76
P.P. Molloy et al.
This is expected in pair-spawners but not multi-male group-spawners (Molloy et al.
2007 ) . It is noteworthy, then, that the only groupers for which there is evidence that
some individuals are gonochoristic (i.e. they do not change sex and only ever
function as one sex) are group-spawning Nassau grouper (Sadovy and Colin 1995 ) ,
leopard grouper (Erisman et al. 2007 ) and kelp bass, Paralabrax clathratus (Sadovy
and Domeier 2005b ) , where selection for large male size is relaxed. Further comparative analyses should reveal that female-fi rst sex change is more common among
those species that spawn in pairs or single-male groups than those that spawn in
multi-male groups.
Highly tuned mate choice by females and intense male-male competition,
particularly in pair-spawning species, lead to highly skewed male reproductive success. Not only do the largest males often defend the best resources or spawning
locations, they are also disproportionately selected as mates (e.g. threadfi n wrasse
Cirrhilabrus temminckii , Kohda et al. 2005 ; Mediterranean parrotfi sh Sparisoma
cretense , Afonso et al. 2008b ; Andersson 1994 ) . Skewed male reproductive success
has rarely been quantifi ed in aggregation spawners. One exception is an example of
the Atlantic cod (North Sea population). Using microsatellite-based parentage
analysis, Bekkevold et al. ( 2002 ) showed that male reproductive success followed a
U-shaped distribution, indicating that most males achieved either very high or very
low reproductive success. A subsequent study showed that 80% of the offspring
were sired by only 2–7 males out of groups of 52–93 individuals (Rowe et al. 2008 ) .
Furthermore, larger males with longer fi ns used in courtship had the highest reproductive success, implying strong sexual selection in this species. Similar analyses
have not been performed on other aggregation spawners but based on the discussion
above relating to pair- and group-spawning males, one can predict that male reproductive success will be more heavily skewed in pair- than group-spawners (see
also Chap. 8 ).
As more species and aggregations are studied, it is likely that within-species
variation in mating systems and tendency to spawn in aggregations will be
recorded in many other species. Such fl exibility may prove important in relation
to their ability to cope with heavy fi shing pressure (Colin 1982a ) : if heavy fi shing
mortality at aggregations makes it unprofi table for individuals to spawn in groups,
non-aggregated resource- or female- defence mating systems may prevail. Other
species without such phenotypic plasticity may be more vulnerable to fi sheriesmediated extirpation. Whether or not this is the case, and whether fi sh have such
behavioural fl exibility, are pertinent questions for fi sheries conservation. As far as
we are aware, no studies have attempted to test such issues. These questions could
be explored using modelling techniques to predict relative vulnerability of species
with and without different spawning tactics, or with varying degrees of behavioural plasticity. More empirical evidence could be obtained by monitoring relative frequency of different spawning tactics on and away from known spawning
sites. Many of the methods that would be involved in the design and implementation of the fi eldwork required to test the predictions described above are detailed
in Chaps. 9 and 10 .
P.P. Molloy et al.
This is expected in pair-spawners but not multi-male group-spawners (Molloy et al.
2007 ) . It is noteworthy, then, that the only groupers for which there is evidence that
some individuals are gonochoristic (i.e. they do not change sex and only ever
function as one sex) are group-spawning Nassau grouper (Sadovy and Colin 1995 ) ,
leopard grouper (Erisman et al. 2007 ) and kelp bass, Paralabrax clathratus (Sadovy
and Domeier 2005b ) , where selection for large male size is relaxed. Further comparative analyses should reveal that female-fi rst sex change is more common among
those species that spawn in pairs or single-male groups than those that spawn in
multi-male groups.
Highly tuned mate choice by females and intense male-male competition,
particularly in pair-spawning species, lead to highly skewed male reproductive success. Not only do the largest males often defend the best resources or spawning
locations, they are also disproportionately selected as mates (e.g. threadfi n wrasse
Cirrhilabrus temminckii , Kohda et al. 2005 ; Mediterranean parrotfi sh Sparisoma
cretense , Afonso et al. 2008b ; Andersson 1994 ) . Skewed male reproductive success
has rarely been quantifi ed in aggregation spawners. One exception is an example of
the Atlantic cod (North Sea population). Using microsatellite-based parentage
analysis, Bekkevold et al. ( 2002 ) showed that male reproductive success followed a
U-shaped distribution, indicating that most males achieved either very high or very
low reproductive success. A subsequent study showed that 80% of the offspring
were sired by only 2–7 males out of groups of 52–93 individuals (Rowe et al. 2008 ) .
Furthermore, larger males with longer fi ns used in courtship had the highest reproductive success, implying strong sexual selection in this species. Similar analyses
have not been performed on other aggregation spawners but based on the discussion
above relating to pair- and group-spawning males, one can predict that male reproductive success will be more heavily skewed in pair- than group-spawners (see
also Chap. 8 ).
As more species and aggregations are studied, it is likely that within-species
variation in mating systems and tendency to spawn in aggregations will be
recorded in many other species. Such fl exibility may prove important in relation
to their ability to cope with heavy fi shing pressure (Colin 1982a ) : if heavy fi shing
mortality at aggregations makes it unprofi table for individuals to spawn in groups,
non-aggregated resource- or female- defence mating systems may prevail. Other
species without such phenotypic plasticity may be more vulnerable to fi sheriesmediated extirpation. Whether or not this is the case, and whether fi sh have such
behavioural fl exibility, are pertinent questions for fi sheries conservation. As far as
we are aware, no studies have attempted to test such issues. These questions could
be explored using modelling techniques to predict relative vulnerability of species
with and without different spawning tactics, or with varying degrees of behavioural plasticity. More empirical evidence could be obtained by monitoring relative frequency of different spawning tactics on and away from known spawning
sites. Many of the methods that would be involved in the design and implementation of the fi eldwork required to test the predictions described above are detailed
in Chaps. 9 and 10 .