However, the differences in nodule physiology among RILs(Fig. 1A) may offer an important clue in investigating the key pro-cesses that influence nodule N2fixation activity under P deficiency.Thus, the increase in nodule O2permeability under P deficiency(Ribet and Drevon, 1995; Vadez et al., 1996; Alkama et al., 2012)would predict a higher discrimination against15N entry withinnodules as N2molecules with either one or two15N atoms. How-ever, this was not confirmed in our work, since we could not findany significant correlation between nodule O2permeability (datanot shown) and the15N/Nt ratio in nodules, regardless of the P sup-ply. Indeed, isotopic discrimination can occur when the N substrateis not consumed to completion (Högberg, 1997). In contrast, highrates of N2fixation and use of the fixed N for plant growth couldlead to low levels of15N discrimination, possibly because most ofthe N2that enters into the nodule would be accumulated via N2fixation, masking any effect of the O2-barrier against15N duringN2diffusion through the nodule cortex. Our results thus open upthe possibility for further investigations about possible interactionsbetween the genotypic and physiological regulation of P use effi-ciency on one hand, and nodule N metabolism and discriminationagainst15N on the other hand