Get PDF - Life Sciences

doi:10.1111/j.1420-9101.2004.00823.x
The influence of territoriality and mating system on the evolution of
male care: a phylogenetic study on fish
M. AH-KING, C. KVARNEMO & B. S. TULLBERG
Department of Zoology, Stockholm University, Stockholm, Sweden
Keywords:
Abstract
certainty of paternity;
paternal care;
sneaking;
sperm competition;
Teleostei;
territory.
Evolution of male care is still poorly understood. Using phylogenetically
matched-pairs comparisons we tested for effects of territoriality and mating
system on male care evolution in fish. All origins of male care were found in
pair-spawning species (with or without additional males such as sneakers) and
none were found in group-spawning species. However, excluding group
spawners, male care originated equally often in pair-spawning species with
additional males as in strict pair-spawning species. Evolution of male care was
also significantly related to territoriality. Yet, most pair-spawning taxa with
male care are also territorial, making their relative influence difficult to
separate. Furthermore, territoriality also occurs in group-spawning species.
Hence, territoriality is not sufficient for male care to evolve. Rather, we argue
that it is the combination of territoriality and pair spawning with sequential
polygyny that favours the evolution of male care, and we discuss our results in
relation to paternity assurance and sexual selection.
Introduction
Parental care is common among animal taxa and has
originated numerous times. Particularly the evolution of
male care has attracted much interest (e.g. Trivers, 1972;
Williams, 1975; Dawkins & Carlisle, 1976; Maynard
Smith, 1977; Ridley, 1978; Perrone & Zaret, 1979;
Werren et al., 1980; Baylis, 1981; Gross & Shine, 1981;
Gross & Sargent, 1985; Wright, 1998; Tallamy, 2000) and
several factors promoting the evolution of male care have
been proposed. Features that should increase the benefit
of caring include, for example, a harsh environment,
competition for resources, and high predation pressure
(Clutton-Brock, 1991). However, caring for young may
be costly and these costs involve reduced foraging
opportunities (Trivers, 1972; Williams, 1975), increased
adult mortality (Magnhagen & Vestergaard, 1991) and,
most importantly for males, lost mating opportunities
because of an often impaired ability to attract new mates
(Trivers, 1972; Williams, 1975; Balshine-Earn & Earn,
1998). Indeed, most models investigating the evolution
Correspondence: Malin Ah-King, Department of Zoology, Stockholm
University, SE-106 91 Stockholm, Sweden.
Tel.: +46 (0) 8 16 43 98; fax: +46 (0) 8 16 77 15;
e-mail: malin.ah-king@zoologi.su.se
of parental care have assumed that caring for young
restricts males from gaining additional matings (e.g.
Wade & Shuster, 2002).
In this study, we found at least 22 origins of male care
in fishes. The question arises why this is so common.
First, in many fish species the assumption that giving care
limits the ability to attract new mates may simply not be
borne out. For example, for males that hold a territory in
which several females may spawn in succession, the cost
of lost mating opportunities is probably insignificant
(Blumer, 1979). Therefore male spawning territoriality is
commonly suggested to be a prerequisite for male care
evolution (Williams, 1975; Clutton-Brock, 1991). Secondly, it is often assumed that a male’s average paternity,
i.e. certainty of paternity of a particular offspring, will be
lower than the female’s average maternity (e.g. Kokko &
Jennions, 2003). However, in fish with male care in
nesting territories and several female clutches, each
female’s maternity may often be lower than the total
paternity of the caring male (C. Kvarnemo, unpublished). Territoriality has been suggested to be a precursor
to male care, as the former has been assumed to ensure
very high paternity (Perrone & Zaret, 1979; Baylis, 1981).
However, as pointed out already by Keenleyside (1981)
this assumption is often inaccurate, as in many fishes that
provide care in territories, paternity is often considerably
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
371
372
M. AH-KING ET AL.
reduced by sneakers intruding into the male-guarded
territory. Nevertheless, territoriality is likely to increase
paternity assurance compared with nonterritoriality.
Moreover, sexual selection can also affect male care
evolution. Females have been shown to prefer males that
can demonstrate care-taking abilities, either through
courtship behaviours (Östlund & Ahnesjö, 1998), nestbuilding ability (Soler et al., 1998; Jones & Reynolds,
1999) or the presence of eggs in their nests (e.g. Ridley &
Rechten, 1981; Marconato & Bisazza, 1986; Knapp &
Sargent, 1989). Males have also been shown to use
parental behaviours as a courtship strategy (e.g.
Pampoulie et al., 2004). As caring becomes attractive to
females, this adds another benefit of care to the male, by
not only increasing offspring survival, but also increasing
his mating opportunities. In fact, Tallamy (2000) has
proposed that male care could imply a benefit in terms of
increased mating success because of female preference,
instead of a cost (after male care has evolved). Available
data on arthropods support the hypothesis that sexual
selection has dominated the evolution of exclusive male
care (Tallamy, 2000). Data on fish in which males guard
their offspring also support this hypothesis, but natural
selection has played a primary role in species that carry,
mouth- or pouch-brood their young (M. Ah-King,
unpublished).
Although the evolution of male care has received
considerable interest, to our knowledge no study has yet
investigated if the factors proposed to influence the origin
of male care do indeed predate it. Here we test hypotheses concerning territoriality and mating system. Fish are
an ideal group for testing these hypotheses phylogenetically, because both paternal care and territoriality have
evolved a number of times and a wide range of mating
systems can be found in this taxon. We test the
hypotheses by comparing male care evolution in phylogenetic lineages contrasted with respect to differences in
territoriality and mating system, respectively.
Material and methods
The data
We used data on exclusive male care, male spawning
territoriality and mating system, in combination with
phylogenetic information, focusing in particular on
origins of exclusive male care (from no male care). We
searched the literature for descriptions of paternal care,
spawning territoriality and mating system and assigned
them to categories, as described below and as summarized in Appendix 1. Our main sources were Breder &
Rosen (1966) and the databases Fishbase, ISI (Institute
for Scientific Information) and ASFA (Aquatic Sciences
and Fisheries Abstracts). To infer origins of male care, we
collected data on species with exclusive male care and
their closest relatives lacking male care, as judged by the
phylogeny of the group. We then tried to find data on
territoriality and mating system for these species. However, in many cases the sister group to a clade with male
care was not well studied and we had to use more distant
relatives for which data were available.
We defined parental care as all behaviours that might
increase the fitness of the offspring, which includes for
example nest building, bearing, guarding and fanning of
eggs. However, merely building a nest could be regarded
as mating effort rather than parental investment. Therefore we present the results both including and excluding
the species in which males only perform care in the form
of nest building.
We categorized the mating systems into three groups:
group spawning, pair spawning, pair spawning with
additional males. Group spawners spawn in a school with
several females and males. Pair spawners include both
monogamous spawnings and sequentially polygynous
spawnings (i.e. matings with multiple females one at a
time). Additional males are usually sneakers or satellites.
However, this group also includes cases in which two
males join a female on each side, so called trio spawnings.
Species that are reported to be broadcasting have been
categorized as group spawners. Data and references are
presented in Appendix A1. To test the effect of mating
system on male care we first contrasted lineages with
group spawning to all varieties of pair spawning. We then
compared lineages with pair spawning, by contrasting
pure pair spawning to spawning where one male and one
female are accompanied by one or several additional
males.
We recorded a species as territorial if at least some
males hold spawning territories, as many species have
both territorial and sneaker males. However, we did not
include pure feeding territories. We assumed that a male
that builds a nest must be territorial during spawning
even if we did not find any references on territoriality in
these species. However, male parental care as such does
not necessarily imply spawning territoriality, as caring
does not always involve site defence. A male can, for
instance, carry the eggs with him (seahorses and pipefish,
Breder & Rosen, 1966) or splash water on the eggs above
the surface (Copella arnoldi, O’Neil & Dunham, 1972).
Phylogenetic tree reconstruction
We constructed a composite tree (Fig. 1) that as far as
possible is based on studies that have used modern
phylogenetic methods. When several phylogenies were
available for a group, we chose the two with the best
resolution and which included the largest number of
species for which we could obtain behavioural data.
Alternative phylogenies (Fig. 2) were used to test whether our results were consistent between phylogenies. As
the different alternative subtrees are independent, we
used the combination of alternative subtrees that produced the minimum and maximum number of changes
respectively in each trait.
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Male care evolution in fish
373
Fig. 1 Composite tree. (a) Optimization of
male care. Black lines, male care; white lines,
no male care; hatched lines, equivocal.
(b) Optimization of male spawning
territoriality. Black lines, male spawning
territoriality; white lines, no male spawning
territoriality; hatched lines, equivocal.
There are relatively many phylogenies that focus on
intra-familial relationships. We found a number of these
in Systematics, Historical Ecology, and North American Freshwater Fishes (Mayden, 1992): Smith (1992)(Catostomidae), Cavender & Coburn (1992) and Coburn &
Cavender (1992) (Cyprinidae), and Lundberg (1992)
(Ictaluridae). The phylogeny for Gasterosteidae was
obtained from McLennan et al. (1988), and alternative
phylogenies for Characidae were found in Lucena (1993)
and Uj (1990) cited in Tree of life (Ortı́ & Vari, 1997). For
the relationships among darters (Percidae) we used
Turner (1997) and Song et al. (1998). The relationships
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
374
M. AH-KING ET AL.
Fig. 1 Continued.
between families and orders are less well resolved. We
used Fink & Fink (1981) for the relationships within
Ostariophysi. Clupeomorpha was positioned as a sister
group of Ostariophysi according to Lecointre & Nelson
(1996). Salmonids were placed together with osmerids as
a sister group of esocoids and neoteleosts (Johnson &
Patterson, 1996). For the relationships within Osteoglossiformes we used Lauder & Liem (1983), Nelson (1994),
Kumazawa & Nishida (2000) and Lavoué et al. (2000).
Alternative phylogenies of Gasterosteiformes were found
in Pietsch (1978), Johnson & Patterson (1993) and
Nelson (1994). In addition, Smith & Stearley (1989)
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Male care evolution in fish
375
Fig. 2 Alternative composite tree. (a) Optimization of male care. Black lines, male care;
white lines, no male care; hatched lines,
equivocal. (b) Optimization of male spawning territoriality. Black lines, male spawning
territoriality; white lines, no male spawning
territoriality; hatched lines, equivocal.
was used for the relationships within Salmoniformes,
Parker & Kornfield (1995) for the relationships within
Cyprinodontiformes and Grande & Bemis (1996) for the
relationships within Acipenseriformes. Cobitidae was
included in Cypriniformes (Nelson, 1994). Acipenseri-
formes, Amia, Lepisosteidae and Teleostei were positioned according to Nelson (1994) and Bemis et al.
(1997), and finally Johnson & Patterson (1993) proposed
a scheme for the relationships within Percomorpha,
which we have used.
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
376
M. AH-KING ET AL.
Characiiformes
Cyprinidae
Fig. 2 Continued.
Optimization of characters
The characters were optimized using parsimony in
MacClade 4.0 (Maddison & Maddison, 2000). We optimized mating system as a two state-character, group
spawning and pair spawning with or without sneakers/
additional males. We dealt with equivocal branches by
using two extreme solutions, either maximizing or
minimizing male care branches over the whole tree.
The same method was used for equivocal optimizations
of mating system and territoriality. The number of origins
shown in the results are the minimum and maximum
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Male care evolution in fish
numbers, which depend on which particular optimization and phylogeny that was used. When a character
optimization for a polytomy (i.e. when the phylogenetic
relationships are not resolved for three or more species)
was equivocal, we used the most parsimonious optimization. For example, in a polytomy of three species for
which no care was inferred to be the ancestral state, and
where male care was present in two and absent in one
species, we assumed that male care had evolved only
once.
Matched pairs comparisons
As explained above, we collected data on male care,
territoriality and mating system and constructed two
alternative composite trees, based on the species with
such information. Territoriality and mating system were
optimized using maximum parsimony and these variables were considered independent in separate analyses.
The optimizations were used to identify independent
matched pairs contrasted by alternate states, following
the logic first presented by Burt (1989) (see also Møller &
Birkhead, 1992; Wickman, 1992; Reed & Nee, 1995;
Lindenfors & Tullberg, 1998; Goodwin et al., 2002). For
each comparison, the paths linking the taxa do not cross
the path of another comparison, and thus the comparisons are phylogenetically separate (Felsenstein, 1985;
Burt, 1989; Reed & Nee, 1995; Purvis & Bromham, 1997;
Maddison, 2000). Once a matched pair has been identified it cannot be accounted for again and comparisons
are sought among taxa that are as closely related as
possible in the remaining phylogeny (see for example
Lindenfors et al., 2003). Therefore a matched pair need
not be sister clades. Thereafter, instances of transitions to
male parental care were scored within each contrast,
under the null hypothesis that transitions from nonparental (the presumed ancestral state) to male care would
be independent of the state for either territoriality or
mating system. A simple binomial test was used to test
the null hypothesis. One important confounding factor in
this analysis is the size of the contrasting clades, such that
more speciose clades would have a higher background
probability of transitions to male care. To partially
compensate for the lack of data on species number we
have compared whether or not male care has evolved in
each contrasting clade instead of the number of times this
has occurred, using a sign test.
Results
Male care has evolved in a few nonterritorial taxa such as
Copella arnoldi and Syngnathidae. However in the majority of cases territoriality has preceded male care (Figs 1
and 2) and in an analysis based on phylogenetically
independent comparisons male care is significantly more
likely to evolve in territorial than nonterritorial lineages
(C 14 ¼ 0, P < 0.001; Table 1). Because nest building can
377
Table 1 Matched-pairs with respect to the presence/absence of
territoriality. The presence of male care origins is compared within
each pair. Ranges in male care origins and alternative matched pairs
are due to alternative phylogenies and/or to equivocal parts of the
tree resulting in alternative optimizations. Male care originates more
often in the territorial group in each of the 15 comparisons involving
a difference in male care origins (sign test: C 14 ¼ 0, P < 0.001).
Post-ovipositional male care (excluding cases with nest-building
only, indicated by *) originates more often in the territorial group in
each of the 12 pairs with a difference in origins of post-ovipositional
male care (sign test: C 11 ¼ 0, P < 0.01).
Paired taxa
No. of origins of male care
1. Territorial: Protopterus–Neoceratodus
Nonterritorial: Acipenser–Polyodon
2. Territorial: Amia calva
Nonterritorial: Lepisosteus osseus
3. Territorial: Scleropages–Mormyrops
Nonterritorial: Clupea
4. Territorial: Silurus–Eigenmannia
Nonterritorial: Aphylocharax–Pristobrycon
5. Territorial*: Hoplias
Nonterritorial*: Copella
6. Territorial: Moxostoma, Erimyzon
Nonterritorial: Catostomus
7 or 8 and 9:
7. Territorial: Rhodeus, Pseudorasbora
Nonterritorial: Ptycocheilus
8. Territorial: Rhodeus
Nonterritorial: Tinca
9. Territorial: Pseudorasbora
Nonterritorial: Ptychocheilus
10. Territorial: Zacco
Nonterritorial: Ctenopharyngodon
11. Territorial: Abramis
Nonterritorial*: Notemigonus or Scardinius
12. Territorial: Rhinichthys, Agosia
Nonterritorial: Gila
13. Territorial: Exoglossum–Margariscus
Nonterritorial: Phoxinus
14. Territorial: Campostoma, Dionda
Nonterritorial: Hybognathus
15. Territorial: Notropis
Nonterritorial: Ericymba
16. Territorial: Luxilus–Pimephales
Nonterritorial: Richardsonius, Clinostomus
17. Territorial: Thymallus
Nonterritorial: Coregonus
18. Territorial: Porichthys
Nonterritorial: Percopsis or Lota, Gadus
19. Territorial: Aulorhynchus–Gasterosteus
Nonterritorial: Syngnathus–Fistularia
20. Territorial: Cyprinodon, Jordanella
Nonterritorial: Menidia–Mugil
21. Territorial: Stizostedion lucioperca
Nonterritorial: S. vitreum or S. canadense
22. Territorial: Etheostoma nigrum–E. zonale
Nonterritorial: E. spectabile
23. Territorial: Etheostoma flabellare
Nonterritorial: E. caeruleum
1
0
1
0
2–3
0
1
0
0
0
1 0
1
0
0
0
1
0
0
0
0
0
1 0
2–3
0
1 0
0
0
1 + 1 0
0
0
1
0
1–2
1
1
0
1
0
1
0
1
0
*Male care originates in the ancestor to Hoplias and Copella.
Origin of male nest building (no post-ovipositional care).
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
378
M. AH-KING ET AL.
be regarded as merely mating effort we also carried out
an analysis based on male post-oviposition care only and
found that male care thus defined was also significantly
more likely to evolve in territorial lineages (C 11 ¼ 0,
P < 0.01; Table 1).
In our data set we found no case where male parental
care has evolved in a group spawning species, and when
contrasting group spawning with different kinds of pair
spawning in a matched-pairs comparison, we found that
male care evolution is significantly related to pair
spawning (C 15–16 ¼ 0, P < 0.001; Table 2). This was
also the case when only taxa with post-oviposition male
care were included (C 11–12 ¼ 0, P < 0.01; Table 2).
More interesting, perhaps, is the question of whether
the occurrence of male parental care in pair-spawning
species is influenced by the occurrence of additional
males such as sneakers. Therefore, we also conducted a
test where we excluded all group-spawning taxa and
contrasted pair spawning with and without sneakers in
phylogenetically matched pairs. In our data set there
were only six such contrasts, and in two of these male
care originated in the lineage without sneakers whereas
in four it evolved in the lineage with sneakers (C 5 ¼ 2,
n.s.; Table 3). Only considering post-ovipositional care
(i.e. excluding nest building only) produced the same
result (C 7 ¼ 3, n.s.; Table 3). Thus, judged from this
limited set of observations we conclude that the evolution of male care is independent of whether a male can
monopolize a female during spawning or not.
Discussion
In this study, we found a strong effect of both mating
system and territoriality on the evolution of paternal care
in fishes. Basically, paternal care does not evolve under
group spawning, but only in species with pair spawning,
even when this involves additional males such as
sneakers. That male care evolves more often in territorial
species has often been assumed, but to our knowledge
not previously been tested. Furthermore, Ridley (1978)
pointed out that territoriality could have evolved secondarily after male care, as well as before, so that
territoriality could be a result of male care rather than a
cause of it. However, our results show that the majority
of male care origins were preceded by spawning territoriality. In the following we discuss our results in relation
to parental certainty and sexual selection.
The low degree of paternal certainty in group spawnings is probably the reason why paternal care does not
evolve under this mating system, and because parental
certainty is low for both sexes the same result is expected
for all kinds of parental care. However, our results show
clearly that parental certainty need not be complete, such
as when a female is monopolized, but that male care can
evolve also under pair spawning where there are additional males present. In species with male care and
sneakers, however, the paired male has been found to
Table 2 Male care origins in matched-pair comparisons of mating
systems. Group spawning (G) is contrasted against pair spawning
with or without sneakers/additional males (P). The presence of male
care origins are compared within each pair. Ranges in male care
origins and alternative matched pairs are due to alternative phylogenies and/or to equivocal parts of the tree resulting in alternative
optimizations. Male care originates in the pair spawning group with
or without sneakers/additional males in each of the 16–17 comparisons where there is a difference in male care origins (sign test: C 15–
16 ¼ 0, P < 0.001). Post-ovipositional male care (excluding taxa in
which male care consists of only nest-building, indicated by asterisks)
originates in the pair spawning species with or without sneakers/
additional males in each of the 12–13 comparisons where there is a
difference in male care origins (sign test: C 11–12 ¼ 0, P < 0.01).
Paired taxa
1 or 2 and 3:
1. P: Protopterus–Neoceratodus
G: Acipenser–Polyodon
2. P: Amia calva
G: Lepisosteus osseus
3. P: Scleropages–Mormyrops
G: Acipenser–Polyodon
4. P: Silurus glanis, Eigenmannia
G: Clupea
5. P: Hoplias–Copella
G: Aphyocharax–Pristobrycon
6. P: Erimyzon, Moxostoma, Catostomus
G: Brachydanio–Cyprinus
7 or 8 and 9:
7. P: Rhodeus ocellatus, Pseudorasbora parva
G: Notemigonus–Clinostomus
8. P: Rhodeus ocellatus
G: Tinca tinca
9. P: Pseudorasbora parva
G: Scardinius–Clinostomus
10. P: Zacco temminicki
G: Ctenopharyngodon idella
11. P: Agosia
G: Rhinichthys
12. P: Exoglossum
G: Phenacobius
13. P: Campostoma
G: Dionda
14. P: Nocomis
G: Hybognathus
15. P: Semotilus
G: Couesius
16. P: Luxilus
G: Lythrurus or Hybopsis–Notropis
17 or 18 and 19:
17. P: Cyprinella spiloptera,
C. lutrensis, Pimephales promelas
G: Lythurus
18. P: Cyprinella spiloptera
G: C. lutrensis
19. P: Pimephales promelas
G: Lythurus
20. P: Thymallus thymallus
G: Coregonus lavaretus, C. autumnalis, C artedii
No. of origins
of male care
1
0
1
0
2–3
0
1
0
1
0
1*
0
1
0
0
0
1
0
0
0
1*
0
1
0
1*
0
1
0
1
0
1*
0
1
0
0
0
1
0
0
0
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Male care evolution in fish
Table 2 Continued.
No. of origins
of male care
Paired taxa
21. P: Mallotus
G: Hypomesus
22. P: Pleuronectes
G: Esox
23. P: Porichthys notatus
G: Percopsis omiscomaycus or Gadus morhua
24. P: Lota lota lota
G: Lota lota maculosa
25. P: Aulorhynchus–Jordanella
G: Menidia menidia, Mugil cephalus
26. P: Sander lucioperca
G: Stizostedion vitreum or S. canadense
27. P: Etheostoma nigrum, E. zonale,
E. spectabile, E. flabellare
G: E. caeruleum
28. P: Percina maculata, Percina caprodes
G: Perca
0
0
0
0
1
0
0
0
3–4
0
1
0
2
0
0
0
*Origin of male nest building (no post-ovipositional care).
Table 3 Male care origins in matched-pair comparisons of pair
spawning groups with or without sneakers/additional males. Pair
spawning (P) is contrasted against pair spawning with sneakers (S)
or trio spawning (T). Ranges are due to alternative phylogenies or
optimizations. In two of the comparisons male care origin is more
common without sneakers and in four comparisons it is more
common in the presence of sneakers (sign test: C 5 ¼ 2, n.s.). The
corresponding numbers for origins of post-oviposition care (excluding taxa in which male care consists of only nest-building, indicated
by asterisks) are three and five comparisons, respectively (sign test:
C 7 ¼ 3, n.s.).
Paired taxa
No. of origins of male care
1. P: Misgurnus
T/S: Moxostoma–Zacco
2. P: Exoglossum, Nocomis
S: Campostoma
3. P: Luxilus
S: Semotilus
4. P: Cyprinella spiloptera
S: Pimephales
5. P: Lota lota lota
S: Porichthys
6. P : Culaea
S : Pungitius
7. P: Syngnathus–Eurypegasus
S: Macroramphosus
8. P: Pleuronectes
S: Aulorhynchus–Gasterosteus
9. P: Jordanella
S: Cyprinodon
10. P: Percina maculata
S: P. caprodes
0
1 + 1*
1–2
1*
1*
1
0
1
0
1
0
0
1
0
0
1–2
1
0
0
0
*Origin of male nest building (no post-ovipositional care).
Male care is already present in the ancestor to this group.
379
fertilize a majority of the eggs under natural spawning
conditions: 81% in bluegill sunfish (Fu et al., 2001) and
89% in sand gobies (Jones et al., 2001). Nevertheless,
sneakers might compensate their lower fertilization
probability in a particular spawning by participating in
a larger number of spawnings. In bluegill sunfish, for
example, sneakers and guarding males have similar
individual fitnesses (Neff, 2001). In the few contrasts
with pair spawning that differed with respect to additional males (Table 3), the difference in male care origins
was not significant. In other words, at this point we
cannot say that paternity is important, within the range
of variation that occurs in pair spawning, for the
evolution of male care. However, this question needs
further investigation using an extended data set.
The male-caring species in our analysis are both pair
spawning and (with a few exceptions such as pipefishes
and seahorses, Appendix 1, Table 2) territorial. Thus, it
could be argued that it is pair spawning as such that is
important for the evolution of paternal care and not
territoriality. Indeed, because many group spawners are
territorial as well, and paternal care never evolves in
group spawners, we hold that territoriality is not a
sufficient condition for male care to evolve. Instead, we
reckon that it is the combination of territoriality with pair
spawning that has favoured the evolution of paternal
care. Then, why are these conditions important for male
care?
We believe that paternity is likely to increase above a
certain minimum threshold under a combination of pairspawning and territoriality. This is so because under
these circumstances a male may both be better able to
guard his mate against other males, and to spawn
sequentially with multiple females, in which case the
cost of male care in terms of missed opportunities to mate
becomes very low (Loiselle, 1978; Gross & Sargent,
1985). In our sample of care-giving species such males
were reported to mate with several, in some cases up to
seven, females as in Spinachia spinachia (Jones et al.,
1998), and in Hypoptychus dybowskii males have been
reported to guard 30 egg masses (Narimatsu & Munehara, 2001). Thus, care giving under pair spawning and
territoriality is fully compatible with a high degree of
polygyny.
Some of our analyses included only post-mating care,
i.e. we excluded species that only built nests. For these
species, too, territoriality was a strong predictor of male
care. In a recent model Wade & Shuster (2002)
concluded that if only some males are successful in
mating, these males will have higher average reproductive rate than females and will thereby gain more by
deserting the young. Wade & Shuster (2002) also argued
that their analysis supports the hypothesis that there is a
trade-off between pre-mating investment in competition
and post-mating investment in offspring care (Wade,
1979). Most ESS models of the evolution of uniparental
care assume that caring is incompatible with continued
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
380
M. AH-KING ET AL.
mating (Maynard Smith, 1977; Wade & Shuster, 2002).
As this is not the case in fishes, our results are not in
accordance with Wade’s hypothesis, because then we
would predict that territorial males are less likely to
invest in the care of young. Instead, as mentioned,
territoriality probably lowers the cost of care (Loiselle,
1978; Gross & Sargent, 1985) and the ability to guard
multiple clutches eliminates the conflict between mating
effort and parental effort.
In addition, if caring becomes attractive to females,
male care could imply a benefit in terms of increased
mating success because of female preference, instead of a
cost (Tallamy, 2000). Possibly, certain aspects of male
care can also improve paternity. For example, a nest
structure may help the male to defend his brood, not
only against predators, but also against sneaker males.
Consistent with this, sand goby males have been shown
to build extra small nest openings before mating when
sneaker males are present (Svensson & Kvarnemo,
2003). Once parental care has evolved, there is probably
a feedback between the levels of care and the intensity of
sexual selection (Reynolds, 1996). For example, the
variance in male reproductive success (and thus the
intensity of sexual selection) could increase when males
care, especially if females prefer to mate with males that
already guard eggs from other females in their nests (e.g.
Ridley & Rechten, 1981; Marconato & Bisazza, 1986;
Knapp & Sargent, 1989). Indeed, a comparative study on
fish suggests that female choice of caring males has been
important for the evolution of male care among
egg-guarding species (M. Ah-King unpublished). Species
that care for eggs by carrying them on the body or in
mouth- or brood pouches, on the other hand, have
limited space for the eggs, which suggests that sexual
selection may have been less important for the evolution
of male care in these species.
Male care has evolved many times in fish and we
conclude that it has been favoured by a combination of
territoriality and pair-spawning with multiple females
sequentially. Probably these traits have the joint effect of
lowering sperm competition (increased paternity security) and lowering the cost of missed mating opportunities. These traits have promoted male care in a group of
organisms where parental care may not be very costly in
the first place.
Acknowledgments
We thank P. Lindenfors for discussion and help with the
early analyses, S.O. Kullander for providing us with
references on fish phylogenies, L. Beesley for some initial
literature search on male care. We also thank C. Sargent,
K. Lindström, R. Fuller and L. Page for valuable information on darters, A. Gronell for information on Aulostomus,
O. Svensson for information on cichlids, K. Lindström for
information on Jordanella floridae and M. Gage for data on
zebrafish and phylogenetic information. S. Balshine-Earn,
Jacob Höglund, Arne Mooers, John Reynolds, Tom
Tregenza and Lars Werdelin provided very helpful
comments on earlier versions of the manuscript. This
work was financially supported by the Swedish Research
Council (grants to CK and to BT) and the Foundation of
Emil and Lydia Kinander, and was in part done while CK
was on a post-doc at the University of Jyväskylä, Finland.
Supplementary material
The following material is available from http://
www.blackwellpublishing.com/products/journals/suppmat/
jeb/jeb823/jeb823sm.htm
Appendix A1 Mating systems, parental care, male
territoriality and references used.
References
Balshine-Earn, S. & Earn, D.J.D. 1998. On the evolutionary
pathway of parental care in mouth-brooding cichlid fish. Proc.
R. Soc. Lond. B. 265: 2217–2222.
Baylis, J.R. 1981. The evolution of parental care in fishes, with
reference to Darwin’s rule of male sexual selection. Env. Biol.
Fish. 6: 223–251.
Bemis, W.E., Findeis, E.K. & Grande, L. 1997. An overview of
Acipenseriformes. Env. Biol. Fish. 48: 25–71.
Blumer, L.S. 1979. Male parental care in the bony fishes. Q. Rev.
Biol. 54: 149–161.
Breder, C.M. & Rosen, D.E. 1966. Modes of Reproduction in Fishes.
The Natural History Press, Garden City.
Burt, A. 1989. Comparative methods using phylogenetically
independent contrasts. In: Oxford Surveys in Evolutionary
Biology, Vol. 6 (P. H. Harvey & L. Partridge, eds), pp. 33–53.
Oxford University Press, Oxford.
Cavender, T.M. & Coburn, M.M. 1992. Phylogenetic relationships of North American Cyprinidae. In: Systematics, Historical
Ecology and North American Freshwater Fishes (R. L. Mayden,
ed.), pp. 293–327. Stanford University Press, Stanford.
Clutton-Brock, T.H. 1991. The Evolution of Parental Care. Monographs in Behavior and Ecology. Princeton University Press,
Princeton, NJ.
Coburn, M.M. & Cavender, T.M. 1992. Interrelationships of
North American cyprinid fishes. In: Systematics, Historical
Ecology and North American Freshwater Fishes (R. L. Mayden,
ed.), pp. 328–373. Stanford University Press, Stanford.
Dawkins, R. & Carlisle, T.R. 1976. Parental investment, mate
desertion and a fallacy. Nature 262: 131–133.
Felsenstein, J. 1985. Phylogenies and the comparative method.
Am. Nat. 125: 1–15.
Fink, S.V. & Fink, W.L. 1981. Interrelationships of the
ostariophysan fishes (Teleostei). Zool. J. Linn. Soc. 72: 297–353.
Fu, P., Neff, B. & Gross, M.R. 2001. Tactic-specific success in
sperm competition. Proc. R. Soc. Lond. B. 268: 1105–1112.
Goodwin, N.B., Dulvy, N.K. & Reynolds, J.D. 2002. Life-history
correlates of the evolution of live bearing in fishes. Phil. Trans.
R. Soc. Lond. B. 357: 259–267.
Grande, L. & Bemis, W.E. 1996. Interrelationship of Acipenseriformes, with comments om ‘‘Chondrostei’’. In: Interrelationships of Fishes (M. L. J. Stiassny, L. R. Parenti & G. D. Johnson,
eds), pp. 85–116. Academic Press Inc., San Diego, CA.
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
Male care evolution in fish
Gross, M.R. & Sargent, R.C. 1985. The evolution of male and
female parental care in fishes. Am. Zool. 25: 807–822.
Gross, M.R. & Shine, R. 1981. Parental care and mode of
fertilization in ectothermic vertebrates. Evolution 35: 775–793.
Johnson, G.D. & Patterson, C. 1993. Percomorph phylogeny: a
survey of acanthomorphs and a new proposal. Bull. Mar. Sci.
52: 554–626.
Johnson, G. D. & Patterson, C. 1996. Relationships of lower
euteleostean fishes. In: Interrelationships of Fishes (M. L. J.
Stiassny, L. R. Parenti & G. D. Johnson, eds), pp. 251–332.
Academic Press Inc., San Diego, CA.
Jones, J.C. & Reynolds, J.D. 1999. The influence of oxygen stress
on female choice for male nest structure in the common goby.
Anim. Behav. 57: 189–196.
Jones, A.G., Östlund-Nilsson, S. & Avise, J.C. 1998. A microsatellite assessment of sneaked fertilizations and egg thievery
in the fifteen-spine stickleback. Evolution 52: 848–858.
Jones, A.G., Walker, D., Lindström, K., Kvarnemo, C. & Avise,
J.C. 2001. Surprising similarity of sneaking rates and genetic
mating patterns in two populations of sand goby experiencing
disparate sexual selection regimes. Mol. Ecol. 10: 461–469.
Keenleyside, M. 1981. Parental care patterns of fishes. Am. Nat.
117: 1019–1022.
Knapp, R. & Sargent, R. 1989. Egg-mimicry as a mating strategy
in the fantail darter, Etheostoma flabellare-females prefer males
with eggs. Behav. Ecol. Sociobiol. 25: 321–326.
Kokko, H. & Jennions, M. 2003. It takes two to tango. Trends
Ecol. Evol. 18: 103–104.
Kumazawa, Y. & Nishida, M. 2000. Molecular phylogeny of
Osteoglossoids: a new model for gondwanian origin and plate
tectonic transportation of the Asian arowana. Mol. Biol. Evol.
17: 1869–1878.
Lauder, G.V. & Liem, K.F. 1983. The evolution and interrelationships of the Actinopterygian fishes. Bull. Mus. Comp. Zool. 150:
95–197.
Lavoué, S., Bigorne, R., Lecointre, G. & Agnèse, J.-F. 2000.
Phylogenetic relationships of mormyrid electric fishes (Mormyridae; Teleostei) inferred from cytochrome b sequences.
Mol. Phylogenet. Evol. 14: 1–10.
Lecointre, G. & Nelson, G. 1996. Clupemorpha, sistergroup of
Ostariophysi. In: Interrelationships of Fishes (M. L. J. Stiassny,
L. R. Parenti & G. D. Johnson, eds), pp. 193–208. Academic
Press Inc., San Diego, CA.
Lindenfors, P. & Tullberg, B.S. 1998. Phylogenetic analyses of
primate size evolution: the consequences of sexual selection.
Biol. J. Linn. Soc. 64: 413–447.
Lindenfors, P., Székely, T. & Reynolds, J.D. 2003. Phylogenetic
analyses of sexual size dimorphism in shorebirds, gulls and
alcids. J. Evol. Biol. 16: 930–938.
Loiselle, P.V. 1978. Prevalence of male brood care in teleosts.
Nature 276: 98.
Lucena, C.A.A.D. 1993. Estudio filogenético da famı́lia Characidae com
uma discussão dos grupos naturais propostos (Teleostei, Ostariophysi,
Characiformes). Universidade de São Paulo, Brazil, São Paulo.
Lundberg, J.G. 1992. The phylogeny of Ictalurid catfishes: a
synthesis of recent work. In: Systematics, Historical Ecology and
North American Freshwater Fishes (R. L. Mayden, ed.), pp. 392–
420. Stanford University Press, Stanford.
McLennan, D.A., Brooks, D.R. & McPhail, J.D. 1988. The
benefits of communication between comparative ethology
and phylogenetic systematics: a case study using gasterosteid
fishes. Can. J. Zool. 66: 2177–2190.
381
Maddison, W.P. 2000. Testing character correlation using pairwise comparisons on a phylogeny. J. Theor. Biol. 202: 195–204.
Maddison, W.P. & Maddison, D.R. 2000. MacClade 4: Analysis of
Phylogeny and Character Evolution. Sinauer Associates, Sunderland, MA.
Magnhagen, C. & Vestergaard, K. 1991. Risk taking in relation to
reproductive investments and future reproductice opportunities field experiments on nest-guarding common gobies,
Pomatoschistus microps. Behav. Ecol. 2: 351–359.
Marconato, A. & Bisazza, A. 1986. Males whose nests contain
eggs are preferred by female Cottus gobio L. (Pisces, Cottidae).
Anim. Behav. 34: 1580–1582.
Mayden, R.L. (ed.) 1992. Systematics, Historical Ecology, and North
American Freshwater Fishes. Stanford University Press, Stanford.
Maynard Smith, J. 1977. Parental investment: a prospective
analysis. Anim. Behav. 25: 1–9.
Møller, A.P. & Birkhead, T.R. 1992. A pairwise comparative
method as illustrated by copulation frequency in birds. Am.
Nat. 139: 644–656.
Narimatsu, Y. & Munehara, H. 2001. Territoriality, egg desertion
and mating success of a paternal care fish, Hypoptychus
dybowskii (Gasterosteiformes). Behaviour 138: 85–96.
Neff, B.D. 2001. Genetic paternity analysis and breeding success
in bluegill sunfish (Lepomis macrochirus). Am. Genet. Assoc. 92:
111–119.
Nelson, J.S. 1994. Fishes of the World, Vol. 3. Wiley, New York.
O’Neil, K.C. & Dunham, D.W. 1972. Preliminary observations
on the reproductive and parental behavior of the spraying
characid Copeina arnoldi Regan. Z. Tierpsychol. 31: 419–437.
Ortı́, G. & Vari, R.P. 1997. Characiformes. The Tree of Life Web
Project
[WWW
document].
URL
http://tolweb.org/
tree?group¼Characiformes&contgroup
Östlund, S. & Ahnesjö, I. 1998. Female fifteen-spined sticklebacks prefer better fathers. Anim. Behav. 56: 1177–1183.
Pampoulie, C., Lindström, K. & St. Mary, C.M. 2004. Have
your cake and eat it too: male sand gobies show more
parental care in the presence of female partners. Behav. Ecol.
15: 199–204.
Parker, A. & Kornfield, I. 1995. Molecular perspective on
evolution and zoogeography of cyprinodontid killifishes
(Teleostei; Atherinomorpha). Copeia 1995: 8–21.
Perrone, M.J. & Zaret, T.M. 1979. Parental care patterns in
fishes. Am. Nat. 113: 351–361.
Pietsch, T.W. 1978. Evolutionary relationships of the sea moths
(Teleostei: Pegasidae) with a classification of gasterosteiform
families. Copeia 1978: 517–529.
Purvis, A. & Bromham, L. 1997. Estimating the transition/
transversion ratio from independent pairwise comparisons
with an assumed phylogeny. J. Mol. Evol. 44: 112–119.
Reed, A.F. & Nee, S. 1995. Inference from binary comparative
data. J. Theor. Biol. 173: 99–108.
Reynolds, J.D. 1996. Animal breeding systems. Trends Ecol. Evol.
11: 68–72.
Ridley, M. 1978. Paternal care. Anim. Behav. 26: 904–932.
Ridley, M. & Rechten, C. 1981. Female sticklebacks prefer to
spawn with males whose nests contain eggs. Behaviour 76:
152–161.
Smith, G. 1992. Phylogeny and biogeography of the Catostomidae, freshwater fishes of North America and Asia. In:
Systematics, Historical Ecology, and North American Freshwater
Fishes (R. L. Mayden, ed.), pp. 788–826. Stanford University
Press, Stanford.
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY
382
M. AH-KING ET AL.
Smith, G.R. & Stearley, R.F. 1989. The classification and
scientific names of rainbow and cutthroat trouts. Fisheries 14:
4–10.
Soler, J.J., Cuervo, J.J., Moller, A.P. & De Lope, F. 1998. Nest
building is a sexually selected behaviour in the barn swallow.
Anim. Behav. 56: 1435–1442.
Song, C.B., Near, T.J. & Page, L.M. 1998. Phylogenetic relationships among percid fishes as inferred from mitochondrial
cytochrome b DNA sequence data. Mol. Phylogenet. Evol. 10:
343–353.
Svensson, O. & Kvarnemo, C. 2003. Sexually selected nestbuilding – Pomatoschistus minutus males build smaller nestopenings in the presence of sneaker males. J. Evol. Biol. 16:
896–902.
Tallamy, D.W. 2000. Sexual selection and the evolution of
exclusive paternal care in arthropods. Anim. Behav. 60: 559–567.
Trivers, R.L. 1972. Parental investment and sexual selection. In:
Sexual Selection and the Descent of Man (B. Campbell, ed.), pp.
136–179. Aldine, Chicago, IL.
Turner, T.F. 1997. Mitochondrial control region sequences and
phylogenetic systematics of darters (Teleostei: Percidae).
Copeia 1997: 319–338.
Uj, A. 1990. Etude comparative de l’ostologie cranienne des poissons de
la famille des Characidae et son importance phylogenetique.
Université de Geneve, Geneve.
Wade, M.J. 1979. Sexual selection and variance in reproductive
success. Am. Nat. 114: 742–747.
Wade, M.J. & Shuster, S.M. 2002. The evolution of parental care
in the context of sexual selection: a critical reassessment of
parental investment theory. Am. Nat. 160: 285–292.
Werren, J.H., Gross, M.R. & Shine, R. 1980. Paternity and
the evolution of male parental care. J. Theor. Biol. 82: 619–
631.
Wickman, P.-O. 1992. Sexual selection and butterfly design – a
comparative study. Evolution 46: 1525–1536.
Williams, G.C. 1975. Sex and Evolution. Princeton University
Press, Princeton, NJ.
Wright, J. 1998. Paternity and paternal care. In: Sperm Competition and Sexual Selection (T. R. Birkhead & A. P. Møller, eds),
pp. 117–145. Academic Press, Cambridge.
Received 9 March 2004; revised 14 August 2004; accepted 16 August
2004
J. EVOL. BIOL. 18 (2005) 371–382 ª 2004 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY