Submarine density flows dominate sediment transport into many parts of the ocean and form submarine fans, which are some of the largest sediment accumulations on Earth. Previous studies often assumed that the distal fringes of submarine fans would be dominated by extensive sheet-like deposits from dilute and expanded turbidity currents, because only turbidity currents would transport large volumes of sediment for long distances across such low gradients. Deposits with a remarkable frondlike shape occur at the furthest fringe of the Mississippi submarine fan. Understanding the emplacement process of these deposits is important because their frondlike shape suggests deposition from debris flows rather than turbidity currents. Previous analyses concluded that these deposits comprise a complex arrangement of thin interbedded turbidity current and debris flow deposits. Here we propose a different internal geometry for the frondlike deposits. Intervals previously described as in place turbidites are interpreted to be clasts within a single debrite interval that is ~ 1 to 2 m thick. This debrite interval is underlain by clean sand. The debrite interval could have been emplaced by multiple debris flows, or one event comprising multiple pulses. Clasts in the debrite interval are much bigger than previously thought, including boulders with diameters that can exceed 50 cm. We show that thin (~ 2 m) and extremely fluidal debris flows could have carried these boulders across gradients of only 0.06° to the fringes of one of the world's largest submarine fans, without hydroplaning. This long transport distance could occur if the density of the clasts was less than that of the surrounding debris flow. Emplacement of the basal clean sand layer appears closely linked to debris flow deposition because the clean sand layer pinches out in a location similar to that of the overlying debrite. We therefore suggest that the basal clean sand most likely settled out from the overlying debris flows, and that the clean sand is not a fore running turbidity current deposit. Shearing at the base of a mud-rich debris flow would most likely produce a muddy basal sand interval, rather than the clean sand that is observed. This basal sand interval is therefore less likely to be a record of a lubricating layer with high-pore-fluid-pressure on which the debris flows moved. This study shows how thin and highly mobile debris flows can redistribute large volumes of sediment including bouldersized clasts long distances into the deep ocean across remarkably low gradients. Copyright © 2010, SEPM (Society for Sedimentary Geology).