Introduction. Experiments in JET with the ITER-like Be wall and W MkII-HD divertor (JET-ILW) have highlighted a sensitivity of L-H transitions to low-Z plasma impurity composition: reduced effective charge, Zeff, from JET-C (line averaged Zeff ~ 1.6­2.0, C as main intrinsic impurity) to JET-ILW (Zeff ~ 1.0­1.2, Be as main intrinsic impurity) and concomitant reduction in L-H power threshold, PL-H, in the high density branch by 30-40%. The transition to H-mode also occurs at lower edge temperature than in JET-C. PL-H denotes here the net power across the separatrix, Psep = Ploss ­ Prad,bulk. In addition, a minimum in PL-H versus density is observed in divertor configurations with the outer strike point on the horizontal target (V5 configuration), whereas matched discharges in JET-C show PL-H monotonically increasing with density, thus following the 2008 ITPA L-H scaling law. The reduction in PL-H in JET-ILW projects favourably to ITER, however the underlying physics mechanism is not yet understood. Moreover, the density of minimum PL-H, ne,min, increases with BT4/5 in JET-ILW (at ~ constant q95 = 3.3­3.7 and IP varying from 1.7 to 2.75MA). In JET-C a minimum in PL-H with ne had been previously observed with the more closed MkII-GB divertor geometry. All these findings show that divertor/SOL physics are strong players. A recent, physics based scaling for ne,min, proposed in [3] (based on the ion energy channel being the important one for the L-H transition) is not compatible with the full set of JET observations described above.