Intensifying livestock production by integrating perennial forages has great potential to contribute to sustainable development and livelihoods in the Mekong region. However, the approach taken must be informed by the environmental, social, and cultural context of the region. Accordingly, we review published research papers and reports from relevant research for development projects to identify options for sustainably integrating forages into farming systems, with a focus on sand-dominant soils of southern Laos and Cambodia. First, we examine existing livestock management practices to determine the compatibility of forages as an option to intensify livestock production. Second, we review the environmental properties of rainfed lowland rice systems with sandy soils and their implications for forage growing. Third, we identify and compare the suitability of existing forage genetics that is adapted to these environmental properties. Fourth, we propose adapted varieties, outline appropriate management options, and discuss the sustainable engagement of smallholders in the production of forages. The key findings are as follows: (1) Forages appear compatible with the sociocultural properties of smallholder farming systems in southern Laos and Cambodia because there is an awareness of existing limitations to livestock production, widespread desire to possess livestock for cultural reasons, and mounting pressure to improve the productivity of grazing areas and the efficiency of labor.(2) The limiting properties of the environment are drought, soil acidity, flooding, and soil infertility, which must be addressed in the selection and management of forage genetics. (3) Broadly adapted perennial tropical grasses and herbaceous legumes exist, but these are unlikely to thrive in lowland ecosystems of southern Laos and Cambodia that are prone to both annual flooding and drought. (4) Variations in surface hydrology at the farm scale often result in differentiated environments suitable for differing varieties. Brachiaria sp. hybrid "Mulato II," Panicum maximum, and Stylosanthes guianensis are recommended for droughtprone, acidic sands that are safe from prolonged submergence and would require the least additional management, whilst Paspalum atratum is recommended for low-lying areas with access to irrigation. (5) The transition to perennial forage integration appears to be accessible to farmers and can allow them to rapidly accumulate benefits in terms of saved labor; however, efforts to intensify animal production have been slow and must contend with multiple challenges: poor understanding of animal husbandry and health, cultural views relating to the role of animals in production systems, and poor access to forage and livestock services. These must all be addressed if sustainably intensified animal production is to be achieved in these and similar regions.

SUMMARYShortages of hired labour are leading to greater interest in mechanisation for crop establishment in smallholder agriculture. Due to small field sizes, mechanised planters mounted on four-wheel tractors are not a suitable technology. The Versatile Multi-crop Planter (VMP) was developed for zero tillage (ZT), strip planting (SP) or single pass shallow tillage (SPST) on flat land and for forming and planting on tops of beds, each in a single pass operation, when mounted on a two-wheel tractor (2WT). The aim of the present study was to evaluate the field performance of the VMP in comparison to conventional broadcast seeding and full rotary tillage (2 to 4 passes; called CT) for establishing chickpea (Cicer arietinum L.), jute (Corchorus olitorius L.), lentil (Lens culinaris Medikus), maize (Zea mays L.), mung bean (Vigna radiata L. R. Wilczek), rice (Oryza sativa L.) and wheat (Triticum aestivum L.) in 15 locations of Bangladesh. Plant populations emerging from all single pass operations viz. SP, ZT, and bed planting (BP) were generally satisfactory and in 12 out of 15 experiments plant populations after SP were similar to or greater than after CT. In addition, SP gave comparable or greater plant populations than SPST and BP planting methods. Overall, the SP planting achieved comparable yields and lower costs of establishment than CT. We conclude that effective and reliable planters are now available for sowing a range of crop species on small fields with minimum soil disturbance. This opens up realistic options for the development of mechanised conservation agriculture suited to small field sizes.

Summary. The effects of surface soil applications of organic matter and calcium salts on the subsoil (200–280 mm) of a red-brown earth (Natrixeralf) were investigated in a field experiment. The calcium treatments included gypsum, lime or nil-calcium added to the surface soil. To these calcium treatments, organic matter or nil-organic matter was added. Three years after final applications of organic matter and calcium salts to the surface soil, the physical and chemical properties of the subsoil were measured. The bulk physical measurements of the subsoil (penetrometer resistance, bulk density, macroporosity and plastic limit) were not significantly influenced by the treatments. The electrical conductivity of the subsoil was greater in the gypsum-treated soil than in the lime- or nil-calcium-treated soils. Subsoil pH values were lower in the gypsum-treated soil than in the lime- or the nil-calcium-treated soil. Dispersion (spontaneous and mechanical) of the clay subsoil decreased with gypsum-treated surface soil. In treatments where both organic matter and gypsum were added to the surface soil, the subsoil showed an increased concentration of exchangeable calcium and decreased spontaneous dispersion of the clay.

The time of sowing chickpea (Cicer arietinum L.) in the High Barind Tract of north-west Bangladesh is critical to crop success. To ensure adequate emergence and subsequent crop growth, chickpea relies on residual soil moisture stored in the profile after rice (Oryza sativa L.) cultivated in the preceding rainy season. With the development of mechanised, one-pass minimum tillage sowing, the time between rice harvest and chickpea sowing is decreased, and temperature constraints that limit biomass and/or pod formation and filling may be avoided. Minimum tillage may also limit evaporation from the soil surface compared with traditional, full cultivation procedures. The objective of this study was to identify the optimum sowing time to achieve adequate crop establishment and limit exposure of the chickpea crop to terminal drought and heat stress later in the growing season. Over three experimental seasons, chickpea sowing dates were spread from 22 November to 22 December. Soil water content, crop growth and temperature were monitored to determine the optimum sowing time. Over all seasons and sowing dates, the volumetric soil water content in the seedbed under minimum tillage remained within 17–34%, a range non-limiting for chickpea establishment in glasshouse and field experiments. Late planting (after 10 December) exposed seedlings to low temperatures (<15°C), which limited biomass formation and extended the vegetative growth phase into periods with high maximum temperatures (>35°C), resulting in unfilled pods and depressed grain yield. The preferred sowing time was determined to be 30 November to 10 December to reduce the risk of high temperatures and low soil water content during chickpea reproductive growth causing terminal heat and drought stress, respectively. Mechanised sowing in one operation allows farmers to optimise their time of sowing to match seed requirements for soil water at emergence and may assist farmers to avoid temperature stresses (both low and high) that constrain chickpea vegetative and reproductive growth.

Sowing of chickpea in the heavy-textured soils of north-west Bangladesh with minimum tillage technology aims to increase the timely planting of large areas during a relatively short sowing window before soil water deficit limits germination and emergence. However, the seedbed conditions into which chickpea is sown need to be better quantified, so that limiting factors which affect germination and emergence can be identified.