We use tank experiments to measure settling rates of deep-sea volcaniclastic material recovered from the Arctic (85 degrees E Gakkel Ridge) and Pacific (Juan de Fuca Ridge, Loihi seamount) Oceans. We find that clast size and shape exert a strong influence on settling velocity, with velocities of similar to 30 cm/s for large (similar to 8 mm), blocky clasts, compared to velocities of similar to 2.5 cm/s for small (<0.5 mm), sheet-like clasts. We fit our observations to the generalized model of Ferguson and Church (2004) to establish empirical scaling laws for settling velocity, and then use these results to test the hypothesis that entrainment in a buoyant plume of hot seawater is an important dispersal mechanism for volcaniclastic material in the deep-sea (Clague et al., 2009). We superpose the observed settling rates on velocity fields generated with the Morton et al. (1956) model for turbulent plumes in stratified media to estimate the rise height of the clastic material under water column conditions corresponding to the Gakkel and Juan de Fuca (JdFR) Ridges, and then estimate dispersal distances assuming the grains settle to the seafloor while being advected in lateral currents. Dispersal distances in our model are a function plume strength (i.e., buoyancy flux), lateral current speeds, and clast settling velocity. Our model demonstrates that large (30 GW) eruption ‘megaplumes’ can loft volcaniclastic material more than a kilometer above the seafloor where entrainment in deep-sea currents can advect dominant clast types (similar to 1 mm, blocky grains) up to a few hundred meters from a source vent. Small bubble-wall fragments (e.g., limu o Pele) entrained in a megaplume could be advected as far as a few kilometers from a source region. These results indicate that entrainment in buoyant seawater plumes during an eruption may play an important role in clast dispersal, but it is not clear if this mechanism can explain the distribution of volcaniclastic material at the sites on the Gakkel and Juan de Fuca Ridges where our samples were acquired. In order to understand the dispersal of volcaniclastic material in the deep-sea it will be necessary to rigorously characterize existing deposits, and develop models capable of incorporating explosive gas phases into the eruption plume. (C) 2011 Elsevier B.V. All rights reserved.