We compare the physical reach of a Neutrino Factory in the 2+2 and 3+1 fourfamily models, with similar results in the two schemes; in both cases huge CPviolating effects can be observed with a near detector in the ν µ → ν τ channel. We also study the capability of long baseline experiments (optimized for the study of the three-family mixing parameter space) in distinguishing a three (active) neutrino model from a four-family scenario.Key words: NUFACT01, neutrino, sterile, oscillations, CP-violation.The present experimental solar and atmospheric neutrino data give strong indications in favour of neutrino oscillations. In addition, the results of the LSND experiment [1] would imply the existence of a puzzling fourth, sterile, neutrino state. There are two very different classes of spectra with four massive neutrinos: two pairs of almost degenerate neutrinos divided by the large LSND mass gap (the 2+2 scheme); or three almost degenerate neutrinos and an isolated fourth one (the 3+1 scheme). The former gave a better fit, as was shown in [2], but the recent SNO results [3] will certainly restrict the allowed parameter region. The latter is at present only marginally compatible with the data [4].An experimental set-up capable of precision measurement of the whole three-neutrino mixing parameter space (including the CP violating phase δ) is under study. This experimental programme consists of the development of a "Neutrino Factory" (highenergy muons decaying in the straight section of a storage ring and producing a very pure and intense neutrino beam [5,6]) and of suitably optimized detectors. We shall consider in what follows a neutrino beam resulting from the decay of n µ = 2 × 10 20 unpolarized positive and/or negative muons per year. The collected muons have energy E µ in the range 10 − 50 GeV.