408
Craig M. YOUNG
evidence to suggest that the opportunistic r-strategy is
well represented in the deep sea. As in shallow water,
the deep sea contains enough spatial and temporal
variability to allow exploitation by species with various
strategies of energy allocation.
Age distribution of reproductive effort
Background
In his classic paper linking life-history traits with
population growth, Cole (1954) focused on the demographic advantages and disadvantages of one-time
reproduction (semelparity) and repeated reproduction
(iteroparity). His major finding, often known as “Cole’s
result” was that semelparous animals that reproduce
early in life produce as many offspring over many
generations (i.e., are “as fit”) as iteroparous animals
that produce more offspring over the course of an
individual’s life, but begin reproduction later in life.
This result has been challenged as an oversimplification
(Charnov and Schaffer, 1973) but the fact remains
that both kinds of organisms are successful in the real
world. Todd (1986) noted that virtually all intertidal
invertebrates are iteroparous and used this observation
in support of the bet-hedging models that predict
iteroparity for habitats where organisms have variable
survival in the early life-history stages.
Predictions
“Bet-hedging” models predict that semelparity should
increase as a function of environmental stability; thus,
“normal” deep-sea habitats should select for semelparity, particularly among species that do not produce
pelagic larvae. The r/K-selection theory, by contrast,
predicts a predominance of long-lived, iteroparous
species in stable deep-sea habitats.
Evidence
The most extensive work on reproductive patterns
in the deep sea is that of Tyler, Gage and their
colleagues in the Rockall Trough. Of the many
species of echinoderms, molluscs, cnidarians and
other invertebrates they have studied, virtually all
reproduce over a wide range of body sizes, a feature
indicative of iteroparity. Moreover, the majority of
deep-sea species for which seasonal samples have been
obtained appear to reproduce more or less continuously,
probably over a number of annual cycles. The large
number of observations on echinoderm reproduction
on the Bahamian Slope also indicate iteroparity in all
species. However, Turner’s observations on xylophagid
bivalves living in wood are in stark contrast, as these
species attain reproductive maturity quickly and must
reproduce before they consume their own habitat. Not
all xylophagids seem to follow this pattern. Knudsen
(1961) found juveniles attached in some way to the
adults of nine species. Although this was interpreted as
evidence of brooding, the young might also have settled
on the adults after a period of dispersal. Thus, although
at first sight it seems that some xylophagids might not
be opportunistic and iteroparous, definitive information
is not available. Other kinds of opportunistic species
which probably have semelparous reproduction include
some cocculinid limpets which colonize palm fronds
and wood (Young and Tyler, unpublished data), and
sipunculans which recruit opportunistically into fibrous
substrata (M.A. Rice, P.A. Tyler and C.M. Young,
unpublished data). The latter appear to live longer than
a year and can reproduce several times in laboratory
culture. However, sipunculans differ from xylophagids
ecologically in that they do not destroy their own
habitat. Before concluding that iteroparity is more
common than semelparity in the deep sea, it should
be noted that semelparity is much more difficult
to document than iteroparity in infrequent deep-sea
sampling programs; indeed, it is probably common for
semelparous species to have life spans shorter than
typical sampling intervals.
Not all small-bodied animals are semelparous in
the deep sea. Wolff (1956a) found that the females
of at least two species of hadal tanaids “. . . pass
through several egg-bearing periods lasting probably
at least three months” and that “these periods may
occur only once every second or third year, provided
that the female reaches an age of, say 15–20 years.”
Wolff (1962) also found evidence of iteroparity in
asellote isopods, and Bishop reported multiple cohorts
of gametes in the gonads of abyssal cumaceans, a clear
indication of iteroparity.
If Jumars and Gallagher (1983) are correct about
the importance of juvenile predation, then the “bethedging” models correctly predict iteroparity; if not,
then the predominance of iteroparity in the deep
sea is best explained as a K-selected life-history
trait. Empirical evaluation of these models requires
demographic data, which, with the exception of a
few echinoderms and bivalves in the Rockall Trough,
remain scarce for deep-sea animals (Gage and Tyler,
1991).
At vents and seeps, many of the larger animals,
including alvinellid polychaetes, siboglinid tube worms
Craig M. YOUNG
evidence to suggest that the opportunistic r-strategy is
well represented in the deep sea. As in shallow water,
the deep sea contains enough spatial and temporal
variability to allow exploitation by species with various
strategies of energy allocation.
Age distribution of reproductive effort
Background
In his classic paper linking life-history traits with
population growth, Cole (1954) focused on the demographic advantages and disadvantages of one-time
reproduction (semelparity) and repeated reproduction
(iteroparity). His major finding, often known as “Cole’s
result” was that semelparous animals that reproduce
early in life produce as many offspring over many
generations (i.e., are “as fit”) as iteroparous animals
that produce more offspring over the course of an
individual’s life, but begin reproduction later in life.
This result has been challenged as an oversimplification
(Charnov and Schaffer, 1973) but the fact remains
that both kinds of organisms are successful in the real
world. Todd (1986) noted that virtually all intertidal
invertebrates are iteroparous and used this observation
in support of the bet-hedging models that predict
iteroparity for habitats where organisms have variable
survival in the early life-history stages.
Predictions
“Bet-hedging” models predict that semelparity should
increase as a function of environmental stability; thus,
“normal” deep-sea habitats should select for semelparity, particularly among species that do not produce
pelagic larvae. The r/K-selection theory, by contrast,
predicts a predominance of long-lived, iteroparous
species in stable deep-sea habitats.
Evidence
The most extensive work on reproductive patterns
in the deep sea is that of Tyler, Gage and their
colleagues in the Rockall Trough. Of the many
species of echinoderms, molluscs, cnidarians and
other invertebrates they have studied, virtually all
reproduce over a wide range of body sizes, a feature
indicative of iteroparity. Moreover, the majority of
deep-sea species for which seasonal samples have been
obtained appear to reproduce more or less continuously,
probably over a number of annual cycles. The large
number of observations on echinoderm reproduction
on the Bahamian Slope also indicate iteroparity in all
species. However, Turner’s observations on xylophagid
bivalves living in wood are in stark contrast, as these
species attain reproductive maturity quickly and must
reproduce before they consume their own habitat. Not
all xylophagids seem to follow this pattern. Knudsen
(1961) found juveniles attached in some way to the
adults of nine species. Although this was interpreted as
evidence of brooding, the young might also have settled
on the adults after a period of dispersal. Thus, although
at first sight it seems that some xylophagids might not
be opportunistic and iteroparous, definitive information
is not available. Other kinds of opportunistic species
which probably have semelparous reproduction include
some cocculinid limpets which colonize palm fronds
and wood (Young and Tyler, unpublished data), and
sipunculans which recruit opportunistically into fibrous
substrata (M.A. Rice, P.A. Tyler and C.M. Young,
unpublished data). The latter appear to live longer than
a year and can reproduce several times in laboratory
culture. However, sipunculans differ from xylophagids
ecologically in that they do not destroy their own
habitat. Before concluding that iteroparity is more
common than semelparity in the deep sea, it should
be noted that semelparity is much more difficult
to document than iteroparity in infrequent deep-sea
sampling programs; indeed, it is probably common for
semelparous species to have life spans shorter than
typical sampling intervals.
Not all small-bodied animals are semelparous in
the deep sea. Wolff (1956a) found that the females
of at least two species of hadal tanaids “. . . pass
through several egg-bearing periods lasting probably
at least three months” and that “these periods may
occur only once every second or third year, provided
that the female reaches an age of, say 15–20 years.”
Wolff (1962) also found evidence of iteroparity in
asellote isopods, and Bishop reported multiple cohorts
of gametes in the gonads of abyssal cumaceans, a clear
indication of iteroparity.
If Jumars and Gallagher (1983) are correct about
the importance of juvenile predation, then the “bethedging” models correctly predict iteroparity; if not,
then the predominance of iteroparity in the deep
sea is best explained as a K-selected life-history
trait. Empirical evaluation of these models requires
demographic data, which, with the exception of a
few echinoderms and bivalves in the Rockall Trough,
remain scarce for deep-sea animals (Gage and Tyler,
1991).
At vents and seeps, many of the larger animals,
including alvinellid polychaetes, siboglinid tube worms