50
OOKUON A. RILEY
passing materials. The experiments were not published at that time,
pending further study of the problem.
Sheldon et al. (1967) were the first to publish an account of the early
events in this aggregation process. They followed the course of particle
formation in filtered sea water with repeated Coulter Counter measurements of the total volume of particles in a graded size series between
1-58 and 5 p. Initial particle formation was rapid, leveling off after a
few days. The smallest sizes were most abundant at first, and there
was a gradual shift toward a maximum concentration at a particle
size of about 4 p. The process could occur in the light or in the dark ;
it was stimulated by addition of glycine and inhibited, at least temporarily, by mercuric chloride. They suggested that bacterial action
probably was implicated in the formation, but this could not be
established beyond question. When the sample was filtered after
equilibrium concentration had been raached, another cycle of production occurred, and as many as four cycles were induced in one sample
by repeated filtration. They estimated that the concentration of
particulate organic carbon might be of the order of 25 pg/litre at the
time of equilibrium.
The writer has followed the course of particle formation in filtered
sea water by means of carbon determinations and to a lesser extent by
visual examination. The initial increase described by Sheldon et al.
(1967) is near the limits of sensitivity of the carbon method, and during
the first 2-3 days the results generally are not significantly higher than
those obtained by immediate refiltration. However, after about five
days a difference can generally be notc3d, and the quantity is more or
less in agreement with the estimate made by Sheldon et al. (1967).
The quantity of particulate organic carbon continues to increase
in these experiments, commonly reaching a peak in 10-20 days, and
at that time the carbon concentration is likely to be at least an order
of magnitude higher than the equilibrium value estimated by Sheldon
et al. (1967). Visual examination shows that the most important
constituents are hazy, amorphous masses of material in the general
size range of about 25 p, which stain heavily with Schiff reagent. The
protein staining reaction is generally less marked and may be absent.
Bacteria are associated with this material ; their protein staining
reaction makes them clearly visible.
Presumably these are two chapters of the same story. The so-called
equilibrium stage described by Sheldon et al. ( 1 967) probably is merely
an equilibrium of small particles within the size range that he was
particularly concerned with, but these particles gradually coalesce
into larger masses, are worked over by bacteria, and are gradually
Précédent

- 63/505

Suivant