75
3 Why Spawn in Aggregations?
3.6 Sexual Selection in Spawning Aggregations
Many aggregative spawners also spawn in pairs outside of aggregations (Aguilar-Perera
1994 ) . For example, the two male phases (initial and terminal) of many parrotfi shes
and wrasses (Fig. 3.5 ) show contrasting spawning behaviours: initial-phase males
often spawn in aggregations and have group spawning (Fig. 12.45) whereas terminal-phase males usually spawn in pairs outside of aggregations and in defended
territories. In the bluehead wrasse, which similarly has two male spawning strategies (Figs. 12.38, 12.39), the success of the two male types, and hence of the prevailing mating system, depends largely on population density: when it is economically
viable to defend a territory, terminal-phase males do well and female- or resourcedefence polygyny prevails, but when population density is high, territoriality is too
costly and initial-phase group-spawning males predominate (Warner and Hoffman
1980a, b ) . Terminal phase queen parrotfi sh, Scarus vetula , also spawn in pairs and
groups. In this species, males usually defend large permanent territories containing
several females, with whom they pair-spawn (i.e. female-defence polygyny)
(Robertson and Warner 1978 ) . But Clavijo ( 1983 ) also observed lek polygyny in a
Puerto Rican population of queen parrotfi sh: males aggregated along the shelf edge
where they defended small, closely packed territories. They courted visiting females,
which inspected several males before selecting one with which they spawned. The
underlying mechanism for this mating-system shift was not investigated.
Selection pressures acting on males will differ according to whether males spawn
in multi- or single-male spawns. In contrast to males that spawn in multi-male
groups, males that spawn alone with a female compete physically and benefi t from
being large. Therefore, among aggregation spawners (and more generally among all
fi shes), we should expect to see a larger ratio of male:female sizes in species that
pair spawn than in species that spawn in multi-male groups. Intraspecifi c evidence
for this prediction from parrotfi sh is strong: group-spawning males are similar in
size to females while pair-spawning males are larger (e.g. Randall and Randall
1963 ; Choat and Robertson 1975 ; Robertson and Warner 1978 ) . Interspecifi c
evidence is less convincing but also in the expected direction. For example, a qualitative comparison of the degree of sexual dimorphism in group- and pair-spawning
groupers offers support for the prediction: generally, group-spawning Nassau
grouper (Sadovy and Colin 1995 ) show complete overlap in size distribution
between the sexes, while males of pair-spawning species such as the tiger grouper,
Mycteroperca tigris (White et al. 2002 ) , red grouper, Epinephelus morio (Brulé
et al. 1999 ; note: the species is not a confi rmed aggregation spawner), and the red
hind (Sadovy et al. 1994 ; Nemeth 2005 ) are larger, on average, than females.
Similarly, pair-spawning, terminal-phase male wrasses are larger than females while
conspecifi c group-spawning, initial-phase, males are similar in size to females
(Warner and Robertson 1978 ) .
When males gain a greater reproductive advantage with growth than females,
female-fi rst sex-change is predicted (Warner 1975 ; Leigh et al. 1976 ; Charnov 1982 ) .
3 Why Spawn in Aggregations?
3.6 Sexual Selection in Spawning Aggregations
Many aggregative spawners also spawn in pairs outside of aggregations (Aguilar-Perera
1994 ) . For example, the two male phases (initial and terminal) of many parrotfi shes
and wrasses (Fig. 3.5 ) show contrasting spawning behaviours: initial-phase males
often spawn in aggregations and have group spawning (Fig. 12.45) whereas terminal-phase males usually spawn in pairs outside of aggregations and in defended
territories. In the bluehead wrasse, which similarly has two male spawning strategies (Figs. 12.38, 12.39), the success of the two male types, and hence of the prevailing mating system, depends largely on population density: when it is economically
viable to defend a territory, terminal-phase males do well and female- or resourcedefence polygyny prevails, but when population density is high, territoriality is too
costly and initial-phase group-spawning males predominate (Warner and Hoffman
1980a, b ) . Terminal phase queen parrotfi sh, Scarus vetula , also spawn in pairs and
groups. In this species, males usually defend large permanent territories containing
several females, with whom they pair-spawn (i.e. female-defence polygyny)
(Robertson and Warner 1978 ) . But Clavijo ( 1983 ) also observed lek polygyny in a
Puerto Rican population of queen parrotfi sh: males aggregated along the shelf edge
where they defended small, closely packed territories. They courted visiting females,
which inspected several males before selecting one with which they spawned. The
underlying mechanism for this mating-system shift was not investigated.
Selection pressures acting on males will differ according to whether males spawn
in multi- or single-male spawns. In contrast to males that spawn in multi-male
groups, males that spawn alone with a female compete physically and benefi t from
being large. Therefore, among aggregation spawners (and more generally among all
fi shes), we should expect to see a larger ratio of male:female sizes in species that
pair spawn than in species that spawn in multi-male groups. Intraspecifi c evidence
for this prediction from parrotfi sh is strong: group-spawning males are similar in
size to females while pair-spawning males are larger (e.g. Randall and Randall
1963 ; Choat and Robertson 1975 ; Robertson and Warner 1978 ) . Interspecifi c
evidence is less convincing but also in the expected direction. For example, a qualitative comparison of the degree of sexual dimorphism in group- and pair-spawning
groupers offers support for the prediction: generally, group-spawning Nassau
grouper (Sadovy and Colin 1995 ) show complete overlap in size distribution
between the sexes, while males of pair-spawning species such as the tiger grouper,
Mycteroperca tigris (White et al. 2002 ) , red grouper, Epinephelus morio (Brulé
et al. 1999 ; note: the species is not a confi rmed aggregation spawner), and the red
hind (Sadovy et al. 1994 ; Nemeth 2005 ) are larger, on average, than females.
Similarly, pair-spawning, terminal-phase male wrasses are larger than females while
conspecifi c group-spawning, initial-phase, males are similar in size to females
(Warner and Robertson 1978 ) .
When males gain a greater reproductive advantage with growth than females,
female-fi rst sex-change is predicted (Warner 1975 ; Leigh et al. 1976 ; Charnov 1982 ) .