Seawater cultures were conducted in large volume (36 l) gas impermeable tri
-laminate bags for the purpose of empirically deriving bacterial growth eff
iciency (BGE) and carbon conversion factors (CCF) in the south central Ross
Sea. This experimental design allowed for concomitant measurements of meta
bolic reactants (loss of total and dissolved organic carbon [TOC and DOG])
and products (gain of total carbon dioxide [TCO2] and bacterial biomass) to
be made from a single incubation vessel. Some previous studies have relied
on proxy measurements (e.g. O-2, H-3-thymidine incorporation and cell abun
dance) to determine BGE and CCF rather than direct carbon measurements. Our
experimental design enabled a complete carbon budget to be constructed and
eliminated variability associated with normally employed parallel bottle i
ncubations. Utilization of TOC was well balanced by the production of TCO2,
in 7 of 8 experiments, validating the use of tri-laminate bags for measuri
ng microbial respiration. In 3 experiments, where TOC, DOC, TCO2 and bacter
ial biovolume were directly measured, carbon mass balance yielded BGE estim
ates of 12, 32 and 38 % and bacterial CCF of 77, 95 and 134 fg C mu m(-3) I
n experiments where independent DOC measurements were not made we used our
empirically derived CCF values to determine bacterial carbon production and
calculated DOC concentrations and BGE for these remaining experiments. The
BGE derived from all the bag experiments conducted throughout the austral
spring and summer 1995-1997 ranged from 9 to 38 %. Our experimental design
and carbon mass balance approach could be applied to other aquatic systems
to empirically derive the BGE and CCF, factors essential for determining ca
rbon flux through bacterioplankton.