A 200 mm laboratory-scale atmospheric bubbling fluidised bed reactor has been used to obtain experimental data for the air/steam gasification of eucalyptus red gum wood chips and commercial wood pellets. The unique feature of this gasifier is the ability to examine the variations to axial gas composition along the bed height. At present no such data is available in the literature for biomass gasification. Gasification tests were performed using beds of, silica sand, char or clay to determine the effect of bed type on the gas composition. The behaviour of the major gas species (CO, H-2, CO2) were observed to be strongly influenced by the water-gas shift reaction within the freeboard of the gasifier resulting in the exit gas being relatively similar in composition as compared to the in-bed variations. These small differences in gas composition for all bed types tested are the result of the achievement of equilibrium in the water-gas shift reaction. The influence of bed type exerted the most impact on the C-2-C-3 emissions (tar proxy) with the char bed found to best aid in their breakdown and to limit the amount of hydrocarbons surviving into the freeboard. The reduction of iron oxide (FE2O3) content in the clay to a more reactive form of magnetite Fe3O4 by CO and H-2 in the product gas resulted in the clay bed to also exhibit a reduction in C2-C3 emissions compared to silica sand but less then char. The clay bed produced the highest calorific values for the producer gas. However, operation of the clay bed above 800 degrees C exhibits the potential for over reduction to form iron with subsequent agglomeration of the bed. Changing the fuel type to a biomass pellet resulted in higher emissions Of C-1-C-3 hydrocarbons and in part its contribution is the result of primary particle fragmentation during screw feed conveying to the bed. Feeder location and bed design (conical or cylindrical) also exhibit an influence on hydrocarbon emissions. (c) 2006 Elsevier Ltd. All rights reserved.