“…Several groups have studied the charm particle production by CRs [4,5,6], which actually requires to take into account next to leading order (NLO) processes, which increase the charm production by more than a factor of two, and the cross sections are sensitive to the values of the partonic distribution functions at very small values of the scaling variable x, what introduces further uncertainties in the results, due to the need to extrapolate beyond the measured range. Even if the charmed mesons (D ± , D 0 , D s , Λ c , ...) are produced at a rate ∼ 10 −4 with respect to the non-charmed ones, their very short decay time, with L D ≃ 2 km(E D /10 PeV), implies that the prompt neutrino spectral slope just follows the CR slope, and hence the prompt neutrinos eventually dominate the atmospheric neutrino fluxes above a PeV in the vertical direction, and somewhat above that energy at large zenith angles 1 .…”