The multiple environmental issues of loss of forest cover due to cattle farming combined with pasture degradation leading to low levels of production, consequent extensification and therefore to more deforestation, are serious concerns in Costa Rica. To test the feasibility of countering these by combining a more productive pasture system with indigenous tree species, a silvopastoral experiment was established on a farm in the seasonally dry lowlands of Cañas, Guanacaste Province. A rapidly growing pasture species (Brachiaria brizantha) was tested against a traditional pasture dominated by Hyparrhenia rufa. Three indigenous tree species were established: Pithecellobium saman, Diphysa robinioides and Dalbergia retusa. Plots were grazed by cattle for 4 or 5 days with one to 2 month intervals between grazing episodes. After 51 months, D. robinioides was the fastest growing species, and P. saman the slowest, while B. brizantha produced three times the above ground and twice the below ground biomass as H. rufa, and trees had no effect upon grass yield. Contrary to competition theory, there was no effect of pasture species upon the two faster growing tree species. The carbon in above and below ground phytomass varied between 3.5 and 12.5 Mg C ha−1 in treeless pasture controls and silvopastoral systems, respectively, and total soil organic carbon (TSOC) in the upper 0.6 m averaged 110 Mg ha−1. B. brizantha appeared to stimulate tree root production, which in turn was highly correlated with TSOC, resulting in annual increments in TSOC of up to 9.9 Mg ha−1 year−1. These early results indicate the promising potential of this silvopastoral system for combining cattle production, and increasing tree cover and carbon sequestration.