Submarine lobe dimensions from six different systems are compared: 1) the exhumed Permian Fan 3 lobe complex of the Tanqua Karoo, South Africa; 2) the modern Amazon fan channel-mouth lobe complex, offshore Brazil; 3) a portion of the modern distal Za??re fan, offshore Angola/Congo; 4) a Pleistocene fan of the Kutai basin, subsurface offshore Indonesia; 5) the modern Golo system, offshore east Corsica, France; and 6) a shallow subsurface lobe complex , offshore Nigeria. These six systems have significantly different source-to-sink configurations (shelf dimension and slope topography), sediment supply characteristics (available grain size range and supply rate), tectonic settings, (palaeo) latitude, and delivery systems. Despite these differences, lobe deposits share similar geometric and dimensional characteristics. Lobes are grouped into two distinct populations of geometries that can be related to basin floor topography. The first population corresponds to areally extensive but thin lobes (average width 14km??length 35km??thickness 12m) that were deposited onto low relief basin floor areas. Examples of such systems include the Tanqua Karoo, the Amazon, and the Za??re systems. The second population corresponds to areally smaller but thicker lobes (average width 5km??length 8km??thickness 30m) that were deposited into settings with higher amplitude of relief, like in the Corsican trough, the Kutai basin, and offshore Nigeria. The two populations of lobe types, however, share similar volumes (a narrow range around 1 or 2km3), which suggests that there is a control to the total volume of sediment that individual lobes can reach before they shift to a new locus of deposition. This indicates that the extrinsic processes control the number of lobes deposited per unit time rather than their dimensions. Two alternative hypotheses are presented to explain the similarities in lobe volumes calculated from the six very different systems. The first states that the wide range of starting flow volume and grain size across all systems enters the basin floor as a narrow range due to slope 'filtering' via more overspill and intra-channel deposition in larger systems. The second hypothesis is a result of the gradual decrease in downstream gradient from the distributive channel base to the lobe top during lobe growth. This is not sustainable as the channel will start to aggrade, and when a steeper lateral gradient is present, an avulsion will occur to an adjacent depositional low, which will be used for flows to fill and build a new lobe. This analysis of submarine lobe volumes indicates that the basin floor topography influences lobe geometry, but the fact that lobe volumes have a narrow range indicates a strong influence of intrinsic processes. ?? 2010 Elsevier B.V.