Nowadays, fossil fuels represent more than half the total energy consumption in developed countries. As biomass is a renewable resource and fi xes carbon when growing, there is a strong interest to use it as energy, as bio-fuel, and in construction. In some countries like France, ambitious plans for increasing harvest have been made until 2020 (namely +35% of annual wood harvesting). When dealing with such stakes, an accurate and reliable estimation of forest resource is needed. When they are built from empirical approaches, traditional biomass equations are hampered by diffi culties to extrapolate to other species and site conditions. To build biomass models that are as robust as possible, it is important to explore the variability for many species on a whole country. We gathered biomass measurements from the main seven French forest organizations into a unique database. This dataset includes volume and/or biomass for around 150,000 trees from 1920 to 2009, half of them having been precisely measured up to 0- or 4-cm diameter in stems and branches, and more than 1 million trees from the National Forest Inventory. Here we show interesting results and analyses, linked to theoretical considerations of biomass distribution in trees.