METHODS
Bacterial counts
Nitrogen-fixing Enterobacteriaceae (NFEs) were enumerated and isolated using Most
Probable Number (MPN) methodology. This technique, although less accurate than
other viable counts, is particularly suitable for nitrogen fixation studies as it enables the
use of acetylene reduction activity (ARA) as the «end point» for counting and does not
require the picking off of individual colonies for subsequent tests for nitrogen fixation. It
also serves as an excellent method for selecting and isolating nitrogen-fixing bacteria
from both aquatic and terrestrial samples.
Water samples were collected in sterile bottles and analysed the same day. 1 cm 3 samples
from decade dilutions (up to 10 -8 ) in Ringers solution were inoculated into 6 screwcapped
Bijou bottles (7 cm 3 ) containing nitrogen-free glucose medium to a final salinity of just
above 0‰ in 3 bottles and 35°/°° into the other 3. The atmosphere in the bottles was
replaced with nitrogen and they were incubated anaerobically at 25°C. After 5 days 1 cm 3
acetylene was injected into each of the bottles and, after a further 24 h incubation, 1 cm 3
samples of the bottles’ atmosphere were removed and analysed for ethylene production.
Positive results for ARA were taken as positive for nitrogen fixation (Stewart et ai, 1967)
and inocula from these bottles were grown aerobically on McConkey agar plates prior to
re-testing for ARA in nitrogen-free glucose broth under anaerobic conditions. MPN
tables were consulted (Collins, 1967) and the results expressed as the number of freshwater (for those positive at 0°/°°) or salinity-tolerant (for those positive at 35‰) NFEs
100 cm -3 water).
Sediment samples were analysed in a similar way except that they ( 10 g wet weight) were
mixed with Ringers solution and the volume made up to 100 cm 3 . The resulting suspension was stirred vigorously for 20 minutes prior to being serially diluted and inoculated
into nitrogen-free glucose broth. The results are expressed as the number of NFEs per
gram dry weight of sediments.
Deposition experiments were carried out by placing sterile filter papers (3 sheets thick)
onto the surface of the sediment and fastening them to the sediments with pins. After tidal
cover the top filter paper was placed in a sterile petri dish for transport to the laboratory.
5 cm 2 portions were shaken in bottles containing 10 cm 3 Ringers solution and then serially
diluted for MPN estimations.
Identification
NFEs were isolated from positive ARA assay bottles and cultured aerobically on
McConkey bile salts agar. Pure cultures were identified using the API 20E analytical
profile index (API Laboratory Products Ltd., Basingstoke, UK.).
RESULTS AND DISCUSSION
NFEs in Rivers
The main rivers flowing into Morecambe Bay and the Gipping, which flows via the Orwell
into the North Sea, were analysed for the occurrence of NFEs (table 1 ). In all of the rivers
substantial numbers of both freshwater and salinity-tolerant strains were detected with
the freshwater strains, on average, twice as plentiful. Similar data were obtained for the
Lune and Orwell estuaries (Tab. 1) with the freshwater strains either exceeding or equal in
number to the salinity-tolerant strains.
526
Bacterial counts
Nitrogen-fixing Enterobacteriaceae (NFEs) were enumerated and isolated using Most
Probable Number (MPN) methodology. This technique, although less accurate than
other viable counts, is particularly suitable for nitrogen fixation studies as it enables the
use of acetylene reduction activity (ARA) as the «end point» for counting and does not
require the picking off of individual colonies for subsequent tests for nitrogen fixation. It
also serves as an excellent method for selecting and isolating nitrogen-fixing bacteria
from both aquatic and terrestrial samples.
Water samples were collected in sterile bottles and analysed the same day. 1 cm 3 samples
from decade dilutions (up to 10 -8 ) in Ringers solution were inoculated into 6 screwcapped
Bijou bottles (7 cm 3 ) containing nitrogen-free glucose medium to a final salinity of just
above 0‰ in 3 bottles and 35°/°° into the other 3. The atmosphere in the bottles was
replaced with nitrogen and they were incubated anaerobically at 25°C. After 5 days 1 cm 3
acetylene was injected into each of the bottles and, after a further 24 h incubation, 1 cm 3
samples of the bottles’ atmosphere were removed and analysed for ethylene production.
Positive results for ARA were taken as positive for nitrogen fixation (Stewart et ai, 1967)
and inocula from these bottles were grown aerobically on McConkey agar plates prior to
re-testing for ARA in nitrogen-free glucose broth under anaerobic conditions. MPN
tables were consulted (Collins, 1967) and the results expressed as the number of freshwater (for those positive at 0°/°°) or salinity-tolerant (for those positive at 35‰) NFEs
100 cm -3 water).
Sediment samples were analysed in a similar way except that they ( 10 g wet weight) were
mixed with Ringers solution and the volume made up to 100 cm 3 . The resulting suspension was stirred vigorously for 20 minutes prior to being serially diluted and inoculated
into nitrogen-free glucose broth. The results are expressed as the number of NFEs per
gram dry weight of sediments.
Deposition experiments were carried out by placing sterile filter papers (3 sheets thick)
onto the surface of the sediment and fastening them to the sediments with pins. After tidal
cover the top filter paper was placed in a sterile petri dish for transport to the laboratory.
5 cm 2 portions were shaken in bottles containing 10 cm 3 Ringers solution and then serially
diluted for MPN estimations.
Identification
NFEs were isolated from positive ARA assay bottles and cultured aerobically on
McConkey bile salts agar. Pure cultures were identified using the API 20E analytical
profile index (API Laboratory Products Ltd., Basingstoke, UK.).
RESULTS AND DISCUSSION
NFEs in Rivers
The main rivers flowing into Morecambe Bay and the Gipping, which flows via the Orwell
into the North Sea, were analysed for the occurrence of NFEs (table 1 ). In all of the rivers
substantial numbers of both freshwater and salinity-tolerant strains were detected with
the freshwater strains, on average, twice as plentiful. Similar data were obtained for the
Lune and Orwell estuaries (Tab. 1) with the freshwater strains either exceeding or equal in
number to the salinity-tolerant strains.
526