Variations in the accumulation rate of barium in marine sediments are thoug
ht to be indicative of variations in marine biological productivity through
time. However, the use of Ba as a proxy for paleoproductivity is partly de
pendent upon its being preserved in the sediment record in a predictable or
consistent fashion. Arguments in favor of high Ba preservation are partly
based on the assumption that sediment porewaters are generally at saturatio
n with respect to pure barite. The idea is that because nondetrital sedimen
tary Ba predominantly exists as barite, porewater saturation would promote
burial. We present sediment porewater, sediment solid phase, and benthic in
cubation chamber data suggesting that solid-phase Ba preservation may be co
mpromised in some geochemical settings. We propose that under suboxic diage
netic conditions, characterized by low bottom water oxygen and high organic
carbon respiration rates, Ba preservation may be reduced. Independent of t
he mechanism, if this assertion is true, then it becomes important to know
when the Ba record is unreliable. We present evidence demonstrating that th
e sedimentary accumulation of authigenic U may serve as a proxy for when th
e Ba record is unreliable. We then provide an example from the Southern Oce
an during the last glacial period where high authigenic U concentrations co
incide with high Pa:Th ratios and high accumulation rates of biogenic opal,
but we find low accumulation rates of sedimentary Ba. Thus, for the study
sites presented here during the last glacial, we conclude that Ba is an unr
eliable productivity proxy. Copyright (C) 1998 Elsevier Science Ltd.