Abstract: Background and aims: Biological fixation of atmospheric nitrogen (N 2 ) is the main pathway for introducing N into unmanaged ecosystems. While recent estimates suggest that free-living N fixation (FLNF) accounts for the majority of N fixed in mature tropical forests, the controls governing this process are not completely understood. The aim of this study was to quantify FLNF rates and determine its drivers in two tropical pristine forests of French Guiana. Methods: We used the acetylene reduction assay to measure FLNF rates at two sites, in two seasons and along three topographical positions, and used regression analyses to identify which edaphic explanatory variables, including carbon (C), nitrogen (N), phosphorus (P) and molybdenum (Mo) content, pH, water and available N and P, explained most of the variation in FLNF rates. Results: Overall, FLNF rates were lower than measured in tropical systems elsewhere. In soils seasonal variability was small and FLNF rates differed among topographies at only one site. Water, P and pH explained 24% of the variation. In leaf litter, FLNF rates differed seasonally, without site or topographical differences. Water, C, N and P explained 46% of the observed variation. We found no regulatory role of Mo at our sites. Conclusions: Rates of FLNF were low in primary rainforest on poor soils on the Guiana shield. Water was the most important rate-regulating factor and FLNF increased with increasing P, but decreased with increasing N. Our results support the general assumption that N fixation in tropical lowland forests is limited by P availability. © 2019, The Author(s).