Printer Friendly

Evolutionary associations of brood parasitic finches (Vidua) and their host species: analyses of mitochondrial DNA restriction sites.

In avian interspecific brood parasitism, a female of the parasitic species does not build her own nest, but lays eggs in the nests of other species, and the parasite parents neither feed nor care for their offspring. Instead, the "host" species adults raise the parasite young along with their own (Payne 1977a, 1997, 1998; Rothstein 1990). In general, birds are known for their parental care with a family bond that may last long after young are independent (Gill 1994), and parental investment of time and energy can be large. Food and care provided by host parents to parasite offspring is care denied their own young. Parasitism could thus have a severe negative impact on reproductive success of the host parents. Selection for the ability to avoid such wasted parental care effort by hosts is expected to be strong, as is selection in parasites for the ability to deceive.

Some of the known avian brood parasitic associations are specific, with a single parasite species utilizing a single host species, while others are general, such that a single parasite species utilizes many host species (Payne 1977a, 1997; Rothstein 1990). Evolutionary associations of species-specific parasites and their hosts may originate through various processes, including (1) cospeciation of the parasites with their hosts, and (2) migration and colonization by the parasites onto new species of hosts. Which mechanism has dominated the history of a parasite/host association is of inherent interest to students of such associations as well as for those interested in mechanisms of evolutionary diversification.

In this paper, we investigate the origins of species-specific parasite-host associations in a group of African finches. The parasites in this study are in the genus Vidua, and their hosts include various estrildid finch genera found in the Old World. The Vidua are the most species-specific of brood-parasitic birds. Each species normally is associated with one host species of estrildid finch; only two of the 19 species regularly use more than one host (Payne 1997, in press). The Vidua finches are considered most closely related to the Estrildidae, which includes the host species of the viduas (Sibley and Ahlquist 1990).

The parasite-host associations of the Vidua finches are known through field studies of behavior, song mimicry, and mouth mimicry. In most species, each male Vidua mimics the songs of only one kind of estrildid finch. For each of these species of Vidua whose behavior has been determined in the field, the finch whose song is mimicked is the host species that raises the young Vidua (C. J. Skead 1957; D. M. Skead 1975; Friedmann 1960; Nicolai 1964, 1973; M.-Y. Morel 1973; Payne 1973a, 1977a,b, 1982, 1985a,b, 1990; Payne and Payne 1994). Vidua nestlings typically mimic the mouth colors and pattern of nestlings of their host species and the two kinds of young often are reared together in the nest (Nicolai 1964; M.-Y. Morel 1973; Payne 1973a,b, 1982; Payne and Payne 1994). However, a few Vidua do not mimic their host mouths and these Vidua may be only recently differentiated species (Payne and Payne 1994). Nestling mimicry may allow the young Vidua to escape recognition and negative discrimination by their hosts, or to at least be cared for by the foster parents along with the host young. Nicolai (1964, 1974) suggested that it was necessary for a Vidua parasite population to match the nestling colors of its host due to selection over many generations and across speciation events in a process of cospeciation and coevolution. Nevertheless, coevolution can occur without cospeciation and brood parasites may have independently evolved their mimicry. This could have occurred if they colonized their hosts after host speciation and with each colonization were selected to match the mouth colors of the host nestlings. For example, parasitic cuckoos have independently evolved egg colors that match their hosts (Southern 1954; Higuchi and Sato 1984; Davies et al. 1989; Rothstein 1990), mimetic insects have independently evolved colors that match other aposematic insects (Plowright and Owen 1980; Brower 1996), and specialist herbivorous insects have switched from one host plant to another (Singer et al. 1992; Ronquist 1994; Funk et al. 1995; Radtkey and Singer 1995). Some hosts have also adapted to parasitism independently of the phylogeny of their parasites (Moore and Gotelli 1996).

Variation in the songs of the brood parasitic Vidua suggests the conditions that could lead to a switch of hosts. The young parasites learn the songs of their host species while in the care of their foster parents, and the adult parasites mimic the songs of their hosts (Nicolai 1964, 1973; Payne 1973a, 1990). A few males give the song of an alternate host species, rather than that of the normal host species. This implies that young in natural conditions are sometimes raised by an alternate host species and the potential exists to establish new breeding populations (Payne et al. 1993; Payne and Payne 1994, 1995).

The focus of this paper is to determine whether diversification within Vidua and evolution of their parasitic relationships with the estrildid finch hosts has occurred mainly through a process of host-parasite cospeciation or through independent colonizations of host species by their brood parasites.

Predictions of the Two Models

Mitter and Brooks (1983) describe two models of evolutionary association of parasites and hosts: cospeciation and independent colonization. For the brood parasitic finches, we develop each model with exclusive and strong predictions that allow its rejection, and thus a test by strong inference (Platt 1964; Hilborn and Mangel 1997).

Prediction set (1): Cospeciation Model. - (a) If the parasites and their host species are associated through cospeciation, then their evolutionary branching diagrams will be congruent because the host and parasite species diverged in parallel. (b) Brood parasites that utilize conspecific host populations in different geographic regions will be more closely related to each other than to brood parasites that parasitize different species of host in the same geographic region. (c) Pairwise genetic distances between related species of parasites will be similar in magnitude to pairwise genetic distances between their host species. (d) Pairwise genetic distances between species of parasites that each use a different host species within a region will be greater than genetic distances between populations that live in different regions but parasitize the same species of host.

Prediction set (2): Independent Colonization Model. - (a) If parasites and their host species are associated through colonization, then the evolutionary branching diagrams of parasite and host species will not be congruent. (b) The brood parasites that live in one geographic region and parasitize different species of hosts will be more closely related to each other than to brood parasites that parasitize the same species of host in different regions. (c) Interspecific pairwise genetic distances will on average be smaller between the parasite species than between their host species (where a parasite lineage has switched from one host species to another). (d) Pairwise genetic distances between species of parasites that each use a different host species within a geographic region will be less than between brood parasites that live in different regions but parasitize the same host. This would occur where the conspecific host populations were colonized independently by different lineages of brood parasites.

Although molecular evolutionary rate heterogeneity independent of time since divergence could potentially also explain differences in genetic distance between taxa under a model of cospeciation, the comparisons of Vidua with their host species are unlikely to be dramatically influenced by this process. Host and parasite groups in this system are relatively closely related and similar in all the features typically associated with variation in molecular rates (e.g., generation times, effective population sizes, body size as it correlates with metabolic rate; Goodwin 1982; Payne 1997). Genetic distance comparisons are thus appropriate in a test of the origin of parasite-host associations in this group of birds.

Here we test the two models of evolutionary association between the brood parasites and their host species by means of phylogenetic analyses of mapped restriction sites of the mitochondrial genome and by comparison of interspecific genetic divergences in the hosts and parasites. The models were tested in two species groups of Vidua finches, the indigobirds and the paradise whydahs, and in their host species.

METHODS

Species and Populations

Parasitic indigobirds (V. chalybeata and others), most paradise whydahs, and their host species were collected by RBP and Laura Payne in Zimbabwe and Malawi, and by RBP, Laura Payne, and NKK in Cameroon (Payne et al. 1992, 1993; Payne and Payne 1994). As many as five species of indigobirds live in a single area with little or no interbreeding between them; in Malawi four species were taken within 50 km of each other. In addition to these field-collected birds, live finches were obtained from importers and avicultural sources. We refer to the area of Zimbabwe and Malawi as "S-C Africa," and Cameroon and the source area of captives for other taxa known to be from western Africa as "western Africa."

In the field, each male indigobird was tape-recorded to determine its mimicry songs. It was then captured when it came to a mist net in response to a playback of its song; birds that did not come to the net were collected with a shotgun. Paradise whydabs, the species group that parasitizes the pytilias, are not readily tape-recorded and caught in the field, so we shot male V. paradisaea and V. obtusa in Malawi. We also obtained captive birds of uncertain geographic provenance and recorded their mimicry songs before sacrificing them to obtain tissue samples: V. paradisaea reportedly from Tanzania; V. orientalis aucupum, which occurs in sub-Saharan West Africa; and V. interjecta (including a female, Payne 1991; no songs were heard from female Vidua). Museum study skins were prepared for all birds that were used in the molecular genetic analysis. Specimens were identified by comparison with other museum collections (American Museum of Natural History, AMNH; Field Museum of Natural History, FMNH; British Museum of Natural History, BMNH; National Museum of Natural History, USNM; and Museum National d'Histoire Naturelie, MNHN, Paris). Specimens are in the University of Michigan Museum of Zoology (UMMZ).
TABLE 1. Brood parasites/song mimics (Vidua species) and their
corresponding host species.

Brood parasite/song mimic                   Host

Indigobirds                Firefinches and twinspots

Vidua chalybeata           Lagonosticta senegala
V. funerea nigerrima       L. rubricata
V. purpurascens            L. rhodopareia
V. maryae(*)               L. sp. nov.(*)
V. larvaticola             L. larvata
V. wilsoni                 L. rufopicta
V. camerunensis            L. rubricata
                           L. rara
                           Euschistospiza dybowskii(*)
                           Clytospiza monteiri
V. codringtoni             Hypargos niveoguttatus
V. raricola                Amandava subtiara(*)
V. nigeriae                Ortygospiza atricollis(*)

Paradise whydahs           Pytilias

V. interjecta              Pytilia phoenicoptera
V. togoensis(*)            P. hypogrammica
V. obtusa                  P. afra
V. orientalis aucupum      P. melba citerior
V. paradisaea              P. m. percivali
V. paradisaea              P. m. grotei

* Species lacking in phylogenetic and genetic distance analyses.


To indicate the associations of brood parasites and their host species, we list each species of Vidua and its host (Table 1), and the origin of specimens used in the genetic analyses (Appendix 1). Additional information on variation within some species is provided below.

Village indigobirds, V. chalybeata, show morphological variation (e.g., bill color) across geography. This species is widespread across sub-Saharan Africa and mimics the song of red-billed firefinch, Lagonosticta senegala, throughout its range (G. R. Morel 1959; Nicolai 1964; M.-Y. Morel 1973; Payne 1973a, 1990). One individual included in this study (#A115, see Appendix 1) was recorded singing the song of an alternate host, L. rubricata.

Vidua codringtoni normally mimics the song of Hypargos niveoguttatus. One individual (#A08, see Appendix 1) was recorded singing the song of an alternate host, L. rubricata.

Vidua camerunensis uses four host species (L. rubricata, L. rara, Clytospiza monteiri, and Euschistospiza dybowskii; Payne and Payne 1994, 1995). In addition to males whose songs we taped, we collected in Cameroon a juvenile indigobird at a call-site of an adult V. camerunensis that mimicked the song of the brown twinspot, Clytospiza monteiri. In Cameroon we found no wild male indigobirds that mimicked black-bellied firefinch, L. rara (these indigobirds occur in Ghana; Payne 1982; Payne and Payne 1994) or that mimicked Dybowski's twinspot, Euschistospiza dybowskii (these indigobirds occur in Sierra Leone; Payne and Payne 1995).

Molecular Genetic Analyses

After birds were recorded and sacrificed, their liver, lung, heart, and pectoral muscle tissues were removed and frozen in liquid nitrogen or maintained at ambient temperature in 0.25 M EDTA/20% DMSO buffer (Seutin et al. 1991). The frozen tissues were later stored at - 80 [degrees] C in ultracold freezers.

Isolation, purification, and restriction endonuclease digestion of mitochondrial DNA (mtDNA) followed methods outlined in Lansman et al. (1981), Dowling et al. (1990), and Klein and Brown (1994). Purified mtDNA was isolated from liver, heart, or pectoral muscle. The amount of tissue used ranged from 0.03 g to 0.3 g; the homogenization buffer consisted of one part 0.5 M sucrose in TE to five parts 200 mM EDTA, 10 mM NaCl, and 10 mM Tris.

Purified mtDNA was digested with 17 restriction endonucleases characterized by six-base recognition sequences: ApaI, BamHI, BclI, BglII, BstEII, ClaI, DraI, EagI, EcoRI, HindIII, KpnI, NcoI, NdeI, NheI, PvuII, SalI, and XbaI. DNA was digested to completion (2-14 h) with an excess of enzyme under conditions recommended by the suppliers (Boehringer-Mannheim and New England Biolabs). Fragments were end-labeled with 32P, run in 1X TBE buffer on both agarose (0.8-1.2%) and polyacrylamide (3.5-5.0%) vertical gels, and visualized by autoradiography (Brown 1980). A size standard of lambda phage DNA digested with HindIII mixed with [Phi]X174 phage DNA digested with HaeIII was included on each gel. Fragment sizes were estimated from calibration curves plotted from log fragment size versus distance migrated of size-standard fragments. The mean size estimate of the mtDNA molecule for all species (calculated from the sizes estimated from the mtDNA fragments generated by each enzyme) was 17.0 kb. Size determination of fragment lengths and localization of the restriction sites is estimated to be accurate within 40-150 base pairs (Nei 1987; Dowling et al. 1990, 1996).

Cleavage sites for each taxon were independently mapped (Appendix 2) (but with only two independent maps representing indigobirds due to the high genetic similarity among all indigobirds) using double and triple digests (Brown and Vinograd 1974; Dowling et al. 1990, 1996); 27 independent cleavage maps were generated. Restriction site homologies and restriction enzyme cleavage site losses (among individuals within a species or among indigobird individuals) relative to the mapped sites were inferred from fragment pattern comparisons with the mapped haplotypes (Vawter and Brown 1986). The positions of restriction enzyme cleavage sites that were gained relative to the mapped haplotypes were determined with additional double digests. Additional double digests were also used to verify positions of synapomorphic restriction sites in unmapped individuals.

The mapping strategy employed was to use initial double digests of all enzymes with BglII and with ClaI to align maps to two common restriction sites (BgllI site A, ClaI site A) (Klein and Brown 1994). All mtDNA samples contained one to three BglII sites and one to three ClaI sites. ClaI site A appears to be conserved in finches and other songbirds examined in this lab, including the New World honeycreepers and warblers Coereba, Dendroica, Setophaga, Geothlypis, Parula, and Basileuterus (Klein and Brown 1994; Seutin et al. 1994), as well as in chicken, Gallus gallus, where the entire mitochondrial genome has been sequenced (Desjardins and Morais 1990). Homology of the ClaI site A in finches was confirmed by its constant position relative to the two Sac II sites that mark a 1.72-kb fragment and are conserved among vertebrate mtDNAs (Brown 1985; Carr et al. 1987; Moritz et al. 1987; Desjardins and Morais 1990). One SacII site is located within the 12s ribosomal RNA (rRNA) gene, the other is in the 16s rRNA gene (Hixson and Brown 1986; Desjardins and Morais 1990). ClaI cleaved this 1.72-kb fragment into 1.4-kb and 0.32-kb fragments in all but one of the finches examined (this ClaI site A was absent in one L. rara (#o36), for which the map was aligned with other maps using BgllI site A and the two Sac II sites). A ClaI site occurs in the chicken mtDNA sequence between the SacII sites, 332 bp (within the range of measurement error of the 320-bp fragment determined in the finch mtDNAs) from the Sac II site in the 12s rRNA gene. Additional double and triple digests with other enzyme combinations were then used to determine more precisely the map positions of sites not determined by double digests with BglII and ClaI. Restriction site data were incomplete in some cases (Appendix 3).

In total, 200 restriction sites were mapped in the finches surveyed (Appendix 2), including captives not used in further analyses. This sample accounts for approximately 7.1% of the 17-kb mitochondrial genome.

Analyses of Phylogenetic Relationships and Genetic Distances

The phylogenetic estimation program PAUP version 3.1.1 (Swofford 1993) was used to generate hypotheses of relationship from the matrix of restriction site presence/absence. Genetic distances were calculated with [PAUP.sup.*] version 4.0d49. For phylogeny estimation, characters were treated as unordered. Haplotypes defined as unique associations of restriction sites were the units of analysis (taxa) except that an additional analysis of relationships within Vidua was done treating individual indigobirds as taxa.

Phylogenetic Analysis of Relationships within Vidua. - To test the relationships within the species complexes of the indigobirds and the paradise whydahs, and whether these two species groups each were monophyletic, we estimated the phylogenetic relationships among all Vidua. We used the entire species assemblage of Vidua as available (we lacked two of the 19 species: an indigobird V. maryae and a paradise whydah V. togoensis).

The following estrildid finches were included as outgroups for rooting the phylogenetic tree: (1) cut-throat finch, Amadina fasciata, a species that in another molecular genetics study (Kakizawa and Watada 1985) was determined to represent the basal split of the set of African estrildid finches that includes all host species of the viduas; (2) orange-winged pytilia, Pytilia afra, a host of the paradise whydahs; (3) redbilled firefinch, Lagonosticta senegala, a host of the indigobirds; and (4) green twinspot, Mandingoa nitidula, which is not known to be a host.

We carried out heuristic searches using the random addition sequence and tree bisection-reconnection branch swapping options. Because there were so many taxa, we completed 30,000 heuristic searches in which only one tree of 273 steps or less was saved from each search. This search strategy allowed us to visit many islands of trees (Swofford 1993, p. 34). Heuristic (50 replicates) and exhaustive searches were also done for the paradise whydahs using V. macroura as the outgroup. This resulted in a nearly identical topology as that generated in the heuristic search that included all Vidua.

Phylogenetic Analysis of Relationships within the Host Groups. - We completed 50 heuristic searches for each host group using the tree bisection-reconnection and random addition sequence options. Mandingoa nitidula was included as the outgroup for the analysis of relationships within the firefinch-twinspot host group, and Amadina fasciata was included as the outgroup in the analysis of relationships within the Pytilia host group. We also completed an exhaustive search for the Pytilia analysis, and a branch-and-bound search for the firefinches-twinspots. The same sets of trees were found as in the heuristic searches.

Because large amounts of missing data can yield many equally most-parsimonious trees, we repeated the analysis excluding those taxa with missing data for more than one enzyme. For the Vidua analysis this also resulted in a large number of equally most-parsimonious trees and essentially the same consensus tree as when all taxa were included. For the firefinch-twinspot analysis, 16 shortest trees were generated and the consensus of these showed the same relationships among the remaining taxa as did the consensus of shortest trees generated when all taxa were included.

Statistical Tests of Differences between the Shortest Vidua Tree and Constrained Trees. - Three types of constraint analyses were done with heuristic searches, all using the complete Vidua dataset: (1) paradise whydah relationships constrained to match the topology of their Pytilia hosts; (2) each individual indigobird included as a taxon on the Vidua tree and constrained such that each species in S-C Africa was monophyletic; and (3) each individual indigobird included as a taxon on the Vidua tree and constrained such that each species in Cameroon and each song form of V. camerunensis was monophyletic. Differences between the shortest Vidua tree and the constrained trees were tested with the Wilcoxon signed-rank test (Templeton 1983). The analysis was conducted as a one-tailed test and the test statistic used was T-, the sum of the negative ranks. Tree number one from each search was arbitrarily chosen as the one to use in the statistical comparisons.

Analysis of Genetic Distances. - The genetic distance between pairs of haplotypes was estimated from the proportion of shared restriction sites, as shown in the matrix of adjusted mean pairwise differences in the PAUP analyses. This adjusted mean difference value excluded sites where presence or absence was not determined for both members of the pain Genetic distances in terms of restriction sites that are shared between species were estimated from the mean distances of all haplotypes identified in each of the two species compared (mean of the values for all haplotypes for both species; Table 2). [TABULAR DATA FOR TABLE 2 OMITTED] Because some species had more than one haplotype, and some haplotypes were shared among species, we adjusted the mean between-species variation by the mean genetic distance within a species, as indicated by the two-parameter model of Nei (1987, p. 223). Mean net interspecific distances were calculated from all pairwise interspecific comparisons of haplotypes within each parasite group and each host group, where each was compared with all other species in the corresponding group. For the indigobirds and their hosts the comparisons were made only within the same geographic region of Africa. The distance estimates included pairwise [d.sub.i,j] = 0 where haplotypes did not differ between two species. Because of the small sample (most haplotypes were represented by only one individual per species, though haplotypes were often shared among species), we did not adjust the distances between species for the frequency of haplotypes within each species, but rather we weighted each haplotype equally in the estimate of genetic distances within and between species. We did not estimate an error term (Nei and Tajima 1983; Nei 1987) for distances between species, as sample sizes of individuals within a species were small.

Genetic distances in terms of nucleotides were estimated by dividing the restriction site distance generated in PAUP version 4.0d49 by the number of bases involved in each restriction enzyme (n = 6 for all restriction enzymes in the survey). This estimate allows a transformation of the restriction site distance to an estimate of nucleotide sequence distance (Nei 1987). The estimate assumes a single nucleotide difference when a restriction site is gained or lost from an ancestral condition, and we restricted the analysis to birds that are thought to be closely related to avoid the complication of multiple changes within a site (Nei 1987; Dowling et al. 1996). As the likelihood of multiple substitutions of nucleotides within a site is higher within the species groups that are less closely related and have higher interspecific genetic distances, the values of estimated genetic distance will underestimate the distances between the host species, which were greater than those between the brood parasite species.

RESULTS

Phylogenetic Relationships within the Vidua Finches

The 12,675 shortest trees of 270 steps each are summarized in the strict consensus tree [ILLUSTRATION FOR FIGURE 1 OMITTED], which illustrates that Vidua are monophyletic with respect to the estrildids included as outgroups and are thus more closely related to each other than any is to the corresponding host species groups. The indigobirds and the paradise whydahs also each comprise a monophyletic group within the Vidua assemblage. The demonstration of monophyly for each of the two species groups of interest allows us to test the relationships within each group. However, examination of branching diagrams depicting relationships within the paradise whydahs and the indigobirds reveals a general lack of species monophyly of mtDNA lineages, especially in the indigobirds [ILLUSTRATION FOR FIGURE 2, 3 OMITTED].

Phylogenetic Comparisons of Brood Parasite Species and Their Hosts

The branching diagrams of the pytilias and of the paradise whydahs are not congruent (Fig. 2). The whydahs, V. orientalis, that are associated with melba finches in western Africa are not most closely related to the whydahs, V. paradisaea, that are associated with melba finches in eastern and southern Africa. Each is instead more closely associated with another species of paradise whydah in the same geographic region, V. orientalis with V. interjecta, and V. paradisaea with V. obtusa. In contrast, the melba finches, Pytilia melba, of southern and eastern Africa (P.m. grotei, P. m. percivali hosts of V. paradisaea) and of western Africa (P. m. citerior hosts of V. orientalis) are each other's closest relatives, and the geographic replacements P. afra in S-C Africa and the two P. phoenicoptera and P. hypogrammica in western Africa are sister taxa. The differences between a shortest Vidua tree (270 steps) and one of the trees constrained so that whydah branching patterns matched those of their hosts (283 steps) were statistically significant (n = 19, T- = -30.0, p [less than] 0.005). In addition to the incongruence of sister group relationships in the paradise whydahs and their hosts, the whydahs V. orientalis and V. interjecta do not have mutually exclusive mitochondrial lineages. Also, in the analysis that included all Vidua and outgroup species the lineages of V. paradisaea and V. obtusa are not fully resolved to species. The phylogenetic estimates thus indicate that the whydahs have not cospeciated with their host species.

The trees for indigobirds and their host species are even more compelling in their lack of evidence for cospeciation of hosts and brood parasites [ILLUSTRATION FOR FIGURE 3 OMITTED]. For the indigobirds, the associations of brood parasites are with geographic regions rather than with their host species. It is also not possible to differentiate the morphologically distinct indigobird species by their restriction site profiles, even though they differ consistently in plumage (except V. funerea and V. purpurascens), song, and mouth colors of the young (Payne 1973a; Payne et al. 1992, 1993; Payne and Payne 1994). The inability to distinguish indigobird species based on mtDNA haplotypes is due to the extreme similarity of haplotypes (mean difference between haplotypes within a geographic area was gain/loss of two restriction sites), and to a nonhierarchical pattern of restriction site gains and losses.

Within the Vidua indigobirds, two main mtDNA clades are apparent, one for the four species in Malawi and Zimbabwe (S-C Africa) and one for the four species (excluding V. chalybeata) in Cameroon. Within each of these regions, the mitochondrial haplotypes do not separate by species of Vidua. Instead, some restriction site haplotypes are shared across species. In S-C Africa the 12 haplotypes form a lineage that is shared among four species of indigobirds, one haplotype is shared by all four species, and two haplotypes are shared by two species. In Cameroon 12 haplotypes form a lineage that includes all species except V. chalybeata, one haplotype is shared by three species, and three haplotypes are shared by two species.

Within the hosts of the indigobirds, Peters' twinspot, Hypargos niveoguttatus, is less closely related to the firefinches than they are to each other, and brown twinspot, Clytospiza monteiri, is more closely associated with the firefinches than with the other twinspot (though its restriction sites were incompletely sampled). Where two or more haplotypes were sampled within a species (L. senegala, L. rara, L. rubricata, L. rhodopareia, H. niveoguttatus), the sets of haplotypes were not shared between species, and geographically replacing subspecific forms within the species L. senegala and L. rubricata each were each others' closest relatives.

Shared Haplotypes among Species

The extensive sharing of haplotypes among species of indigobirds might be due.either to shared ancestral polymorphisms, to hybridization and introgression of mtDNA between species, or to an independent origin of restriction sites within each species. We evaluated the origin of the shared polymorphisms versus independent origin of restriction sites by testing the effect on the number of steps that would be involved in each model. The model of shared ancestral polymorphisms involved the minimal number of steps estimated in the distribution of haplotypes in the sample; the model of independent origins of the same restriction site haplotypes added to the number of steps that would be involved in describing their distribution when we constrained each species to be monophyletic in terms of its mtDNA haplotypes. The minimal number of steps in the tree for the 22 S-C African birds was 18. When individuals in S-C Africa were constrained to cluster as monophyletic species groups the tree was 28 steps, an increase in length of 56%. This difference was statistically significant (n = 10, T = 4.5, P [less than] 0.01) The minimal number of steps for the 23 birds taken in Cameroon was 26; when the species were constrained and the song forms of V. camerunensis were considered as separate entities, the resulting tree was 38 steps, an increase of 46%. This difference was also statistically significant (n = 17, T = 36.5, P [less than] 0.05). The large increase in the estimate of the number of steps when not allowing ancestral polymorphisms to be retained leads us to consider a retention of ancestral polymorphisms as the more parsimonious explanation (relative to independent origin of restriction sites within species) for the shared haplotypes among the species of indigobirds.

Genetic Distances

Mean genetic distances between parasite species in both groups were less than mean distances between their host species (Table 2). Within the paradise whydahs the mean was 42% of that between the pytilias. In the indigobirds there were larger between-haplotype genetic distances within a species than between species in six of the nine regional populations where two or more haplotypes were sampled. Within the indigobird group from Malawi and Zimbabwe, the estimated mean genetic distance between species [Mathematical Expression Omitted] was thus sometimes less than zero because the within-species distance was greater than the between-species distance. Where estimates of [Mathematical Expression Omitted] are adjusted to 0.001, the mean genetic distance between indigobird species in Malawi and Zimbabwe was 0.9% of the mean distance between their host species in that region. In Cameroon, the mean genetic distance between five indigobird populations was 13% that of their five host species; the larger [Mathematical Expression Omitted]-estimates for these indigobirds relative to those in Malawi and Zimbabwe involved Cameroon V. chalybeata, which did not share haplotypes with the other species. Combining estimates, the mean genetic distance between the indigobirds was only 7.3% as large as the mean genetic distance between their host species in the same regions.

In all 13 pairwise comparisons of brood parasite and their host species, the genetic distance between the parasite species (for the indigobirds, compared within a region) is less than the genetic distance between their corresponding host species. This was true both for the estimates uncorrected for within-species variation [Mathematical Expression Omitted] and for the net estimates that take into account the within-species variation [Mathematical Expression Omitted] (Table 2). The probability that the observed distribution is explained by an equal proportion of hosts and parasites with the larger genetic distance can be estimated with a binomial distribution, where the factorial expression gives an estimate P = 0.00012. The low probability allows us to reject the model of cospeciation and to accept the alternative model of colonization of the host species after the host species had differentiated and speciation of parasites after hosts.

In addition, the genetic distances between geographic replacements of indigobirds that parasitize the same host species in different regions of Africa were about the same or greater than the genetic distances between different indigobird species (which each parasitizes a different host) within a region. The [Mathematical Expression Omitted] between V. chalybeata (song mimics and brood parasites of red-billed firefinch, L. senegala), in Cameroon and those in S-C Africa was 0.0019, whereas between V. chalybeata and the other indigobird species within a region [Mathematical Expression Omitted] was 0.0063 in Cameroon and 0.0004 in S-C Africa. [Mathematical Expression Omitted] between the Vidua song mimics of African fire finch, L. rubricata, in Cameroon and the song mimics of this firefinch in S-C Africa was 0.0092, whereas between different song mimics within a region [Mathematical Expression Omitted] was -0.0005 in S-C Africa and 0.0009 in Cameroon. The greater genetic distances between Vidua that are associated with the same host in different regions than between local Vidua with different species of hosts within a region is consistent with a model of independent colonizations of the host species from a local source, in particular with the indigobirds that are associated with African firefinch, L. rubricata.

DISCUSSION

Cospeciation, Colonization, and the Origins of Brood Parasite-Host Associations

A preliminary comparison of the Vidua brood parasite and host species (Klein et al. 1993) used restriction fragment length polymorphisms rather than the mapped restriction sites, which allow a spatial criterion of homology. In both, the estimates of relationships among the brood parasites are not congruent with the estimates of relationships among the host species. The lack of parallel speciation is also apparent in a quantitative comparison of the trees of these brood parasites and their hosts (Page 1994). Although the restriction fragment studies did not take into account the within-species genetic variation, both studies gave similar results: the brood parasites were genetically more similar to each other than were their host species and the differences were an order of magnitude lower in the brood parasitic indigobirds.

The molecular genetic analyses thus support the colonization model rather than the cospeciation model. Each of the predictions of the model of independent colonization was supported by the molecular results. (a) The branching diagrams of the Vidua mtDNAs do not parallel the branching diagrams of their host species' mtDNAs. (b) The branching sets for indigobirds are more closely associated with geographic regions. (c) Similarity of mtDNAs is much greater among the brood parasites than among their host species. (d) Genetic distances between species of brood parasite sampled from within a geographic region were smaller than distances between parasitic individuals that use the same species of host in different geographic regions. The much smaller genetic distances between species of brood parasite than between host species suggest a more recent speciation in the Vidua parasites than in their hosts. The results indicate that the brood parasites have associated with new host species by colonization, learned their songs, and then later matched the colors and patterns of the host nestlings' mouths only after many generations of selection for mimicry during periods of competition between the parasite and host nestlings.

The diversity of host species of estrildid finches that are associated with the indigobirds is consistent with a colonization model of association. Early observations indicated that the indigobirds were associated only with the firefinches Lagonosticta (Nicolai 1964; Payne 1973a, 1982). They are now known also to be associated with other species groups. Certain indigobirds mimic the songs and are associated with twinspots in three other estrildid genera (Hypargos, Clytospiza, and Euschistospiza; Payne et al. 1993; Payne and Payne 1994, 1995). In western Africa including Cameroon, one indigobird species is associated with Amandava subflava goldbreast and another is associated with Ortygospiza atricollis quail-finch (Payne and Payne 1994); neither host is closely related to firefinches or twinspots (Goodwin 1982; Kakizawa and Watada 1985; Wolters 1987). Some populations of V. camerunensis are associated with brown twinspot or Dybowski's twinspot, and others are associated with the fire finches L. rara or L. rubricata (Payne and Payne 1994, 1995). The greater variation in behavior and morphology among the host species than among the indigobirds (Payne 1973a; Goodwin 1982; Payne and Payne 1994) also suggests that a series of colonizations occurred well after the time of the host species divergence.

The behavior of the Vidua is consistent with a colonization model. In the field, occasional males (1% of 484 males, in areas where two or more species of indigobirds live together) have songs mimicking a species of estrildid that is not the normal host of this species of indigobird (Payne et al. 1993). Two of those birds were included in our restriction sites analysis: a V. codringtoni that mimicked the songs of African firefinch L. rubricata instead of the usual host, the twinspot H. niveoguttatus, and a V. chalybeata that mimicked songs of L. rubricata rather than the usual host, L. senegala (Appendix 1). Neither indigobird mimicked any of the songs of its usual host species. Two other indigobirds with songs of alternate host species were both V. chalybeata that mimicked songs of Jameson's firefinch, L. rhodopareia (Payne 1973a; Payne et al. 1993). Their songs indicate that these males were reared by the alternate host species and not by their normal host.

Second, in fostering experiments the indigobirds that were raised by an alternate species, the Bengalese finch, Lonchura striata, copied the Lonchura song and not that of their normal firefinch host (Payne et al., in press). An implication is that this behavior allows a switch from one host species to another. The switch of a brood-parasitic female could found a population where descendant males mimic the new host species, females are attracted to males with this song, and females are imprinted on their foster parents and return to lay their eggs in the new foster species' nests (Payne 1973a, 1982).

Vidua nestlings in the nest of such a new host may be disadvantaged in receiving parental care, but mouth mimicry of their foster species' nestlings is not necessary insofar as their survival in the brood may vary with the social and feeding conditions. Nicolai (1964) found that nesting estrildids in captivity often do not rear the young of species except their own, but he noted that sometimes they accept or adopt the young of other species. Goodwin (1960, 1982) noted that some parents desert their young, whereas others rear not only their own but also other species. Immelmann et al. (1977) compared the growth of nestling zebra finches, Taeniopygia guttata, of two kinds: (1) nestlings (normal plumage) with pigmented mouth markings; and (2) nestlings that lack the markings. Normal nestlings received more food from the parents, had priority to first feeds of the day, grew faster, and had higher survival. In two additional studies comparing normal and unmarked nestlings, the unmarked young grew more slowly when food was limited, but there was no difference when food was abundant and there was no difference in survival (Skagen 1988); the survival of unmarked nestlings was lower when food was limited, but equal when food was abundant (Reed and Freeman 1991). The experiments in this finch suggest the conditions when a nestling brood parasite will survive in the brood of a new species of host whose own nestlings have a different mouth pattern.

Species Trees, Gene Trees, and the Distribution of mtDNA Haplotypes

Mitochondrial DNA haplotypes are shared among indigobirds that are recognizable both as distinct morphologically diagnostic phylogenetic species and as biological species or intrabreeding populations (Payne et al. 1993). A lack of difference or a very low genetic distance ([less than] 1%) in molecular genetic profiles between species has also been reported in a few other birds (Kessler and Avise 1984; Shields and Helm-Bychowski 1988; Avise et al. 1990; Zink et al. 1991; Seutin et al. 1995). The sharing of haplotypes can be interpreted as due to (1) genetic polymorphisms that are retained from an ancestral population (Tajima 1983; Moran and Kornfield 1993; Avise 1994; Moore 1995); (2) independent gains or losses of certain restriction sites in different lineages (Aquadro and Greenberg 1983; Templeton 1983; Moritz et al. 1987); or (3) hybridization and introgression of mtDNA between species (Moritz et al. 1992; G. R. Smith 1992; Avise 1994; Moore 1995).

The large amount of genetic polymorphism within a species and the distribution of shared haplotypes among species of indigobirds within a geographic region is consistent with a history of retained ancestral polymorphisms within very recently diverged descendant species (Golding 1992; Avise 1994). A hypothesis of recently diverged indigobird species is also supported by their low between-species genetic variation when compared with the paradise whydahs, the estrildids included in this study, and other songbirds (Edwards and Wilson 1990; Johnson and Cicero 1991; E. F. G. Smith et al. 1991; Zink et al. 1991; Richman and Price 1992; Seutin et al. 1995). The lack of congruence between species trees and the mitochondrial-gene trees, the parsimonious accounting for shared haplotypes in a model of ancestral polymorphisms, and the occurrence of shared haplotypes among species within a geographic region all indicate that these species retain a set of ancestral polymorphisms that have not had time to become differentially lost through the stochastic lineage sorting process.

As an alternative to colonization, a hypothesis of cospeciation and subsequent hybridization could account for the observed sharing of haplotypes among the species of indigobirds. This hypothesis might be supported if haplotypes were shared between phylogenetically remote lineages of species (Moritz et al. 1992; G. R. Smith 1992; Moore 1995), that is, if there were any cases of large genetic distances (similar to levels found in host species) between mtDNA haplotypes within the indigobirds, as these might trace ancient speciation events. In some lizards, the remote relationships between species that later hybridized to form parthenogens are reflected in mitochondrial markers of distant past speciation and differentiation that are carried by the parthenogens (Moritz et al. 1992). However, within the indigobirds all mitochondrial haplotypes were very similar in restriction site profiles and none involved genetic distances comparable to those observed between the host species, as would be expected if there were survivors of past ancient cospeciations with host species. This suggests that none of the current parasite lineages of indigobirds diverged as long ago as did the host species. Although the occurrence of host switches may provide an opportunity for introgression of mtDNA across species boundaries, the lack of any genetic divergence greater than 1% between indigobird haplotypes suggests this phenomenon is not masking ancient splitting events that would have occurred under a model of cospeciation with hosts.

The cospeciation and subsequent hybridization hypothesis might also be supported if there were morphological intermediates due to hybridization and introgression between the species. Morphologically intermediate males are quite uncommon (Payne et al. 1992, 1993; Payne and Payne 1994). No hybridization is apparent in size, plumage, or colors of the individual indigobirds used in the molecular samples, or in larger samples of museum specimens from the same regions (Payne et al. 1992, 1993).

Genetic distance comparisons assume similar rates of mutational change, but rates may vary among lineages (Gillespie 1991; Martin and Palumbi 1993; Hafner et al. 1994; Mindell et al. 1996). An assumption of similar rates is appropriate in the brood-parasitic finches and their host species, because in addition to being closely related (Bentz 1979; Sibley and Ahlquist 1990), Vidua and their estrildid hosts are similar in body size (10-20 g) and generation time (females breed at one year of age) (M.-Y. Morel 1973; Payne 1973a). Genetic distances between species of Vidua other than those within the paradise whydah and indigobird species complexes are comparable to distances between the estrildid finch species (2-4%). This similarity suggests similar rates of molecular evolution in these two clades, rather than a slowdown of rate within the Vidua finches. Some variation is expected in the rate of molecular change in different clades, but the number of nucleotide substitutions between a pair of species should be positively correlated with time since divergence (Wilson et al. 1977; Nei 1987; Avise 1994). For these reasons, the much smaller genetic distances between species of brood parasites than between their hosts is consistent with a model of early speciation of the hosts and later colonization and differentiation of the parasites.

Parallel evolution (homoplasy of restriction gains/losses) could also explain why morphologically and behaviorally distinct parasite species were not distinguished in the molecular genetic results. Similarities between species that share identical haplotypes could be due to independent gains and losses of the same set of restriction sites. An independent evolution of identical restriction site profiles is unlikely, due to low rates of mutation, although this can be difficult to track with phylogenetic methodology. The statistically significant increase (nearly 50%) in the number of steps required to describe a monophyletic origin of the haplotypes within each species argues against a model of parallel evolution of restriction sites.

Both molecular genetic evidence and morphological and behavioral comparisons suggest that the brood parasites have colonized their host species well after the host species had diverged, rather than having cospeciated with them. This has profound implications for our understanding of the ecological and evolutionary contexts of host-brood parasite associations and of the relative rapidity with which some morphological changes (e.g., mimicry of nestling mouth patterns) can take place.

ACKNOWLEDGMENTS

For help in the field we thank L. L. Payne, K. Hustler, and M. E. D. Nhlane. For permits in Zimbabwe we thank the National Museum of Natural History and the Chief Wildlife Officer, Department of National Parks and Wildlife. For permits in Malawi we thank the National Research Council of Malawi; the Chief Wildlife Officer, Department of Wildlife and National Parks; and the Officer in Charge, Lengwe National Park. The Museums of Malawi provided transport and assistance. For permits in Cameroon we thank the Government of the Republic of Cameroon, IRZ Institute for Zoological Research, and the Ministry of Tourism. NKK carried out the laboratory work in the University of Michigan Museum of Zoology, where W. M. Brown made available the facilities of the Laboratory of Molecular Systematics. D. L. Swofford allowed use of PAUP 4.0d49 before its release, D. P. Mindell provided computer facilities and programs, and M. D. Sorenson advised on the analyses. The original recordings of songs by RBP and the specimens are in UMMZ. We thank the curators at AMNH, FMNH, USNM, BM(NH), and NMNH for access to their museum collections. For comments on the manuscript we thank J. L. Cracraft, D. P. Mindell, W. S. Moore, R. F. Rockwell, M. D. Sorenson, R. M. Zink, and two anonymous reviewers. Research was supported by the National Science Foundation (BSR 89-14890 and IBN-9412399).

LITERATURE CITED

AQUADRO, C. F., AND B. D. GREENBERG. 1983. Human mitochondrial DNA variation and evolution: analysis of nucleotide sequences from seven individuals. Genetics 103:287-312.

AVISE, J. C. 1994. Molecular markers, natural history and evolution. Chapman and Hall, New York.

AVISE, J. C., C. D. ANKNEY, AND W. S. NELSON. 1990. Mitochondrial gene trees and the evolutionary relationship of mallard and black ducks. Evolution 44:1109-1119.

BENTZ, G. D. 1979. The appendicular myology and phylogenetic relationships of the Ploceidae and Estrildidae (Aves: Passeriformes). Bull. Carnegie Mus. Nat. Hist. 15:1-25.

BROWER, A. V. Z. 1996. Parallel race formation and the evolution of mimicry in Heliconius butterflies: a phylogenetic hypothesis from mitochondrial DNA sequences. Evolution 50:195-221.

BROWN, W. M. 1980. Polymorphism in mitochondrial DNA of humans as revealed by restriction endonuclease analysis. Proc. Nat. Acad. Sci. USA 77:3605-3609.

-----. 1985. The mitochondrial genome of animals. Pp. 95-130 in R. J. MacIntyre, ed. Molecular evolutionary genetics. Plenum, New York.

BROWN, W. M., AND J. VINOGRAD. 1974. Restriction endonuclease cleavage maps of animal mitochondrial DNAs. Proc. Nat. Acad. Sci. USA 71:4617-4621.

CARR, S. M., A. J. BROTHERS, AND A. C. WILSON. 1987. Evolutionary inferences from restriction maps of mitochondrial DNA from nine taxa of Xenopus frogs. Evolution 41:176-190.

DAVIES, N. B., A. F. G. BOURKE, AND M. DE L. BROOKE. 1989. Cuckoos and parasitic ants: interspecific brood parasitism as an evolutionary arms race. Trends Ecol. Evol. 4:274-278.

DESJARDINS, P., AND R. MORALS. 1990. Sequence and gene organization of the chicken mitochondrial genome: a novel gene order in higher vertebrates. J. Mol. Biol. 212:599-634.

DOWLING, T. E., C. MORITZ, AND J. D. PALMER. 1990. Nucleic acids. II. Restriction site analysis. Pp. 250-317 in D. M. Hillis and C. Moritz, eds. Molecular systematics. Sinauer, Sunderland, MA.

DOWLING, T. E., C. MORITZ, J. D. PALMER, AND L. H. RIESEBERG. 1996. Nucleic acids. III. analysis of fragments and restriction sites. Pp. 249-320 in D. M. Hillis, C. Moritz, and B. K. Mable, eds. Molecular systematics. 2d ed. Sinauer, Sunderland, MA.

EDWARDS, S. V., AND A. C. WILSON. 1990. Phylogenetically informative length polymorphism and sequence variability in mitochondrial DNA of Australian songbirds (Pomatostomus). Genetics 126:695-711.

FRIEDMANN, H. 1960. The parasitic weaverbirds. U.S. Nat. Mus. Bull. 223:1-196.

FUNK, D. J., D. J. FUTUYMA, G. ORTI, AND A. MEYER. 1995. A history of host associations and evolutionary diversification for Ophraella (Coleoptera: Chrysomelidae): new evidence from mitochondrial DNA. Evolution 49:1008-1017.

GILL, F. B. 1994. Ornithology. 2d ed. W. H. Freeman, New York.

GILLESPIE, J. H. 1991. The causes of molecular evolution. Oxford Univ. Press, New York.

GOLDING, B. 1992. The prospects for polymorphisms shared between species. Heredity 68:263-276.

GOODWIN, D. 1960. Observations on avadavats and golden-breasted waxbills. Avic. Mag. 66:174-199.

-----. 1982. Estrildid finches of the world. Cornell Univ. Press, Ithaca, NY.

HAFNER, M. S., P. D. SUDMAN, F. X. VILLABLANCA, T. A. SPRADLING, J. W. DEMASTES, AND S. A. NADLER. 1994. Disparate rates of molecular evolution in cospeciating hosts and parasites. Science 265:1087-1090.

HIGUCHI, H., AND S. SATO. 1984. An example of character release in host selection and egg colour of cuckoos Cuculus sp. in Japan Ibis 126:398-404.

HILBORN, R., AND M. MANGEL. 1997. The ecological detective Monographs in population biology no. 28. Princeton Univ. Press, Princeton, NJ.

HIXSON, J. E., AND W. M. BROWN. 1986. A comparison of small ribosomal RNA genes from the mitochondrial DNA of the great apes and humans: sequence, structure, evolution, and phylogenetic implications. Mol. Biol. Evol. 3:1-18.

IMMELMANN, K., A. PILTZ, AND R. SOSSINKA. 1977 Experimentelle Untersuchungen zur Bedeutung der Rachenzeichnungen junger Zebrafinken. Z. Tierpsychol. 45:210-218.

JOHNSON, N. K., AND C. CICERO. 1991. Mitochondrial DNA sequence variability in two species of sparrows of the genus Amphispiza. Pp. 600-610 in Acta XX Congressus Internationalis Ornithologici, Christchurch, New Zealand

KAKIZAWA, R., AND M. WATADA. 1985. The evolutionary genetics of the Estrildidae. J. Yamashina Inst. Ornithol. 17:143-158.

KESSLER, L. G., AND J. C. AVISE. 1984. Systematic relationships among waterfowl (Anatidae) inferred from restriction endonuclease analysis of mitochondrial DNA. Syst. Zool. 33:370-380.

KLEIN, N. K., AND W. M. BROWN. 1994. Intraspecific molecular phylogeny in the yellow warbler (Dendroica petechia), and implications for avian biogeography in the West Indies Evolution 48:1914-1932.

KLEIN, N. K., R. B. PAYNE, AND M. E. D. NHLANE. 1993. A molecular perspective on speciation in the brood parasitic Vidua finches. Pp. 29-39 in R. T. Wilson, ed. Proceedings of the VIII pan-African ornithological congress, Bujumbura.

LANSMAN, R. A., R. O. SHADE, J. F. SHAPIRA, AND J. C. AVISE. 1981. The use of restriction endonucleases to measure mitochondrial DNA sequence relatedness in natural populations. J. Mol. Evol. 17:214-226.

MARTIN, A. P., AND S. R. PALUMBI. 1993. Body size, metabolic rate, generation time, and the molecular clock. Proc. Nat. Acad. Sci. USA 900:4087-4091.

MINDELL, D. P., A. KNIGHT, C. BAER, AND C. J. HUDDLESTON. 1996. Slow rates of molecular evolution in birds and the metabolic rate and body temperature hypotheses. Mol. Biol. Evol. 13:422-426.

MITTER, C., AND D. R. BROOKS. 1983. Phylogenetic aspects of coevolution. Pp. 65-98 in D. J. Futuyma and M. Slatkin, eds. Coevolution. Sinauer, Sunderland, MA.

MOORE, J., AND N. J. GOTELLI. 1996. Evolutionary patterns of altered behavior and susceptibility in parasitized hosts. Evolution 50:807819.

MOORE, W. S. 1995. Inferring phylogenies from mtDNA variation: mitochondrial gene-trees versus nuclear-gene trees. Evolution 49: 718-726.

MORAN, P., AND I. KORNFIELD. 1993. Retention of an ancestral polymorphism in the mbuna species flock (Teleostei: Cichlidae) of Lake Malawi. Mol. Biol. Evol. 10:1015-1029.

MOREL, G. R. 1959. Le parasitisme de Lagonosticta senegala (L.) par Hypochera chalybeata (Muller). Pp. 157-159 in M. K. Rowan, ed. Proceedings of the I pan-African ornithological congress, Livingstone.

MOREL, M.-Y. 1973. Contribution a l'etude dynamique de la population de Lagonosticta senegala L. (Estrildides) a Richard-Toll (Senegal). Interrelations avec le parasite Hypochera chalybeata (Muller) (Viduines). Memoires Museum National d'Histoire Naturelle, serie A, Zoologie 78.

MORITZ, C., T. E. DOWLING, AND W. M. BROWN. 1987. Evolution of animal mitochondrial DNA: relevance for population biology and systematics. Annu. Rev. Ecol. Syst. 18:269-292.

MORITZ, C., J. W. WRIGHT, AND W. M. BROWN. 1992. Mitochondrial DNA analyses and the origin and relative age of parthenogenetic Cnemidophorus: phylogenetic constraints on hybrid origins. Evolution 46:184-192.

NEI, M. 1987. Molecular evolutionary genetics. Columbia Univ. Press, New York.

NEI, M., AND F. TAJIMA. 1983. Maximum likelihood estimation of the number of nucleotide substitutions from restriction site data. Genetics 105:207-217.

NICOLAI, J. 1964. Der Brutparasitismus der Viduinae als ethologisches Problem. Z. Tierpsychol. 21:129-204.

-----. 1973. Das Lernprogramm in der Gesangsausbildung der Strohwitwe Tetraenura fischeri Reichenow. Z. Tierpsychol. 32: 113-138.

-----. 1974. Mimicry in parasitic birds. Sci. Am. 231:92-98.

PAGE, R. D. M. 1994. Parallel phylogenies: reconstructing the history of host-parasite assemblages. Cladistics 10:155-173.

PAYNE, R. B. 1973a. Behavior, mimetic songs and song dialects, and relationships of the parasitic indigobirds (Vidua) of Africa. Ornithological Monographs, no. 11. American Ornithologists' Union, Washington, DC.

-----. 1973b. Vocal mimicry of the paradise whydahs (Vidua) and response of female whydahs to the songs of their hosts (Pytilia) and their mimics. Anim. Behav. 21:762-771.

-----. 1977a. The ecology of brood parasitism in birds. Annu. Rev. Ecol. Syst. 8:1-28.

-----. 1977b. Clutch size, egg size, and the consequences of single vs. multiple parasitism in parasitic finches. Ecology 58:500-513.

-----. 1982. species limits in the indigobirds (Ploceidae, Vidua) of West Africa: mouth mimicry, song mimicry, and description of new species. Miscellaneous publications of the University of Michigan Museum of Zoology, no. 162, Ann Arbor.

-----. 1985a. Behavioral continuity and change in local song populations of village indigobirds, Vidua chalybeata. Z. Tierpsychol. 70:1-44.

-----. 1985b. Song populations and dispersal in steelblue and purple widowfinches. Ostrich 56:135-146.

-----. 1990. Song mimicry by the village indigobird (Vidua chalybeata) of the red-billed firefinch (Lagonosticta senegala). Vogelwarte 35:321-328.

-----. 1991. Female and first-year male plumages of paradise whydahs Vidua interjecta. Bull. Br. Ornithol. Club 111:95-100.

-----. 1997. Avian brood parasitism. Pp. 338-369 in D. H. Clayton and J. Moore, eds. Host-parasite evolution. Oxford Univ. Press, Oxford.

-----. 1998. Brood parasitism in birds: strangers in the nest. Bioscience 48:377-386.

-----. In press. A new species of firefinch Lagonosticta from northern Nigeria, and its association with the Jos Plateau indigobird Vidua maryae. Ibis 140.

PAYNE, R. B., AND L. L. PAYNE 1994. Song mimicry and species status of the indigobirds Vidua: associations with quail-finch Ortygospiza atricollis, goldbreast Amandava subflava and brown twinspot Clytospiza monteiri. Ibis 136:291-304.

-----. 1995. Song mimicry and association of brood-parasitic indigobirds (Vidua) with Dybowski's twinspot (Euschistospiza dybowskii). Auk 112:649-658.

PAYNE, R. B., L. L. PAYNE, AND M. E. D. NHLANE. 1992. Song mimicry and species status of the green widowfinch Vidua codringtoni. Ostrich 63:86-97.

PAYNE, R. B., L. L. PAYNE, M. E. D. NHLANE, AND K. HUSTLER. 1993. Species status and distribution of the parasitic indigo-birds Vidua in east and southern Africa. Pp. 40-52 in R. T. Wilson, ed. Proceedings of the VIII pan-African ornithological congress, Bujumbura.

PAYNE, R. B., L. L. PAYNE, AND J. L. WOODS. In press. Song learning in brood parasitic indigobirds Vidua chalybeata: song mimicry of the host species. Anim. Behav. 55. PLATT, J. R. 1964. Strong inference. Science 146:347-353.

PLOWRIGHT, R., AND R. E. OWEN. 1980. The evolutionary significance of bumble bee color patterns: a mimetic interpretation. Evolution 34:622-637.

RADTKEY, R. R., AND M. C. SINGER. 1995. Repeated reversals of host preference evolution in a specialist insect herbivore. Evolution 49:351-359.

REED, H. J., AND N. H. FREEMAN. 1991. Does an absence of gape markings affect survival of leucistic young in the zebra finch? Bird Behav. 9:58-63.

RICHMAN, A. D., AND T. PRICE. 1992. Evolution of ecological differences in the Old World leaf warblers. Nature 355:817-821.

RONQUIST, F. 1994. Evolution of parasitism among closely related species: phylogenetic relationships and the origin of inquilinism in gall wasps (Hymenoptera, Cynipidae). Evolution 48:241-266.

ROTHSTEIN, S. I. 1990. A model system for coevolution: avian brood parasitism. Annu. Rev. Ecol. Syst. 2:481-508.

SEUTIN, G., B. N. WHITE, AND P. T. BOAS. 1991. Preservation of avian blood and tissue samples for DNA analyses. Can. J. Zool. 69:82-90.

SEUTIN, G., N. K. KLEIN, R. E. RICKLEFS, AND E. BERMINGHAM. 1994. Historical biogeography of the bananaquit (Coereba flaveola) in the Caribbean region: a mitochondrial DNA assessment. Evolution 48:1041-1061.

SEUTIN, G., L. M. RATCLIFFE, AND P. T. BOAG. 1995. Mitochondrial DNA homogeneity in the phenotypically diverse redpoll finch complex (Aves: Carduelinae: Carduelis flammea - hornemanni). Evolution 49:962-973.

SHIELDS, G. F., AND K. M. HELM-BYCHOWSKI. 1988. Mitochondrial DNA of birds. Curr. Ornithol. 5:273-195.

SIBLEY, C. G., AND J. E. AHLQUIST. 1990. Phylogeny and classification of birds: a study in molecular evolution. Yale Univ. Press, New Haven, CT.

SINGER, M. C., D. NG, D. VASCO, AND C. D. THOMAS. 1992. Rapidly evolving associations among oviposition preferences fail to constrain evolution of insect diet. Am. Nat. 139:9-20.

SKAGEN, S. K. 1988. Asynchronous hatching and food limitation: a test of Lack's hypothesis. Auk 105:78-88.

SKEAD, C. J. 1957. Parasitism of the common waxbill, Estrilda astrild, by the pin-tailed widow-bird, Vidua macroura. Ostrich 28: 214-216.

SKEAD, D. M. 1975. Ecological studies of four estrildines in the central Transvaal. Ostrich Suppl. 11:1-55.

SMITH, E. E G., P. ARCTANDER, J. FJELDSA, AND O. G. AMIR. 1991. A new species of shrike (Laniidae: Laniarius) from Somalia, verified by DNA sequence data from the only known individual. Ibis 133:227-235.

SMITH, G. R. 1992. Introgression in fishes: significance for paleontology, cladistics, and evolutionary rates. Syst. Biol. 41:41-57.

SOUTHERN, H. N. 1954. Mimicry in cuckoos' eggs. Pp. 257-270 in J. Huxley, A. C. Hardy, and E. B. Ford, eds. Evolution as a process. Collier, New York.

SWOFFORD, D. L. 1993. PAUP: phylogenetic analysis using parsimony. Vers. 3.1. Illinois Natural History Survey, Champaign.

TAJIMA, F. 1983. Evolutionary relationship of DNA sequences in finite populations. Genetics 105:437-460.

TEMPLETON, A. R. 1983. Phylogenetic inference from restriction endonuclease cleavage site maps with particular reference to the evolution of humans and the apes. Evolution 37:221-244.

VAWTER, L., AND W. M. BROWN. 1986. Nuclear and mitochondrial DNA comparisons reveal extreme rate variation in the molecular clock. Science 234:194-196.

WILSON, A. C., S. S. CARLSON, AND T. J. WHITE. 1977. Biochemical evolution. Annu. Rev. Biochem. 46:573-639.

WOLTERS, H. E. 1987. Zur Stammesgeschichte der afrikanischen Prachtfinken. Trochilus 8:37-76.

ZINK, R. M., D. L. DITTMANN, AND W. L. ROOTES. 1991. Mitochondrial DNA variation and the phylogeny of Zonotrichia. Auk 108:578-584.

APPENDIX 1

Species and geographic sources of specimens used in the mitochondrial DNA analyses, and haplotype identities for each specimen. Haplotype numbers for indigobirds are in boldface, whether unique or shared with another bird.

Vidua: Indigobirds

Vidua camerunensis, n = 10, mimics of L. rubricata, Cameroon: Tibati, A194 (#A194), A217 (#A217), A225 (#A210), A244 (#A217); Wakwa, A138 (#A138); mimics of C. monteiri, Cameroon: Tibati, A203 (#A210), A210 (#A210), A215 (#A194), A229 (#A229), A230 (#A230) independent juvenile male at call-site of males A203, A210, and A229, no song.

V. codringtoni, n = 4, Zimbabwe: Premier Estate, A07 (#A07), A14 (#A14), A08 (#A31) song mimic of L. rubricata; Malawi: Lengwe NP, A33 (#A33).

V. funerea nigerrima, n = 4, Zimbabwe: Chipinge, A18 (#A31); Jersey Tea Estate, A15 (#A31); Malawi: Chididi, A100 (#A100); Pwezi, A71 (#A71).

V. larvaticola, n = 1, Cameroon: Garoua, A164 (#A164).

V. nigeriae, n = 2, Cameroon: Garoua, A170 (#A170), A173 (#A173).

V. purpurascens, n = 9, Zimbabwe: Beatrice, A01 (#A01); Eiffel Blue, A02 (#A30); Jabulisa, A22 (#A31); Kadoma, A04 (#A03); Malawi: Khondowe, A72 (#A72); Lengwe NP, A30 (#A30); Mwezisi, A73 (#A31); Rumphi, A67 (#A67); Tomali, A62 (#A03).

V. raricola, n = 5, Cameroon: Ngaoundere, A130 (#A210), A161 (#A170), A200 (#A170); Tibati, A202 (#A202), A234 (#A234).

V. wilsoni, n = 2, Cameroon: Ngaoundere, A139 (#A173), Tcheboa, A186 (#A186).

V. chalybeata amauropteryx, n = 5, Zimbabwe: Eiffel Blue, A03 (#A03); Gwaai River, A21 (#A03); Malawi: Lengwe, A31 (#A31); Limbe, A115 (#A33) song mimic of L. rubricata; Mwezisi, A88 (#A88).

V. chalybeata neumanni, n = 3, Cameroon: Garoua, A177 (#A177), A179 (#A179), A189 (#A189).

Vidua: Paradise Whydahs

Vidua paradisaea, n = 2, Malawi: Rumphi, A81 (#A81); captive ex Tanzania, 025 (#o25).

V. obtusa, n = 2, Malawi: Rumphi, A82 (#A82), A87 (#A87).

V. interjecta, n = 2, captive, o511 (#o511) female, o512 (#o511) male.

V. orientalis aucupum, n = 2, captive, 053 (#o53), o561 (#o561).

Vidua: Other Whydahs

Vidua macroura, n = 4, Cameroon: Ngaoundere, A147 (#o19); Malawi, Mwezisi, A89 (#o18); captive, o18 (#o18), o19(#o19).

V. hypocherina, n = 2, captive, o31 (#o31), 032 (#o31).

V. fischeri, n = 2, captive, 050 (#o50), 062 (#o62).

V. regia, n = 2, captive, 038 (#o38), 039 (#o38).

Estrildidae Finches

Amadina fasciata, n = 2, captive, o81 (#o81), 0422 (#o81).

Clytospiza monteiri, n = 1, Cameroon: Ngaoundere, A132 (#A132).

Hypargos niveoguttatus, n = 2, Zimbabwe: Gwaai River, A24 (#A24); captive, 099 (#o99).

Lagonosticta larvata, n = 1, Cameroon: Ngaoundere, A145 (#A145).

L. rara rara, n = 1, Cameroon: Tibati, A248 (#A248).

L. rara forbesi, n = 1, captive, 036 (#036).

L. rhodopareia, n = 5, Zimbabwe: Jabulisa, A23 (#A23); Malawi: Lengwe NP, A41 (#A41), A42 (#A41), A43 (#A41); captive, #A107 (#A23).

L. rubricata haematocephala, n = 2, Malawi: Limbe, A114 (#A114), Mwezisi, A75 (#A75).

L. r. congica, n = 1, Cameroon: Ngaoundere, A131 (#A131).

L. r. poliocephala, n = 1, captive, o13 (#o13).

L. rufopicta, n = 1, captive, #030 (#o30).

L. senegala rendalii, n = 3, Malawi: Mwezisi, A74 (#A74), Lengwe NP, A56 (#A56), A104 (#A104).

L. s. rhodopsis, n = 2, Cameroon: Garoua, A167 (#A167), A171 (#A171).

Mandingoa nitidula, n = 1, Malawi: Limbe, A113 (#A113).

Pytilia afra, n = 3, Malawi: Lengwe NP, A46 (#A46), A51 (#A46); captive, o107 (#A46).

P. hypogrammica, n = 2, captive, o108 (#o108), 098 (#o108).

P. melba grotei, n = 3, Malawi: Lengwe NP, A35 (#A35), A36 (#A35), A57 (#A57).

P. m. percivali, n = 2, captive, 056 (#o56), o510 (#056).

P. m. citerior, n = 2, captive, o06 (#006), o52 (#o52).

P. phoenicoptera, n = 2, captive, o54 (#o61), o61 (#o61).
APPENDIX 2

Restriction site map positions for all taxa and haplotypes used in
this study. Map position is in kilobase pairs from the conserved
ClaI site. Enzyme abbreviations are: Ap = ApaI, Ba = BamHI, Bc
=BclI, Bg = BglII, Bs =
BstEII, Cl = ClaI, Dr = DraI, Ea = EagI,
Ec = EcoRI, Hi = HindlII, Kp = KpnI, Nc = NcoI, Nd = NdeI, Nh
=NheI, Pv = PvulI, Sa = SalI,
Sc = SaclI, Xb = XbaI. The letters after
the enzyme abbreviations refer to the particular restriction site.
Where more than one map position is listed for a site, the exact
position could not be determined due to a lack of intervening sites
for comparison. The map is presented in the direction depicted by
the chicken gene map (Desjardins and Morais 1990), rather than the
direction of the chicken sequence.

ClA       0.0         HiI         6.7         NhG         11.6
ApD       0.15        BsA         6.75        BcL         11.7
BgA       0.195       BcQ         6.78        SaB         11.8
KpF       0.30        BcB         6.83        NhK         11.9
Sc(*)     0.32        NhO         6.9         HiG         11.95
EaB       0.6         BcF         7.14        KpD         12.0
DrE       0.7         BcV         7.15        ApF         12.05
BcG       0.73        ClH         7.4         BcR         12.1
EaE       0.8         XbH         7.5         HiM         12.1
DrB       0.9         NdF         7.55        ClF         12.2
NdU       1.05        HiB         7.65        HiK         12.35
EcE       1.05        NdP         7.67        BcT         12.4
NdC       1.2         NcK         7.69        XbI         12.4
BaA       1.3         XbD         7.72        BcM         12.6
NdJ       1.5         BcP         7.83        NhD         12.9
EcB       1.6         DrD         7.9         NcE         13.05
KpA       1.65        BaG         7.95        NhI         13.1
EaF       1.7         NdA         8.0         PvH         13.12
KpH       1.75        NhH         8.1         PvF         13.4
NdB       1.85        BaE         8.16        XbF         13.5
BgD       1.92        DrM         8.2         ApE         13.5
BgE       1.97        NcM         8.2         KpB         13.62
EcN       2.0         BcS         8.3         ApB         13.7
ApH       2.0         BaB         8.4         PvY(**)
13.77/13.2
NdV       2.05        HiH         8.6         PvN         13.8
EcXY      2.05        PvAA        8.75        PvZ         13.9
NdW       2.1/2.8     BsG         8.8         EcA         14.0
NhN       2.2         EaD         8.85        BaC         14.1
BaD       2.25        PvS         8.85        NhM         14.15
EaK       2.4/1.0     XbE         8.85        PvE         14.2
DrC       2.45        NdH         8.9         BgC         14.3
EaI       2.6         DrP         8.96/10.4   BsC         14.35
XbC       2.68        HiC         9.0         NdD         14.45
PvK       2.7         KpC         9.03        BcC         14.48
EaJ       2.8         BsB         9.2         ApG         14.5
NhC       2.88        BaP         9.2/12.05   NhJ         14.51
CID       2.9         XbK         9.3         BaJ         14.55
BaM       3.05        NcF         9.3         NhE         14.6
EcF       3.23        ApJ         9.35        NcG         14.7
PvQ       3.3         PvL         9.4         ApL
14.75/14.9
EaC       3.6         C/B         9.5         BcI         14.8
NdE       3.7         NcC         9.55        NhA         14.9
NdQ       3.72        BaO         9.6         DrH         15.0
ClC       3.75        HiE         9.65        XbJ         15.1
NhB       3.8         HiL         9.7         PvW         15.1
SaE       3.9         EcL         9.7         BcA         15.25
BcJ       4.0         BcD         9.75        XbB         15.3
EcH       4.4         BaI         9.8         DrK         15.5
HiF       4.5         ApK         9.82        EaA         15.55
DrG       4.6         SaC         10.0        Sc(***)     15.6
BaL       4.7         BgB         10.1        KpE         15.7
NdK       4.8         EaG         10.15       HiA         15.8
ApI       4.85        BaK         10.2        ApA         15.95
NcH       4.9         NhL         10.2        DrA         16.0
PvC       5.0         BcN         10.25       DrN         16.2
NdL       5.2         SaA         10.3        PvG         16.6
PvD       5.25        EcC         10.3        XbA         16.85
EaH       5.3         PvB         10.35       EcJ         16.95
DrF       5.6         NcB         10.4        DrQ
16.95/0.02
BcE       5.65        BsF         10.4        PvDE        ?
HiJ       5.7         NcI         10.43       PvDF        ?
EaX       5.7/13.4    EcP         10.6        PvDG        ?
NdG       5.95        PvJ         10.8
ClE       6.0         EcG         10.85
DrL       6.05        NdI         10.9
BcH       6.17        HiD         10.92
NhF       6.2         NcA         11.1
EcK       6.3         PvA         11.2
NcD       6.5         NdBB        11.4
BaF       6.7         BsE         11.5

* The SacII site, conserved in vertebrates, in the 12S rRNA gene.
** PvuIIY and PvuIIN could be the same site.
*** The SacII site, conserved in vertebrates, in the 16s rRNA gene.

These two SacII sites were used only as reference points for
aligning restriction site maps. They
were not included in the phylogenetic analysis, nor in estimates of
sequence divergence.


APPENDIX 3

Presence (1) and absence (0) of 200 restriction sites in estrildid and Vidua finches. Numbers next to species names indicate the haplotype numbers; some were shared among individuals and (in the Vidua indigobirds) among species (cf. Table 1). Restriction enzymes are identified by standard abbreviations; for restriction site map positions on the mitochondrial DNA molecule, see Appendix 2.

ApaIA ApaIB ApaID ApaIE ApaIF ApaIG ApaIH ApaII ApaIJ ApaIK ApaIL BamHIA BamHIB BamHIC BamHID BamHIE BamHIF BamHIG BamHII BamHIJ BamHIK BamHIL BamHIM BamHIO BamHIP BclIA BclIB BclIC BclID BclIE BclIF BclIG BclIH BclII BclIJ BclIL BclIM BclIN BclIP BclIQ BclIR BclIS BclIT BclIV BglIIA BglIIB BglIIC BglIID BglIIE BstEIIA BstEIIB BstEIIC BstEIIE BstEIIF BstEIIG ClaIA ClaIB ClaIC ClaID ClaIE ClaIF ClaIH DraIA DraIB DraIC DraID DraIE DraIF DraIG DraIH DraIK DraIL DraIM DraIN DraIP DraIQ EagIA EagIB EagIC EagID EagIE EagIF EagIG EagIH EagII EagIJ EagIK EagIX EcoRIA EcoRIB EcoRIC EcoRIE EcoRIF EcoRIG EcoRIH EcoRIJ EcoRIK EcoRIL EcoRIN EcoRIP EcoRIXY HindIIIA HindIIIB HindIIIC HindIIID HindIIIE HindIIIF HindIIIG HindIIIH HindIIII HindIIIJ HindIIIK HindIIIL HindIIIM KpnIA KpnIB KpnIC KpnID KpnIE KpnIF KpnIH NcoIA NcoIB NcoIC NcoID NcoIE NcoIF NcoIG NcoIH NcoII NcoIK NcoIM NdeIA NdeIB NdeIC NdeID NdeIE NdeIF NdeIG NdeIH NdeII NdeIJ NdeIK NdeIL NdeIP NdeIQ NdeIU NdeIV NdeIW NdeIBB NheIA NheEB NheIC NheID NheIE NheIF NheIG NheIH NheII NheIJ NheIK NheIL NheIM NheIN NheIO PvuIIA PvuIIB PvuIIC PvuIID PvuIIE PvuIIF PvuIIG PvuIIH PvuIIJ PviIIK PvuIIL PvuIIN PvIIQ PvuIIS PvuIIW PvuIIY PvuIIZ PvuIIAA PvuIIDE PvuIIDF PvuIIDG SalIA SalIB SalIC Salle XbaIA XbaB XbaIC XbaID XbaIE XbaIF XbaIH XbaII XbaIJ XbaIK
COPYRIGHT 1998 Society for the Study of Evolution
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Klein, Nedra K.; Payne, Robert B.
Publication:Evolution
Date:Apr 1, 1998
Words:11460
Previous Article:Gene flow and ineffective antipredator behavior in a stream-breeding salamander.
Next Article:Virulence of mixed-clone and single-clone infections of the rodent malaria Plasmodium chabaudi.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters