepted in revised form 4 June 2003

Keywords: Biological nitrogen fixation, Inorganic N, Mucuna, N-balance, Soil fertility

Abstract Research was conducted at two sites located in medium and low altitude zones in eastern Uganda. The aim of the study was to evaluate the benefit of Velvet bean 共Mucuna pruriens) and inorganic N fertilizer in improving maize production in contrasting agro-ecological zones over two seasons. The medium altitude zone 共Bulegeni兲 is a high-potential agricultural zone, with much more reliable rainfall and soils with high-productivity rating. The opposite is true for the low-altitude zone 共Kibale兲. The soils were fertile for the site in the high-potential zone and poor in the low-potential zone. Over 22 weeks of fallow or relay with maize, Mucuna produced on average 8.2 t ha ⫺ 1 dry matter, accumulating 170 kg N ha ⫺ 1, with 57% of the N derived from the atmosphere in the low-potential zone, compared to 11.6 t ha ⫺ 1 dry matter, 350 kg N ha ⫺ 1, with 43% of the N derived from air, in the high-potential zone. Between 77 and 97% of the Mucuna-accumulated N was released over a period of 25 weeks, at a rate of 0.081 and 0.118 week ⫺ 1 in the high- and low-potential zones, respectively. The N-balance study shows that 93% of the applied N was accounted for in the high-potential zone, compared to 61% in the low-potential zone, due to differences in soil texture, soil fertility and maize biomass production at the two sites. As much as 44–73% of the N remained in the soil in the high-potential zone, compared to 39–53% in the lowpotential zone, which might benefit the subsequent crops. There was a significant increase in maize yield in response to the added N, both from urea or Mucuna. The average increment above the control 共continuous maize兲 was 3.2 t ha ⫺ 1 in the high-potential zone and 1.0 t ha ⫺ 1 in the low-potential zone. The maize yield increase over two seasons added up to 3.1 t ha ⫺ 1 with the application of inorganic fertilizers, and 1.9 t ha ⫺ 1 with a preceding Mucuna–maize relay in the high-potential zone, compared to an average of, 1.7 t ha ⫺ 1 with application of inorganic fertilizers and with Mucuna–maize relay in the low-potential zone. Application of P fertilizers with either N supply strategy significantly increased maize yield in the low-potential zone only, resulting in an additional 0.8 t ha ⫺ 1 for the inorganic N fertilizers and 1.3 t ha ⫺ 1 for a preceding Mucuna–maize relay. Apparently, P fertilizers are needed on poor soils. Clearly farmers stand to gain in terms of maize production from fertilizers as well as from the use of Mucuna, with more benefits from inorganic fertilizers in the high-potential zone.

Introduction Cereals are important crops for the smallholder farmers in sub-Saharan Africa. However, crop yields per

unit area of production are declining 共IBSRAM 1994; Sanchez et al. 1996兲. The main contributing biophysical factors are low inherent soil fertility, particularly N and P deficiencies 共Bekunda et al. 1997兲, exacer-

60 bated by nutrient/soil fertility depletion 共Vlek 1993; Sanchez et al. 1997兲. Loss of nutrients in crop harvests as well as through runoff and soil erosion is on the increase in many farming systems. Yet, many farmers are unable to compensate for these losses, resulting in the negative nutrient balances at the national level for sub-Saharan Africa countries 共Stoorvogel and Smaling 1990兲 and at the regional scale for the farming systems of Eastern and Central Uganda 共Wortman and Kaizzi 1998兲. Replenishing and enhancing soil N, P and K is essential for sustained productivity and for the rehabilitation of eroded and depleted soils. This can be achieved through the use of inorganic fertilizers, organic fertilizers or a combination of both. Inorganic fertilizers are the only option available to improve and balance the loss of P and K. For N it can be achieved through the use of inorganic fertilizers and Biological Nitrogen Fixation 共BNF兲. Unfortunately, social and economic factors do not favour the use of inorganic fertilizers by the smallholder farmers. In sub-Saharan Africa, inorganic fertilizers cost two to six times as much as in Europe 共Sanchez 2002兲, mainly due to transport costs, and other charges. In addition, the profitability of fertilizer use is highly variable and dependent on agro-climatic and economic conditions at local and regional levels 共Vlek 1990兲. There are also constraints limiting the use of organic materials, including labour for collecting and applying the materials 共Ruhigwa et al. 1995兲, limited quantities and variation in quality 共Palm et al. 1997兲, and the demand for crop residues as fuel and fodder 共Palm 1995兲. In the case of green manure or in situ biomass production, farmers have to keep land out of food production 共Giller et al. 1997兲, a luxury especially in densely populated areas. Biological nitrogen fixation 共BNF兲 as a nitrogen source can be exploited for increased productivity through the use of legumes in the farming systems. Giller and Cadisch 共1995兲 reported that BNF contributes to productivity both directly, when the fixed N is harvested in grain or other food for human or animal consumption, or indirectly by adding N to the soil and thus contributing to the maintenance or enhancement of soil fertility. Under favourable environmental conditions with a good supply of nutrients, moisture availability and good pH 共Peoples et al. 1995兲 BNF can meet N requirements and sustain tropical agriculture 共Giller et al. 1994, 1997兲. Economic considerations make BNF an attractive N source for resourcepoor farmers 共Giller and Wilson 1991兲 and the most

practical solution for the low-input agriculture typical in sub-Saharan Africa 共Van Cleemput 1995兲. However, for biological N fixing systems to provide a substantial amount of N to the system, it is essential to ensure good legume growth, which also may require the use of fertilizers 共Giller and Cadisch 1995兲. Leguminous green manures are promoted on the assumption that they are fixing atmospheric N. However, little research has been conducted to quantify N fixation by legume green manure 共Giller 2001兲. Under favourable conditions, it is estimated that herbaceous legumes can accumulate 100 to 200 kg N ha ⫺ 1 in 100 to 150 days in the tropics with a significant portion derived from BNF 共Giller et al. 1994兲. Evaluating the BNF activity of Mucuna, the N-release pattern and crop response to Mucuna-derived N as compared to inorganic N on contrasting soils and agro-ecological zones would help target its deployments. In this study the use of inorganic N fertilizers was compared with the exploitation of BNF through the use of Velvet bean 共Mucuna pruriens兲, either as relay crops, or as an improved fallow. The objectives of the study were to determine on contrasting soils and in contrasting agro-ecological zones: 共i兲 Mucuna biomass production and BNF with and without P fertilizers, 共ii兲 the decomposition and N release pattern of Mucuna residues, 共iii兲 the N balance following the application of Mucuna and inorganic N, and 共iv兲 maize yield response to inorganic N, and to a preceding Mucuna fallow or relay crop on contrasting soils.

Materials and methods The trials were conducted at Kibale Technology Verification Centre 共TVC兲 and Bulegeni Agricultural Research and Development Centre 共ARDC兲, in contrasting agro-ecological zones in eastern Uganda. Bulegeni is in a high-potential agricultural area, and Kibale TVC in a low-potential agricultural area. The agricultural potential is determined primarily by the quantity and variability in rainfall and the parent materials of the soils. Prior to the trials composite soil samples were collected for analysis. The characteristics of the sites are indicated in Table 1. Mucuna biomass production and maize response The experiments were conducted over two seasons, during the short rains 共August–December兲 of 2000

61 Table 1. Site characteristics for Kibale and Bulegeni. Location

Location Altitude 共m a.s.l.兲 Latitude Longitude Mean annual precipitation 共mm兲 Agro-ecological zone FAO-UNESCO classificationa Productivity ratingb Parent materialb

Kibale

Bulegeni

1132 1°12' N 33°47' E 1370 Jinja and Mbale farmlands Ferralsols Low–medium B.C. gneiss and granite

1430 1°18' N 34°20' E 1850 Southern and eastern Lake Kyoga basin Andosols High Volcanic ash and rocks

Source: aSsali 共2000兲; bHarrop 共1970兲.

Table 2. Treatments during the 2000b and 2001a seasons. Number

i ii iii iv v vi vii viii

Season 2000b

2001a

Maize Maize Maize Maize Maize Mucuna relay Mucuna relay⫹ 25 kg P ha ⫺ 1 Mucuna fallow

Maize-control 共without inputs兲 Maize⫹40 kg N ha ⫺ 1 Maize⫹80 kg N ha ⫺ 1 Maize⫹ 共40 kg N⫹25 kg P兲 ha ⫺ 1 Maize⫹ 共80 kg N⫹25 kg P兲 ha ⫺ 1 Maize Maize Maize

共2000b兲 and long rains 共March–June兲 of 2001 共2001a兲. The field was prepared according to local practise, using ox-ploughs at Kibale, whereas a tractor was used at Bulegeni. The individual plot size was 6 ⫻ 4.5 m, with six rows each 6 m long. The trials were laid out in a randomized complete block design using three replicates at Kibale and four replicates at Bulegeni. The treatments during the 2000b season are listed in Table 2. ‘Longe 1’, an open pollinated composite maize variety, was planted at the recommended spacing of 75 ⫻ 60 cm on 8 August 2000 at Bulegeni and on 12 August 2000 at Kibale and thinned after germination, leaving two plants per hill. Phosphorus fertilizer 共25 kg ha ⫺ 1兲 was applied at planting as TSP. Mucuna was planted on 14 September 2000 at Bulegeni and on 15 September 2000 at Kibale, at a spacing of 75 ⫻ 60 cm under sole crop production, or in an intercrop between two maize rows. During the 2001a season, the treatments listed in Table 2 were superimposed on those of the previous season. The fields were prepared using hand hoes.

The Mucuna residues were chopped into pieces 12 cm long and placed on the soil surface at rates of 12 t ha ⫺ 1 at Bulegeni and 8 t ha ⫺ 1 at Kibale, corresponding to the site’s mean biomass yield of Mucuna. A better adapted maize hybrid 共PANNAR 67兲 was planted at Bulegeni ARDC, receiving more rainfall. ‘Longe 1’, an open pollinated variety 共OPV兲, was planted at Kibale TVC, which receives less rainfall. Maize was planted in the thick mulch of Mucuna on 17 March 2001 at Kibale and on 23 March 2001 at Bulegeni. Phosphorus 共25 kg P ha ⫺ 1兲 in the form of TSP, and the first split of ammonium-sulphate N 共20 or 40 kg N ha ⫺ 1兲 were applied to the targeted plot at planting, and the second N split 共20 or 40 kg N ha ⫺ 1兲 was applied when the maize was 1 m high. Weeds were controlled using hand hoes. Beta-cyfluthrin 0.05–2.5% 共Bulldock兲 was applied 3–4 weeks after maize had germinated to prevent damage by the maize stalk borer and Chloropyrifos 5% 共Dursban兲 was used for termite control in both seasons. Data collection Mucuna biomass production was determined after 22 weeks 共2000b兲 by harvesting an area equivalent to 3 m2 using a 1 m2 quadrant placed randomly at three different places within a plot. All the materials within the quadrant, including litter, were collected and weighed. Sub-samples 共including leaves, pods and vines兲 were dried at 70 °C to a constant weight for moisture determination and ground for N, P and K determination at the Kawanda Agricultural Research Institute 共KARI兲, national soils and plant-tissue laboratory. Maize 共grain and stover兲 yield was determined by harvesting a 3 ⫻ 4.2 m area 共middle 4 rows, 4.2 m long兲 at maturity 共130 and 165 days for ‘Longe 1’ and

62 ‘PANNAR’, respectively兲. The maize stover was left in the field. Sub-samples of the grain were collected for moisture determination and the grain yield was adjusted to 14% moisture content.

% Ndff ⫽

Estimation of biological nitrogen fixation A separate experiment was conducted using a main plot size of 6 ⫻ 4.5 m, with micro plots of 3 ⫻ 2.4 m and laid out in a randomized complete block design with three replicates. The treatments were: 共i兲 Mucuna, 共ii兲 Mucuna⫹25 kg P ha ⫺ 1, and 共iii兲 Luffa 共Luffa cylindrical 共L.兲 Roem.兲. Luffa was used as reference plant. Mucuna and Luffa were planted at a spacing of 75 ⫻ 60 cm on 14 September 2000 and 16 September 2000 at Kibale and Bulegeni, respectively. Twenty kg N ha ⫺ 1 with 5 at.% excess 15N was applied to Mucuna, while 100 kg N ha ⫺ 1 with 1 at.% excess 15N was applied to Luffa. The labeled fertilizers were prepared from a stock of ammonium sulphate fertilizer with 10.19% 15N abundance. The fertilizers were calculated using the equation 共IAEA 1990兲: m1 ⫽

冉

冊

共m1 ⫹ m2兲a⬘ a⬘1

共1兲

where: m1⫽quantity of fertilizer with 10.19% 15N abundance; m2⫽quantity of fertilizer at natural abundance; a⬘1⫽% 15N excess of material of higher 15N enrichment 共9.824兲; and a ' ⫽% 15N excess desired in the final mixture. Phosphorus fertilizer was applied at planting. Labeled N fertilizers were applied to the soil in solution in four equal splits at two-week intervals with the first dose at planting. The percentage of N derived from the air 共% Ndfa兲 was calculated using Equations 2–5 below 共IAEA 2001兲:

冉

冊

冉 冊

%NdffF 1 +%NdffF % Ndfa=100 1⫺ n%NdffNF n⫺1

n⫽

冉

The percentage of N in the plant derived from fertilizers 共%Ndff兲 according to Equation 4 共IAEA 2001兲:

amount of fertiliser applied to a fixing crop

amount of fertiliser applied to a non-fixing crop

冊

共2兲

共3兲

冉

atom%15N excess of the plant atom%15N excess of fertiliser

冊

× 100 共4兲

The amount N fixed 共Ndfa兲 and N yield were calculated according to Equations 5–8 below 共IAEA 2001兲: Ndfa 共kg ha⫺1兲 ⫽

冉

冊

% Ndfa×total N in plant 100

共5兲

N yield 共kg ha⫺1兲 of each plant part = Dry matter yield of each plant part×%N

冉

冊

100

N fertiliser yield 共kg ha⫺1兲 = 共N yield 共kg ha⫺1兲 × %Ndff

冉

冊

100

%Ndff 共weighted average兲 = Total N fertiliser yield × 100 Total N yield

冉

冊

共6兲

共7兲

共8兲

where Ndfa ⫽ N derived from air 共kg ha ⫺ 1兲; %Ndff ⫽ percentage of N in plant derived from fertilizer; %NdffF ⫽ percentage of N derived from fertilizer by fixing plant; and %NdffNF ⫽ percentage of N derived from fertilizer by non-fixing plant. Data collection Plants were sampled after 22 weeks using a 1 m2 quadrant placed in the centre of the micro plot. The biomass in the quadrant was separated into three components 共vines, leaves and pods兲, and weighed in the field. Sub-samples were collected, dried at 70 °C, weighed, and ground for 15N analysis by continuous flow mass spectrometry at the University of Bonn, Germany. The biomass produced by Mucuna and Luffa was determined in a similar way from the designated harvest areas.

63 Mucuna decomposition and N release Mucuna residues 共100 g, oven dry weight兲 were placed in 30 cm by 30 cm polyethylene litterbags with a mesh size of 5 mm, allowing access by soil mesoand micro fauna 共Swift et al. 1979兲. The litterbags were randomly placed between rows of maize on the soil surface in treatments 共vi兲 to 共viii兲 previously under Mucuna. Four litterbags were retrieved at each sampling time per site. The contents were cleaned by hand to remove roots and mineral soil, weighed, ground and total N content was determined by Kjeldahl digestion 共Anderson and Ingram 1993兲. Sub-samples were combusted in a muffle furnace at 550 °C for 4 h to correct for mixing with mineral soil. Ash-free dry weights were determined, and subtracted from the original sub-sample dry weights to determine the amount of plant material in the sub-sample. The amount of N remaining was determined as the weight of the material multiplied by the N content. The single exponential equation, Y ⫽ e⫺kt, was used to calculate the decomposition and N release rate constants, k, at each site, where Y is the percentage of initial weight of material, or N remaining at time t in weeks 共Wieder and Lang 1982兲. N balance study (fate of applied N) Production of 15N labelled Mucuna 15 N-labelled Mucuna was produced at Kawanda Agricultural Research Institute 共KARI兲, Ssenge farm 共0°24' N, 32°31' E兲 at an altitude of 1200 m above sea level. The soil at the site is classified as Rhodic Kandhapludalf, with the following surface 共0–20 cm兲 soil properties: pH of 5.0; organic carbon 1.7%; available P 2.25 mg kg ⫺ 1 soil; exchangeable Ca 3.18 cmolc kg ⫺ 1 soil, and exchangeable K 0.20 cmolc kg ⫺ 1 soil. Mucuna was planted on 3 September 2000 in a field of 26 ⫻ 25 m, split into five equal strips and received 100 kg N ha ⫺ 1 with 5 at.% excess 15N prepared from ammonium sulphate fertilizer stock of 10.19% 15N abundance. The fertilizer was applied in solution in two equal splits at two-week intervals. Mucuna was harvested after 16 weeks, dried in the field and stored until the time for its application. To check for the uniformity of 15N labelling, the Mucuna field was split into five strips, and three samples were collected from each strip, giving a total of 15 samples for the entire field.

The fate of applied inorganic and Mucuna N in the soil–plant system was determined by applying 40 kg N ha ⫺ 1 either as ammonium sulphate labeled with 5 at.% excess 15N or as 400 kg N ha ⫺ 1 共at Bulegeni兲 and 290 kg N ha ⫺ 1 共at Kibale兲 15N-labelled Mucuna residues to micro plots of 3.0 ⫻ 2.4 m, installed in the center of the main plots for treatments 共ii兲, 共iv兲, 共vii兲 and 共viii兲 of the main experiment. The micro plots were enclosed in aluminium-sheet borders driven 50 cm deep into the soil, with 10 cm remaining above the ground to prevent lateral movement of labeled N and to confine the maize roots within the micro plots 共Stumpe et al. 1989兲. The area outside the micro plots received unlabelled fertilizer at the equivalent N rate 共40 kg N ha ⫺ 1兲, while those for Mucuna treatments received equivalent quantities of unlabelled Mucuna residues. The inorganic N was applied in two splits, first at planting and again when the maize was 1 m high. The labeled fertilizer was applied in a furrow 共8 cm deep兲 and 10 cm from the maize row and covered with soil. The Mucuna residues from the previous crop were removed from the micro plot areas only and replaced with the 15N-labelled residues, which was also applied as mulch at planting. The percentage N in maize derived from Mucuna residues % 共Ndfr兲 was calculated according to Equation 9 共IAEA 2001兲:

冉

% Ndfr =

atom%15N excess in the crop 15

atom% N excess in the Mucuna residues added

冊

× 100

共9兲

The quantity of N derived from the Mucuna residues 共%Nrec兲 was calculated according to Equation 10 共IAEA 2001兲: Nrec. 共kg兲 ⫽

冉

冊

%Ndfr × total N in maize 100

共10兲

Percentage recovery of N from Mucuna residues 共%Nrec兲 was calculated according to Equation 11 共IAEA 2001兲:

冉

% Nrec.=

Ndfr 共kg兲

amount of N added in Mucuna residues

冊

× 100

共11兲

64 The percentage of N in maize derived from fertilizers 共%Ndff兲 was calculated according to Equation 12 共IAEA 2001兲: % Ndff ⫽

冉

atom%15N excess in the maize

atom%15N excess of the fertiliser

冊

× 100

共12兲

The quantity of N derived from fertilizers 共Ndff兲 was calculated according to Equation 13 共IAEA 2001兲: Ndff共kg兲 ⫽

冉

冊

%Ndff×total N in maize 100

共13兲

Percentage recovery of fertilizer N was calculated according to Equation 14 共IAEA 2001兲: Nrec. ⫽

冉

Ndff 共kg兲

amount of fertiliser N added

冊

× 100 共14兲

where Ndfr ⫽ amount of N derived from Mucuna residues, and Ndff ⫽ amount of N derived from fertilizer. Equivalent calculations were done to trace 15N in the soil 共IAEA 2001兲. Data collection At maturity, eight maize plants in the center of the micro plot 共four plants from two central maize rows兲 were harvested and separated into grain and stover, dried at 70 °C, weighed, and ground for 15N analysis. Soil samples were taken from the center of the micro plot in an area of 1 m2. The entire top layer 共0–15 cm兲 of soil was removed and mixed, and a composite sample was taken for 15N analysis. Samples were taken from the 15–30 cm and 30–60 cm soil layers by auguring 共five 2.5 cm Ø cores for each layer兲. Soil bulk density was determined for each of the soil layers. The 15N was used to estimate plant uptake of inorganic fertilizer N, of N mineralised from Mucuna residues, and the amount of N remaining in the soil at harvest and the N balance according to IAEA 共2001兲. Soil and plant analysis Soil samples were air-dried and ground to pass a 2 mm sieve and analysed according to Foster 共1971兲. Extractable P, K and Ca were measured in a single

ammonium lactate/acetic acid extract buffered at pH 3.8. Soil pH was measured using a soil to water ratio of 1:2.5. Plant samples were dried in an oven at 70 °C, ground to pass a 0.5 mm sieve and analysed for total N, P and K by Kjeldahl digestion with concentrated sulphuric acid 共Anderson and Ingram 1993兲. Posphorus was determined colorimetrically, and K by flame photometry 共Foster 1971兲. Analysis of total N and 15N for plant and soil samples from the BNF and N balance studies was carried out using a mass spectrometer 共an ANCA-SL coupled to a 20–20 stable isotope analyser IRMS – PDZ Europa兲 at the Institute of Agricultural Chemistry, University of Bonn. Statistical analysis Data was examined by ANOVA using a general linear model and comparisons of treatments were made by the least significant difference 共LSD兲 using Statistix V. 2.0 共Statistix for Windows, Analytical Software 1998兲.

Results and discussion Soil characteristics The mean values of selected soil properties at the sites are listed in Table 3. The results indicate that the soils at Bulegeni are fertile, hence will give good crop yields under good management. The values at Kibale indicate deficiency conditions 共Foster 1971兲; poor crop yields are expected at this site. The results from chemical analysis are in agreement with the productivity rating of the soils 共Harrop 1970兲. Maize and Mucuna biomass production The mean maize yield and Mucuna dry matter production during the 2000b season are presented in Table 4. There was no significant difference 共P⫽0.05兲 in Mucuna dry matter production between the relay with and without P, implying that P fertilizers did not have an effect on Mucuna dry matter production at both sites. The dry matter production of Mucuna was also not affected by inter-planting with maize. Because of the aggressive nature of Mucuna it competed favourably with maize for nutrients and other requirements. There was a significant reduction 共P⫽0.05兲 in maize yield in the intercrop compared to the sole crop

65 Table 3. Selected soil properties at Kibale and Bulegeni. Property

Critical valuesa

Location

pHb OMc 共%兲 Extractable P 共mg kg ⫺ 1兲d Extractable K 共cmolc kg ⫺ 1兲d Extractable Ca 共cmolc kg ⫺ 1兲d Bulk density 共kg m ⫺ 3兲 Sand 共%兲 Silt 共%兲 Clay 共%兲 Texture class

Kibale

Bulegeni

4.8 2.1 2.5 0.3 0.6 1470 60 12 28 Sandy clay loam

5.6 5.6 14.9 1.2 7.7 1160 16 26 58 Clay

5.2 3.0 5.0 0.4 0.9 na na na na na

a Below these values soils are deficient or poor 共Foster 1971兲; na⫽not applicable; b Measured in 1:2.5 共soil:water兲 suspension; c Walkley– Black method, modified according to Foster 共1971兲; d Measured in single ammonium lactate/acetic acid extract 共pH 3.8兲 according to Foster 共1971兲.

Table 4. Mucuna and maize yield 共t ha ⫺ 1兲 at Bulegeni and Kibale during the 2000b season. Treatment

Maize Maize⫹Mucuna relay⫹25 kg P ha ⫺ 1 Maize⫹Mucuna relay Mucuna fallow Mean LSD5%

Mucuna dry matter

Maize grain

Maize stover

Bulegeni

Kibale

Bulegeni

Kibale

Bulegeni

Kibale

na 12.4 10.6 11.8 11.6 ns

na 7.5 7.9 9.0 8.2 ns

3.0 1.4 1.3 na 1.8 1.5

0.9 1.2 0.7 na 0.9 ns

3.8 1.9 1.7 na 2.4 1.3

1.4 2.2 1.1 na 1.6 ns

na⫽not applicable; ns⫽not significantly different at the 5% level.

at Bulegeni, attributed to competition for resources between Mucuna and maize, and also to the smothering of maize by Mucuna climbers. Moreover, efforts to reduce the smothering effect through frequent physical removal, and cutting the vines was not effective at Bulegeni because of the vigorous growth, probably due to the high fertility of the soil. On the other hand, at Kibale the maize grain yield was hardly affected by the Mucuna relay, showing a loss of 200 kg ha ⫺ 1 only. The addition of P in this system largely benefited the maize and caused an increase of 300 kg ha ⫺ 1 in maize grain and 800 kg ha ⫺ 1 in straw yield. Thus, on the basis of a single season it appears that Mucuna/maize relay cropping should be restricted to the poorer environments. Mucuna N, P, and K yield The N, P and K yields by Mucuna after 22 weeks are given in Table 5. There was a significant increase 共P⫽0.05兲 in the amount of N and P accumulated by Mucuna relay in response to P fertilizers, possibly due to an increased uptake of nutrients due to a better root

system. Phosphorus is known to improve root growth 共Tisdale et al. 1999兲. In general, Mucuna accumulated disproportionally more N, P and K at Bulegeni than at Kibale, yielding higher quality Mucuna residues at Bulegeni. This is attributed to the more fertile soils at Bulegeni. The N, P and K accumulation by Mucuna shows its potential in nutrient recycling. Biological nitrogen fixation The amount of N fixed by Mucuna with and without inorganic P fertilizer is presented in Tables 6a and 6b for Bulegeni and Kibale, respectively. The results indicate that Mucuna was effective in fixing N at both sites, highlighting the potential of Mucuna as a source of N in the low-input agriculture common for the smallholder farmers in Uganda. The percentage of N fixed was within the range 共40–86%兲 reported in similar studies in central Uganda by Wortman et al. 共2000兲, and in West Africa 共Sanginga et al. 1996; Ibewiro and Sanginga 2000兲.

66 Table 5. Mucuna N, P and K yield 共kg ha ⫺ 1兲 at Bulegeni and Kibale during the 2000b season. Treatment

N yield

P yield

K yield

Bulegeni

Kibale

Bulegeni

Kibale

Bulegeni

Kibale

Mucuna relay⫹25 kg P ha ⫺ 1 Mucuna relay Mucuna fallow Mean LSD5%

430 290 320 350 120

190 150 170 170 ns

40 25 28 31 11

8.0 8.3 9.5 8.6 ns

370 260 290 310 ns

90 90 100 93 ns

Table 6a. Nitrogen fixation by Mucuna at Bulegeni 共means of three replicates兲. Dry matter yield 共t ha ⫺ 1兲 Mucuna 0 kg P ha ⫺ 1 Vines 2.7 Leaves 2.7 Pods 2.8 Total 8.2 Mucuna 25 kg P ha ⫺ 1 Vines 2.8 Leaves 3.0 Pods 2.1 Total 7.9 LSD5% Luffa Vines Leaves Pods Total

3.5 2.4 1.1 7.1

N 共%兲

N yield 共kg ha ⫺ 1兲

Atom 15N excess 共%兲

1.8 3.0 2.9

60 80 80 220

0.069 0.084 0.075

60 100 70 230

0.093 0.072 0.064

80 60 30 170

0.131 0.122 0.116

2.1 3.3 3.2

2.2 2.7 2.9

Ndff 共%兲

N fert. yield 共kg ha ⫺ 1兲

Ndfa 共%兲

2.05 1.67 1.49 1.71a

1.23 1.34 1.19 3.76

39

86

1.85 1.44 1.29 1.50a

1.11 1.48 0.89 46 ns

106

13.1 12.2 11.6 12.5a

Fixed N 共kg ha ⫺ 1兲

10.5 7.3 3.5 21.4

a

Weighted average.

There was a significant increase 共P⫽0.05兲 in N fixation in response to P fertilizers at Kibale, the site with poor soils, indicating that P was a limiting factor for N fixation. This is in agreement with the reports by several investigators that P fertilizers increase BNF under P limiting conditions 共Ssali and Keya 1986; Thomas 1995; Wani et al. 1995兲. Interestingly, the absolute amounts of fixed N were similar for the two sites 共within 7–12% of each other兲. Yet, the fraction of Mucuna-N derived from BNF is substantially 共30–38%兲 higher when grown on the poorer soils of Kibale. This reflects the lower availability of soil N at this site. Mucuna decomposition and nitrogen release The mass loss 共km兲 and N release constants 共kN兲 for Mucuna under field conditions at Bulegeni and Kibale

are presented in Table 7. Mucuna decomposed and released N rapidly at the two sites, due to the high total N 共3.5%兲 of the substrate, well above the 1.5– 2.0% critical level for net mineralisation 共Heal et al. 1997兲. The relatively lower temperatures at Bulegeni and the absence of termites explain the difference in the mass loss and N-release rates between Bulegeni and Kibale. The litterbags had a 5 mm mesh size, which allowed termite access to the Mucuna substrate. The N release rate constants can be used to estimate the amount of N released after a specified time. The percentage of N released is within the range 共70–95%兲 reported for tropical conditions 共Giller and Cadisch 1995兲. The decomposition and N release patterns were described well by the single exponential function 共Table 7兲. The decomposition rate constants 共km and kN兲 are in the range reported for materials of similar composition in the tropics 共Tian et al. 1992;

67 Table 6b. Nitrogen fixation by Mucuna at Kibale 共means of three replicates兲. Dry matter yield 共t ha ⫺ 1兲 Mucuna 0 kg P ha ⫺ 1 Vines 1.3 Leaves 1.1 Pods 4.0 Total 6.4 Mucuna 25 kg P ha ⫺ 1 Vines 1.9 Leaves 1.5 Pods 4.6 Total 8.1 LSD5% Luffa Vines Leaves Pods Total a

0.9 0.4 0.5 1.7

N 共%兲

N yield 共kg ha ⫺ 1兲

Atom 15N excess 共%兲

Ndff 共%兲

N fert. yield 共kg ha ⫺ 1兲

Ndfa 共%兲

Fixed N 共kg ha ⫺ 1兲

2.5 2.3 2.8

30 20 120 170

0.129 0.172 0.129

2.58 2.77 2.4 2.61a

0.83 0.65 2.95 4.44

54

92

40 30 130 200

0.118 0.114 0.095

2.35 2.28 2.08 2.16a

1.0 0.8 2.5 60 5

120

20 10 10 40

0.264 0.187 0.161

26.4 18.7 16.1 21.8a

5.28 1.87 1.61

2.2 2.3 2.8

2.7 2.8 2.9

Weighted average.

Table 7. Mucuna mass loss 共km兲 and N release constant 共kN 兲, as well as percentage loss of original quantities remaining after 25 weeks in the field at Bulegeni and Kibale. Site

Mass loss km 共week

Bulegeni ARDC Kibale TVC LSD1%

N release

⫺1

兲

0.081 0.118 0.041

*R

2

0.960 0.938

共%兲

kN 共week ⫺ 1兲

*R2

共%兲

80 95

0.065 0.130 0.064

0.908 0.989

77 97

* Indicate the fit of the decomposition data to the single exponential function.

Fosu 1999; Kaizzi and Wortmann 2000兲. The N release patterns following a single exponential function imply that large amounts of N are released during the early phase of decomposition. This N is liable to loss, especially if applied to crops whose demand is not synchronised with the release. Maize response to alternative N treatments Bulegeni ARDC Maize yield in response to the different treatments at Bulegeni during the 2001a season is presented in Table 8. There was a significant increase 共P⫽0.05兲 in maize yield of 2.5–3.5 t ha ⫺ 1 in response to the application of inorganic N fertilizers, and to a preceding Mucuna fallow or relay as compared to the control of continuous maize without inputs. Thus, both inorganic N fertilizers and Mucuna green manure served as effective N sources for maize. The

average increase in maize grain yield 共during the 2001a season兲 due to a preceding Mucuna was on average 80%, which is in agreement with results reported by several investigators 共Versteeg et al. 1998; Fischler and Wortmann 1999; Tian et al. 2000兲. On average the Mucuna-derived N was applied at a rate of 350 kg N ha ⫺ 1 共Table 5兲, and the results from the decomposition study 共Table 7兲 indicated that 270 kg N ha ⫺ 1 共77%兲 would have been released and available for uptake by maize. However, increasing inorganic N from 40 to 80 kg N ha ⫺ 1 did not result in an increase in maize yield, implying that either the lower N rate was sufficient to meet the maize N requirements at the site or another nutrient or the environment became limiting. The better response to Mucuna, containing a sweep of other nutrients, may confirm this notion. The lack of a significant 共P⫽0.05兲 increase in maize yield in response to application of 25 kg P ha ⫺ 1 in combination with either

68 Table 8. Maizea yield 共t ha ⫺ 1兲 at Bulegeni during the 2000b and 2001a seasons, and the sum of the two seasons 共means of four replicates兲. Grain Treatment

Control 共no input兲 40 kg N ha ⫺ 1 80 kg N ha ⫺ 1 共40 kg N⫹25 kg P兲 ha ⫺ 1 共80 kg N⫹25 kg P兲 ha ⫺ 1 Preceding Mucuna relay Preceding Mucuna relay⫹25 kg P ha ⫺ 1 Preceding Mucuna fallow LSD5%

Stover

Season

Total for

Season

2000b

2001a

year

2000b

2001a

Total for year

3.0 3.0 3.0 3.3 3.7 1.3 1.4 na 1.5

4.5 7.3 7.1 7.5 7.5 8.0 8.0 8.2 1.2

7.5 10.3 10.1 10.8 11.2 9.3 9.4 8.2 1.6

3.8 3.8 3.9 4.4 4.2 1.7 1.9 na 1.3

5.3 10.0 12.6 14.1 10.1 13.4 15.4 14.5 3.3

9.1 13.8 16.5 18.5 14.3 15.1 17.3 14.5 5.9

na⫽not applicable; a ‘Longe 1’ maize variety during the 2000b season and ‘PANNAR 67’ maize variety during the 2001a season.

Table 9. Maizea yield 共t ha ⫺ 1兲 at Kibale during the 2000b and 2001a seasons, and the sum of the two seasons 共means of three replicates兲. Grain Treatment

Control 共no input兲 40 kg N ha ⫺ 1 80 kg N ha ⫺ 1 共40 kg N⫹25 kg P兲 ha ⫺ 1 共80 kg N⫹25 kg P兲 ha ⫺ 1 Preceding Mucuna relay Preceding Mucuna relay⫹25 kg P ha ⫺ 1 Preceding Mucuna fallow LSD5%

Stover

Season

Total for

Season

Total for

2000b

2001a

year

2000b

2001a

year

0.9 1.1 1.1 1.2 1.5 0.7 1.5 na ns

0.8 1.1 1.7 1.7 2.3 1.5 2.3 1.9 0.5

1.7 2.2 2.8 2.9 3.8 2.2 3.5 1.9 0.6

1.4 1.2 1.4 1.4 1.7 1.1 2.2 na ns

1.2 1.3 2.1 2.3 3.3 1.8 3.7 2.8 1.0

2.6 2.5 3.5 3.7 5.0 2.9 5.9 2.8 1.6

na⫽not applicable; ns⫽not significantly different. a‘Longe 1’ maize variety.

40 or 80 kg N ha ⫺ 1 indicates that P is not the limiting factor. Because the use of Mucuna in a fallow results in the loss of a complete crop season, and reductions in maize yield were observed in a relay, it is necessary to evaluate the benefits of Mucuna over two seasons so as to include its effect on yield during the preceding season. The combined yield of maize for the two seasons indicates a significant 共P⫽0.05兲 increase of 2.6–3.7 t ha ⫺ 1 above the control in response to inorganic N fertilizers, and on average 1.85 t ha ⫺ 1 to a preceding Mucuna relay. This implies that both N supply strategies are effective in increasing maize yield at the site, but inorganic fertilizers give an additional 1.3 t ha ⫺ 1 of grain, on average. However, the maize grain yield in response to preceding Mucuna fallow was not significantly different 共P⫽0.05兲 from

the control, indicating that Mucuna fallow compensated for the maize yield lost. Kibale TVC Maize yield in response to the different treatments at Kibale is presented in Table 9. Application of a combination of 40 kg N ha ⫺ 1 and 25 kg P ha ⫺ 1 resulted in a significant increase 共P⫽0.05兲 in maize yield of 0.9 t ha ⫺ 1 above the control. This is in contrast to the lack of response to 40 kg N ha ⫺ 1, indicating that P is a limiting factor at this site. There was a significant increase 共P⫽0.05兲 in maize yield of 0.7–1.5 t ha ⫺ 1 in response to the application of 80 kg N ha ⫺ 1 and to a preceding Mucuna fallow or relay compared to the control. Maize response to the alternative N strategies was similar, indicating that they were effective in supplying N. The Mucuna-derived N was applied at about a rate of 170 kg N ha ⫺ 1 共Table 5兲,

69 Table 10. Maize yield 共t ha ⫺ 1兲 for two seasons 共2000b and 2001a兲 at Kibale and Bulegeni. Treatment

Control 共no input兲 40 kg N ha ⫺ 1 80 kg N ha ⫺ 1 共40 kg N⫹25 kg P兲 ha ⫺ 1 共80 kg N⫹25 kg P兲 ha ⫺ 1 Preceding Mucuna relay Preceding Mucuna relay⫹25 kg P ha ⫺ 1 Preceding Mucuna fallow Mean LSD5%

Grain at site

Stover at site

Bulegeni

Kibale

Prob

7.5 10.3 10.1 10.8 11.2 9.3 9.4 8.2 9.6 1.6

1.7 2.2 2.8 2.9 3.8 2.2 3.5 1.9 2.6 0.6

*** *** *** *** *** *** *** *** ***

and the results from the decomposition study 共Table 7兲 indicated that 165 kg N ha ⫺ 1 共97%兲 N would have been released. Some of this was taken up by maize, resulting in the observed response. Application of 25 kg P ha ⫺ 1 to 80 kg N ha ⫺ 1 resulted in an additional 0.6 t of grain and 0.8 t in case of the preceding Mucuna relay. Phosphorus was a limiting factor to maize production at this site. The results are in agreement with the findings of earlier investigators who reported that N and P are limiting cereal production in Uganda 共Stephen 1970; Foster 1980a, b兲. The combined yield over two seasons adds up to a significant 共P⫽0.05兲 increment of 1.2 t ha ⫺ 1 of grain above the control, in response to 40 kg N with 25 kg P ha ⫺ 1. The increase with 80 kg N ha ⫺ 1 was 1.1 t ha ⫺ 1, with the application of 25 kg P ha ⫺ 1 resulting in an additional 1.0 t ha ⫺ 1 of grain. Mucuna relay plus 25 kg P ha ⫺ 1 resulted in a significant increase 共P⫽0.05兲 of 1.8 t ha ⫺ 1 of grain. There was no significant increase 共P⫽0.05兲 in maize grain yield in response to the application of 40 kg N ha ⫺ 1 and preceding Mucuna fallow or relay without P. The increase in maize grain in response to a preceding Mucuna fallow was not significantly different 共P⫽0.05兲 from the control treatment, indicating that it enhanced the yield during the second season and made up for the yield loss when the fields were under fallow. The results show that the N supply sources are effective on poor soils too, but P becomes limiting after overcoming the N problem. Therefore, application of both nutrients is necessary for better results.

1

Bulegeni

Kibale

Prob1

9.1 13.8 16.5 18.5 14.3 15.1 17.3 14.5 14.9 5.9

2.6 2.5 3.5 3.7 5.0 2.9 5.9 2.8 3.8 1.6

*** *** *** *** *** *** *** *** ***

Comparing the N sources at Bulegeni and Kibale The results in the previous section indicate that inorganic fertilizers and a preceding Mucuna are effective N sources at Bulegeni and Kibale. The two sites have soils of contrasting fertility 共Table 3兲 and are located in contrasting agro-ecological zones. The total maize production over two seasons at the two sites is presented in Table 10. There was a highly significant difference 共P ⬍ 0.001兲 in maize yield between the two sites, largely attributed to the differences in soil fertility. The soils at Bulegeni are of above average fertility, wheras those at Kibale are deficient in plant nutrients as observed from Table 3 and their productivity rating 共Table 1兲. Although ‘PANNAR’ has a higher yield potential than ‘Longe 1’, most likely, it would have given poor yields on the low fertility soils of Kibale. The significant increase 共P⫽0.05兲 in maize yield over the two seasons indicates that N was limiting maize production. However, the N sources are more effective on the high fertility soil at Bulegeni than on the poor soils at Kibale. Application of inorganic N resulted in an average increase of 2.7 t ha ⫺ 1 of grain at Bulegeni compared to 0. 5 for 40 kg N ha ⫺ 1, and 1.1 t ha ⫺ 1 for 80 kg N ha ⫺ 1 at Kibale. A preceding Mucuna relay increased yield by 1.8 t ha ⫺ 1 at Bulegeni, compared to a mere 0.5 t ha ⫺ 1 at Kibale. Application of P fertilizers at Kibale resulted in an additional 1.3 t ha ⫺ 1 of grain for the Mucuna relay, 0.7 t ha ⫺ 1 for 40 kg N ha ⫺ 1 and 1.0 t ha ⫺ 1 for 80 kg N ha ⫺ 1, which was significant 共P⫽0.05兲. At Bulegeni the increases were not significant.

70 Table 11a. Nitrogen balance 共% recovery兲 during the 2001a season at Bulegeni 共means of four replicates兲. 40 kg N ha ⫺ 1 P kg ha

Mucunaa

⫺1

0

25

0

25

Plant recovery Grain Stover Total 共LSD5%⫽9.9兲

30 13 43

29 18 47

8 13 21

8 9 17

Soil recovery Soil layer 共cm兲 0–15 15–30 30–60 Total 共LSD5%⫽9.4兲 Total recovery

27 12 14 53 96

28 8 8 44 91

46 13 13 71 92

33 13 28 74 91

Mucuna-derived N applied at a rate equivalent to 400 kg N ha ⫺ 1.

a

Table 11b. Nitrogen balance 共% recovery兲 during the 2001a season at Kibale 共means of three replicates兲. 40 kg N ha ⫺ 1

Mucunaa

P kg ha ⫺ 1 0

25

0

25

Plant recovery Grain Stover Total 共LSD5%⫽10兲

15 9 24

16 7 23

5 5 10

3 2 5

Soil recovery Soil layer 共cm兲 0–15 15–30 30–60 Total 共LSD5%⫽27兲

10 5 24 39

9 11 25 45

11 18 24 53

13 11 23 47

Overall recovery

63

68

63

53

Mucuna-derived N applied at a rate equivalent to 290 kg N ha ⫺ 1.

a

Nitrogen balance/fate of applied N Plant and soil recovery based on 15N derived from fertilizer 共ammonium sulphate兲 or Mucuna is presented in Tables 11a and 11b for Bulegeni and Kibale, respectively. Between 91 and 96% of the applied fertilizer N was accounted for at Bulegeni, with 17–47% and 44–74% recovered by plants and in the soil, respectively. The total recovery at Kibale was in the range 53–68%, with plant and soil recovery of 5–24% and 39–53%, respectively. The percent recovery of

Mucuna-derived N by the plants 共5–21%兲 is within the range 共6–28%兲 reported for leguminous materials in the tropics 共Giller and Cadisch 1995兲. The low recovery by plants is partly attributed to the excessive amounts of green manure N 共approximately 290 and 400 kg N ha ⫺ 1兲 added, with 280 kg N ha ⫺ 1 expected to have been released during the season at Kibale and 300 kg N ha ⫺ 1 at Bulegeni. The results from the decomposition study indicated rapid release of Mucuna-derived N, for instance 110 kg N ha ⫺ 1 共calculated from N release rate constants兲 was released during the first 8 weeks at Kibale and 142 kg N ha ⫺ 1 at Bulegeni. These amounts exceed the needs of the young maize plants, leading to its immobilisation by the microbial biomass and loss through leaching below 60 cm and volatilisation. Similar amounts of applied inorganic N were recovered by plants and in the soil at Bulegeni, whereas more was recovered in the soil than taken up by plants at Kibale. This is partly attributed to lower plant demand at the less favourable Kibale site. More Mucuna-derived N was recovered in the soil than in plants at both sites due to excessive amounts added. Most of the added N remaining in the soil 共62–83%兲 at Bulegeni was found in the 0–30 cm layer, indicating that N loss through leaching might not have been significant. In contrast, relatively more of the applied N was found in the 30–60 cm depth at Kibale, implying that leaching was significant at this site. Therefore, some of the N was possibly lost through this process, resulting in unaccounted N fractions of 32–47%. The soils at the site are sandy. Any mineral N present in the soil and not taken up by the plants was liable to being lost with the excess rainwater draining through the sandy soils at Kibale. Nitrogen losses are high in permeable, coarse textured soils, as reported by Vlek et al. 共1980兲 and Singh et al. 共1991兲. The higher amount of Mucuna-derived N and inorganic N recovered by plants at Bulegeni compared to Kibale is partly attributed to the higher N demand by maize to produce 21 t ha ⫺ 1 of dry matter compared to 4.2 t ha ⫺ 1 produced at Kibale. Consequently, the total N recovery is higher at Bulegeni than at Kibale. The results stress the need for designing better management strategies to improve or increase the amount of Mucuna-derived N taken up by the plant. One option is to apply the Mucuna biomass over a larger area so as to reduce the excessive N applied. Since a significant fraction of the applied N remains in the soil at the end of the season, the next season

71 crop should be planted early to utilise the N before it is lost through leaching, as the rains intensify with the progress of the season.

Conclusions Mucuna accumulates large amounts of N in contrasting environments, with 43–57% derived from the atmosphere through BNF. The Mucuna-accumulated N is largely released during decomposition with 7.5 to 19% taken up by the subsequent crop, resulting in 25 to 68% increase in yield. Mucuna, like inorganic N fertilizers, increases maize yield because N is limiting maize production. Either source can be used for increased maize production in the region, in order to improve food security, a major concern of the smallholder farmers. The N supply strategies, though effective across the two environments, perform better in the more favourable environment 共good soils and more rainfall兲, with 40 kg N ha ⫺ 1 being sufficient on the better soils compared to both N and P being needed for the less favourable environment 共poor soils兲. Therefore, efforts to promote fertilizer use are more likely to succeed in favourable environments, and Mucuna-based technologies can be promoted across the environments. The excessive amounts of Mucuna-derived N applied and that which remains in the soil at the end of the season calls for better management practices such as applying the green manure over a larger area and early planting of the subsequent season crop to exploit this N. Though the use of green manures for soil fertility maintenance has declined in many countries where N fertilizers are widely available 共Giller and Cadisch 1995兲, it has a potential in the agriculture of the smallholder farmers of east Africa since it is typically based on low external input, due to the economic constraints limiting the use of inorganic fertilizers. On-farm evaluation of these strategies of nutrient management is discussed elsewhere 共Kaizzi et al., in press兲.

Acknowledgements The authors are grateful to Mrs. L. Lembusi and Mr. Okurut, the field assistants at Bulegeni and Kibale, respectively, and Ms. A. Nansamba, Research Assistant, for the good work done, and to The International Atomic Energy Agency 共IAEA兲 for providing the 15N

labeled fertilizers through TC project UGA/05/020. This work was part of the project on policies for improved land management in Uganda with the following collaborating institutions; International Food Policy Research Institute 共IFPRI兲, Center for Development Research 共ZEF兲, Agricultural Policy Secretariat 共APSEC兲 and National Agricultural Research Organization 共NARO兲, sponsored by the Federal Ministry for Development Cooperation 共BMZ-GTZ兲, Germany.

References Anderson J.M. and Ingram S.J. 1993. Tropical Soil Biology and Fertility: A Handbook of Methods. CAB International, Wallingford, UK, 221 pp. Bekunda M.A., Bationo A. and Ssali H. 1997. Soil fertility management in Africa. A review of selected research trials. In: Buresh R.J., Sanchez P.A. and Calhoun F. 共eds兲, Replenishing Soil Fertility in Africa. SSSA Special Publication 51. SSSA, Madison, WI, pp. 63 –79. Fischler M. and Wortmann C.S. 1999. Green manures maize–bean systems in eastern Uganda: Agronomic performance and farmers’ perceptions. Agrofor. Syst. 47: 123–138. Foster H.L. 1971. Rapid routine soil and plant analysis without automatic equipment. I. Routine soil analysis. E. Afric. Agric. For. J. 37: 160–170. Foster H.L. 1980a. The influence of soil fertility on crop performance in Uganda. II. Groundnut. Trop. Agric. 57: 29–42. Foster H.L. 1980b. The influence of soil fertility on crop performance in Uganda. III. Finger millet and maize. Trop. Agric. 57: 123–132. Fosu M. 1999. The role of Cover Crops and their Accumulated N in Improving Cereal Production in Northern Ghana. Ph.D. Thesis No. 135, University of Göttingen, Germany. Giller K.E. 2001. Nitrogen Fixation in Tropical Cropping Systems. 2nd edn, CAB International, Wallingford, UK, 423 pp. Giller K.E. and Cadisch G. 1995. Future benefits from biological nitrogen fixation: An ecological approach to agriculture. Plant Soil 174: 255–277. Giller K.E. and Wilson K.J. 1991. Nitrogen Fixation in Tropical Cropping Systems. CAB International, Wallingford, UK, 313 pp. Giller K.E., McDonagh J.F. and Cadisch G.1994. Can biological nitrogen fixation sustain agriculture in the tropics? In: Syers J.K. and Rimmer D.L. 共eds兲, Soil Science and Sustainable Land Management in the Tropics. CAB International, Wallingford, UK, pp. 173–191. Giller K.E., Cadisch G., Ehakuitusm C., Adams E., Sakala W.D. and Mafongoya P.L.1997. Building soil nitrogen capital in Africa. In: Buresh R.J., Sanchez P.A. and Calhoun F. 共eds兲, Replenishing Soil Fertility in Africa. SSSA Special Publication 51. SSSA, Madison, WI, pp. 151–192. Harrop J. 1970. Soils. In: Jameson J.D. 共ed.兲, Agriculture in Uganda. Oxford University Press, Oxford, UK, pp. 43–71. Heal O.W., Anderson J.W. and Swift M.J. 1997. Plant litter quality and decomposition: An historical overview. In: Cadisch G. and

72 Giller K.E. 共ed.兲, Driven by Nature: Plant Litter Quality and Decomposition. CAB International, Wallingford, UK, pp. 3–45. IAEA 1990. Stable and radioactive isotopes. In: Hardason G. 共ed.兲 Use of Nuclear Techniques in Studies of Soil–Plant Relationships. IAEA-Training Course Series No. 2. International Atomic Energy Agency, Vienna, Austria. IAEA 2001. Use of Isotope and Radiation Methods in Soil and Water Management and Crop Nutrition. Manual, Training Course Series No. 14. International Atomic Energy Agency, Vienna, Austria. Ibewiro B. and Sanginga N. 2000. Transformations and recovery of residue and fertilizer nitrogen-15 in a sandy N Lixisol of West Africa. Biol. Fertil. Soils 31: 261–269. IBSRAM 1994. IBSRAM position paper. In: Greenland D.J., Bowen G.D., Eswaran H., Rhoades R. and Valentin C. 共eds兲, Soil, Water and Nutrient Management Research – A New Agenda. Bangkok, Thailand. Kaizzi C.K. and Wortmann C.S. 2000. Plant materials for soil fertility management in sub humid tropical areas. Agron. J. 93: 929–935. Kaizzi C.K., Ssali S. and Vlek P.L.G. 2004. Differential use and benefits of Velvet bean 共Mucuna pruriens兲 and fertilizers in maize production in contrasting agroecological zones of E. Uganda. Agric. Systems. 共In press兲. Palm C.A. 1995. Contribution of agroforestry trees to nutrient requirements of intercropped plants. Agrofor. Syst. 30: 105–124. Palm C.A., Myers R.J.K. and Nandwa S.M. 1997. Combined use of organic and inorganic nutrient sources for soil fertility maintenance and replenishment. In: Buresh R.J., Sanchez P.A. and Calhoun F. 共eds兲, Replenishing Soil Fertility in Africa. SSSA Special Publication 51. SSSA, Madison, WI, pp. 193–217. Peoples M.B., Herridge D.F. and Ladha J.K. 1995. Biological nitrogen fixation: an efficient source of nitrogen for sustainable agriculture?. Plant Soil 174: 3–28. Ruhigwa B.A., Gichuru M.P., Spencer D.S.C. and Swennen R. 1995. Economic analysis of cut-and-carry and alley cropping systems of mulch production for plantations in south-eastern Nigeria. IITA Res. 11: 11–14. Sanchez P.A. 2002. Soil fertility and hunger in Africa. Science 295: 2019–2020. Sanchez P.A., Izac A.-M.N., Valencia I. and Pieri C. 1996. Soil fertility replenishment in Africa; A concept note. In: Breth S.A. 共ed.兲, Achieving Greater Impact from Research Investments in Africa. Sasakawa Africa Association, Mexico City, pp. 200–207. Sanchez P.A., Sheperd K.D., Soulé M.J., Place F.M., Buresh R.J., Izac A.-M.N., Mokwunye A.V., Kwesiga F.R., Ndiritu C.G. and Woomer P.L. 1997. Soil fertility replenishment in Africa: An investment in natural resource capital. In: Buresh R.J., Sanchez P.A. and Calhoun F. 共eds兲, Replenishing Soil Fertility in Africa. SSSA Special Publication 51. SSSA, Madison, WI, pp. 1–46. Sanginga N., Ibewiro B., Houngnandan P., Vanlauwe B., Okogon J.A., Akobundu I.O. and Versteeg M. 1996. Evaluation of symbiotic properties and nitrogen contribution of Mucuna to maize grown in the derived savanna of West Africa. Plant Soil 179: 119–129. Singh B., Singh Y., Khind C.C. and Meelu O.P. 1991. Leaching losses of urea-N applied to permeable soils under lowland rice. Fert. Res. 28: 179–184. Ssali H. 2000. Soil Resources of Uganda and their relationship to major farming systems. Resource paper, Soils and Soil Fertility Management Programme, Kawanda, NARO.

Ssali H. and Keya S.O. 1986. The effects of phosphorus and nitrogen fertilizer level on nodulation, growth and dinitrogen fixation of three bean cultivars. Trop. Agric. 63: 105–109. Statistix for Windows, 1988. Analytical Software, Tallahassee, FL. Stephen D. 1970. Soil fertility. In: Jameson J.D. 共ed.兲, Agriculture in Uganda. Oxford University Press, Oxford, UK, pp. 72–89. Stoorvogel J.J. and Smaling E.M.A. 1990. Assessment of soil nutrient depletion in sub-Saharan Africa, 1983–2000. Report 28, Winand Staring Centre for Integrated Land, Soil and Water Research, Wageningen, The Netherlands. Stumpe J.M., Vlek P.L.G., Mughogho S.K. and Ganry F. 1989. Micro plot size requirements for measuring balances of fertilizer nitrogen-15 applied to maize. Soil Sci. Soc. Am. J. 53: 797–800. Swift M.J., Heal O.W. and Anderson J.M. 1979. Decomposition in Terrestrial Ecosystems. Studies in Ecology, Vol. 5. University of California Press, Berkeley, CA. Tian G., Kolawole G.O., Kang B.T. and Kirchhof G. 2000. Nitrogen fertilizer replacement indexes of legume cover crops in the derived savanna of West Africa. Plant Soil 224: 287–296. Tian G., Kang B.T. and Brussaard L. 1992. Biological effects of plant residues with contrasting chemical compositions under humid tropical conditions – decomposition and nutrient release. Soil Biol. Biochem. 24: 1051–1060. Tisdale S.L., Havlin J.L., Beaton J.D. and Nelson W.L. 1999. Soil Fertility and Fertilizers: An Introduction to Nutrient Management. Prentice Hall, Upper Saddle River, NJ. Thomas R.J. 1995. Role of legumes in providing N for sustainable tropical pasture systems. Plant Soil 174: 103–118. Van Cleemput O. 1995. Fertiliser, sustainable agriculture and preservation of the environment. In: IAEA 共ed.兲, Nuclear Methods in Soil–Plant Aspects of Sustainable Agriculture. Proceedings of an FAO/IAEA Regional Seminar for Asia and Pacific held in Colombo, Sri Lanka, 5–9 April 1993. IAEA-TECDOC-785. Vienna, Austria, pp. 7–16. Versteeg M.N., Amadji F., Eteka A., Gogan A. and Koudokpon V. 1998. Farmers’ adoptability of Mucuna fallowing and agroforestry technologies in the coastal savanna of Benin. Agric. Syst. 56: 269–287. Vlek P.L.G. 1990. The role of fertilisers in sustaining agriculture in sub-Saharan Africa. Fert. Res. 26: 327–339. Vlek P.L.G. 1993. Strategies for sustaining agriculture in sub-Saharan Africa. In: Rogland J. and Lal R. 共eds兲, Technologies for Sustaining Agriculture in the Tropics. ASA Special Publication 56. ASA, CSSA, and SSSA, Madison, WI, pp. 265–277. Vlek P.L.G., Byrnes B.H. and Crasswell E.T. 1980. Effect of urea placement on leaching losses of nitrogen from flooded rice soils. Plant Soil 54: 441–449. Wani P.S., Rupela O.P. and Lee K.K. 1995. Sustainable agriculture in the semi-arid tropics through biological nitrogen fixation in grain legumes. Plant Soil 174: 29–49. Wieder R. and Lang G. 1982. A critique of the analytical methods in examining data obtained from litterbags. J. Ecol. 63: 1636– 1642. Wortman C.S., McIntyre B.D. and Kaizzi C.K. 2000. Annual soil improving legumes: agronomic effectiveness, nutrient uptake, nitrogen fixation and water use. Fields Crops Res. 68: 75–83. Wortman C.S. and Kaizzi C.K. 1998. Nutrient balances and expected effects of alternative practices in farming systems of Uganda. Agric. Ecosyst. Environ. 71: 115–129.