In recent years, reconstituted small samples have often been used to assess the performance of radial consolidation due to prefabricated vertical drains (PVDs), but the permeability and compressibility of samples of undisturbed soil often differ from those of the remoulded ones. The problem seems more complex in marine environment due to the presence of random coarse particles including gravels, shells and natural partings. Performing small-scale laboratory experiment with reconstituted samples, especially in marine environment, cannot predict the exact soil behaviour in the field. This paper describes an experimental programme that measures radial consolidation using a conventional Rowe cell and a large-scale consolidometer, where the samples of undisturbed soil obtained from a site along the Pacific Highway (north of Sydney) were compared using measured settlements and excess pore pressures. Moreover, this paper highlights the implications of the smear effect and sample size influence, which are imperative in translating the laboratory testing practices to actual real-life behaviour. The effect of vacuum pressure on the coefficient of radial consolidation of a large-scale undisturbed test specimen is also discussed. The paper demonstrates that the extent of smear zone in the field can be very similar to the large-scale laboratory consolidation test using a scaled-down drain and mandrel, but considerably different from the data obtained for small laboratory specimens.