small cells would pass through the 3 µm pore diameter membrane fitler, the smallest size
used in the reverse flow concentrators. Thus, despite the concentration of larger particles,
80 % of the cells were found in the <3 µm size fraction.
Although the amount of suspended particulate matter in the Sargasso Sea is low,
particulates are a likely, if not essential, habitat for active, fermentative hydrogenproducing bacteria. Since we are most likely dealing with fermentative bacteria, anaerobic microenvironments must be present within the aerobic water column for in situ
hydrogen production to occur and particulate matter is a candidate for such microenvironments. At a station in the eastern equatorial Atlantic, bacteria accounted for up to
40% of the organic carbon in particulates (Bishop et al., 1977). The bacteria that were
concentrated in the >30 µm size fraction at Station 18 clearly were associated with
particulate matter, since the small free bacteria should easily pass through the 35 µm
mesh. Station 4 samples, containing unconcentrated water, had relatively high amounts
of particulate matter within which bacterial hydrogen production could occur.
It is difficult to evaluate the role of particulate matter with respect to microbial activities
in the water column because most sampling techniques are not capable of collecting intact
particulates. Trent et al. (1978) found that marine waters contain fragile aggregate
particles called marine snow, up to 9 cm long, that are destroyed by conventional
sampling methods. The aggregates are diverse microhabitats containing bacteria, ciliates,
dinoffagellates, and diatoms. Nutrients are more concentrated in aggregates than in the
surrounding water and it has been suggested that microbial activity in the nutrient rich
aggregates could alter the microenvironment (Shanks and Trent, 1979). In terms of
hydrogen production, this means that intensive microbial activity could deplete oxygen to
a level at which facultative anaerobes could begin fermentative metabolism, releasing
hydrogen. Production of hydrogen in the <3 µm size fraction at Station 11 indicates that
facultative anaerobes are free in the water column, as well as attached to particles, and
could rapidly take advantage of any developing microscale anaerobic habitats.
Our incubation technique, using serum vials, does not permit a quantitative estimate of
the number of hydrogen producers in the water; it is designed to detect the potential for
hydrogen production and allow isolation of the gas-producing organisms. Thus, our
hydrogen production figures must be considered maximum potential production value.
The maximum amount of hydrogen produced from each station’s samples was relatively
constant at about 10 nl l -1 . The added nutrients did not affect either the rate of
production or total amount of hydrogen (Fig. 2). The other organic supplements, yeast
extract and cysteine, which all of the vials received, provided adequate nutrients for
hydrogen production without the addition of glucose or lactate. The relatively constant
amount of hydrogen produced indicates that the hydrogen itself may inhibit further
production. In several species of bacteria, increasing partial pressures of hydrogen inhibit
hydrogen evolution (Zajic et ai, 1978), so this is a reasonable mechanism to explain the
constant upper limit to our measured hydrogen concentrations.
Our serum vial sample incubation technique has been further modified to permit the use
of unconcentrated water and to allow estimation of the initial rates of hydrogen gas
production. When water samples from various depths were incubated individualy, differences in initial rates of gas production were observed as shown in Figure 4. Production
rates were greatest and exhibited a marked maximum at the base of the mixed layer
(45-55m). This is the same depth range where the greatest number of hydrogen-producing
bacteria were enumerated through the use of fluorescent antibody techniques (Figure 4,
FA). Thus, it appears that hydrogen gas-producing bacteria are associated with the
particulates found at the base of the mixed layer, have the highest gas production
potential there, and are present in the greatest numbers there.
206
used in the reverse flow concentrators. Thus, despite the concentration of larger particles,
80 % of the cells were found in the <3 µm size fraction.
Although the amount of suspended particulate matter in the Sargasso Sea is low,
particulates are a likely, if not essential, habitat for active, fermentative hydrogenproducing bacteria. Since we are most likely dealing with fermentative bacteria, anaerobic microenvironments must be present within the aerobic water column for in situ
hydrogen production to occur and particulate matter is a candidate for such microenvironments. At a station in the eastern equatorial Atlantic, bacteria accounted for up to
40% of the organic carbon in particulates (Bishop et al., 1977). The bacteria that were
concentrated in the >30 µm size fraction at Station 18 clearly were associated with
particulate matter, since the small free bacteria should easily pass through the 35 µm
mesh. Station 4 samples, containing unconcentrated water, had relatively high amounts
of particulate matter within which bacterial hydrogen production could occur.
It is difficult to evaluate the role of particulate matter with respect to microbial activities
in the water column because most sampling techniques are not capable of collecting intact
particulates. Trent et al. (1978) found that marine waters contain fragile aggregate
particles called marine snow, up to 9 cm long, that are destroyed by conventional
sampling methods. The aggregates are diverse microhabitats containing bacteria, ciliates,
dinoffagellates, and diatoms. Nutrients are more concentrated in aggregates than in the
surrounding water and it has been suggested that microbial activity in the nutrient rich
aggregates could alter the microenvironment (Shanks and Trent, 1979). In terms of
hydrogen production, this means that intensive microbial activity could deplete oxygen to
a level at which facultative anaerobes could begin fermentative metabolism, releasing
hydrogen. Production of hydrogen in the <3 µm size fraction at Station 11 indicates that
facultative anaerobes are free in the water column, as well as attached to particles, and
could rapidly take advantage of any developing microscale anaerobic habitats.
Our incubation technique, using serum vials, does not permit a quantitative estimate of
the number of hydrogen producers in the water; it is designed to detect the potential for
hydrogen production and allow isolation of the gas-producing organisms. Thus, our
hydrogen production figures must be considered maximum potential production value.
The maximum amount of hydrogen produced from each station’s samples was relatively
constant at about 10 nl l -1 . The added nutrients did not affect either the rate of
production or total amount of hydrogen (Fig. 2). The other organic supplements, yeast
extract and cysteine, which all of the vials received, provided adequate nutrients for
hydrogen production without the addition of glucose or lactate. The relatively constant
amount of hydrogen produced indicates that the hydrogen itself may inhibit further
production. In several species of bacteria, increasing partial pressures of hydrogen inhibit
hydrogen evolution (Zajic et ai, 1978), so this is a reasonable mechanism to explain the
constant upper limit to our measured hydrogen concentrations.
Our serum vial sample incubation technique has been further modified to permit the use
of unconcentrated water and to allow estimation of the initial rates of hydrogen gas
production. When water samples from various depths were incubated individualy, differences in initial rates of gas production were observed as shown in Figure 4. Production
rates were greatest and exhibited a marked maximum at the base of the mixed layer
(45-55m). This is the same depth range where the greatest number of hydrogen-producing
bacteria were enumerated through the use of fluorescent antibody techniques (Figure 4,
FA). Thus, it appears that hydrogen gas-producing bacteria are associated with the
particulates found at the base of the mixed layer, have the highest gas production
potential there, and are present in the greatest numbers there.
206