Behavioral Ecology Vol. 12 No. 1: 98-102
© 2001 International Society for Behavioral Ecology
Vireo song repertoires and migratory distance: three sexual selection hypotheses fail to explain the correlation
Nebraska Behavioral Biology Group, Department of Psychology, University of Nebraska—Lincoln, Lincoln, NE 68588-0308, USA
Address correspondence to D. J. Mountjoy, who is now at the Department of Biology, Franklin and Marshall College, PO Box 3003, Lancaster, PA 17604, USA. E-mail: j_mountjoy{at}acad.fandm.edu .
Received 12 January 2000; revised 29 June 2000; accepted 5 July 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Previous studies have found a relationship between migration and the degree of elaboration of sexually selected traits, but investigators have differed in the mechanisms they proposed to account for this association. We examined the relationship between song repertoire size and distance migrated among birds in the genus Vireo. There is a strong positive relationship between migratory distance and repertoire size in this genus, but our data do not support the specific predictions of any of the three proposed mechanisms (the "rapid pairing," "good migrations," and "territory lottery" hypotheses). Migration distance is presumably correlated with other life-history characteristics that influence the development of sexually selected traits.
Key words: bird song, migration, sexual selection, song repertoires, vireos.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In recent years a tremendous research effort documented that sexual selection is a widespread and influential force on the evolution of morphology and behavior, leading to a clearer understanding of its operation. One aspect that remains poorly understood is why there is tremendous variation in the development of sexually selected traits, even among closely related species with superficially similar life histories (Andersson, 1994
).
Mating systems have long been considered to be a likely correlate of
variation in sexually selected traits, as polygamous species are expected to
show greater variation in mating success
(Darwin, 1871;
Kirkpatrick, 1987). The
results of recent comparative studies of birds have been mixed, however; some
studies have shown that particular traits are correlated strongly with
polygyny (plumage conspicuousness: Baker
and Parker, 1979; size dimorphism:
Payne, 1984), and others have
found weak correlations (tail dimorphism:
Winquist and Lemon, 1994), or
no significant correlation at all (size and plumage dimorphism:
Møller, 1986; song
repertoires: Read and Weary,
1992). Höglund
(1989) found no significant
association between size or plumage dimorphism and the occurrence of lekking,
and Trail (1990) pointed out
that about 25% of lek-breeding bird species were monomorphic in size and
plumage, although lek breeding has traditionally been viewed as the most
extreme form of polygyny. Oakes
(1992) did find more frequent
sexual size dimorphism among lekking species (with probability values ranging
from 0.064 to 0.003) when the data were analyzed using Maddison's concentrated
changes test. Two studies have found that variation within the broader
traditional categories of mating systems is important. Møller and
Birkhead (1994) found that
plumage brightness was more closely related to extrapair paternity than to
social mating system, although this correlation was still relatively weak
(r2 =.109). Scott and Clutton-Brock
(1989) found that males were
brighter in waterfowl species that paired frequently and suggested that such
life-history variation among monogamous birds might be one factor obscuring
the expected relationship between polygyny and exaggerated sexually selected
traits.
Another suggested correlate of the development of sexually selected traits
is migratory behavior. There are several explanations for why migration should
be correlated with sexual selection. Catchpole
(1982) suggested that migratory
species seem to have larger song repertoires than resident species (a
relationship confirmed for within-genera comparisons by
Read and Weary, 1992).
Catchpole proposed that migratory species would have less time available for
pairing and breeding and that this would favor rapid pairing and increase
sexual selection pressure in the migratory species relative to residents. We
refer to this idea as the "rapid pairing" hypothesis.
Fitzpatrick's (1994)
"good migrations" hypothesis suggests that if the development of
sexually selected traits is influenced by the suitability of the wintering
grounds (through their effects on the condition of the birds), then females in
migratory species might be able to choose mates with genes that would help
guide their offspring to suitable wintering areas. A third hypothesis is based
on the idea that resident and migratory species may use different strategies
to acquire territories and breeding opportunities. Migratory species are
likely to face a situation where all potential territories are available when
the first individuals begin to return to the breeding grounds, and acquisition
of territories may subject most individuals to intense competition, which will
favor those sexually selected traits used in intrasexual competition. Resident
species, in contrast, cannot rely on a large number of territories being
vacant at any particular time, and success in acquiring a territory may depend
primarily on being the first individual to detect a vacancy and to occupy the
area. This "territory lottery" hypothesis suggests that chance is
more important in acquiring a territory in resident species and that traits
used in intrasexual competition will not be as strongly emphasized. Mate
choice may also be restricted in resident species if year-round pair bonds are
normally maintained and opportunities for pairing usually arise after a member
of an established pair dies. In such systems there will be few potential mates
available for comparison at any time and therefore less incentive to advertise
individual quality. A concept similar to the territory lottery hypothesis is
discussed by Fitzpatrick
(1998) with respect to
intrasexual selection alone.
This article examines the correlation between migratory behavior and song
repertoire size among species of the genus Vireo and tests these
three hypotheses. The vireos are an excellent group for such a study because
there is a great deal of variation in song repertoire size and migratory
behavior in what is in most respects a uniform genus. Repertoires range from 1
to at least 95 song types, and there are resident species in both tropical and
temperate climates as well as short- and long-distance migrants. Vireos are
small (8.5-20 g), primarily insectivorous passerines that live in forested or
scrub habitats. Most species are rather drab in plumage and all are nominally
monogamous, although an isolated case of polygyny and frequent extrapair
fertilizations are known to occur at least in the red-eyed vireo (Vireo
olivaceus; Morton et al.,
1998; Mountjoy,
1997). We tested the rapid pairing hypothesis by comparing
sympatric pairs of vireo species to test the prediction that later-arriving
species should possess larger repertoires. The good migrations and territory
lottery hypotheses are similar in that both predict a dichotomy between
resident species (with low levels of sexual selection and therefore small song
repertoires) and migratory species (with more intense sexual selection and
larger repertoires). We tested these hypotheses by asking whether migratory
species, examined separately, still show a relationship between migration
distance and song repertoire size. No such relationship is predicted by the
good migrations or territory lottery hypotheses, so any correlation within the
subset of migratory species would imply that alternative hypotheses are
required to explain a link between migration and sexual selection.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Estimates of song repertoire size were obtained for 28 of the 31 species recognized by Sibley and Monroe (1990
) in the genus
Vireo (Table 1). We
used published estimates where available, using the maximum values for
individual repertoires when a range was given or when more than one estimate
was available. Repertoire estimates for other species were based on analysis
of commercially available recordings
(Barlow, 1995). We prepared
spectrograms for all songs analyzed using Canary software
(Charif et al., 1995) and
classified songs based on visual assessment. Two species were omitted from
most analyses (V. gilvus and V. swainsonii) because no
precise literature estimates of repertoire size were available, and the
available recordings were too short to make an estimate of the repertoire
sizes, which are apparently quite large (V. gilvus:
Howes-Jones, 1985; the
possibly conspecific V. swainsonii is probably similar). A minimum
limit for the repertoire size of V. gilvus was obtained by analyzing
a small sample of song from one individual. This recording was made at
Burchard Lake, Pawnee County, Nebraska, USA, on 18 July 1996. We prepared
spectrograms for 36 consecutive songs of V. gilvus using Canary
software (Charif et al., 1995)
and assigned them to song types based on visual inspection. The Noronha vireo
(Vireo gracilirostris) was also omitted from all analyses because the
published analysis of its song (Barlow and
Bartolotti, 1988) is believed to be inaccurate
(Olson, 1994).
|
We estimated migration distance for each species as the difference between
the latitude of the northern edge of the breeding range and the northern edge
of the wintering range, based on published maps and range descriptions
(American Ornithologists' Union,
1983; Godfrey,
1986; Howell and Webb,
1995; Scott, 1987;
Sibley and Monroe, 1990).
To control for the lack of independence of related species, we analyzed the
data using the Comparative Analysis by Independent Contrasts (CAIC) package
(Purvis and Rambaut, 1995).
The CAIC software calculates standardized linear contrasts, which represent
differences between two species or groups of species at each node in the
phylogeny. The contrasts are standardized by dividing the raw contrasts by
their expected standard deviation in order to eliminate heterogeneity of
variance. The standardized contrasts were then analyzed using standard linear
regression with the regression line forced through the origin. The phylogeny
used in these analyses (Figure
1) was based on the single most parsimonious tree found by Murray
et al. (1994) using cytochrome
b sequence variation. Murray et al. analyzed only 11 species; the other
species in the genus were added to this framework based on super-species
groupings and traditional taxonomic views regarding relationships within the
genus (American Ornithologists' Union,
1983; Sibley and Monroe,
1990) and some additional molecular evidence
(Johnson, 1995). For the
purpose of analysis in CAIC, we assumed equal branch lengths in the phylogeny
(equivalent to assuming a punctuational model of evolution). Distance migrated
and repertoire size are continuous variables, so we used the recommended
algorithm in CAIC ("Crunch") for continuous variables.
|
Arrival dates on the breeding grounds for species which regularly nest north of Mexico were compared by identifying independent pairs of species within the phylogeny, with the constraint that both members of a pair had to breed within the same geographic region. Five pairs were identified for which arrival date information could be obtained from regional annotated checklists.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The repertoire size of vireos is well correlated with the distance migrated by the species (r =.741, n = 18 contrasts, p =.0003; Figure 2). This correlation is not a result of any assumptions in the traditional classification incorporated in the phylogeny used here; the correlation remains highly significant when the analysis is restricted to the species represented in a phylogeny based on molecular data only (r =.842, n = 8 contrasts, p =.0044). The correlation also does not depend on the repertoire estimates derived from commercial recordings; the correlation is still significant if the analysis is restricted to species for which repertoire estimates are available in the literature (r =.852, n = 9 contrasts, p =.0017).
|
The correlation between repertoire size and migration distance does not appear to be due to correlations between repertoire size and either the northernmost breeding latitude or the northernmost wintering latitude. Repertoire size and northernmost wintering latitude are uncorrelated (r = -.028, n = 18 contrasts, p =.908), and although the correlation between repertoire size and northernmost breeding latitude is significant (r =.598, n = 18 contrasts, p =.0069) it is not as strong as the correlation with migration distance.
Catchpole's (1982) rapid
pairing hypothesis predicts that species that return to the breeding grounds
later should have larger repertoires. We compared the arrival dates and
repertoire sizes in five species pairs. Data were available from more than one
geographic area for four of these pairs. V. gilvus is included in
this analysis based on a lower limit for its repertoire obtained from a
Nebraska bird that sang 36 song types in 36 consecutive songs. The prediction
that later-arriving species should have larger repertoires is not supported
among the five species pairs analyzed; in only one pair does the
later-arriving species have the larger repertoire, while in the other four
pairs the earlier-arriving species has the larger repertoire (sign test:
p >.8; Table
2).
|
Fitzpatrick's (1994) good
migrations hypothesis and the territory lottery hypothesis both predict that
sexual selection should be more intense in migratory species compared to
resident species, but they predict that all migratory species should
experience similar sexual selection pressure regardless of the distance
migrated. To test this prediction we restricted the analysis to those species
that migrate. Even among all species that migrate, there is a significant
correlation between song repertoire size and the distance migrated (r
=.812, n = 10 contrasts, p <.005).
Recently, concerns have been expressed about problems in the reconstruction
of ancestral characters (Cunningham et
al., 1998) and other problems that may affect comparative analyses
that use independent contrasts. It has been suggested that species-level
analyses may, in some cases, be as appropriate a method for analysis of
comparative data as is the use of independent contrasts
(Losos, 1999;
Ricklefs and Starck, 1996).
Although we believe that the use of independent contrasts or other methods to
control for phylogenetic effects is important, it is worth noting that our
results do not depend on the use of such analyses. The correlation between
migration and repertoire size is significant when analyzed as a simple linear
regression using species-level data (r =.589, n = 31
species, p =.0005). Similarly, the correlation remains significant
when only migratory species are considered (r =.580, n = 14
species, p =.030).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
There is a robust correlation between repertoire size and distance migrated in vireos. This correlation explains almost 55% of the substantial variation in repertoire size in this genus, which seems to be a higher proportion of variation accounted for by a single variable than has been found in any other comparative study of a sexually selected trait. However, none of the three proposed hypotheses that might explain this correlation appears to be supported by the data.
Catchpole's (1982) rapid
pairing hypothesis is the only one of the three hypotheses considered here to
have been previously applied to interspecific variation in song repertoires.
Read and Weary (1992)
suggested that their finding of larger repertoires in migratory species than
in residents supported the rapid pairing hypothesis, but they did not consider
any alternative explanations for an association between migration and
repertoire size. Although only five pairs of species were tested, the fact
that the large repertoire species returned earlier in four of the five pairs,
counter to the prediction of the hypothesis, seems to argue strongly against
this hypothesis.
Fitzpatrick's (1994) good
migrations hypothesis was originally proposed to explain variation in plumage
brightness, but it would seem to apply as well to other sexually selected
traits such as bird song that may be affected by the condition of individuals
as a result of the suitability of their wintering grounds. Although there is
some evidence that condition affects the repertoire size of individual birds
(Mountjoy and Lemon, 1996),
there is no information on whether song repertoires are particularly sensitive
to condition in winter or any other season. Fitzpatrick
(1994) did use comparisons
between more and less migratory species when testing the good migrations
hypothesis, but it would seem that this hypothesis does not make any
predictions regarding the strength of selection between short- and
long-distance migrants. Females in all migratory species would benefit from
having a mechanism to assess the suitability of a male's wintering grounds, if
obtaining good migratory genes is an important fitness benefit as the
hypothesis suggests. The strong correlation between migration distance and
repertoire size when only migrants are considered is thus not predicted by the
good migrations hypothesis and suggests that some other explanation is
required.
The territory lottery hypothesis is, to our knowledge, a novel hypothesis for explaining interspecific variation in the elaboration of sexually selected traits. Like the good migrations hypothesis, it predicts a dichotomy between resident and migratory species and does not predict any difference between short- and long-distance migrants. It is therefore also rejected on the basis of the correlation between migratory distance and repertoire size among migratory species.
Having rejected all three of the hypotheses available that predict an
association between migration and sexual selection, there is a need to
generate additional hypotheses that can provide an explanation for these data.
It would also be worth-while to test whether current hypotheses can explain
the results of previous studies
(Fitzpatrick, 1994;
Read and Weary, 1992) that
have claimed to find an association between migration and sexual selection. It
is unlikely that migration distance itself is in any sense a causative factor.
Rather, it seems that there must be some other life-history correlates of
migration that promote or allow the exaggeration of sexually selected
traits.
|
ACKNOWLEDGEMENTS |
|---|
We thank four anonymous referees for helpful comments. This work was supported by the Nebraska Behavioral Biology Group through National Science Foundation grant OSR 9255225.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
American Ornithologists' Union, 1983. Check-list of North American Birds, 6th ed. Washington, DC: American Ornithologists' Union.
Andersson MB, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Baker RR, Parker GA, 1979. The evolution of bird coloration. Phil Trans R Soc Lond B 287: 63-130.
Barlow JC, 1995. Songs of the vireos and their allies, 2nd ed. Gainesville, Florida: ARA Records.
Barlow JC, Bortolotti GR, 1988. Adaptive divergence in morphology and behavior in some New World island birds, with special reference to Vireo altiloquus. In: Acta XIX Congressus Internationalis Ornitologici (Ouellet H, ed). Ottawa: University of Ottawa Press; 1535-1549.
Borror DJ, 1972. Yellow-green vireo in Arizona, with notes on vireo songs. Condor 74: 80-86.
Borror DJ, 1987. Song in the white-eyed vireo. Wilson Bull 99: 377-397.
Brandt H, 1951 Arizona and its bird life. Cleveland, Ohio: Bird Research Foundation.
Catchpole CK, 1982. The evolution of bird sounds in relation to mating and spacing behavior. In: Acoustic communication in birds, vol. 1 (Kroodsma DE, Miller EH eds). New York: Academic Press; 297-319.
Charif RA, Mitchell S, Clark CW, 1995. Canary 1.2 user's manual. Ithaca, New York: Cornell Laboratory of Ornithology.
Cunningham CW, Omland KE, Oakley TH, 1998. Reconstructing ancestral character states: a critical reappraisal. Trends Ecol Evol 13: 361-366.
Darwin C, 1871. The descent of man, and selection in relation to sex. London: J. Murray.
Davis JN, 1995. Hutton's vireo (Vireo huttoni). In: The Birds of North America, no. 189 (A Poole, F Gill, eds). Philadelphia, Pennsylvania, and Washington, DC: The Academy of Natural Sciences and American Ornithologists' Union.
Fitzpatrick S, 1994. Colourful migratory birds: evidence for a mechanism other than parasite resistance for the maintenance of `good genes' sexual selection. Proc R Soc Lond B 257: 155-160.
Fitzpatrick S, 1998. Intraspecific variation in wing length and male plumage coloration with migratory behaviour in continental and island populations. J Avian Biol 29: 248-256.
Godard R, 1993. Red-eyed vireos have difficulty recognizing individual neighbors' songs. Auk 110: 857-862.
Godfrey WE, 1986. The birds of Canada, revised ed. Ottawa: National Museum of Natural Sciences.
Grinnell J, Miller AH, 1944. The distribution of the birds of California. Pacific Coast Avifauna no. 27. Berkeley, California: Cooper Ornithological Club.
Höglund J, 1989. Size and plumage dimorphism in lek-breeding birds: a comparative analysis. Am Nat 134: 72-87.
Howell SNG, Webb S, 1995. The birds of Mexico and northern Central America. Oxford: Oxford University Press.
Howes-Jones D, 1985. The complex song of the warbling vireo. Can J Zool 63: 2756-2766.
James RD, 1981. Factors affecting variation in the primary song of North American solitary vireos (Aves: Vireonidae). Can J Zool 59: 2001-2009.
James RD, 1984. Structure, frequency of usage, and apparent learning in the primary song of the yellow-throated vireo, with comparative notes on solitary vireos (Aves: Vireonidae). Can J Zool 62: 468-472.
James RD, 1991. Annotated checklist of the birds of Ontario. Toronto: Royal Ontario Museum.
Janssen RB, 1987. Birds in Minnesota. Minneapolis: University of Minnesota Press.
Johnson NK, 1995. Speciation in vireos. I. Macrogeographic patterns of allozymic variation in the Vireo solitarius complex in the contiguous United States. Condor 97: 903-919.
Kirkpatrick M, 1987. Sexual selection by female choice in polygynous animals. Annu Rev Ecol Syst 18: 43-70.[ISI]
Leberman RC, 1976. The birds of the Ligonier Valley. Pittsburgh: Carnegie Museum of Natural History.
Losos JB, 1999. Uncertainty in the reconstruction of ancestral character states and limitations on the use of phylogenetic comparative methods. Anim Behav 58: 1319-1324.[ISI][Medline]
Møller AP, 1986. Mating systems among European passerines: a review. Ibis 128: 234-250.
Møller AP, Birkhead T, 1994. The evolution of plumage brightness in birds is related to extra-pair paternity. Evolution 48: 1089-1100.[ISI]
Morton ES, Stutchbury BJM, Howlett JS, Piper WH, 1998.
Genetic monogamy in blue-headed vireos and a comparison with a sympatric vireo
with extrapair paternity. Behav Ecol 9:
515-524.
Mountjoy DJ, 1997. A probable case of polyterritorial polygyny in the red-eyed vireo, Vireo olivaceus. Can Field-Nat 111: 628-630.
Mountjoy DJ, Lemon RE, 1996. Female choice for complex song in the European starling: a field experiment. Behav Ecol Sociobiol 38: 65-71.
Murray BW, McGillivray WB, Barlow JC, Beech RN, Strobeck C, 1994. The use of cytochrome B sequence variation in estimation of phylogeny in the Vireonidae. Condor 96: 1037-1054.
Oakes EJ, 1992. Lekking and the evolution of sexual dimorphism in birds: comparative approaches. Am Nat 140: 665-684.
Olson SL, 1994. The endemic vireo of Fernando de Noronha (Vireo gracilirostris). Wilson Bull 106: 1-17.
Payne RB, 1984. Sexual selection, lek and arena behavior, and sexual size dimorphism in birds. Ornithol Monogr 33: 1-53.
Pulich WM, 1988. The birds of North Central Texas. College Station: Texas A & M University Press.
Purvis A, Rambaut A, 1995. Comparative analysis by
independent contrasts (CAIC): an Apple Macintosh application for analysing
comparative data. Comput Appl Biosci 11:
247-251.
Read AF, Weary DM, 1992. The evolution of bird song: comparative analyses. Phil Trans R Soc Lond B 338: 165-187.
Ricklefs RE, Starck JM, 1996. Applications of phylogenetically independent contrasts: a mixed progress report. Oikos 77: 167-172.
Robbins MB, Easterla DA, 1992. Birds of Missouri. Columbia: University of Missouri Press.
Robbins SD Jr, 1991. Wisconsin birdlife. Madison: The University of Wisconsin Press.
Robinson JC, 1990. An annotated checklist of the birds of Tennessee. Knoxville: The University of Tennessee Press.
Scott SL, 1987. Field guide to the birds of North America, 2nd ed. Washington, DC: National Geographic Society.
Scott DK, Clutton-Brock TH, 1989. Mating systems, parasites and plumage dimorphism in waterfowl. Behav Ecol Sociobiol 26: 261-273.
Sibley CG, Monroe, BL, Jr, 1990. Distribution and taxonomy of the birds of the World. New Haven, Connecticut: Yale University Press.
Stevenson HM, Anderson BH, 1994. The birdlife of Florida. Gainesville: University Press of Florida.
Sutton GM, 1967. Oklahoma birds. Norman: University of Oklahoma Press.
Trail PW, 1990. Why should lek-breeders be monomorphic? Evolution 44: 1837-1852.
Wauer RH, 1973. Birds of Big Bend National Park and vicinity. Austin: University of Texas Press.
Wayne AT, 1910. Birds of South Carolina. Charleston, South Carolina: Daggett Printing Co.
Winquist T, Lemon RE, 1994. Sexual selection and exaggerated male tail length in birds. Am Nat 143: 95-116.
Zeranski JD, Baptist TR, 1990. Connecticut Birds. Hanover, New Hampshire: University Press of New England.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||