predators, but also grazers, so apparent growth increases again, and that effect is even
stronger with the 1 μm filter. Thus, there must be a “trophic cascade” of at least three
trophic levels among organisms less than 20 μm, even those less than 8 μm. Calbet et
al. (2001) have pursued such evaluations further. A trophic cascade involves
progressive interactions among trophic levels. For example, one way to remove
planktonic algae from a lake is to introduce piscivorous fish to eat planktivorous fish,
which then do not eat zooplankton, which increase and graze down the algae.
Sometimes that works.
(^) A summary (Fig. 5.11; Strom 2000) of studies determining the relative rates of
bacterivory and bacterial production shows that at low production rates, that is, in
oligotrophic or oceanic habitats, there is near balance between bacterial growth and
protist grazing. In more eutrophic, essentially nearshore, habitats, grazing lags
somewhat at almost all sites; bacteria are not eaten as fast as they grow. The methods
are imperfect, and sometimes imbalance up to a factor of two or so can be simply an
artifact. Note that viruses are present in dilution experiments, further complicating
interpretations. However, the result is consistent enough that another bacterial control
mechanism must usually operate in inshore waters.
Fig. 5.11 Comparison of bacterial production and grazing rates on bacteria based on a
literature survey. Grazing balances production in oligotrophic, mostly oceanic waters
(inset), while production, when it is high, usually exceeds grazing somewhat.
(^) (After Strom 2000.)