Seismic refraction and single-channel reflection data taken along 0.5-, 2.5-, and 4.5-m.y.-age isochrons near the East Pacific Rise during Project ROSE are used to determine if a systematic change in the P velocity-depth function with age can be resolved. Inversion of these data suggests that any change in crustal P velocity structure related to age is smaller than variability in the seismic velocity-depth function along an isochron. The emergence of a 'normal' crust-mantle transition by 4.5-m.y.-age is seen in these data. Crust and crust-mantle transition zone heterogeneity along these isochrons may be related to the along strike variability in processes at the ridge crest. The velocity-depth functions for the threee split profile refractions lines are compared with velocity-depth functions for the Samail ophiolite, which is thought to represent oceanic crust of similar age. The velocity-depth functions for the ROSE data are bounded by different velocity-depth models for the Samail ophiolite; this suggests that those models are not in disagreement but represent the lateral heterogeneity that can be expected in young oceanic crust.