Biol Fertil Soils (2004) 40: 20–27 DOI 10.1007/s00374-004-0740-8 ORIGINAL PAPER T. S. Chirwa . P. L. Mafongoya . D. N. M. Mbewe . B. H. Chishala Changes in soil properties and their effects on maize productivity following Sesbania sesban and Cajanus cajan improved fallow systems in eastern Zambia Received: 14 July 2003 / Accepted: 16 February 2004 / Published online: 3 April 2004 # Springer-Verlag 2004 Abstract Improved fallows with leguminous trees have been developed in Southern Africa as a viable alternative to inorganic fertilizers but the changes in soil properties that are responsible for crop productivity improvement and implications of mixing litter and fresh leaves from the same tree species on soil fertility are not fully understood. Our objectives were to quantify (1) some changes in soil properties that are responsible for crop production improvement under improved fallow systems; (2) the N mineralization patterns of mixtures of litter and fresh leaves from the same tree species. The treatments used in the study were 2-year planted Sesbania sesban (sesbania) and Cajanus cajan (cajanus) and controls of natural fallow, continuous fertilized and unfertilized maize. At fallow clearing sesbania contributed 56 kg N ha -1 through litter and fresh leaves. Sesbania (fresh leaves + litter) showed high N mineralization after 10 weeks compared to the mixture of cajanus fresh leaves with litter. Maize yields were significantly correlated with preseason NO 3 - -N and total inorganic-N content of the top 20-cm soil layer. Soil penetrometer resistance at 4 weeks after planting was lowest in the sesbania land-use system (2.2 Mpa), whereas the highest percentage of water-stable aggregates at fallow clearing and crop harvest was in sesbania (83%) and cajanus (77%), respectively. The improved soil conditions and N contribution of sesbania and cajanus fallows to the subsequent maize crop was evidenced by increased maize yields of between 170–200% over maize without fertilizer. Keywords Improved fallows infiltration rate . Nitrogen mineralization . Stable aggregates . Penetration resistance Introduction Under traditional farming methods, farmers have relied on short natural or shrub fallows to grow maize and other crops. In eastern Zambia this fallow system is locally known as “cisala” (Kwesiga, personal communication). Nye and Greenland (1960) also reported that natural fallows have long been used to overcome soil fertility depletion that results from continuous cropping with no nutrient inputs. The fallow period may vary from 5 to 20 years. However, long fallow periods have become impractical because of increasing human and livestock populations. Losses of mineral nutrients during the cultivation phase, through runoff, erosion, leaching and crop removal, can no longer be restored by short 1–5 year periods of natural bush fallow (Brady 1996). This has necessitated the need for improved fallows with legume species, planted deliberately in order to achieve the aims of natural fallow within a short time or a smaller area (Prinz 1986). Nitrogen limits maize crop production over large areas of Zambia. The main sources of plant-available N are mineralization of soil organic matter (SOM), biological N 2 fixation (BNF), fertilizers and organic inputs (e.g., plant residues, composts and manures (Giller et al. 1997). An improvement in soil physical properties could be another reason for yield increase but little quantitative data exist on these changes. At fallow clearing there is a mixture of fresh leaves and litter. The N release from organic residues of litter + fresh leaves has not been studied even though these are the main source of N at fallow clearance. Studies by Mafongoya and Nair (1997) under field conditions have shown that lignin, polyphenols and N content have a significant effect on N release and maize yield. Research on mixing of legume tree prunings from different species of high quality with low quality has been done by some researchers (Handayanto et al. 1995; Mafongoya and Neir 1997). High quality materials are defined as those materials with N concentrations greater than 25 g kg -1 , lignin values less than 150 g kg -1 and soluble polyphenols less than T. S. Chirwa (*) . P. L. Mafongoya Msekera Research Station, ICRAF-Zambia Agroforestry Project, P.O. Box 510089 Chipata, Zambia e-mail: tedchirwa@yahoo.com Tel.: +26-062-21725 Fax: +26-062-21404 D. N. M. Mbewe . B. H. Chishala School of Agricultural Sciences, University of Zambia, P.O. Box 32379 Lusaka, Zambia