Abstract: Abstract: The gasification behavior of agriculture waste was investigated using a Top-lit Up Draft (TLUD) fixed-bed gasifier system which consists of four subsystems (gasifier reaction system, temperature monitor system, tar sampling system, and gas measurement system), in an attempt to find out the influence of air flow on the gasifier's temperature distribution, gas component, and heating value, and discuss the gasification mechanism of biomass. In the study, peanut shells and rice husks were chosen as experiment material. The reactor is made from Ф219 mm×6 mm stainless steel pipe with length of 800 mm. After the reactor filled up with sample materials (about 3.5 kg), a certain amount of air (range from 2.5 m3/h to 4.0 m3/h) was pumped in at the bottom, and the sample was lit at the top. The temperature was then tested by a temperature monitoring system, and the gas product was detected by a Gasboard-3100P-type multi-component portable infrared gas analyzer (Quartet Photoelectric Technology Co., Ltd. Wuhan, China), and the tar was collected by a tar sampling system. From the results of the investigation, the gasification process lasted about 100 min, and was divided into two stages. First, the main reaction was biomass drying, volatile releasing, and char gasification, and the main gas products were CO2, CO, H2, and CH4. About 50 min later, it went into the second stage, and the oxidization of solid char residue played a major role in the gasifier with a large amount of CO production. Throughout the whole reaction, the heating value of gas products was a minimum of 3 MJ/m3, and the evolution of the gasfier temperature profile was very similar with that of wave propagation. During the first stage of gasification, the maximum temperature of the gasifier was 730℃; when it switched into the second stage, the main reaction occurred at the bottom, and the max temperature would rise up to 1000℃. The airflow rate can be converted to the Air/Fuel equivalence ratio (ER), the most important aspect of gasifier operation. ER had great influence on biomass gasification property, and higher ER was favorable for the enhancing of gas yield, however, it had different effects on the releasing of flammable gas (CO and H2). The optimum gasfication condition was achieved when ER is 0.31 and 0.35 for peanut shell and rice husk, respectively, and the tar content of gas is very low (as low as 0.25g/m3 and 0.49g/m3, respectively). This study would be helpful to the design and operation of TLUD fixed-bed gasifiers.