Quantitative changes of SOM are a ramification oftwo effects of fire: i) the removal of most or all of the litter layer and, possibly, some organics from the upper few centimetres of mineral soil, the only ones that really experience high temperatures;
and ii) the input of charred materials from the scorched stems and crowns (Alexis et al., 2007). Net fire-induced gain or loss of SOM results from the dominance of one of these processes. The C balance of any study focused to assess such gain or loss is highly affected by the sampling procedure. Due to fire induced heterogeneity it requires: i) collecting a high number of soil samples in both unburned and burned areas to have a sufficiently large database, ii) sampling by volume, because fire may change the bulk density of soil, and iii) taking into account the litter layer at the unburned area. The litter layer is in fact often ignored although it is an integral part of forest soils. Delayed investigations should account for post-fire erosion, which raises SOM loss by removing the barren topsoil. Finally, when studies are performed in forests burned by wildfires rather than controlled fires, the comparison between the pre-fire and post-fire soil is necessarily done on adjacent plots, which must display uniformity in terms of topographic, vegetation and pedologic characteristics.