46
QORDON A. RILEY
apparatus was constructed consisting of a flat polyethylene pan with
inlet and bubbling stone at one end and an outlet at the other at a level
permitting a 2-litre capacity with a surface area of about 550 om2.
Water from the outlet passed through a glass fibre filter into a small
collecting bottle and thence was pumped at a rate of 2 litre/hr back
through the bubbling chamber. Yields obtained with this device were
compared with the results of alternate bubbling and filtering of an
aliquot of the same sample. I n all cases continuous harvesting increased
the yield. The results of one such experiment, conducted with Nova
Scotia coastal water, are shown in Fig. 3. Each point represents the
time when a filter was removed and the total yield up to that time;
the latter obtained by adding all of the carbon analyses together. The
form of the curve is typical. The quantities obtained were higher than
average, and the sample presumably had an unusually rich supply of
organic matter. Unfortunately, however, no data on dissolved organic
matter are available for this series of experiments.
The work was not completed and published at that time, because
there was evidence that organic contaminants in air supplies could be
serious enough to produce spurious results, and further examination
of possible errors was necessary.
Menzel (1966) found that bubbling with his laboratory air supply
for three hours increased the dissolved organic carbon by a factor of
2-4. He therefore developed a method for combusting his air supply in
order to get rid of airborne organic material. He also developed an all
glass apparatus. With this he conducted some twenty-five experiments
using local coastal water (Woods Hole), surface and deep water from
the Sargasso Sea, and culture media in which algae had been grown.
The water was triple-filterzd through 0-8 p silver filters prior to the
experiments, and he then compared the results of 3-6 hours of bubbling
with unbubbled controls. In all cases small quantities of carbon (1836 pg C/litre) werz retained 011 the filters, both experiments and controls, but the difference between them was statistically insignificant.
Menzel’s work was done with considerable car3 with r3spect t o
cleanliness in all materials used arid statistical design. His bubbling
chamber was a glass cylinder, which is relatively inefficient for particle
production, and a longer bubbling time might have been used ; however,
these factors in themselves should not have been sufficient to prevent
significant particle production. The discrepancy between his results
and earlier ones that have been described required re-examination of
various factors which might affect the bubbling process. I n addition
to questions of technique and artifact, the role of bacteria also needed
to be examined, for Barber (1966) had reported that he was unable to
QORDON A. RILEY
apparatus was constructed consisting of a flat polyethylene pan with
inlet and bubbling stone at one end and an outlet at the other at a level
permitting a 2-litre capacity with a surface area of about 550 om2.
Water from the outlet passed through a glass fibre filter into a small
collecting bottle and thence was pumped at a rate of 2 litre/hr back
through the bubbling chamber. Yields obtained with this device were
compared with the results of alternate bubbling and filtering of an
aliquot of the same sample. I n all cases continuous harvesting increased
the yield. The results of one such experiment, conducted with Nova
Scotia coastal water, are shown in Fig. 3. Each point represents the
time when a filter was removed and the total yield up to that time;
the latter obtained by adding all of the carbon analyses together. The
form of the curve is typical. The quantities obtained were higher than
average, and the sample presumably had an unusually rich supply of
organic matter. Unfortunately, however, no data on dissolved organic
matter are available for this series of experiments.
The work was not completed and published at that time, because
there was evidence that organic contaminants in air supplies could be
serious enough to produce spurious results, and further examination
of possible errors was necessary.
Menzel (1966) found that bubbling with his laboratory air supply
for three hours increased the dissolved organic carbon by a factor of
2-4. He therefore developed a method for combusting his air supply in
order to get rid of airborne organic material. He also developed an all
glass apparatus. With this he conducted some twenty-five experiments
using local coastal water (Woods Hole), surface and deep water from
the Sargasso Sea, and culture media in which algae had been grown.
The water was triple-filterzd through 0-8 p silver filters prior to the
experiments, and he then compared the results of 3-6 hours of bubbling
with unbubbled controls. In all cases small quantities of carbon (1836 pg C/litre) werz retained 011 the filters, both experiments and controls, but the difference between them was statistically insignificant.
Menzel’s work was done with considerable car3 with r3spect t o
cleanliness in all materials used arid statistical design. His bubbling
chamber was a glass cylinder, which is relatively inefficient for particle
production, and a longer bubbling time might have been used ; however,
these factors in themselves should not have been sufficient to prevent
significant particle production. The discrepancy between his results
and earlier ones that have been described required re-examination of
various factors which might affect the bubbling process. I n addition
to questions of technique and artifact, the role of bacteria also needed
to be examined, for Barber (1966) had reported that he was unable to