410
Craig M. YOUNG
variability in salinity, diet and energy availability
(Giese and Pearse, 1974). Most of the periodic cues
known to control reproduction in shallow water (e.g.,
daylength, sunrise, temperature, lunar period, salinity,
wave action) are not available in the deep sea. Nevertheless, there are seasonal changes even at abyssal
depths that could possibly entrain reproductive cycles
(reviewed by Tyler, 1988). Some of these factors, all of
which remain untested as controllers of gametogenesis
or spawning, include eddy kinetic energy, spring or
summer falls of phytodetritus, and turbulence during
benthic storms.
Predictions
Phytodetritus on the sea floor might qualify as a
reliable cue for the entrainment of gametogenesis,
particularly in temperate habitats where food may not
only entrain reproduction, but also provide the energy
needed for gamete production. A number of workers
have suggested that phytoplankton detritus is probably
the main controller of gametogenesis (Tyler, 1988;
Tyler et al., 1994; Campos-Creasey et al., 1994).
The phytodetrital entrainment hypothesis gives rise
to three predictions about the distribution of seasonal
or periodic reproduction among deep-sea invertebrates:
1) If phytoplankton entrains gametogenesis, then one
would expect more seasonality at the sea floor underlying highly productive temperate seas than in
oligotrophic tropical regions, where the seasonal signal
should be less intense; 2) one would expect more
seasonality at bathyal depths than at abyssal depths;
and 3) seasonal reproduction should be more common
among animals living in non-chemosynthetic systems
than in vent or seep animals, since the latter should
have food available to them throughout the year.
Evidence
Latitudinal and depth comparisons of reproductive
periodicity are easiest with the Echinodermata, since
this phylum has been most extensively studied. At
least one common deep-sea group, the echinoid family
Echinothuriidae, reproduce continuously at all depths
and latitudes where they occur. This suggests that
reproductive periodicity, as well as mode of larval
development, may be phylogenetically constrained for
this group. The data show that seasonal breeders are
present among echinoderms at both temperate and
tropical latitudes and to a depth of about 3000 m.
However, the few data available for greater depths
suggest that seasonal reproduction may indeed become
less common. When tropical and temperate latitudes
are compared, it appears that seasonal breeders are
more common in tropical seas than in the temperate
ones, a pattern not predicted by the phytodetrital
entrainment hypothesis. However, the comparison is
not entirely valid, since the only data set for tropical
seas is for Bahamian Slope echinoids at depths
less than 1000 m, whereas most data for temperate
species are from greater depths. Greater geographic and
bathymetric coverage is needed before these correlative
tests are fully satisfactory.
In the past few years, my laboratory has undertaken
experimental studies (as yet unpublished) to determine
if detrital food entrains reproductive cycles in bathyal
echinoids. Although the data remain preliminary,
laboratory experiments and also experiments in field
enclosures suggest that the timing of reproduction can
be shifted at least slightly by providing excess food at
certain times of year. Food quantity and quality control
fecundity and egg quality in bathyal echinoids (S.
George and Young, unpublished data), but can also shift
the speed of the gametogenic process. Thus, evidence
is beginning to accrue to support the hypothesis that the
timing of food falls can control the onset or completion
of gametogenesis.
Tyler et al. (1994) showed that Plutonaster bifrons, a
seasonally breeding sea star from the Rockall Trough,
shifts its diet to phytodetritus when this food source
is available, whereas a continuously breeding starfish,
Bathybiaster vexillifer, is predatory and therefore
one trophic link away from the direct influence of
the seasonal phytodetrital pulse. Echinus affinis, an
echinoid from lower bathyal depths, shows a seasonal
change in gut contents which reflects the composition
of the sea-floor phytodetritus (Campos-Creasey et al.,
1994). This species breeds annually in mid-winter,
releasing eggs several months before phytodetritus
becomes available to plankton at bathypelagic depths
(Tyler and Gage, 1984a). If larval development is very
slow, it is conceivable that planktotrophic larvae may be
using this detritus as a food source. However, one does
not know how long larvae take to reach the feeding
stage in this species.
Fertilization success and the control of spawning
Background
Successful reproduction requires that eggs and sperm
be mature and present at exactly the same time.
Sperm generally have a short swimming time, and
must occur at relatively high concentrations in order
Craig M. YOUNG
variability in salinity, diet and energy availability
(Giese and Pearse, 1974). Most of the periodic cues
known to control reproduction in shallow water (e.g.,
daylength, sunrise, temperature, lunar period, salinity,
wave action) are not available in the deep sea. Nevertheless, there are seasonal changes even at abyssal
depths that could possibly entrain reproductive cycles
(reviewed by Tyler, 1988). Some of these factors, all of
which remain untested as controllers of gametogenesis
or spawning, include eddy kinetic energy, spring or
summer falls of phytodetritus, and turbulence during
benthic storms.
Predictions
Phytodetritus on the sea floor might qualify as a
reliable cue for the entrainment of gametogenesis,
particularly in temperate habitats where food may not
only entrain reproduction, but also provide the energy
needed for gamete production. A number of workers
have suggested that phytoplankton detritus is probably
the main controller of gametogenesis (Tyler, 1988;
Tyler et al., 1994; Campos-Creasey et al., 1994).
The phytodetrital entrainment hypothesis gives rise
to three predictions about the distribution of seasonal
or periodic reproduction among deep-sea invertebrates:
1) If phytoplankton entrains gametogenesis, then one
would expect more seasonality at the sea floor underlying highly productive temperate seas than in
oligotrophic tropical regions, where the seasonal signal
should be less intense; 2) one would expect more
seasonality at bathyal depths than at abyssal depths;
and 3) seasonal reproduction should be more common
among animals living in non-chemosynthetic systems
than in vent or seep animals, since the latter should
have food available to them throughout the year.
Evidence
Latitudinal and depth comparisons of reproductive
periodicity are easiest with the Echinodermata, since
this phylum has been most extensively studied. At
least one common deep-sea group, the echinoid family
Echinothuriidae, reproduce continuously at all depths
and latitudes where they occur. This suggests that
reproductive periodicity, as well as mode of larval
development, may be phylogenetically constrained for
this group. The data show that seasonal breeders are
present among echinoderms at both temperate and
tropical latitudes and to a depth of about 3000 m.
However, the few data available for greater depths
suggest that seasonal reproduction may indeed become
less common. When tropical and temperate latitudes
are compared, it appears that seasonal breeders are
more common in tropical seas than in the temperate
ones, a pattern not predicted by the phytodetrital
entrainment hypothesis. However, the comparison is
not entirely valid, since the only data set for tropical
seas is for Bahamian Slope echinoids at depths
less than 1000 m, whereas most data for temperate
species are from greater depths. Greater geographic and
bathymetric coverage is needed before these correlative
tests are fully satisfactory.
In the past few years, my laboratory has undertaken
experimental studies (as yet unpublished) to determine
if detrital food entrains reproductive cycles in bathyal
echinoids. Although the data remain preliminary,
laboratory experiments and also experiments in field
enclosures suggest that the timing of reproduction can
be shifted at least slightly by providing excess food at
certain times of year. Food quantity and quality control
fecundity and egg quality in bathyal echinoids (S.
George and Young, unpublished data), but can also shift
the speed of the gametogenic process. Thus, evidence
is beginning to accrue to support the hypothesis that the
timing of food falls can control the onset or completion
of gametogenesis.
Tyler et al. (1994) showed that Plutonaster bifrons, a
seasonally breeding sea star from the Rockall Trough,
shifts its diet to phytodetritus when this food source
is available, whereas a continuously breeding starfish,
Bathybiaster vexillifer, is predatory and therefore
one trophic link away from the direct influence of
the seasonal phytodetrital pulse. Echinus affinis, an
echinoid from lower bathyal depths, shows a seasonal
change in gut contents which reflects the composition
of the sea-floor phytodetritus (Campos-Creasey et al.,
1994). This species breeds annually in mid-winter,
releasing eggs several months before phytodetritus
becomes available to plankton at bathypelagic depths
(Tyler and Gage, 1984a). If larval development is very
slow, it is conceivable that planktotrophic larvae may be
using this detritus as a food source. However, one does
not know how long larvae take to reach the feeding
stage in this species.
Fertilization success and the control of spawning
Background
Successful reproduction requires that eggs and sperm
be mature and present at exactly the same time.
Sperm generally have a short swimming time, and
must occur at relatively high concentrations in order