Experiments on methanogenesis from horse dung were conducted in laboratory-scale batch reactors in order to determine the substrate performance in a solid-phase digestion process, more specifically in terms of potential energy recovery and suitable process technology. Dung from a horse stable with straw bedding was used. The temperature was kept in the mesophilic range. In the percolation process (with process water sprinkled over the stacked biomass) a proportion of 10-20% of solid inoculum (pre-digested horse dung) was found to be suitable. Comparative experiments with both percolation and flooding revealed a higher biogas production per volume for the flooded process, as no addition of solid inoculum was necessary. Methane yield from fresh material was similar in both processes: around 170 L(N) CH(4) per kg VS added was obtained in six-week cycles with untreated material under optimized conditions. Methane production was increased after chopping the substrate. Pre-aeration resulted in decreased methane production.