Weeds are a perennial problem in coconut plantations and cause significant losses in terms of nut yield. The occurrence of a wide range of weeds also causes difficulties in their eradication. The influence of buffalo grazing on weed biomass and nut yield was evaluated to determine the economical and effectiveness of this method to control weeds in coconut plantations in the low country wet zone of Sri Lanka. Treatments imposed were: tractor slashing as control (T1), cover cropping with Pueraria phaseoloides (T2) and buffalo grazing (T3). Slashing treatment was applied three times per year and buffalo grazing was practiced once a month to control grasses and weeds. The treatments were arranged in a Randomized Complete Block Design with three replicates. Based on the reduction of the weed biomass, treatments T2 and T3 were found to be significantly effective (P<0.05) over slashing treatment. Coconut yield was increased significantly (P<0.05) in buffalo grazing plots and decreased in cover cropping plots. Control of weeds with buffalo grazing and cover cropping with Pueraria resulted in 7% increase and 6% decrease in nut yield over the slashing plots respectively. Buffalo grazing was found to be the most effective method of controlling weeds in coconut plantations. Cover cropping with Pueraria phaseoloides was effective in controlling weeds in the long term, but was not economically beneficial compared with the buffalo grazing. Establishment of cover crop was helpful to conserve soil moisture but it appeared to compete with palms for soil nutrients. A high soil penetrometer resistance in the grazing plots pointed to a significant change in the soil structure as a result of buffalo traffic in the area. Harrowing the buffalo grazing plots appeared to be the best method to overcome soil compaction and finally increase soil aeration and create more favorable environment for palm growth.

ABSTRACT.The present study was carried out with the objective of develoing a leguminous leaf meal (LM)

The objective of the present study was to find the effect of feeding of Gliricidia (Gliricidia sepium) leaf meal blocks on intake, live weight gain and milk yield of dairy cows. Two types of Gliricidia leaf meal blocks (B1 and B2) were prepared and compared with a control diet of grass and straw. Gliricidia leaf meal blocks were prepared using Gliricidia leaf meal, coconut (Cocos nucifera) poonac and rice (Oryza sativa) bran. A feeding trial was conducted according to 3x3 Latin square design using lactating cows. Treatments were basal diet (T1), basal diet with B1 (T2) and basal diet with B2 (T3). Basal diet consisted of Brachiaria brizantha, Coimbatoor 3 (Pennisetum perpureum x Pennisetum americarnum) and rice straw. Data were statistically analyzed using general linear model procedure in SAS version 8. Average dry matter intake of cows fed with both types of blocks were higher (p<0.05) compared to control diet. Similarly, live weight gain of cows fed with Gliricidia leaf meal blocks were also higher compared to the group fed with the control diet. Milk production was not affected by treatments. However, average milk production of cows fed with T3 (basal feed + B2) was significantly (p<0.05) higher (2.4 L/day/cow) than the other two treatments. Therefore, Gliricidia leaf meal block (B2) prepared with coconut poonac and rice bran can be recommended as the best block having favourable nitrogen and energy balance among the three diets, to be used with a roughage diet for indigenous cross bred cows.

This study evaluates the economic feasibility and biological productivity of coconut-based agroforestry models designed for coconut smallholders. Coconut Research Institute of Sri Lanka has developed and established twenty-six agroforestry models in farmers' fields. Of them, four models were selected for this study. NPVs of agroforestry models were higher than the monocrop, suggesting the higher net worth of the agroforestry models in the long run. Other economic indicators also showed that agroforestry models were more profitable than coconut monoculture. In all models, LER exceeded unity, indicating the yield advantage with agroforestry over coconut monocultures. However, RYT values indicated that there were negative interactions between component crops of certain models, which can be attributed to the competition that occurs between crops.

One hectare of coconut land (156 palms ha-1) planted with gliricidia (2250 trees ha-1 into double rows in avenues of coconut), available natural pasture and outside supplied paddy straw coupled with six buffaloes were mixed into a farming model to examine total productivity and potential of green energy production by wood and biogas.  Six buffaloes were maintained in a shed and the manure was collected for biogas generation. Biogas was purified from H2S and the bio fuel was used to run a 0.75 hp engine. Wood of gliricidia was used to energize a 3.5 kW gassifire-engine system for generating electricity. Productivity of gliricidia in the 3rd year was 4.5 kg wood (at 20% moisture) and 3.5 kg of fresh foliage per tree/year.  Each buffalo produced an average of 622 liters of milk at the first lactation (for a period of 10 months) and value of a calf was Rs. 41244.00 at the age of 17 months.   Soils of the model were enriched by adding the effluent of biogas digester. The soil fertility (N, P, K, Mg, moisture holding capacity) improved significantly over soils sampled outside the model. The effluent of biogas was more fertile than dried buffalo dung. As a result, nut yield of coconut palms increased from 30 to 60 nuts palm-1 year-1 over a period of two years. In green energy production, the dung of six buffaloes passed through a 0.5 kW biogas-engine system generator produced 2 kwh of electricity per day which if generated by diesel driven generator would require 0.9 liters of diesel. The annual generation of electricity from the biogas amounted to 700 kwh. On the other hand, gliricidia wood of one hectare of coconut land was sufficient to energize 3.5 kW gassifire – engine – generator set for 1600 hours/year and this was equivalent to 5000 kwh of electricity (kWh). Thus, one hectare of coconut/gliricidia/natural pasture/paddy straw with six buffaloes was able to produce green energy equivalent to 5700 units of electricity (kWh) or 2,565 liters of diesel, in addition to farm income derived from coconut, buffalo milk and other benefits of buffalo farming. The total return (coconut, selling of calves, buffalo milk and bio fertilizer) added up to Rs. 704,070 ha-1 year-1.  The feasibility of this model for adoption by small farmers of the coconut triangle in Sri Lanka is presented on the basis of this study.