A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN SOIL AND WATER MANAGEMENT OF SOKOINE UNIVERSITY OF AGRICULTURE, MOROGORO, TANZANIADeclining soil fertility continues to be one of the most important challenges in Malawi’s crop production systems, with nitrogen considered as the most limiting nutrient element. However, food legume crops such as pigeon pea (PP) and cowpea (CP) are popular amongst smallholder farmers as they contribute to food security and protein nutrition, source of income to farmers and contribute to soil fertility improvement through biological nitrogen fixation (BNF). They are grown in various cropping systems such as sole cropping, legume-cereal and legume-legume or “doubled-up” intercrops. However, information on BNF, crop productivity, vesicular arbuscular mycorrhizal (VAM) fungal colonisation and contributions of the pigeon pea-cowpea intercrop to subsequent maize (MZ) yields and grain quality is scanty. Therefore, a study to address the aforementioned information gaps was conducted in the 2013/14 and 2014/15 cropping seasons on the Chromic Luvisols of Lilongwe and Dowa districts of Central Malawi. Specific study sites were at the Lilongwe University of Agriculture and Natural Resources (LUANAR), Department of Crop and Soil Sciences Research Farm (14o 11′ S, 33o 46′ E) and at Nachisaka Extension Planning Area (EPA) (13o 37′ S, 33o 56′ E) in Lilongwe and Dowa districts, respectively. The study was aimed at optimizing pigeon pea and cowpea intercropping practices for increased yields of subsequent maize in rotation. Specifically, it included evaluation of the extent of nodulation and BNF by the PP and CP and their productivity in sole cropping, legume-legume and legume-cereal intercropping systems. Additionally, the VAM fungal colonisation of the PP and CP-based systems and rotational maize, and its contribution to the BNF and crop growth was also assessed. Furthermore, N mineralization patterns in the legume-based system plots with residues retained and the rotational maize plant N uptake were assessed. Finally, maize grain and total dry matter (TDM) yields, harvest index percentages (HI %), grain crude protein contents and nitrogen use efficiencies (NUE) in the subsequent season were evaluated. In the first cropping season (2013/14), the experiment was arranged in the randomized complete block design (RCBD) whereby pigeon pea, cowpea and maize were grown as sole crops, legume-cereal and legume-legume intercrops. The split plot design was used in the second cropping season to grow maize in an integrated soil fertility management (ISFM) approach in which the legume-based systems with residues retained in their plots were the main plots and the 0, 45, 90 and 120 kg ha-1 N fertilizer applications formed the sub-plots. Results showed significant effects of the cropping systems on the performance of the PP, CP and MZ crops. Nodulation was significantly increased (P < 0.05) under sole cropping. Sole cropped PP nodule dry weights were significantly higher (P < 0.05) by 25% and 48% than those of PP in intercrops with CP and MZ, respectively, in the Lilongwe site. Similarly, the nodule dry weights were also significantly higher (P < 0.05) by 25% and 46% compared with those in PP in intercrops with CP and MZ, respectively, in the Dowa site. Significant differences in PP nodule numbers were noted for Dowa, with only slight differences in Lilongwe site. Furthermore, sole cropped CP produced significantly higher (P < 0.05) nodule dry weights by 38% and 36% than that in CP in an intercrop with PP or MZ, respectively, in the Lilongwe site. Similarly, intercropping systems decreased the percentage of nitrogen derived from the air (%Ndfa) and the total amount of N2 fixed by each of the two legume species. The highest amount of biologically fixed N or N2 fixed (92.9 kg ha-1), which was significantly higher (P < 0.05) than that by the PP under both the PP-CP and PP-MZ intercrops by 31% and 36%, respectively, was noted in the Dowa site. However, a comparison of the overall cropping system BNF contribution per unit area showed the combined amount of biologically fixed N (82.9 kg ha-1) from the two component legume crops in the PP-CP “doubled-up” was comparable to that by the sole cropped PP, at the Dowa site. Furthermore, the PP-CP doubled up BNF at Dowa was significantly higher (P < 0.05) than the amounts of N2 fixed by the sole cropped CP (62.5 kg N ha-1), pigeon pea in the PP-MZ intercrop (59.9 kg N ha-1) or CP in the CP-MZ intercrop (13.1 kg N ha-1). However, a different trend was noted at the Lilongwe site. Although the biologically fixed N (85.7 kg ha-1) by the sole cropped pigeon pea was similarly the highest, the combined amount of N2 fixed (57.4 kg N ha-1) by the PP and CP in the pigeon pea-cowpea “doubled-up” was significantly lower than that by the sole pigeon pea, by 33%. From this study it was concluded that both legume-legume and legume-cereal intercropping reduces nodulation and BNF per plant but the overall amount of nitrogen fixed per unit area by the PP-CP “doubled up” can be comparable to that by the sole cropped PP depending on environmental conditions. Similar to the BNF, grain and TDM yields per plant and HI%, were also decreased by the intercropping systems. However, the productivity by all the intercropping combinations (PP+MZ, PP+CP and MZ+CP) was higher than under sole cropping as they all resulted in LERs of greater than one and positive monetary advantage index (MAI) values. The PP+MZ intercrop showed to be the most beneficial in terms of both yields and monetary gains as it produced highest LERs and MAI values at both sites of Lilongwe and Dowa. Furthermore, the partial LERs, relative N and P yields showed maize to be the most resilient when intercropped with either PP or CP whereas cowpea was the most suppressed when intercropped with either PP or MZ. Additionally, the VAM fungal colonisation was not affected by the PP and CP-based cropping systems such as sole cropping, legume-cereal and legume-legume intercrops. However, a weak positive relationship was noted between VAM fungal colonisation and yields, P uptake or BNF. Furthermore, all the legume-based cropping systems led to significant increases of the VAM fungal colonisation of the subsequent maize roots by ranges of 39 to 50% and 15 to 36% in the Lilongwe and Dowa sites, respectively, which showed potential of the PP and CP based systems in influencing the P uptake enhancing VAM associations. Furthermore, interactive effect of the legume residues and inorganic fertilizer led to higher maize grain yields by a range of 30% under treatment that was previously CP+MZ intercrop (1689 kg ha-1) to 59% under treatment that was previously sole cropped CP (2864 kg ha-1) at 0 kg N ha-1 fertilizer application than the treatment that was previously sole cropped MZ (1178 kg ha-1), in the Lilongwe site. Similarly, at the highest rate of N application, 120 kg N ha-1, treatments that were previously legume-based produced higher grain yields than the treatment that was previously sole cropped MZ (3277 kg ha-1) by a range of 28% under treatment that was previously CP+MZ intercrop (4525 kg ha-1) to 42% under treatment that was previously sole cropped CP (5665 kg N ha-1), at the Lilongwe site. A similar trend was observed at the Dowa site. Furthermore, from this study it was shown that mixing high quality pigeon pea and cowpea with the low quality maize residues increased mineralization rates, N uptake, and nitrogen use efficiency by the maize grown after the legumes in rotation, with implications on yields. In addition, increasing inorganic N application increased maize grain crude protein content in both study sites, which indicates increased grain quality. Therefore, it was concluded that for smallholder farmers on the Chromic Luvisols of Lilongwe and Dowa districts, central Malawi, an ISFM approach involving PP and CP, either as sole crops, legume-legume or legume-cereal intercrops can substantially increase rotational maize yields, both quantitatively and qualitatively with the implication on reducing the investment costs of inorganic fertilizers.