INTRODUCTION
Estimates of the transfer of photosynthetically fixed carbon through the bacterioplankton community have recently become widespread (Cole et ai, 1982 ; Laake et ai, 1983 ;
Newell and Linley, 1984 ; Lancelot and Billen, 1984 ; Painting et al., 1985 ; Lochte and
Turley, 1985). However, the quantification of carbon transfer and nutrient cycling by
bacteria and through the “microbial loop” by microzooplankton bactivory (Azam et ai.
1983 ; Wambeke and Bianchi, 1985a&b) are very sensitive to accurate measures of
bacterial numbers and biomass, production, net growth yield or carbon conversion
efficiency, the activity of the cells both in time and space and to the grazing impact of
microzooplankton (Bauerfeind, 1985 and Lucas, 1986).
Recently, attempts have been made (Fallon et al., 1983 ; Riemann el ai, 1984 ; Linley and
Newell, 1984) to assess the reliability and agreement between the numerous methods
employed to measure bacterial productivity. The most notable of these techniques include
incorporation of
3 H and
14 C labelled precursors into DNA (Fuhrman and Azam. 1980,
1982) and RNA (Karl, 1982). In recognition of the numerous assumptions required to
convert incorporation of these macromolecules into microbial biomass, more direct
estimates of microbial production have also been used, including the FDC technique
(Hagstrdm et ai, 1979) and direct observations of log phase growth of cells in small
volume enclosed predation free incubations (Meyer-Reil, 1977 ; Linley et ai, 1983) or in
situ using dialysis chambers (Lochte and Turley, 1985). However, some investigators have
found limited bacterial predation by very small micro-flagellates in the <0.6µm fraction
(Fuhrman and Mc Manus, 1984) while others have reported inconsistencies between
many of the methods outlined (Laake et ai, 1983 ; Fallon et ai, 1983 ; Riemann et ai,
1984; Linley and Newell, 1984 and Pollard and Moriarty, 1984).
Estimates of net growth yield, or carbon conversion efficiency required to support
estimates of heterotrophic bacterial production also vary widely ; ranging from 10 -80 %
(Williams, 1981 ; Joint and Pomroy, 1982 ; Linley and Newell, 1984 ; Bauerfeind, 1985)
depending largely on the nature of the substrate being utilised. Clearly, such a range will
significantly affect our estimates of carbon flow through the heterotrophic bacterial
community. Erroneous estimates of net growth yield will be seriously compounded if our
estimates of bacterial production lack precision.
We have attempted to impose more rigorous constraints on our estimates of bacterial
production as measured by 3 H-Thymidine incorporation and compared these with direct
estimates of bacterial growth under reduced predation pressue in <3µm incubations. We
have also demonstrated marked changes in bacterial activity associated with a phytoplankton bloom in an enclosed mesocosm.
SAMPLING AND ANALYTICAL METHODS
Sampling
Hydrographic data and biological samples were obtained three times per day using a
conductivity, temperature and depth (CTD) rosette sampler for profiles down to 300 m
taken from R S Africana during the course of a cruise (14.3.83- 20.3.83) in the S. Benguela
upwelling system off the west coast of the Cape Peninsula, South Africa. The cruise track
was determined by following a drogue deployed into newly upwelled water. Daily
airborne radiation thermometry (ART) flights and CTD profiles confirmed that the
drogue remained in the same cell of water. All sampling profiles were taken adjacent to
the drogue thus giving a good temporal sequence of data. Further bacterial samples were
obtained from a 60L mesocosm containing newly upwelled and nutrient rich water
376
Estimates of the transfer of photosynthetically fixed carbon through the bacterioplankton community have recently become widespread (Cole et ai, 1982 ; Laake et ai, 1983 ;
Newell and Linley, 1984 ; Lancelot and Billen, 1984 ; Painting et al., 1985 ; Lochte and
Turley, 1985). However, the quantification of carbon transfer and nutrient cycling by
bacteria and through the “microbial loop” by microzooplankton bactivory (Azam et ai.
1983 ; Wambeke and Bianchi, 1985a&b) are very sensitive to accurate measures of
bacterial numbers and biomass, production, net growth yield or carbon conversion
efficiency, the activity of the cells both in time and space and to the grazing impact of
microzooplankton (Bauerfeind, 1985 and Lucas, 1986).
Recently, attempts have been made (Fallon et al., 1983 ; Riemann el ai, 1984 ; Linley and
Newell, 1984) to assess the reliability and agreement between the numerous methods
employed to measure bacterial productivity. The most notable of these techniques include
incorporation of
3 H and
14 C labelled precursors into DNA (Fuhrman and Azam. 1980,
1982) and RNA (Karl, 1982). In recognition of the numerous assumptions required to
convert incorporation of these macromolecules into microbial biomass, more direct
estimates of microbial production have also been used, including the FDC technique
(Hagstrdm et ai, 1979) and direct observations of log phase growth of cells in small
volume enclosed predation free incubations (Meyer-Reil, 1977 ; Linley et ai, 1983) or in
situ using dialysis chambers (Lochte and Turley, 1985). However, some investigators have
found limited bacterial predation by very small micro-flagellates in the <0.6µm fraction
(Fuhrman and Mc Manus, 1984) while others have reported inconsistencies between
many of the methods outlined (Laake et ai, 1983 ; Fallon et ai, 1983 ; Riemann et ai,
1984; Linley and Newell, 1984 and Pollard and Moriarty, 1984).
Estimates of net growth yield, or carbon conversion efficiency required to support
estimates of heterotrophic bacterial production also vary widely ; ranging from 10 -80 %
(Williams, 1981 ; Joint and Pomroy, 1982 ; Linley and Newell, 1984 ; Bauerfeind, 1985)
depending largely on the nature of the substrate being utilised. Clearly, such a range will
significantly affect our estimates of carbon flow through the heterotrophic bacterial
community. Erroneous estimates of net growth yield will be seriously compounded if our
estimates of bacterial production lack precision.
We have attempted to impose more rigorous constraints on our estimates of bacterial
production as measured by 3 H-Thymidine incorporation and compared these with direct
estimates of bacterial growth under reduced predation pressue in <3µm incubations. We
have also demonstrated marked changes in bacterial activity associated with a phytoplankton bloom in an enclosed mesocosm.
SAMPLING AND ANALYTICAL METHODS
Sampling
Hydrographic data and biological samples were obtained three times per day using a
conductivity, temperature and depth (CTD) rosette sampler for profiles down to 300 m
taken from R S Africana during the course of a cruise (14.3.83- 20.3.83) in the S. Benguela
upwelling system off the west coast of the Cape Peninsula, South Africa. The cruise track
was determined by following a drogue deployed into newly upwelled water. Daily
airborne radiation thermometry (ART) flights and CTD profiles confirmed that the
drogue remained in the same cell of water. All sampling profiles were taken adjacent to
the drogue thus giving a good temporal sequence of data. Further bacterial samples were
obtained from a 60L mesocosm containing newly upwelled and nutrient rich water
376