We analyze n−n oscillations in generic models with large extra dimensions in which standard-model fields propagate and fermion wavefunctions have strong localization. We find that in these models n −n oscillations might occur at levels not too far below the current limit.11.10. Kk,11.30.Fs,14.20.Dh Although current experimental data are fully consistent with a four-dimensional Minkowski spacetime, it is useful to explore the possibility of extra dimensions, both from a purely phenomenological point of view and because the main candidate theory for quantum gravity -string theory -suggests the existence of higher dimensions. Here we shall focus on theories in which the standard-model (SM) fields can propagate in the extra dimensions and the wavefunctions of the SM fermions have strong localization (with Gaussian profiles) at various points in this extra-dimensional space [1]- [9]. Such models are of interest partly because they can provide a mechanism for obtaining a hierarchy in fermion masses and quark mixing. In generic models of this type, excessively rapid proton decay can be avoided by arranging that the wavefunction centers of the u and d quarks are separated far from those of the e and µ [1]. However, this separation does not, by itself, guarantee adequate suppression of another source of baryon number violation, namely n −n oscillations. Here we shall analyze these oscillations in generic models of this type. Early studies of n −n oscillations in conventional d = 4 dimensional spacetime include [10]-[16]; there is currently renewed experimental and theoretical interest [17,18].Our theoretical framework is as follows. Usual spacetime coordinates are denoted as x ν , ν = 0, 1, 2, 3 and the n extra coordinates as y λ ; for definiteness, the latter are taken to be compact. The framework is such that fermion fields have the form Ψ(x, y) = ψ(x)χ(y). In the extra dimensions the SM fields are assumed tofields thus have Kaluza-Klein (KK) mode decompositions. We shall work in a low-energy effective field theory approach with an ultraviolet cutoff M * . These models provide a possible explanation for the hierarchy in the fermion mass matrices via the localization of fermion wavefunctions with half-width µ −1 << L at various points in the higherdimensional space. We denote ξ = µL; ξ ∼ 30 is chosen for adequate separation of the various fermion wavefunctions while still fitting well within the thick brane. As a result of this localization, the y-dependent part of the wavefunction for a fermion field f has the generic form χ f (y) = Ae 1/2 . For ℓ = 1 and ℓ = 2, this fermion localization can be accomplished in a low-energy fieldtheoretic manner by coupling to a scalar with a kink or vortex solution, respectively [19]. The normalization factor A = (2/π) ℓ/4 µ ℓ/2 is included so that after the integration over the ℓ extra dimensions, the kinetic termψ(x)i∂ /ψ(x) has canonical normalization. Starting from an effective Lagrangian in the d-dimensional spacetime, one obtains the resultant effective theory in four dimensio...