The shift of incubation temperature in anoxic paddy soil from 30°C to 15°C resulted in a reversible decrease of the methane production rate and of the H2 steady state partial pressure. Only at 30°C but not at 17°C, total CH4 production rates were enhanced by the addition of H2, acetate, or cellulose compared to the control. Apparent activation energies which were calculated from the temperature dependence of CH4 production were higher in presence than in absence of excess H2. Decrease of temperature caused a decrease of the H2 turnover rate constant and of the Gibbs free energy of H2‐dependent methanogenesis, and also resulted in a smaller contribution of H2 to total methanogenesis. However, H2‐dependent methanogenesis was significantly stimulated by excess H2 and slightly inhibited by acetate at low as well as high temperature. The results show that H2‐producing bacteria were limited by temperature to a greater extent than the methanogens so that the methanogenic microbial community in paddy soil was limited by the supply of H2. At low as well as high temperatures, excess H2 apparently enabled part of the methanogenic community to shift from acetate‐dependent to H2‐dependent CH4 production. At low temperature, excess H2 had only this effect, but with increasing temperature, excess H2 additionally stimulated total methanogenic activity and eventually even growth.

Dilution of anoxic slurries of paddy soil resulted in a proportional decrease of the rates of total methanogenesis and the rate constants of H2 turnover per gram soil. Dilution did not affect the fraction of H2/CO2‐dependent methanogenesis which made up 22% of total CH4 production. However, dilution resulted in a ten fold decrease of the H2 steady state partial pressure from approximately 4 to 0.4 Pa indicating that H2/CO2‐dependent methanogenesis was more or less independent of the H2 pool. The rates of H2 production calculated from the H2 turnover rate constants and the H2 steady state partial pressures accounted for only < 5% of H2/CO2‐dependent methanogenesis in undiluted soil slurries and for even less after dilution. Upon dilution, the Gibbs free energy available for H2/CO2‐dependent methanogenesis decreased from −28.4 to only −5.6 kJ per mol. The results indicate that methane was mainly produced from interspecies H2 transfer within syntrophic bacterial associations and was not significantly affected by the outside H2 pool.