Evolution, 58(5), 2004, pp. 1122–1130
NEW MITOCHONDRIAL DNA DATA AFFIRM THE IMPORTANCE OF PLEISTOCENE
SPECIATION IN NORTH AMERICAN BIRDS
NED K. JOHNSON1,2,3
1 Museum
AND
CARLA CICERO1,4
of Vertebrate Zoology, University of California, Berkeley, California 94720-3160
of Integrative Biology, University of California, Berkeley, California 94720-3140
4 E-mail: [email protected]
2 Department
Abstract. The timing of origin of modern North American bird species in relation to Pleistocene glaciations has long
been the topic of significant discussion and disagreement. Recently, Klicka and Zink (1997) and Avise and Walker
(1998) enlivened this debate by using calibrated molecular distance values to estimate timing of speciations. Here we
use new molecular studies to test their conclusions. Molecular distance values for 39 pairs of proven sister species,
27 of which are based on new data, alter the currently perceived pattern that avian species splits occurred mainly in
the Pliocene and early–mid-Pleistocene. Mitochondrial DNA divergence values for this set of taxa showed a skewed
distribution pointing toward relatively young speciation times, in contrast to the pattern presented by Klicka and Zink
(1997) for 35 sister plus non-sister species pairs. Our pattern was not significantly different from that of Avise and
Walker (1998) for ‘‘intraspecific phylogroups,’’ some of which are species. We conclude that the entire Pleistocene,
including the last two glacial cycles (,250,000 years ago), was important in speciations of modern North American
birds. A substantial number of speciations were both initiated and completed in the last 250,000 years. Simultaneously,
many taxa began to diverge in the Pleistocene but their speciations are not yet complete (per Avise and Walker 1998).
The suggestion that durations of speciations average two million years is probably a substantial overestimate.
Key words.
Avian speciation, mitochondrial DNA divergences, phylogroups, Pleistocene, sister species.
Received May 12, 2003.
Accepted December 19, 2003.
Over one-half century ago, Rand (1948) proposed a significant role for Pleistocene glaciations in subdividing distributions and promoting speciation in several boreal species
of North American birds. Since then the timing of origin of
current species has been the topic of elaborate discussion and
disagreement. Largely from fossil evidence, Wetmore (1959)
opined that most living species already were present by the
beginning of the Pleistocene; implicit in his opinion was that
modern birds thus were largely immune from glacial effects.
Selander (1971) doubted such a Pliocene origin, as did Brodkorb (1971). The latter author, citing a preponderance of
avian fossils from the Pleistocene and the tenuous identifications of Pliocene fossils, stated that ‘‘the evidence continues to amass that all living species of birds arose during the
Quaternary [5 Pleistocene 1 Holocene]. . . Pending critical
reevaluation, all Pliocene reports of neospecies should be
disregarded’’ (Brodkorb 1971, p. 46–47). Most subsequent
authors, including Mengel (1964, 1970), Selander (1965,
1971), and Selander and Giller (1961) agreed with Brodkorb
that the entire Pleistocene was the predominant period of
speciation for modern North American birds, except that Selander (1965) and Mengel (1970) explicitly doubted involvement of the late Pleistocene. In contrast, Beecher (1955),
Dillon (1956), Brewer (1963), and, most notably, Hubbard
(1973) emphasized the latter half of the Pleistocene, especially the last two glacial intervals.
Using data from both mitochondrial DNA (mtDNA) gene
sequences and restriction fragment length polymorphism
(RFLP), Klicka and Zink (1997) enlivened the debate with
the application of temporally calibrated molecular distances
between 35 pairs of species that were suggested by Hubbard
(1973) and others to have been influenced by late Pleistocene
glaciations (the ‘‘late Pleistocene origins’’ or LPO model).
3
Ned K. Johnson passed away on June 11, 2003.
An additional 13 pairs not specifically theorized to have late
Pleistocene origins also were analyzed to test their results.
These authors based their study on North American songbird
species that traditionally have been regarded as sister species,
although they admitted (Klicka and Zink 1997, p. 1669, footnote 13) that ‘‘recent work has shown that some are not sisters
but rather are members of species complexes.’’ (In fact, our
analysis shows that 24 of their 35 late Pleistocene species
pairs [68.6%] are not sister taxa.) Assuming that mtDNA
evolves at a constant, clocklike rate of 2% per million years,
Klicka and Zink (1997) found a surprisingly high average
divergence of 5.1% (2,550,000 years) between these 35 pairs,
with values ranging from 0.4 to 10.9% (200,000–5,450,000
years). A similar mtDNA divergence value (5.2%) was obtained when comparing the 13 presumably older pairs of
species. These data supported their conclusion that modern
species of songbirds have had a protracted history of speciation throughout the Pleistocene and Pliocene, and that
strong evidence was lacking for significant late Pleistocene
radiations as were specifically invoked by Hubbard (1973).
This theme is repeated in Zink and Slowinski (1995) and
Zink and Klicka (1999).
In a related study, Avise and Walker (1998) also examined
mtDNA differences among North American birds but emphasized dates calculated from ‘‘phylogroups’’, defined as
genetically distinctive and geographically oriented populations within species. Compared with the mtDNA divergence
times calculated by Klicka and Zink (1997), which centered
on the Pliocene and early- to mid-Pleistocene, those of Avise
and Walker (1998) were notably younger and clearly emphasized the entire Pleistocene. The latter authors concluded
that ‘‘Pleistocene biogeographic factors promoted substantial
microevolutionary genetic diversification in birds’’ (Avise
and Walker 1998, p. 457), but that their data were compatible
with those of Klicka and Zink (1997) if one viewed avian
1122
q 2004 The Society for the Study of Evolution. All rights reserved.
DNA DIVERGENCE TIMES OF NORTH AMERICAN BIRDS
speciation as a protracted temporal process. Thus, they regarded Klicka and Zink’s (1997) study as focusing on species’ origins, whereas their results emphasized the Pleistocene as a time when earlier speciations were completed and
new phylogeographic divisions initiated (for further discussion, see Zink and Klicka 1999). In a companion paper, Avise
et al. (1998) extended their comparisons to include mammals,
amphibians, reptiles, and fishes, and concluded again that
Pleistocene conditions were important both in starting phylogeographic splits in living species and in further molding
pre-existing phylogeographic diversity into many current sister species. Finally, these authors opined that data from all
sources pointed to protracted speciation durations averaging
at least two million years.
Klicka and Zink (1999) critically evaluated the methodology and conclusions of Avise and Walker (1998), including
the phylogroup concept, and offered additional mtDNA data
for intraspecific diversification in 14 North American passerine species. Using the standard clock calibration of 2%
per million years, they estimated mtDNA divergence times
for 63 pairs of taxa (Klicka and Zink 1997, table 1 and reference 21; Avise and Walker 1998, table 1; Klicka and Zink
1999, table 1) on the basis of net sequence divergence calculations, that is, mtDNA sequence divergence estimates corrected for intraspecific variation (Avise and Walker 1998).
(Note that they refer to these comparisons as ‘‘phylogroups’’
[axis of figure 2], a combination of intraspecific ‘‘phylogroups’’ and ‘‘sister species’’ [legend in figure 2], and ‘‘sister-phylogroup pairs.’’ The muddled wording in this section,
along with the lack of a list of comparisons included in their
analysis [we tried to replicate their data and were unable to
obtain 63 pairs], made it difficult to follow their exact plotting.) Although Klicka and Zink (1999, p. 698) asserted that
‘‘all taxon pairs depicted are likely phylogenetic species at
different states of evolutionary divergence,’’ they concluded
that 80% of all ‘‘phylogroups (whether within or between
biological species) have divergence dates older than 500,000
years’’ (Klicka and Zink 1999, p. 698). They used these data
to reject again the late Pleistocene origins model (LPO) for
avian speciation, and to argue against Avise and Walker’s
(1998) assertion that the Pleistocene was a profound time for
phylogeographic diversification. Furthermore, they reasoned
that speciation probably occurs much more rapidly than estimated by Avise and Walker (1998); they suggested less
than 500,000 years, although they admit that additional data
are needed.
The contribution of all of these studies lies in their focused
effort to bring molecular evidence to bear on a fundamental
issue in evolutionary biology, namely, the timing of modern
species divergences. Since the appearance of Klicka and
Zink’s (1997) and Avise and Walker’s (1998) widely cited
papers, new studies have emerged that provide additional
mtDNA distance data for pairs of North American bird species. These data allow testing of the aforementioned authors’
conclusions. Because many of the species pairs examined by
Klicka and Zink (1997) have either proven to be non-sisters
or represent species clusters that were sampled incompletely
(i.e., available molecular phylogenies excluded some taxa),
we hypothesized that examination of true sister species would
alter the currently perceived temporal pattern of avian spe-
1123
ciation in North America. Furthermore, we proposed that
younger speciation times in proven sister species could throw
light on Avise and Walker’s (1998) major conclusion of protracted speciation in North American birds. To test these
hypotheses, we analyzed new molecular distance values for
27 pairs of proven sister species, which we combined with
12 valid pairs of sisters already considered by Klicka and
Zink (Table 1). Although Klicka and Zink (1997) limited
their analysis to songbirds, we expanded the comparative data
base by including data from seven pairs of non-passeriform
species, including geese, ducks, rails, grouse, shorebirds,
doves, and woodpeckers. All of these North American species
pairs, except perhaps the doves, have temperate breeding distributions that suggest possible Pleistocene influence.
Taxonomy and species concept clearly affects selection of
taxa for this kind of analysis. Although a discussion and
debate of species concepts (see Johnson et al. 1999) is beyond
the scope of this paper, differing opinions will influence the
point at which divergent taxa are recognized as species. For
example, phylogroups that are recognized as species under
the phylogenetic species concept because of reciprocal monophyly in mtDNA sequences may or may not qualify as biological species depending on the evidence for reproductive
isolation (Johnson et al. 1999). Regardless of these issues,
studies that address questions of timing and duration of speciation must be explicit in their species concept and choice
of taxa. Klicka and Zink (1997, 1999) and Avise and Walker
(1998) analyzed taxon pairs at different stages of speciation,
that is, from well-marked subspecies (or phylogenetic species) to those clearly beyond the biological sister species
stage. In contrast, we focused explicitly on sister taxa that
either (1) are recognized as biological species (American Ornithologists’ Union 1998, and supplements), (2) have historically been considered species, (3) are phenotypic sibling
species (Mayr 1942), or (4) are near the species boundary
but that are probably best treated as species on the basis of
biologic criteria (e.g., differences in suites of traits, including
voice, morphology, and/or genetics; local sympatry with limited interbreeding). Furthermore, because of taxon sampling
issues, we limited our analysis of species groups to those
with molecular phylogenies that include all potential North
American relatives. This required exclusion of numerous species pairs used by Klicka and Zink (1997) that subsequently
have been shown to be non-sisters based on molecular data,
for example, Baeolophus bicolor-B. inornatus (tufted and
‘‘plain’’ titmice), Polioptila melanura-P. nigriceps (blacktailed and black-capped gnatcatchers), Toxostoma lecontei-T.
redivivum (LeConte’s and California thrashers), Dendroica
townsendi-D. virens (Townsend’s and black-throated green
warblers), Oporornis philadelphia-O. agilis (mourning and
Connecticut warblers), Piranga olivacea-P. ludoviciana (scarlet and western tanagers), Pipilo aberti-P. fuscus (Abert’s and
canyon towhees), Spizella pallida-S. breweri (clay-colored
and Brewer’s sparrows), Passerina cyanea-P. versicolor (indigo and varied buntings), and Icterus galbula-I. bullockii
(Baltimore and Bullock’s orioles). Likewise, we excluded
several of Klicka and Zink’s (1997) pairs that represent complexes that have not been adequately surveyed using molecular data and thus have unresolved sister relationships, for
example, Cyanocitta cristata-C. stelleri (blue and Steller’s
1124
N. K. JOHNSON AND C. CICERO
TABLE 1. Estimated mitochondrial DNA (mtDNA) divergence times (mean % uncorrected divergence) for 39 pairs of North American
avian sister species.
Species
pair 1
Chen caerulescens-C. rossi (snow and
Ross’s geese)
Anas discors-A. cyanoptera (bluewinged and cinnamon teal)
Rallus longirostris-R. elegans (clapper
and king rails)
Centrocercus urophasianus versus C.
minimus4 (northern and Gunnison
sage grouse)
Limnodramus griseus-L. scolopaceus
(short-billed and long-billed dowitchers)
Zenaida macrouri-Z. graysoni (mourning and Socorro doves)
Sphyrapicus nuchalis-S. ruber (rednaped and red-breasted sapsuckers)
Empidonax oberholseri-E. affinis (dusky
and pine flycatchers)
Empidonax fulvifrons-E. atriceps (buffbreasted and black-capped flycatchers)
Empidonax alnorum-E. traillii (alder
and willow flycatchers)
Empidonax difficilis-E. occidentalis (Pacific-slope and Cordilleran flycatchers)
Vireo solitarius-V. cassinii (blue-headed
and Cassin’s vireos)
# Baeolophus inornatus-B. ridgwayi
(oak and juniper titmice)
* # Baeolophus bicolor-B. atricristatus
(tufted and black-crested titmice)
Catharus minimus-C. bicknelli (graycheeked and Bicknell’s thrushes)
* Polioptila californica-P. melanura
(California and black-tailed gnatcatchers)
* Toxostoma cinereum-T. bendirei (gray
and Bendire’s thrashers)
% mtDNA
divergence
Estimated
age (years
ago)2
Data type 3
0.8
400,000
RFLP, 14 enzymes
0.3
150,000
0.6
300,000
RFLP, 15 enzymes, 193 sites (0.4%);
sequence, cyt b, ND2, tRNAs, 2147
bp (0.2%)
RFLP, 19 enzymes, 68 restriction sites
1.8
850,000
8.2
4,100,000
0.9
Source
Shields and Wilson
1987
Kessler and Avise
1984; Johnson and
Sorenson 1999
Avise and Zink 1988
sequence, control region I, 141–380
bp
Kahn et al. 1999
RFLP, 19 enzymes, 77 restriction sites
Avise and Zink 1988
450,000
sequence, cyt b, ND2, 3281 bp
0.5
250,000
sequence, cyt b, 711 bp
1.8
900,000
3.8
1,900,000
4.6
2,300,000
0.7
350,000
2.7
1,350,000
sequence,
bp
sequence,
bp
sequence,
bp
sequence,
bp
sequence,
4.2
2,100,000
sequence, cyt b, 900 bp
Johnson and Clayton
2000
Cicero and Johnson
1995
Johnson and Cicero
2002
Johnson and Cicero
2002
Johnson and Cicero
2002
Johnson and Cicero
2002
Cicero and Johnson
1998
Cicero 1996
0.4
200,000
RFLP, 19 enzymes, 83 restriction sites
Avise and Zink 1988
1.8
900,000
RFLP, 13 enzymes
Seutin et al. 1995
4.0
2,000,000
sequence, cyt b, ND2, ND6, 922 bp
Zink and Blackwell
1998
1.6
800,000
Zink et al. 1999
3.5
1,750,000
RFLP, 19 enzymes, 74 restriction sites
(1.4%); sequence, cyt b, ND2, ND6,
944 bp (1.7%)
sequence, cyt b, ND6, 619 bp
1.0
500,000
0.4
200,000
Lovette and Bermingham 2001
Bermingham et al.
1992; Lovette and
Bermingham 1999
* Dendroica townsendi-D. occidentalis
(Townsend’s and hermit warblers)
0.9
450,000
Dendroica nigrescens- D. graciae
(black-throated gray and Graces’s
warblers)
* Oporornis philadelphia-O. tolmiei
(mourning and MacGillivray’s warblers)
Piranga ludoviciana-P. bidentata (western and flame-colored tanagers)
* # Pipilo maculatus-P. erythrophthalmus (spotted and eastern towhees)
* Pipilo crissalis-P. aberti (California
and Abert’s towhees)
1.5
750,000
sequence, cyt b, ND2, CO I, ATPase 6,
ATPase 8, 3639 bp
RFLP, 14 enzymes, 49 restriction sites
(0.6%); sequence, cyt b, ND2,
ATPase 6, ATPase8, COI, 3639 bp
(0.2%)
RFLP, 14 enzymes, 49 restriction sites
(0.7%); sequence, cyt b, ND2,
ATPase 6, ATPase8, COI, 3639 bp
(1.0%)
sequence, cyt b, ND2, ATPase 6,
ATPase8, COI, 3639 bp
Lovette and Bermingham 1999
2.1
1,050,000
sequence, cyt b, 1050 bp
Klicka and Zink 1997
4.5
2,250,000
sequence, cyt b, tRNA, 1073 bp
Burns 1998
0.8
400,000
RFLP, 18 enzymes
Ball and Avise 1992
2.3
1,150,000
RFLP, 16 enzymes, 119 restriction
sites (2.5%); sequence, cyt b, ND2,
744 bp (2.1%)
sequence, cyt b, control region I, 1413
bp
Zink and Dittmann
1991; Zink et al.
1998
Klicka et al. 1999
Toxostoma lecontei-‘‘T. arenicolor’’
(LeConte’s and ‘‘Vizcaino’’ thrashers)
Parula americana-P. pitiayumi (northern and tropical parulas)
* Dendroica coronata-‘‘D. auduboni’’
(myrtle and Audubon warblers)
Spizella breweri-‘‘S. taverneri’’ (Brewer’s and timberline sparrows)
0.07
35,000
cyt b, ND2, ND3, COI, 3069
cyt b, ND2, ND3, COI, 3069
cyt b, ND2, ND3, COI, 3069
cyt b, ND2, ND3, COI, 3069
cyt b, 1143 bp
Zink et al. 1997, 1999
Bermingham et al.
1992; Lovette and
Bermingham 1999
1125
DNA DIVERGENCE TIMES OF NORTH AMERICAN BIRDS
TABLE 1.
Species pair 1
* # Ammodramus maritimus (seaside
sparrow, Gulf and Atlantic coastal
forms)
* # Ammodramus caudacutus-A. nelsoni
(saltmarsh and Nelson’s sharp-tailed
sparrows)
Zonotrichia leucophrys-Z. atricapilla
(white-crowned and golden-crowned
sparrows)
Junco caniceps-J. oreganus (gray-headed and Oregon juncos)
Calcarius ornatus-C. pictus (chestnutcollared and Smith’s longspurs)
Plectrophenax nivalis-P. hyperboreus
(snow and McKay’s buntings)
Passerina versicolor-P. ciris (varied
and painted buntings)
Passerina caerulea-P. amoena (blue
grosbeak and Lazuli bunting)
* Quiscalus major-Q. mexicanus (boattailed and great-tailed grackles)
Icterus galbula-I. abeillei (Baltimore
and Abeille’s orioles)
Loxia ‘‘curvirostra’’5 (‘‘red’’ crossbill,
sympatric call types)
Carduelis flammea-C. hornemanni
(common and hoary redpolls)
Mean value
Continued.
Estimated
age (years
ago)2
Data type 3
1.0
500,000
RFLP, 18 enzymes, 89 restriction sites
Avise and Nelson 1989
1.3
650,000
RFLP, 17 enzymes, 82 restriction sites
Rising and Avise 1993
0.1
50,000
0.0
0
RFLP, 19 enzymes, 122 restriction
sites (0.1%); sequence, cyt b, control
region, 985 bp (0.2%)
sequence, ND6, 723 bp
2.9
1,450,000
sequence, cyt b, 1143 bp
Zink et al. 1991;
Weckstein et al.
2001
G. F. Barrowclough,
unpubl. data
Klicka et al. 2003
0.2
100,000
sequence, cyt b, 1143 bp
Klicka et al. 2003
3.0
1,500,000
sequence, cyt b, 1143 bp
Klicka et al. 2001a
4.9
2,450,000
sequence, cyt b, 1143 bp
Klicka et al. 2001a
2.1
800,000
0.6
300,000
RFLP, 19 enzymes, 80 restriction sites
(1.6%); sequence, cyt b, ND2, 1914
bp (2.5%)
sequence, cyt b, ND2, 2005 bp
Avise and Zink 1988;
Johnson and Lanyon
1999
Omland et al. 1999
0.38
190,000
sequence, cyt b, ND6, 1828 bp
J. Groth, unpubl. data
0.2
100,000
RFLP, 20 enzymes, 124 restriction
sites
Seutin et al. 1995
1.86
930,000
% mtDNA
divergence
Source
1
Names preceded by an asterisk refer to pairs of species included in table 1 and figure 1 of Klicka and Zink (1997); names preceded by a number sign
were analyzed as ‘‘phylogroups’’ by Avise and Walker (1998; Phylogeographic Category I). Five pairs of taxa are treated as subspecies or ‘‘groups’’ by
the American Ornithologists’ Union (1998), but are probably best regarded as species: Dendroica coronata-D. auduboni, Spizella breweri-S. taverneri,
Ammodramus maritimus (Gulf and Atlantic coastal forms), Junco caniceps-J. oreganus, and Loxia curvirostra (sympatric call types). The pairs of Dendroica
and Ammodramus also were included by Klicka and Zink (1997) because they qualified as phylogenetic species. We regard the two forms of Junco as
species because they breed sympatrically in the Grapevine Mountains, Nevada (G. F. Barrowclough and N. K. Johnson, unpubl. data). Of these five pairs
of taxa, the species status of Spizella breweri-S. ‘‘taverneri’’ is perhaps most controversial (Klicka et al. 2001b; Mayr and Johnson 2001); for the purposes
of our analysis they are best categorized as borderline, although they were discussed by Klicka and Zink (1997) as the youngest species pair in North
America.
2 Ages are based on the mtDNA clock calibration of 2% Myr21 (uncorrected for within-lineage divergence). We used this method to allow direct
comparison with the data presented by Klicka and Zink (1997) and Avise and Walker (1998). Problems and caveats associated with use of this calibration
are beyond the scope of the present paper (e.g., see Arbogast and Slowinski 1998). Furthermore, Edwards and Beerli (2000) and Arbogast et al. (2002)
have described potentially serious pitfalls that attend estimates of divergence times from molecular data, especially for recent speciations. A major conclusion
of their population genetic analyses is that molecular data generally overestimate divergence times because gene divergences in ancestors antedate population
divergences (i.e., speciations). Thus, estimated ages (years ago) in this table should be younger by an unknown figure. See text for further discussion.
3 In accordance with methods used by Klicka and Zink (1997) and Avise and Walker (1998), datasets include both RFLP and mtDNA sequences. Data
are presented separately for species where both kinds of sets are available, to enable comparison of percentage divergence values given by different methods
(see Lovette et al. 1999). General similarity in values provides confidence that data presented using one method or another are not biased. Thus, values
were averaged to estimate overall divergence for these taxon pairs.
4 Reported mean sequence divergence 5 17.5%. Kahn et al. (1999) calculated that the control region sequenced evolves approximately 10.1 times faster
than the average mtDNA rate (i.e., 20.2% per my, but see caveats in footnote 1). This rate is similar to that used by Avise and Walker (1998) for hypervariable
portions of the control region. Thus, the percent sequence divergence was adusted accordingly. The estimated age of divergence was obtained directly
from Kahn et al. (1999).
5 The sympatric call types of Loxia ‘‘curvirostra’’ differed at 0–7 sites; here we use the maximum divergence value of 0.38%. Relevant to this comparison
is the study by Piertney et al. (2001), who found no significant allelic differences at microsatellite loci and similarly low levels of divergence in mtDNA
control region sequences in comparisons of the red (5common; Loxia curvirostra), Scottish (L. scotica), and parrot (L. pytyopsittacus) crossbills from Britain
and Europe.
jays; the real sister of strongly polytypic C. stelleri is probably a Mexican or Central American form [N. K. Johnson
and C. Cicero, unpubl. data]), Sialia sialis-S. mexicana (eastern and western bluebirds; a comparison with S. currucoides
is lacking), Amphispiza belli-A. bilineata (sage and blackthroated sparrows; the sister of A. belli lies within this strongly polytypic complex [C. Cicero and N. K. Johnson, unpubl.
data]; furthermore, recent molecular data suggest that these
taxa belong in separate clades at the generic level; Carson
and Spicer 2003), Pheucticus ludovicianus-P. melanocephalus
(rose-breasted and black-headed grosbeaks; a published phylogeny that includes all six potential sisters is lacking), Cardinalis cardinalis-C. sinuatus (northern cardinal and Pyrrhuloxia; the sister of C. cardinalis lies within this strongly polytypic complex, with‘‘C. carneus’’ as a possible candidate;
Monroe and Sibley 1993), Agelaius phoeniceus-A. tricolor
(red-winged and tricolored blackbirds; although Lanyon
[1994] reported these taxa as sisters, his analysis excluded
1126
N. K. JOHNSON AND C. CICERO
A. assimilis [red-shouldered blackbird], a close relative [Garrido and Kirkconnell 1996]; furthermore, A. phoeniceus is
strongly polytypic and two divergent forms, gubernator and
grandis, meet in secondary contact in the Lerma Marshes of
Mexico [Hardy and Dickerman 1965]), and Sturnella magnaS. neglecta (eastern and western meadowlarks; these taxa
need to be compared with the vocally distinct form lillianae,
which is recognized as a species by Monroe and Sibley 1993).
Using the sister species comparisons and molecular data
compiled in Table 1 of this study, we mimicked Klicka and
Zink’s (1997) procedure by plotting a histogram of uncorrected mtDNA divergences to examine the shape of the distribution (Figure 1). For comparison, we also plotted the data
sets from both Klicka and Zink (1997) and Avise and Walker
(1998, Phylogeographic Category I, North American taxa
only). We compared these frequency distributions using Kolmogorov-Smirnov’s two-sample D statistic with STATISTICA for Windows, version 5.1 (StatSoft 1997). In addition,
we tested all three distributions for normality using Kolmogorov-Smirnov’s one-sample D statistic with Lilliefors
(1967) probabilities, which is appropriate when the mean and
standard deviation of the normal distribution is estimated
from the data rather than known a priori. Following Klicka
and Zink (1997) and Avise and Walker (1998), we assumed
that mtDNA evolves in a strictly clocklike manner (2% per
million years).
A confounding issue in these studies is the application of
a correction factor that accounts for ancestral diversity. This
factor corrects for the difference between initial haplotype
divergence (pAB) and lineage sundering (pAB(net)) by estimating net sequence divergence as pAB(net) 5 pAB 2 0.5(pA
1 pB), where pA and pB are values for mean mtDNA sequence
divergence among individuals within groups A and B, respectively, and pAB is the mean sequence divergence between
individuals of these two groups (Avise and Walker 1998;
Klicka and Zink 1999, fig. 1). Klicka and Zink (1997) did
not formally correct their pAB values between ‘‘sister species,’’ although ‘‘evaluation of the distribution of pAB(net)
values formed the basis for [their] rejection of the LPO’’
(Klicka and Zink 1999, p. 696). Avise and Walker (1998)
applied this correction to 14 (34%) of their 41 phylogroup
comparisons, but their plot of mtDNA sequence divergence
does not distinguish between uncorrected versus corrected
data. Avise et al. (1998) found that 71% versus 31% of avian
sister-species pairs (Klicka and Zink 1997) dated to the Pleistocene when corrected by the mean estimated mtDNA divergence of 0.027 (Avise and Walker 1998) between intraspecific phylogroups; this factor is not equivalent to pAB(net)
(Klicka and Zink 1999). Klicka and Zink (1999) criticized
their correction factor of 0.027, stating that this value should
←
FIG. 1. (A) Mitochondrial DNA (mtDNA) divergences between
35 pairs of North American songbird species analyzed by Klicka
and Zink (1997) and specifically theorized by earlier evolutionary
biologists to have late Pleistocene origins; black bars indicate proven sister species used in their analysis (asterisks in Table 1), gray
bars indicate non-sisters or pairs with unresolved sister relationships
(see text). (B) mtDNA divergence times among North American
‘‘phylogroups,’’ defined as genetically distinctive, geographically
oriented populations within species (Phylogeographic Category I,
Avise and Walker 1998); black bars indicate taxa which are either
currently recognized as species or which are probably best treated
as species (see text and Table 1). (C) mtDNA divergence times
between proven sister species, including both songbirds and nonpasserine species pairs (this study). The star notes the speciation
time for the long-billed and short-billed dowitchers. Arrows in all
three graphs denote the Pleistocene/Pliocene boundary.
DNA DIVERGENCE TIMES OF NORTH AMERICAN BIRDS
be reduced to 0.020 (5 pAB(net)) by correcting for intraphylogroup variation (Moore 1995); furthermore, they argued
that the average interphylogroup distance is substantially
lower (0.017 6 0.015) when only North American passerines
are considered. Although Klicka and Zink (1999) plotted
mean corrected mtDNA divergence estimates for North
American songbirds using the method of Avise and Walker
(1998), it is unclear from their wording whether they calculated pAB(net) for all 63 pairwise comparisons or whether
they applied a mean correction factor (and, if so, which correction). In addition, most interspecific studies do not provide
sufficient data to estimate intraspecies variation (and thus
calculate pAB(net)) with confidence. In view of these confounding issues, and because the goal of our study was to
directly compare our data (Table 1) to the plottings of Klicka
and Zink (1997) and Avise and Walker (1998), we did not
correct mtDNA divergence estimates for the sister species
pairs in Table 1. Although this makes it difficult to evaluate
our results in light of Klicka and Zink’s (1999), the use of
raw sequence divergence values should facilitate comparison
for future studies.
The distribution of our dataset was dramatically different
from that of Klicka and Zink (1997) (Fig. 1; D 5 20.555,
P , 0.001). Although Klicka and Zink (1997) could not
distinguish their plotting from a normal distribution (we reanalyzed their data and corroborated that result, D 5 0.125,
P , 0.20), our distribution was significantly non-normal (D
5 0.159, P , 0.05) with values strongly skewed toward
younger (i.e., Pleistocene) mtDNA divergence times. Furthermore, our distribution was not significantly different from
that of North American ‘‘phylogroups’’ per Avise and Walker
(1998) (Fig. 1; D 5 20.212, P . 0.10), which also was
skewed away from normality (D 5 0.207, P , 0.01). The
latter result has a twofold explanation: (1) our emphasis on
proven sister species yields relatively young mtDNA divergence times that overlap with those of phylogeographic lineages enroute to speciation; and (2) some of the ‘‘phylogroups’’ analyzed by Avise and Walker (1998) are either
currently recognized as biological species (e.g., Picoides tridactylus [P. tridactylus-P. dorsalis, Eurasian and American
three-toed woodpeckers], Baeolophus bicolor [B. bicolor-B.
atricristatus, tufted and black-crested titmice], Baeolophus
inornatus [B. inornatus-B. ridgwayi, oak and juniper titmice],
Pica pica [P. pica-P. hudsonia, common and black-billed
magpies], Pipilo erythrophthalmus [P. erythrophthalmus-P.
maculatus, eastern and spotted towhees], Ammodramus caudacutus [A. caudacutus-A. nelsoni, saltmarsh and Nelson’s
sharp-tailed sparrows]; American Ornithologists’ Union 1998
and supplements) or are probably best treated as species (e.g.,
Branta canadensis [Canada goose], Ammodramus maritimus
[seaside sparrow], Passerella iliaca [fox sparrow]).
Of the 39 sister species pairs analyzed in our study, only
one (short-billed and long-billed dowitchers at 8.2% divergence; asterisk in Fig. 1) was of late Pliocene origin. This
result is surprising because these sibling species have extremely similar plumages. However, it is worth noting that
this unusual result is based on RFLP (Avise and Zink 1988)
and has not been corroborated independently with mtDNA
sequencing. Six pairs (15%) showed mtDNA divergence values between 3.8 and 4.9% and thus split in the vicinity of
1127
the late Pliocene-early Pleistocene boundary (1,900,000–
2,400,000 years ago). The remaining 32 pairs (82% of the
total) diverged during the Pleistocene. More important, however, are the numerous mtDNA divergences now evident in
the last quarter of the Pleistocene. Indeed, at least 17 pairs
(43.6%) diverged in the last 450,000 years. Of these, 11 pairs
(28.2% of the total) split less than 250,000 years ago in the
late Pleistocene, and one pair (gray-headed and Oregon juncos) showed no mtDNA divergence and could therefore be
of very recent, even Holocene, origin. It is worth noting here
that application of a correction factor to these estimates (Avise and Walker 1998; Avise et al. 1998; Klicka and Zink
1999) would only serve to skew our data further toward Pleistocene speciation times.
Although we cannot directly compare our data to that of
Klicka and Zink (1999) because of reasons mentioned above,
their distribution of 63 pairwise comparisons showed that
83% have divergence dates older than 500,000 years, and
43% have dates that fall outside of the Pleistocene. These
data are difficult to interpret because they compared taxa at
very different stages of separation (from intraspecific lineages
to non-sister relatives). In contrast, our results support the
conclusion that the Pleistocene was an important time for
divergences of North American sister species, and that the
entire epoch—including the last two glacial cycles
(,250,000 years ago)—was involved. The new data also
demonstrate the need for including all potential relatives
when using molecular divergence values to estimate speciation times, and especially the analysis of proven sister taxa
when investigating the youngest speciation events. This finding by no means excludes the continued relevance of mtDNA
differences for estimating the timing of speciation in nonsisters that occurred earlier in congeneric clades of three or
more species. Overall, the numbers of analyzed sister taxa
are still too small to reveal possible pulses of divergence over
the last three to four million years. Additional data for other
sister species comparisons, and for multilocus as well as single-locus (mtDNA) markers, will continue to shed light on
this debate.
The estimation of speciation times from calibrated molecular distances between sister species is not without its pitfalls.
Five caveats, among others, contribute to imprecision and
ambiguity: (1) The calibration of molecular clocks continues
to be problematic (e.g., Avise 1994, 2000), and different
molecular markers and methods (e.g., RFLP, sequencing)
may yield variable results. (2) Our sister species comparisons
are between living taxa, and recent extinctions easily could
have obliterated some very young sister species splits. (3)
Single-locus estimates of divergence have large standard errors (Edwards and Beerli 2000). (4) Because of ancestral
polymorphisms, mtDNA gene divergences—the basis for calculation of speciation times in most studies—antedate population divergences or speciations (Edwards and Beerli 2000;
Arbogast et al. 2002); thus, actual speciation times for all
pairs of birds considered here are younger by an unknown
figure than indicated by direct molecular distance comparisons. Although the discrepancies between gene and population or species divergences are unknown in single locus
(mtDNA) studies, such discrepancies could be measured and
better estimates obtained of actual species divergence times
1128
N. K. JOHNSON AND C. CICERO
using multilocus nuclear sequence data and new analytical
methods (reviewed in Arbogast et al. 2002; Brumfield et al.
2003); workers should focus further work in this direction.
(5) Finally, many studies use single or few sequences per
species to estimate relationships and mtDNA divergences,
and such limited sampling yields uncertainty in whether these
sequences have achieved reciprocal monophyly (an assumption that is implicit when estimating divergence times); if
points of divergence represent divergences between two alleles that belong to populations of alleles that are not reciprocally monophyletic, gene and species divergences will not
correspond (S. V. Edwards, pers. comm.). Although we regard estimates of speciation times using mtDNA divergences
to be tentative for the above reasons, they nonetheless allow
a first-pass exploration into this important evolutionary issue,
especially when compared directly across studies (e.g.,
Klicka and Zink 1997; Avise and Walker 1998) using similar
methods.
Avise and Walker (1998) presented an original method for
the calculation of speciation durations to which our data can
be referred. They viewed speciation as an extended temporal
process, rather than as a point event, and suggested that upper
and lower bounds of speciation times can be estimated using
molecular data as follows: molecular divergence between extant sister species provides a maximum estimate of the time
when speciation might have been initiated; and molecular
divergence between intraspecific phylogroups provides a
minimum estimate of evolutionary time that has occurred
without completion of speciation. According to their logic,
speciation duration therefore must lie somewhere between
these upper and lower limits, that is, at the midpoint. Applying this method to their own data and that of Klicka and
Zink (1997), Avise and Walker (1998) concluded that the
Pleistocene was important both in completing avian speciations that began in the Pliocene and in initiating phylogeographic subdivisions within species. Furthermore, Avise et
al. (1998) found that the average speciation time in birds as
well as other vertebrate groups, that is, the midpoint between
mtDNA divergences of sister taxa and those of intraspecific
phylogroups, is approximately two million years. In our view,
which agrees with that of Klicka and Zink (1999), but for
other reasons, this estimate of the average duration of speciation may be substantially inflated. With few exceptions
(e.g., Ammodramus), the set of species used to determine
dates of initiation of speciation (or maximum speciation
times) differs from the set of species used to calculate minimum estimates of speciation durations based on phylogroup
mtDNA divergences. Therefore, the overall data reflect a time
span over which many taxa have split, each with a much
briefer duration of speciation than shown by the average of
the combined group, rather than the time span elapsed within
a single species lineage from the initiation of speciation to
its completion. In addition, their estimation of the midpoint
was based on maximum mtDNA divergence times for both
sister and non-sister comparisons (Klicka and Zink 1997);
inclusion of non-sisters clearly would inflate these results.
Our report of relatively recent mtDNA divergences of numerous proven sister species clearly indicates that at least
the youngest species pairs considered here, including teal,
rails, sapsuckers, titmice, warblers, sparrows, juncos, bun-
tings, crossbills and redpolls, split from their ancestors and
completed speciation within the last 250,000 years or less.
These data prove that speciation in birds can be relatively
rapid under certain circumstances.
ACKNOWLEDGMENTS
We are indebted to G. Barrowclough and Jeff Groth for
allowing us to include their unpublished data for juncos and
crossbills, respectively. J. Avise, G. Barrowclough, S. Edwards, and one anonymous reviewer offered many useful
comments that greatly improved the manuscript. We invited
J. Klicka and R. Zink to read the manuscript before submission but they declined, stating conflict of interest. Nonetheless, we appreciate the valuable discussions of this topic that
they offered after our presentation at a meeting and in subsequent e-mails.
LITERATURE CITED
American Ornithologists’ Union. 1998. Check-list of North American birds. 7th ed. American Ornithologists’ Union, Washington,
DC.
Arbogast, B. S., and J. B. Slowinski. 1998. Pleistocene speciation
and the mitochondrial DNA clock. Science 282:1955a.
Arbogast, B. S., S. V. Edwards, J. Wakeley, P. Beerli, and J. B.
Slowinski. 2002. Estimating divergence times from molecular
data on phylogenetic and population genetic timescales. Annu.
Rev. Ecol. Syst. 33:707–740.
Avise, J. C. 1994. Molecular markers, natural history and evolution.
Chapman and Hall, New York.
———. 2000. Phylogeography: the history and formation of species. Harvard Univ. Press, Cambridge, MA.
Avise, J. C., and W. S. Nelson. 1989. Molecular genetic relationships of the extinct dusky seaside sparrow. Science 243:
646–648.
Avise, J. C., and D. Walker. 1998. Pleistocene phylogeographic
effects on avian populations and the speciation process. Proc.
R. Soc. Lond. B 265:457–463.
Avise, J. C., and R. M. Zink. 1988. Molecular genetic divergence
between avian sibling species: king and clapper rails, long-billed
and short-billed dowitchers, boat-tailed and great-tailed grackles, and tufted and black-crested titmice. Auk 105:516–528.
Avise, J. C., D. Walker, and G. C. Johns. 1998. Speciation durations
and Pleistocene effects on vertebrate phylogeography. Proc. R.
Soc. Lond. B 265:1707–1712.
Ball, R. M., and J. C. Avise. 1992. Mitochondrial DNA phylogeographic differentiation among avian populations and the evolutionary significance of subspecies. Auk 109:626–636.
Beecher, W. J. 1955. Late-Pleistocene isolation in salt-marsh sparrows. Ecology 36:23–28.
Bermingham, E. S., S. Rohwer, S. Freeman, and C. Wood. 1992.
Vicariance biogeography in the Pleistocene and speciation in
North American wood warblers: a test of Mengel’s model. Proc.
Nat. Acad. Sci. USA 89:6624–6628.
Brewer, R. 1963. Ecological and reproductive relationships of
black-capped and Carolina chickadees. Auk 80:9–47.
Brodkorb, P. 1971. Origin and evolution of birds. Pp. 20–55 in D.
S. Farner and J. R. King, eds. Avian biology Vol. 1. Academic
Press, New York.
Brumfield, R. T., P. Beerli, D. A. Nickerson, and S. V. Edwards.
2003. The utility of single nucleotide polymorphisms in inferences of population history. Trends Ecol. Evol. 18:249–256.
Burns, K. 1998. Molecular phylogenetics of the genus Piranga:
implications for biogeography and the evolution of morphology
and behavior. Auk 115:621–634.
Carson, R. J., and G. S. Spicer. 2003. A phylogenetic analysis of
the emberizid sparrows based on three mitochondrial genes. Mol.
Phylogenet. Evol. 29:43–57.
DNA DIVERGENCE TIMES OF NORTH AMERICAN BIRDS
Cicero, C. 1996. Sibling species of titmice in the Parus inornatus
complex (Aves: Paridae). Univ. Calif. Publ. Zool. 128:1–217.
Cicero, C., and N. K. Johnson. 1995. Speciation in sapsuckers
(Sphyrapicus): III. Mitochondrial-DNA sequence divergence at
the cytochrome-b locus. Auk 112:547–563.
———. 1998. Molecular phylogeny and ecological diversification
in a clade of New World songbirds (genus Vireo). Mol. Ecol.
7:1359–1370.
Dillon, L. S. 1956. Wisconsin climate and life zones in North America. Science 123:167–176.
Edwards, S. V., and P. Beerli. 2000. Perspective: gene divergence,
population divergence, and the variance in coalescence time in
phylogeographic studies. Evolution 54:1839–1854.
Garrido, O., and A. Kirkconnell. 1996. Taxonomic status of the
Cuban form of the red-winged blackbird. Wilson Bull. 108:
372–374.
Hardy, J. W., and R. W. Dickerman. 1965. Relationships between
two forms of the red-winged blackbird in Mexico. Living Bird
4:107–129.
Hubbard, J. P. 1973. Avian evolution in the aridlands of North
America. Living Bird 12:155–196.
Johnson, K. P., and D. H. Clayton. 2000. A molecular phylogeny
of the dove genus Zenaida: mitochondrial and nuclear DNA
sequences. Condor 102:864–870.
Johnson, K. P., and S. M. Lanyon. 1999. Molecular systematics of
the grackles and allies, and the effect of additional sequence
(cyt b and ND2). Auk 116:759–768.
Johnson, K. P., and M. D. Sorenson. 1999. Phylogeny and biogeography of dabbling ducks (Genus: Anas): a comparison of molecular and morphological evidence. Auk 116:792–805.
Johnson, N. K., and C. Cicero. 2002. The role of ecologic diversification in sibling speciation of Empidonax flycatchers (Tyrannidae): multigene evidence from mtDNA. Mol. Ecol. 11:
2065–2081.
Johnson, N. K., J. V. Remsen, Jr.,and C. Cicero. 1999. Resolution
of the debate over species concepts in ornithology: a new comprehensive biologic species concept. Pp. 1470–1482 in N. H.
Adams and R. H. Slotow, eds. Proceedings of the 22nd International Ornithological Congress, BirdLife South Africa, Johannesburg.
Kahn, N. W., C. E. Braun, J. R. Young, S. Wood, D. R. Mata, and
T. W. Quinn. 1999. Molecular analysis of genetic variation
among large- and small-bodied sage grouse using mitochondrial
control region sequences. Auk 116:819–824.
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.
Klicka, J. and R. M. Zink. 1997. The importance of recent ice ages
in speciation: a failed paradigm. Science 277:1666–1669.
———. 1999. Pleistocene effects on North American songbird evolution. Proc. R. Soc. Lond. B 266:695–700.
Klicka, J., R. M. Zink, J. C. Barlow, W. B. McGillivray, and T. J.
Doyle. 1999. Evidence supporting the recent origin and species
status of the timberline sparrow. Condor 101:577–588.
Klicka, J., A. J. Fry, R. M. Zink, and C. W. Thompson. 2001a. A
cytochrome-b perspective on Passerina bunting relationships.
Auk 118:611–623.
Klicka, J., R. M. Zink, J. C. Barlow, W. B. McGillivray, and T. J.
Doyle. 2001b. The taxonomic rank of Spizella taverneri: a response to Mayr and Johnson. Condor 103:420–422.
Klicka, J., R. M. Zink, and K. Winker. 2003. Longspurs and snow
buntings: phylogeny and biogeography of a high-latitude clade
(Calcarius). Mol. Phylogenet. Evol. 26:165–175.
Lanyon, S. M. 1994. Polyphyly of the blackbird genus Agelaius and
the importance of assumptions of monophyly in comparative
studies. Evolution 48:679–693.
Lilliefors, H. W. 1967. On the Kolmogorov-Smirnov test for normality with mean and variance unknown. J. Am. Stat. Assoc.
64:399–402.
Lovette, I. J., and E. Bermingham. 1999. Explosive speciation in
the New World Dendroica warblers. Proc. R. Soc. Lond. B 266:
1629–1636.
1129
———. 2001. Mitochondrial perspective on the phylogenetic relationships of the Parula wood-warblers. Auk 118:211–215.
Lovette, I. J., E. Bermingham, S. Rohwer, and C. Wood. 1999.
Mitochondrial restriction fragment length polymorphism (RFLP)
and sequence variation among closely related avian species and
the genetic characterization of hybrid Dendroica warblers. Mol.
Ecol. 8:1431–1441.
Mayr, E. 1942. Systematics and the origin of species from the
viewpoint of a zoologist. Columbia Univ. Press, New York.
Mayr, E., and N. K. Johnson. 2001. Is Spizella taverneri a species
or a subspecies? Condor 103:418–419.
Mengel, R. M. 1964. The probable history of species formation in
some northern wood warblers (Parulidae). Living Bird 3:9–43.
———. 1970. The North American Central Plains as an isolating
agent in bird speciation. Pp. 279–340 in W. Dort, Jr., and J. K.
Jones, Jr., eds. Pleistocene and Recent environments of the Central Great Plains. Dept. Geol. Univ. Kansas Spec. Publ. 3, Univ.
Press of Kansas, Lawrence, KS.
Monroe, B. L., Jr., and C. G. Sibley. 1993. A world checklist of
birds. Yale Univ. Press, New Haven, CT.
Moore, W. S. 1995. Inferring phylogenies from mtDNA variation:
mitochondrial-gene trees versus nuclear-gene trees. Evolution
49:718–726.
Omland, K. E., S. M. Lanyon, and S. J. Fritz. 1999. A molecular
phylogeny of the New World orioles (Icterus): the importance
of dense taxon sampling. Mol. Phylogenet. Evol. 12:224–239.
Piertney, S. B., R. W. Summers, and M. Marquiss. 2001. Microsatellite and mitochondrial DNA homogeneity among phenotypically diverse crossbill taxa in the UK. Proc. R. Soc. Lond.
B 268:1511–1517.
Rand, A. L. 1948. Glaciation, an isolating factor in speciation.
Evolution 2:314–321.
Rising, J. D., and J. C. Avise. 1993. An application of genealogical
concordance principles to the taxonomy and evolutionary history
of the sharp-tailed sparrow (Ammodramus caudacutus). Auk 110:
844–856.
Selander, R. K. 1965. Avian speciation in the Quaternary. Pp. 527–
542 in H. E. Wright, Jr. and D. G. Frey, eds. The quaternary of
the United States. Princeton Univ. Press, Princeton, NJ.
———. 1971. Systematics and speciation in birds. Pp. 57–147 in
D. S. Farner and J. R. King, eds. Avian biology. Vol. 1. Academic Press, New York.
Selander, R. K., and D. R. Giller. 1961. Analysis of sympatry of
great-tailed and boat-tailed grackles. Condor 63:29–86.
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 A. C. Wilson. 1987. Calibration of mitochondrial
DNA evolution in geese. J. Mol. Evol. 24:212–217.
StatSoft, Inc. 1997. STATISTICA for Windows. Ver. 5.1. StatSoft,
Inc., Tulsa, OK.
Weckstein, J. D., R. M. Zink, R. C. Blackwell-Rago, and D. A.
Nelson. 2001. Anomalous variation in mitochondrial genomes
of white-crowned (Zonotrichia leucophrys) and golden-crowned
(Z. atricapilla) sparrows: pseudogenes, hybridization, or incomplete lineage sorting? Auk 118:231–236.
Wetmore, A. 1959. Birds of the Pleistocene in North America.
Smithson. Misc. Collect. 138:1–24.
Zink, R. M., and R. C. Blackwell. 1998. Molecular systematics and
biogeography of aridland gnatcatchers (genus Polioptila) and
evidence supporting species status of the California gnatcatcher
(Polioptila californica). Mol. Phylogenet. Evol. 9:26–32.
Zink, R. M., and D. L. Dittman. 1991. Evolution of brown towhees:
mitochondrial DNA evidence. Condor 93:98–105.
Zink, R. M., and J. Klicka. 1999. Avian evolution during the Pleistocene in North America. Pp. 275–288 in M. Kato, ed. The
biology of biodiversity. Springer-Verlag, Tokyo.
Zink, R. M., and J. B. Slowinski. 1995. Evidence from molecular
systematics for decreased avian diversification in the Pleistocene
Epoch. Proc. Natl. Acad. Sci. USA 92:5832–5835.
Zink, R. M., D. L. Dittman, and W. L. Rootes. 1991. Mitochondrial
1130
N. K. JOHNSON AND C. CICERO
DNA variation and the phylogeny of Zonotrichia. Auk 108:
578–584.
Zink, R. M., R. C. Blackwell, and O. Rojas-Soto. 1997. Species
limits in the LeConte’s thrasher. Condor 99:132–138.
Zink, R. M., S. J. Weller, and R. C. Blackwell. 1998. Molecular
phylogenetics of the avian genus Pipilo and a biogeographic
argument for taxonomic uncertainty. Mol. Phylogenet. Evol. 10:
191–201.
Zink, R. M., D. L. Dittman, J. Klicka, and R. C. Blackwell-Rago.
1999. Evolutionary patterns of morphometrics, allozymes, and
mitochondrial DNA in thrashers (genus Toxostoma). Auk 116:
1021–1038.
Corresponding Editor: K. Shaw