PARTlClLlLATE ORGANIC MATTER 1N SEA WATER
25
ately underneath. In deeper water there have been occasional indications of microstratification, as in the work of Holm-Hansen et al. ( I 966).
Most studies of particulate organic carbon have been based on collections in which the bottles were spaced much too far apart for microstratification to be detectable as such. However, concentrations
frequently vary with depth in an irregular manner. Gordon (1970b)
found that most of the observations at his stations in the North Atlantic
were significantly non-random. I n other words, present data indicate
that vertical variations exist at all levels from surface to bottom, even
though the form of the variation remains obscure. In Gordon’s comparison of PVC sampling bottles with the Menzel glass-teflon sampler,
the bottles were spaced 15 m apart. Here, too, there were significant
differences in some of the pairs, although the overall averages were
not significantly different, and this suggests that some of the vertical
variations may be quite abrupt.
Thus the available data on organic carbon provide only hints about
small-scale variations in particulate organic matter. However, at the
present time there is increasing interest in non-random distributions of
filter feeding organisms and their particulate food, so that a discussion
of the problem and an account of indirect evidence on the subject are
in order.
Baylor et al. (1962) and Sutcliffe et al. (1963) brought forward
evidence that organic particles could be produced by adsorption of
organic materials on near-surface bubbles. There were indications that
phytoplankton and bacteria might also be trapped on bubbles. The
net effect would be a concentration of organic materials in the surface,
and this could be at least part of the reason for the near-surface gradients in organic carbon which were discussed above.
This local concentration at the surface is subject to gradual removal
by vertical turbulence and convection cells. These are long cells oriented
more or less downwind, in which the surface layer moves laterally and
downwind into narrow, parallel lines of convergence. Sutcliffe et al.
(1963) measured the sinking speed of the water in these convergences
and obtained values of the order of 3-6 cm/sec. Between the narrow
curtains of descending water, principles of mass continuity require a
corresponding upward movement. This is presumably slow and diffuse,
for the area between the convergences is much broader. The trajectory
of any particular parcel of water in such a cell is probably a spiral
motion downwind, although little is actually known about these cells
except the surface water movements (Woodcock, 1944) and the downdrafts.
In an area such as the Sargasso Sea, where convergences are marked
25
ately underneath. In deeper water there have been occasional indications of microstratification, as in the work of Holm-Hansen et al. ( I 966).
Most studies of particulate organic carbon have been based on collections in which the bottles were spaced much too far apart for microstratification to be detectable as such. However, concentrations
frequently vary with depth in an irregular manner. Gordon (1970b)
found that most of the observations at his stations in the North Atlantic
were significantly non-random. I n other words, present data indicate
that vertical variations exist at all levels from surface to bottom, even
though the form of the variation remains obscure. In Gordon’s comparison of PVC sampling bottles with the Menzel glass-teflon sampler,
the bottles were spaced 15 m apart. Here, too, there were significant
differences in some of the pairs, although the overall averages were
not significantly different, and this suggests that some of the vertical
variations may be quite abrupt.
Thus the available data on organic carbon provide only hints about
small-scale variations in particulate organic matter. However, at the
present time there is increasing interest in non-random distributions of
filter feeding organisms and their particulate food, so that a discussion
of the problem and an account of indirect evidence on the subject are
in order.
Baylor et al. (1962) and Sutcliffe et al. (1963) brought forward
evidence that organic particles could be produced by adsorption of
organic materials on near-surface bubbles. There were indications that
phytoplankton and bacteria might also be trapped on bubbles. The
net effect would be a concentration of organic materials in the surface,
and this could be at least part of the reason for the near-surface gradients in organic carbon which were discussed above.
This local concentration at the surface is subject to gradual removal
by vertical turbulence and convection cells. These are long cells oriented
more or less downwind, in which the surface layer moves laterally and
downwind into narrow, parallel lines of convergence. Sutcliffe et al.
(1963) measured the sinking speed of the water in these convergences
and obtained values of the order of 3-6 cm/sec. Between the narrow
curtains of descending water, principles of mass continuity require a
corresponding upward movement. This is presumably slow and diffuse,
for the area between the convergences is much broader. The trajectory
of any particular parcel of water in such a cell is probably a spiral
motion downwind, although little is actually known about these cells
except the surface water movements (Woodcock, 1944) and the downdrafts.
In an area such as the Sargasso Sea, where convergences are marked