Sharp gradients with depth (z) of fluxes (F) and concentrations (C) of organic particles are present in the upper few 100 m of the open sea. Because of under-determination of the system in the field, the balance of processes that maintain these gradients cannot yet be specified. A new scenario, with both quickly settling ( ∼ 100 m d-1), slowly decaying material and slowly sinking ( ∼ 20-33 m d-1), labile material that is degraded from within, mimics the observed gradients of F and C. A critique of recent ideas, which minimize decomposition of particles from within, notes that unambiguous interpretation in differences of cumulative particle collection between poisoned and live traps is difficult if daily decomposition rates of ∼ 20 to ∼ 30% prevail; these are claimed here for the labile fraction but have occasionally been observed even in bulk material from traps. The gradients with depth of F or C cannot be modeled as due to a first-order reaction. In addition, the description C = a zb may require b to change below the euphotic zone; quantitative and especially qualitative changes of mesozooplankton (0.2-0.3 mm) with depth are inferred as the cause. Finally, as in the euphotic zone, mesozooplankton might generate substantial amounts of dissolved organic matter at depth, which is presumably degraded by free-living bacteria. Attention to animals is urged for flux studies. © 1990.