Download Johnson, K. P. 1999. The evolution of bill coloration and plumage

yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Introduction to evolution wikipedia , lookup

Hologenome theory of evolution wikipedia , lookup

Saltation (biology) wikipedia , lookup

Sexual conflict wikipedia , lookup

Koinophilia wikipedia , lookup

Signalling theory wikipedia , lookup

Extended female sexuality wikipedia , lookup

Sexual selection wikipedia , lookup

Behavioral Ecology Vol. 10 No. 1: 63–67
The evolution of bill coloration and plumage
dimorphism supports the transference
hypothesis in dabbling ducks
Kevin P. Johnson
Bell Museum of Natural History and Department of Ecology, Evolution, and Behavior, University of
Minnesota, St. Paul, MN 55108, USA
Bright coloration in male birds is typically thought to be driven by sexual selection (female choice or male–male competition).
Bird species often vary in the intensity of bright coloration, but few studies have addressed this cross-species variation. Potentially
this variation could result from either variation in female preferences or in the relative costs of male traits. Species of dabbling
ducks vary in the presence of bright male plumage and bill coloration. I tested the transference hypothesis for ornament
evolution in dabbling ducks using a phylogenetic study of character evolution. The transference hypothesis makes three predictions: (1) a costly male ornamental trait is the ancestral condition, (2) a less costly male ornamental trait is the derived
condition, and (3) gains in the less costly trait are associated with losses or absence of the more costly male trait. All three of
these predictions were satisfied in this study of the evolution of plumage dimorphism and bright bill coloration in the dabbling
ducks, given that bright plumage coloration is more costly than bright bill coloration. Key words: Anas, female choice, ornamentation, sexual selection, waterfowl. [Behav Ecol 10:63–67 (1999)]
heoretical and empirical studies of female mate choice
and sexual selection make predictions about why females should prefer ornamental male traits (Andersson, 1994;
Fisher, 1930; Grafen, 1990; Iwasa et al., 1991; Lande, 1981;
Pomiankowski et al., 1991; Ryan et al., 1990; Zahavi, 1975).
However, few workers have examined why male secondary sexual traits take the particular form that they do (Wiley and
Richards, 1978; Zahavi, 1987) or why these traits vary among
species as they do (Hamilton and Zuk, 1982; Read and Weary,
1992). Because many male secondary sexual traits are assumed to be costly, variation among species might arise due
to differences in the cost of these traits (Endler, 1977; Endler
and Houde, 1995).
One hypothesis to explain evolution and evolutionary variation in male traits is ‘‘transference’’ (Borgia, 1993; Gilliard,
1956, 1963; Kusmierski et al., 1993). According to this hypothesis, if males are able to use less costly traits and still
attract females, then these less costly traits should be favored
by natural selection and evolve. This hypothesis was developed
for bowerbirds because it was noted (Borgia, 1993; Gilliard,
1956, 1963) that bowerbird species with bright male plumage
tend to build simple bowers, whereas monomorphic species
with dull males tend to build elaborate bowers with many decorations. Ornamentation may be transferred to the bower if
it is a less costly trait than male plumage because the male
can ‘‘escape’’ the costly ornament if necessary (Gilliard 1956,
1963). Borgia (1993) examined the cost of the bower display
in male satin bowerbirds (Ptilonorhynchus violaceus) and
found no evidence of a high cost for this display; however, the
cost of bright plumage in bowerbird males has not been measured. A phylogenetic test of the transference hypothesis in
bowerbirds was partially consistent with the prediction that
K. P. Johnson is currently at the Department of Biology, University
of Utah, 257 South 1400 East, Salt Lake City, UT 84112-0840, USA.
E-mail: [email protected]
Received 10 February 1998; revised 16 June 1998; accepted 29 June
q 1999 International Society for Behavioral Ecology
bright male plumage and lack of sophisticated bowerbuilding
are the ancestral conditions (Kusmierski et al., 1993).
The transference hypothesis of male trait evolution, which
can be extended to other taxa, makes three major predictions:
(1) elaborate (more costly) ornaments should be the ancestral
condition, (2) less costly ornaments should be a derived condition, and (3) gains in the less costly ornament should be
associated with absence or loss of the more costly ornament.
I tested these predictions in dabbling ducks by examining the
evolution of bill coloration and dimorphic plumage using a
phylogeny derived from mitochondrial DNA sequences
( Johnson, 1997; Johnson and Sorenson, 1998).
Dabbling ducks have worldwide distribution, and they exhibit remarkable variation between species in traits that are
believed to be under sexual selection. Both bill coloration and
plumage characteristics are targets of female choice in dabbling ducks. In mallards (Anas platyrhynchos), Omland
(1996a,b) found that females strongly prefer males with
brightly colored bills and that females also show a preference
for overall plumage condition (Holmberg et al., 1989; Weidmann, 1990). In northern pintails (Anas acuta), Sorenson
and Derrickson (1994) found that females have preferences
for some plumage ornaments but not others. These results,
taken together with the elaborateness of male plumage and
bill coloration in dabbling ducks, suggest that these traits are
likely to be under the influence of sexual selection in many
dabbling duck species. Because bright male plumage often
covers most of the male’s body and is often very striking, it is
likely that bright plumage is more costly (at least in terms of
predator pressure) than a bill spot (which only covers a small
portion of the bill). Metz and Ankney (1991) documented
increased hunter-caused mortality of male ducks with brightly
colored plumage compared to dull individuals. It is currently
unknown whether bright bill coloration or bright plumage is
more costly to produce, so I assumed differences in predation
costs for this analysis. I investigated the evolution and evolutionary correlation of these two traits with respect to the transference hypothesis of female choice with the expectation that
gains in carotenoid bill coloration are associated with monomorphic plumage.
Behavioral Ecology Vol. 10 No. 1
Figure 1
Phylogenetic reconstruction of
carotenoid bill-coloration using unordered parsimony.
Three alternative reconstructions shown as equivocal
branches. A. 5 Anas.
I scored (using colored plates in Madge and Burn, 1988) the
presence or absence of male bill coloration and plumage dimorphism as dichotomous characters for 42 species of dabbling ducks. Species were scored as possessing bill coloration
if the bill of the male has spots or is completely colored with
carotenoid pigmentation. I assumed that red or yellow bills or
spots contained carotenoids. For example, the male whitecheeked pintail (Anas bahamensis) possesses a red bill spot,
while the bill of the male mallard (Anas platyrhynchos) is entirely yellow. Both of these species were considered to have
carotenoid bill coloration. Scoring of plumage dimorphism
and bill coloration by two independent observers produced
identical coding for both characters.
I optimized the dichotomous characters bill coloration and
plumage dimorphism over the phylogenetic tree ( Johnson,
1997; Johnson and Sorenson, 1998) using MacClade (Maddison and Maddison, 1992) with unordered parsimony. Given
the phylogenetic distribution of these two characters in waterfowl, the unordered assumption seems reasonable (but see
Omland, 1997, for an alternative view). I examined all possible reconstructions for each character. I used the concentrated changes test (Maddison, 1990) to test for an association
between gains and losses in one character and a particular
character state in the other character. I did this for all possible
(six in this case) combinations of character reconstructions.
Under the transference hypothesis, gains and losses in bill
coloration should be associated with monomorphic and dimorphic plumage, respectively, because bill coloration is assumed to be less costly than bright plumage. The concentrated changes test examines, in this case, whether more gains in
bright bill coloration (changes from white to black in Figure
1) occur on branches with monomorphic plumage reconstructed (white branches in Figure 2) than expected by
chance. However, I also tested for the alternative association;
that is, whether the changes in plumage were associated with
the state of bill coloration.
Optimization of plumage dimorphism over the tree resulted
in two equally parsimonious reconstructions of plumage dimorphism. In both reconstructions bright male dimorphic
plumage was the ancestral condition. Either six losses and five
gains of dimorphism or five losses and six gains were reconstructed (Figure 2). This is consistent with the first prediction
of the transference hypothesis that the more costly male trait
is ancestral. I found three equally parsimonious reconstructions for bill coloration. Between six and eight gains in bill
coloration and between three and one losses occur in this
character (Figure 1). In all reconstructions, the absence of
bill coloration is the ancestral condition, and bill coloration
evolves independently at least six times. This is consistent with
the second prediction of the transference hypothesis that the
less costly male trait is a derived condition.
When I examined the association of gains in bill coloration
with monomorphic plumage and losses of bill coloration with
dimorphic plumage, I found that there was a strong association (concentrated changes test, p , .01 for all six combinations of reconstructions, range 0.005 to , 0.001). That is, the
several independent gains in bill coloration were concentrated on branches that had monomorphic plumage reconstructed on them. This is consistent with the final prediction of the
transference hypothesis that gains in the less costly ornament
should be associated with absence or loss of the more costly
Alternatively, when I examined the association of changes
in plumage dimorphism with the presence or absence of bill
coloration, there was no association (concentrated changes
test, p . .50 for all six combinations of reconstructions). That
Johnson • Transference hypothesis in dabbling ducks
Figure 2
Phylogenetic reconstruction of
plumage dimorphism using
unordered parsimony. Two
shown as equivocal branches.
A. 5 Anas.
is, changes in the more costly trait are not influenced by the
state of the less costly trait, but once the more costly trait is
lost, origin of the less costly trait is favored. This result is in
accordance with the transference hypothesis, which makes
predictions about the origin of the less costly trait (bright bill
coloration) but not about the origin of the more costly trait
(bright plumage).
The three predictions of the transference hypothesis are supported for bill coloration and plumage dimorphism in dabbling ducks. The more costly male secondary sexual trait (in
terms of its potential obviousness to predators), bright male
plumage, is the ancestral condition in dabbling ducks, and it
has been lost several times. The less costly trait, bill coloration,
is derived several times independently within dabbling ducks.
Gains in bill coloration are associated with the absence of
bright dimorphic plumage. The pattern of changes in these
male traits supports the transference hypothesis (Kusmierski
et al., 1993).
Nearly all species that possess bill coloration have monomorphic plumage. The exception to this pattern is the mallard (A. platyrhynchos). The mallard has dimorphic plumage
and a colored bill. Female mate-choice experiments in this
species have shown that bill coloration and overall plumage
condition are both important predictors of female choice
(Holmberg et al., 1989; Omland, 1996a,b; Weidmann, 1990).
The mallard group has also radiated recently compared to
other dabbling duck groups ( Johnson, 1997; Johnson and Sorenson, 1998), and this may result in an evolutionary time lag
in the changes in plumage and bill coloration.
Support for the transference hypothesis relies on the assumption that bill coloration is less costly than bright plumage. Although the actual costs of bill coloration and bright
plumage in male dabbling ducks have not been measured
(but see Metz and Ankney, 1991, who documented hunter
preference for bright males), there is indirect evidence that
this is the case. Females of species with bright male bill coloration often have colored bills (10 out of 14 species). This
could be due to a genetic correlation between males and females which causes bill coloration to be present in females
because it is being selected for in males (Fisher, 1930; Lande,
1980; Møller, 1994). Typically, colored bills are duller in females than in males ( Johnson KP, personal observation). In
contrast, in species with markedly bright plumage, only the
males possess this coloration in the plumage. Females of dimorphic species are typically mottled brown, presumably an
adaptation for crypsis to avoid predation while sitting on the
nest (Sargeant and Raveling, 1992). This pattern is further
evidence suggesting that bill coloration is not as costly as
bright plumage; presumably selection for females to lose bill
coloration is weak.
While changes in these two male secondary sexual traits
form a strong pattern, it is unknown how female preferences
change in relation to male traits. The transference hypothesis
as developed (Kusmierski et al., 1993) suggests that female
preferences should also change in association with changes in
the male traits. Relatively few mate choice experiments have
been performed using dabbling duck species (mallard: Holmberg et al., 1989; Omland, 1996a, b; Weidmann, 1990; northern pintail: Sorenson and Derrickson, 1994). To understand
the full nature of the pattern of change in male traits, it will
also be important to document how female preferences
change. For example, do females of dimorphic species have
preferences for bright plumage but lack preference for bill
coloration, or do females have a preexisting bias (Basolo,
1990; Ryan et al., 1990) for bill coloration that favors the repeated evolution of this trait in dabbling ducks?
Even though bill brightness is clearly associated with monomorphic plumage, it is unclear from this analysis what forces
are driving changes in plumage brightness. Although the an-
Behavioral Ecology Vol. 10 No. 1
cestral condition of dimorphic plumage is consistent with the
transference hypothesis, the five to six rederivations of plumage dimorphism are not easily explained. Under the transference hypothesis, it is expected that once female preference
has shifted to a less costly male trait, the trait should be stable
against further evolution to a more costly trait. Other factors
(perhaps ecological or social) influencing the cost and/or
benefit of particular male traits might play a role in the evolutionary changes in plumage dimorphism.
In addition to plumage and bill coloration, male dabbling
ducks also use a wide variety of movement and vocal displays
in their courtship (Lorenz, 1941; McKinney, 1975). Large display repertoires are associated with dimorphic plumage
( Johnson, 1997), but few studies have documented female
preferences for displays. Mallard males that display more frequently tend to obtain more mates (Bossema and Kruijt, 1982;
Holmberg et al., 1989; Klint et al., 1989); however, none of
these studies experimentally manipulated display rates, so the
actual role of display performance and intensity in female
choice is still uncertain. Future experimental and phylogenetic studies integrating male variation in plumage, bill coloration, and movement and vocal display with female choice
are needed to fully understand the evolution of secondary
sexual traits in dabbling ducks.
In conclusion, although the transference hypothesis appears to be supported in bowerbirds (Kusmierski et al., 1993)
and dabbling ducks (this study), it is unclear how broadly it
will apply to other taxa. The cost of ornaments is assumed by
the transference hypothesis, but the transference hypothesis
does not make assumptions about the mechanism of sexual
selection (e.g., Fisher, 1930; Ryan et al., 1990; Zahavi, 1975).
The costliness of ornaments is an underlying theme of all theories of sexual selection by female choice. If ornaments are
not costly, then they could be explained by ‘‘ordinary’’ natural
selection, and female choice would not need to be invoked
(Andersson, 1994). The fact that female choice overrides the
negative fitness costs of male ornaments is of primary interest
in theories of sexual selection. The transference hypothesis
could generally operate under any of the models of female
choice. In runaway sexual selection models (Fisher, 1930;
Lande, 1981; Pomiankowski et al., 1991), ornaments with a
smaller fitness penalty (that females prefer equally to more
costly ornaments) should be favored by selection and displace
more costly ornaments. Likewise, under handicap models of
sexual selection (Grafen, 1990; Iwasa et al., 1991; Zahavi,
1975), if a less costly trait provides just as much information
about a male as a more costly trait, it should be favored and
displace the more costly trait (Grafen, 1990). Even under the
preexisting bias models (Ryan et al., 1990), biases for less costly traits are more likely to prevail in a population than biases
of similar magnitude for more costly traits. Variation in costs
and kinds of male ornaments should be of considerable interest in explorations of sexual selection.
In most taxa, courtship and female choice often involve several traits (and often different types of traits, such as morphology versus vocalizations). Presumably some of these traits
are less costly for males to produce than others. If this is the
case, then there should be selection for males to use cheaper
displays, given that females show equal or compensating preferences for these traits. The evolution of less costly traits from
costly traits bears more investigation in other taxa.
I thank F. McKinney, D. Clayton, and two anonymous reviewers who
reviewed earlier versions of this manuscript and made helpful suggestions. Discussions with K. Omland provided valuable insights into
the nature of female choice in dabbling ducks. B. Walther and T.
Jones provided independent scoring of the characters used in this
study. This work was supported financially by a National Science Foun-
dation Doctoral Dissertation Improvement Grant, Frank M. Chapman
Memorial Grant, and Dayton and Wilkie Funds for Natural History
Andersson MB, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Basolo AL, 1990. Female preference for male sword length in the
green swordtail, Xiphophorus helleri (Pisces: Poeciliidae). Anim Behav 40:332–338.
Borgia G, 1993. The cost of display in the non-resource based mating
system of the Satin Bowerbird. Am Nat 141:729–743.
Bossema I, Kruijt JP, 1982. Male activity and female mate acceptance
in the mallard (Anas platyrhynchos). Behaviour 79:313–324.
Endler JA, 1977. Natural selection on color patterns in Poecilia reticulata. Evolution 34:76–94.
Endler JA, Houde AE, 1995. Geographic variation in female preferences for male traits in Poecilia reticulata. Evolution 49:456–468.
Fisher RA, 1930. The genetical theory of natural selection. Oxford:
Clarendon Press.
Gilliard ET, 1956. Bower ornamentation versus plumage characters in
bowerbirds. Auk 73:450–451.
Gilliard ET, 1963. The evolution of bowerbirds. Sci Am 209:38–46.
Grafen A, 1990. Sexual selection unhandicapped by the Fisher process. J Theor Biol 144:473–516.
Hamilton W, Zuk M, 1982. Heritable true fitness and bright birds: a
role for parasites? Science 218:384–387.
Holmberg K, Edsman L, Klint T, 1989. Female mate preferences and
male attributes in mallard ducks (Anas platyrhynchos). Anim Behav
Iwasa Y, Pomiankowski A, Nee S, 1991. The evolution of costly mate
preferences. II. The ‘‘handicap’’ principle. Evolution 45:1431–1432.
Johnson KP, 1997. The evolution of behavior in the dabbling ducks
(Anatini): a phylogenetic approach (PhD dissertation). St. Paul:
University of Minnesota.
Johnson KP, Sorenson MD, 1998. Comparing molecular evolution in
two mitochondrial protein coding genes (cytochrome b and ND2)
in the dabbling ducks (Tribe: Anatini). Mol Phylogenet Evol. 10:
Klint T, Edsman L, Holmberg K, Silverin B, 1989. Hormonal correlates of male attractiveness during mate selection in the mallard
ducks (Anas platyrhynchos). Hormones Behav 23:83–91.
Kusmierski R, Borgia G, Crozier RH, Chan BHY, 1993. Molecular information on bowerbird phylogeny and the evolution of exaggerated male characteristics. J Evol Biol 6:737–752.
Lande R, 1980. Sexual dimorphism, sexual selection and adaptation
in polygenic characters. Evolution 34:292–305.
Lande R, 1981. Models of speciation by sexual selection on polygenic
characters. Proc Natl Acad Sci USA 78:3721–3725.
Lorenz K, 1941. Vergleichende Bewegungsstudien an Anatinen. J Ornithol 89:19–29.
Maddison W, 1990. A method for testing the correlated evolution of
two binary characters: are gains or losses concentrated on certain
branches of a phylogenetic tree? Evolution 44:539–557.
Maddison WP, Maddison DR, 1992. MacClade: analysis of phylogeny
and character evolution. 3.04. Sunderland, Massachusetts: Sinauer
Madge S, Burn H, 1988. Waterfowl: an identification guide to the
ducks, geese, and swans of the world. Boston: Houghton Mifflin.
McKinney F, 1975. The evolution of duck displays. In: Function and
evolution in behavior (Baerends G, Beer C, and Manning A, eds).
Oxford: Clarendon Press; 331–357.
Metz KJ, Ankney CD, 1991. Are brightly coloured male ducks selectively shot by duck hunters? Can J Zool 69:279–282.
Møller AP, 1994. Sexual selection and the barn swallow. Oxford: Oxford University Press.
Omland KE, 1996a. Female mallard mating preferences for multiple
male ornaments. I. Natural variation. Behav Ecol Sociobiol 39:353–
Omland KE, 1996b. Female mallard mating preferences for multiple
male ornaments. II. Experimental variation. Behav Ecol Sociobiol
Omland KE, 1997. Examining two standard assumptions of ancestral
Johnson • Transference hypothesis in dabbling ducks
reconstructions: repeated loss of dichromatism in dabbling ducks
(Anatini). Evolution 51:1636–1646.
Pomiankowski A, Iwasa Y, Nee S, 1991. The evolution of costly mate
preferences. I. Fisher and biased mutation. Evolution 45:1422–
Read AF, Weary DM, 1992. The evolution of bird song: comparative
analyses. Phil Trans R Soc Lond B 338:165–187.
Ryan MJ, Fox JH, Wilczynski W, Rand AS, 1990. Sexual selection for
sensory exploitation in the frog, Physalaemus pustulosus. Nature
Sargeant AB, Raveling DG, 1992. Mortality during the breeding season. In: Ecology and management of breeding waterfowl (Batt et
al., eds). Minneapolis: University of Minnesota Press; 396–422.
Sorenson LG, Derrickson SR, 1994. Sexual selection in the Northern
Pintail (Anas acuta): the importance of female choice versus malemale competition in the evolution of sexually selected traits. Behav
Ecol Sociobiol 35:389–400.
Weidmann U, 1990. Plumage quality and mate choice in mallards
(Anas platyrhynchos). Behaviour 115:127–141.
Wiley RH, Richards DG, 1978. Physical constraints on acoustic communication in the atmosphere: Implications for the evolution of
animal vocalizations. Behav Ecol Sociobiol 3:69–94.
Zahavi A, 1975. Mate selection—a selection for a handicap. J Theor
Biol 53:205–214.
Zahavi A, 1987. The theory of signal selection and some of its implications. Proc Intl Symp Biol Evol; 305–327.