- Research article
- Open Access
Dating the diversification of the major lineages of Passeriformes (Aves)
© Ericson et al.; licensee BioMed Central Ltd. 2014
- Received: 23 July 2013
- Accepted: 2 January 2014
- Published: 15 January 2014
The avian Order Passeriformes is an enormously species-rich group, which comprises almost 60% of all living bird species. This diverse order is believed to have originated before the break-up of Gondwana in the late Cretaceous. However, previous molecular dating studies have relied heavily on the geological split between New Zealand and Antarctica, assumed to have occurred 85–82 Mya, for calibrating the molecular clock and might thus be circular in their argument.
This study provides a time-scale for the evolution of the major clades of passerines using seven nuclear markers, five taxonomically well-determined passerine fossils, and an updated interpretation of the New Zealand split from Antarctica 85–52 Mya in a Bayesian relaxed-clock approach. We also assess how different interpretations of the New Zealand–Antarctica vicariance event influence our age estimates. Our results suggest that the diversification of Passeriformes began in the late Cretaceous or early Cenozoic. Removing the root calibration for the New Zealand–Antarctica vicariance event (85–52 Mya) dramatically increases the 95% credibility intervals and leads to unrealistically old age estimates. We assess the individual characteristics of the seven nuclear genes analyzed in our study. Our analyses provide estimates of divergence times for the major groups of passerines, which can be used as secondary calibration points in future molecular studies.
Our analysis takes recent paleontological and geological findings into account and provides the best estimate of the passerine evolutionary time-scale currently available. This time-scale provides a temporal framework for further biogeographical, ecological, and co-evolutionary studies of the largest bird radiation, and adds to the growing support for a Cretaceous origin of Passeriformes.
- Molecular dating
- Fossil calibrations
- New Zealand–Antarctica vicariance
Passeriformes (passerines) is the largest and most diverse avian order, comprising about 5,700 species and representing almost 60% of all living birds. Due to their ubiquity and enormous diversity, passerines have been the focus of many ecological, behavioral, anatomical and evolutionary studies. The systematic relationships, early evolution and biogeography of passerines have long been debated among avian taxonomists. While disagreeing on many other aspects of passerine systematics and evolution, most ornithologists of the 20th century agreed that the Passeriformes is one of the youngest avian orders. Therefore, it was rather surprising when molecular data proposed that the evolutionary history of these birds dated back to the end of the Cretaceous [1, 2]. Under the current paradigm, passerines arose on the southern supercontinent Gondwana and major passerine lineages became isolated with the continental break-up. This hypothesis rests on an interpretation of the modern distributions of the major passerine clades: the New World suboscines in South and Central America (a dispersal to North America seems to have been facilitated by the formation of the Isthmus of Panama 3 Mya), the Old World suboscines in tropical Africa and Asia (except Sapayoa aenigma, which occurs in west Colombia), and all basal members of the oscines in the Australo–Papuan region. Furthermore, the endemic New Zealand wrens (Acanthisittidae) form the sister group to all other passerines. It has been postulated that the ancestors of the New Zealand wrens became isolated when the Zealandia continental fragment separated from Antarctica 85–82 Mya [3, 4].
The supposed role of Gondwana in the diversification of passerines implies a late Cretaceous origin of this radiation. Several studies have used the 85–82 Mya date for the New Zealand–Antarctica vicariance event to calibrate estimates of the passerine evolutionary time-scale, typically in the absence of suitable fossils. There are two problems with this calibration method. First, it builds on the assumption that the New Zealand wrens were actually present on Zealandia when it separated from Antarctica. Under this assumption, the argument for a Cretaceous origin of passerines becomes circular. Second, it has been suggested that the last land connection is considerably younger than the postulated 85–82 Mya date that has been used to calibrate the passerine tree. Recent geological data suggest that the separation of New Zealand began about 85 mya but was not complete until about 55–52 mya [5–7]. Dispersal possibilities for terrestrial organisms may even have existed as late as the Paleogene . Some geologists have argued that the entire landmass of New Zealand was completely submerged during the marine transgression in the Oligocene [9, 10]. Phylogeographic and paleontological data, however, suggest that endemic terrestrial biota were present in the Oligocene, even though New Zealand was reduced to a few scattered islands or an estimated 18% of the current landmass [11, 12] and papers cited therein].
The purpose of this paper is two-fold. First, we aim to provide an updated time-scale for the evolution of the major passerine clades. Second, we aim to present age estimates for major groups of passerines, which can be used as secondary calibration points in future phylogenetic studies of passerines at lower taxonomic levels.
Our estimates of the passerine phylogeny and evolutionary time-scale are based on 7,193 bp obtained from seven nuclear genes. We calibrate the molecular clock using both fossils and the New Zealand–Antarctica vicariance event described above, but with an updated interpretation of the time frame of this geological event [5–7]. Unlike many previous analyses, we also incorporate the uncertainty in these calibrations, and assess the sensitivity of our age estimates to different interpretations of the New Zealand–Antarctica vicariance event.
Selection of taxa and gene sequences
The ingroup consists of 55 taxa and includes representatives of almost all of the basal lineages of passerines identified in previous molecular phylogenies (e.g., ). Most of the sequence data are from Alström et al. , Barker et al. [1, 13], Ericson et al. [2, 15, 16], Ericson and Johansson , Fjeldså et al. [18, 19], Fuchs et al. [20, 21], Gelang et al. , Irestedt et al. [23–25], Irestedt and Ohlson , Johansson et al. , Jønsson et al. , Ohlson et al. , Zuccon and Ericson  and Zuccon et al. . We have also added 25 novel sequences to the data set, which we have deposited in GenBank (Additional file 1: Table S1). Although some lineages are represented in the analyses by sequences obtained from different individuals, the concatenated sequences are in most cases from conspecific or congeneric individuals (Additional file 1: Table S1). Only for the Meliphagidae, Callaeidae and Petroicidae were we forced to combine sequences from more than one genus.
The total alignment consists of 7,193 bp obtained from seven genes: MOS, oocyte maturation factor mos (622 bp in the alignment/177 parsimony-informative characters), MYC, c-myc proto-oncogene exon 3 (504/83 bp), GAPDH, glyceraldehyde-3-phosphodehydrogenase intron 11 (419/222 bp), MB, myoglobin intron 2 (800/299 bp), ODC1, ornithine decarboxylase introns 6–7 (749/316 bp), RAG1, recombination activating protein 1 (2 947/740 bp), and RAG2, recombination activating protein 2 (1 152/326 bp). Details of extraction, amplification and sequencing procedures are given for MOS by Cooper and Penny , for MYC by Irestedt et al. , for MB by Irestedt et al. , for GAPDH by Fjeldså et al. , and for ODC1 by Allen and Omland .
Bayesian phylogenetic inference
For Bayesian inference of phylogeny and divergence times, the program MrBayes v.3.2  was used to obtain Markov Chain Monte Carlo (MCMC) approximations of posterior tree distributions. Gene partitions were analyzed both separately and in concatenation. In the concatenated analyses, the dataset was partitioned by gene and by codon position (combined first and second versus third codon positions). Nucleotide substitution models were unlinked across partitions, and a reversible-jump MCMC over the space of all GTR sub-models was run for each of them (“nst = mixed” command in MrBayes ). Among-site rate variation was modeled using a discrete gamma distribution with four categories and a proportion of invariant sites . We used partition-specific rate multipliers with a Dirichlet-distributed prior to allow the overall evolutionary rates to differ among partitions.
For all analyses, four Metropolis-coupled chains (temperature constant set to 0.1) were run for a minimum of 30 million generations, sampling every 1,000th generation. Four independent runs for the final analyses were conducted with the preferred clock model (see below), whereas two runs were done for the remaining analyses. Parameter and tree files were analyzed using Tracer  and AWTY  to check for convergence issues and suitable burn-in. Average standard deviations of split frequencies fell below the recommended value of 0.01 after about 5 million generations, indicating good topology convergence. After being scrutinized for convergence, tree and parameter files from separate runs were then concatenated to maximize precision. Calculations of credibility intervals, statistical analyses, and graphical output were generated using the R statistical programming language , utilizing functions in R-packages APE , PHANGORN , and PHYLOCH .
Clock and tree model choice
We used stepping-stone sampling  to obtain estimates of the marginal likelihoods to choose between different clock models: the strict clock, a relaxed, auto-correlated log-normal clock (TK02 ), and a relaxed, uncorrelated gamma clock model (the white-noise or independent gamma rates (IGR) model ). We considered convergence to be acceptable when the standard deviation of split frequencies was below 0.04 in each of the steps. To achieve this, we ran 100,000,000 generations in total, distributed over 49 steps between the posterior and the prior, treating the first step as a burn-in. Within each step, 500,000 generations were discarded as a burn-in, and samples were drawn every 1,000 generations over the remaining 1,500,000 generations. Bayes factor comparisons strongly preferred the uncorrelated gamma relaxed clock model over the auto-correlated and strict clock models (2* ln(Bayes factor) equals 14.3 and 132.5, respectively ), and we used this model for subsequent analyses. In order to enhance convergence of the clock rate as well as tree height, we followed an empirical Bayes approach . We initially ran a strict-clock analysis without calibrations in order to obtain an estimate for the tree height in substitution units. This value was then divided by the average tree height in million years according to the prior on the root (see below) in order to get a first estimate of the clock rate. This estimate was used as the median for a log-normal prior with a large standard deviation of 1, leading to an only weakly informative prior. For the priors on the variance of the relaxed-clock models, we used an exponential (10) prior.
Instead of relying on explicit models of the branching process to obtain a prior for the tree, we used the recently proposed extension of the uniform tree prior to calibrated clock settings . Very little is known about the relative performance of different tree priors (with one notable exception ), but given the incomplete and uneven taxon sampling in our dataset, we preferred to use the uniform prior which is rather uninformative. Furthermore, it does not require the setting of additional parameters as in the birth-death or coalescent priors, and does not require any assumptions to be made about the sampling strategy.
Calibration of the rate of evolution
To calculate absolute ages from the trees, two categories of calibration points can be used: geological vicariance events and fossils. The only geological vicariance event used in our analysis is based on the assumption that the New Zealand wrens became isolated when this continental fragment separated from Antarctica. However, it is now currently understood that separation of New Zealand was not complete until about 55–52 Mya [5–8, 49]. In addition to using this best current estimate in the form of a uniform prior on the root age, we tested the previously used range of 85–82 Mya  and a de facto removal of the root calibration (uniform prior between 4,000 and 20 Mya).
Presence of the six calibrated nodes was enforced using monophyly constraints in the dating analyses. These nodes had high support in a previous, uncalibrated and unconstrained analysis (see electronic supplementary material). In addition to the nodes used for calibration, we invoked a rooting constraint between Acanthisitta and the remaining passerines. This was because a preliminary, uncalibrated relaxed-clock analysis showed the clock model to be insufficient to recover the correct root [35, 57].
The Bayesian analysis using an uncorrelated gamma relaxed clock model (IGR) produced a phylogenetic estimate that agrees strongly with that obtained from an analysis without a clock and with previous studies based on nuclear DNA data (e.g., [13, 58]) (Figure 1, for node numbers see Additional file 2: Figure S1). Most clades in the tree have high posterior probabilities (Additional file 3: Table S2), with the exceptions occurring in the same parts of the tree that have previously been difficult to resolve.
The tree was rooted with the New Zealand wrens (Acanthisittidae), with the remaining taxa forming two clades, the suboscines and the oscines. The suboscines are divided into two reciprocally monophyletic clades consisting of taxa distributed in the New World and Old World, respectively. The New World suboscines are further divided into Furnariida and Tyrannida. Within each of these groups the inferred phylogenetic relationships agree well with those from previous studies (e.g., [29, 59–62]).
Within the oscines, the delimitation of Passerida has been much discussed [13, 17, 26, 27]. The question concerns the relationships between Eupasseri (the core group of Passerida) and the five families Picathartidae (rockfowl), Chaetopidae (rockjumpers), Eupetidae (rail-babblers) (not included in this study), Petroicidae (Australasian robins) and Callaeidae (New Zealand wattlebirds). With the addition of more nuclear markers, Picathartidae and Chaetopidae form a sister pair that is placed in an unresolved clade with Petroicidae, Callaeidae and Eupasseri. This tree topology does not contradict the hypothesis that the insertion of one amino acid in MYC in Picathartidae, Chaetopidae, Callaeidae and Eupasseri, but not in Petroicidae, may be a synapomorphy for Passerida [17, 63]. The well-supported parts of the inferred relationships within Passerida all agree with previous phylogenetic studies of this radiation (e.g., [13, 17]).
Sensitivity of divergence-time estimates to the root prior
The prior specified for the root age had a considerable effect on the posterior of the age of the passerine radiation. As expected, the 95% credibility intervals were narrowest when the most restrictive prior was used, i.e., a uniform distribution between 85 and 82 Mya for the root node (posterior: median = 83.5, 95% credibility interval = 85.0–82.2 Mya). Widening the prior interval to 85–52 Mya, which better reflects our current understanding of the geological events leading to the separation of New Zealand from Antarctica [5–7], we obtained a posterior median around 73.6 Mya (85.0–59.0 Mya). When de facto removing the root calibration, the median estimate increased to 90.8 Mya, but with a very large uncertainty (125.6–49.4 Mya).
Performance of individual genes for divergence-time estimation
An alternative way of examining congruency of estimates among genes, in combination with precision, is to evaluate the extent of overlap in 95% credibility intervals for individual nodes among genes (Additional file 4: Figure S2). Despite the large variation observed, there is a fairly good overlap between individual gene estimates. For some nodes, both congruence and precision are high for the majority of genes.
The tree topology inferred from the Bayesian analysis of the combined data agrees well with previous estimates based on multiple nuclear and mitochondrial markers. The taxa that are difficult to place phylogenetically in the present analysis have also been problematic in other studies. Increasing the number of nuclear genes in the data set did not help to resolve the relative positions of Pitta–Calyptomena–Smithornis cf. , Pipra–Rupicola cf. [29, 61, 64], and Petroicidae vs. Picathartes–Chaetops cf. [26, 28, 65]. Furthermore, as in previous analyses (e.g., [1, 13, 17, 66]), Parus and Bombycilla cannot be confidently placed. This suggests that genome-scale sequence data are necessary to resolve these splits, or that the diversification in certain parts of the tree occurred too rapidly to be resolved with confidence.
Utility of individual genes for divergence-time estimates
The factors determining the utility of a gene for topological inference have been investigated in several studies (e.g., [67–73]), but there has been less attention on branch-length estimation . However, the latter is crucial for understanding the performance of different genes in estimating node ages. In our analyses, we found considerable differences among genes in their ability to provide congruent (Figure 3) and precise (Figure 4) divergence-time estimates. These gene-specific performances also varied over branches in the tree (Additional file 4: Figure S2).
One factor that is likely to affect a gene's performance is the size of the data set. As the size of the data set grows, sampling error and the width of the 95% credibility intervals will tend to decrease (down to a limit set by the prior ). This improvement in precision with an increased amount of data is apparent in our analyses (Figure 4, Additional file 3: Figure S1). Interestingly, lower precision seemed to be confined to a particular time period in the tree (ca. Oligocene; Figure 1). The uncertainty in divergence-time estimates for this time period affected all genes, which is inconsistent with the notion that “some genes are better” for different ages.
Another factor that might affect performance of a gene is its substitution rate. A method for predicting the utility of a gene is to examine its saturation levels. A saturated gene would reach a plateau in pair-wise sequence distances over time. Using this approach, we did not find pronounced saturation in any of the individual genes or the combined data (Additional file 5: Figure S3). As an example, the MYC gene, which seems to be performing less well than other genes, is not the most rapidly evolving among the genes included in our study, but has genetic distances similar to the combined data. Hence, saturation, on its own, does not seem to be a good predictor of gene performance.
Other factors that can affect the performance are clock- and/or substitution-model inadequacy. The models used in the analyses might fail to capture the actual patterns of rate variation and correctly estimate branch lengths and divergence times. This might explain the tendency to over- or underestimate divergence times at different depths in the tree (Figure 3).
Finally, different genes might support different topologies (gene trees), making them less congruent or precise when compared to a concatenated tree (“species tree”). In this study, we assume that the gene trees evolved on the same topology by concatenating all of the genes in a single analysis. However, we could quantify the incongruence by looking at gene-specific estimates and identifying regions of conflict (Additional file 4: Figure S2). This approach allows us to focus on clades for which dates were estimated with high precision, and where the majority of the genes have high overlap in their 95% credibility intervals.
Calibrating the passerine phylogeny
The increased availability of molecular data allows us to estimate phylogenetic relationships and relative branch lengths with unprecedented precision, but dating analyses still mostly depend on how the molecular clock is calibrated [76, 77]. While novel approaches that directly use fossil data are certainly preferable [35, 78], they require high-quality morphological datasets for both fossil and extant taxa which are currently not available for most groups. Divergence-time estimates thus still largely rely on an adequate translation of the fossil record or of certain vicariance hypotheses into priors for node ages .
Our study took a combined approach, using available relevant passerine fossils in combination with an interpretation of the New Zealand-Antarctica vicariance event that is in better accordance with the most recent geological evidence [5–7]. We also used two additional root calibrations to assess sensitivity. The resulting age estimates were sensitive to the vicariance calibration, an outcome that could be expected for a calibration at the root. Additionally, calibrations based on vicariance events, in contrast to fossils, also provide a maximum age constraint . Removing the root calibration altogether led to inflated 95% credibility intervals and older median ages, indicating that the mean used for the fossil calibrations (obtained from 55, see Methods) was not sufficient to provide upper bounds on the node ages. In contrast, when we used the traditional, narrow interpretation of the vicariance event, i.e., an age of the Acanthisitta split of 85–82 Mya, we obtained much more precise age estimates for the passerine radiation. When widening the prior to 85–52 Mya, which more closely matches the current interpretation of the history of the break-up of eastern Gondwana [5–8], divergence-time estimates became somewhat less precise, but younger. Given that divergence-time estimates relying on a narrow interpretation of the New Zealand vicariance event are probably biased, we have focused on the results obtained using the more relaxed prior on the age of the passerine root. By using both the relevant passerine fossils and an up-to-date interpretation of the New Zealand–Antarctica vicariance event, our analysis arguably provides the best estimate of the passerine evolutionary time scale.
Divergence times between major groups of Passerines
Our estimated dates of divergences (Figure 1) are generally younger than those from previous studies based on a large sample of passerine species (e.g. , see also ). Barker et al.  used a fixed age (82 Mya) for the split between New Zealand and Antarctica as a single calibration point for the passerine tree, which may explain the difference between the estimated divergence times compared to our study.
Comparison of divergences times for five major clades of passerines as estimated in our study and in previous studies
A: Present study
B: Present study
C: Pereira & Baker (2006)
D: Barker et al. (2004)
E: Barker et al. (2004)
F: Ericson et al. (2002)
Split Suboscines vs. Oscines (node 2)
84 – 58
123 – 49
103 – 81
81 – 73
80 – 72
79 – 62
Split OW vs. NW Suboscines (node 35)
77 – 50
110 – 44
78 – 65
73 – 61
78 – 65
Oscines (node 3)
71 – 46
104 – 41
70 – 60
67 – 57
70 – 60
Core Corvoidea (node 24)
47 – 28
65 – 25
45 – 35
42 – 32
45 – 35
Eupasseri (node 9)
53 – 33
74 – 30
54 – 15
49 – 36
54 – 15
Seven nuclear genes (7,204 bp)
Seven nuclear genes (7,204 bp)
Two nuclear genes (4,126 bp)
Two nuclear genes (4,126 bp)
Two nuclear genes (1,407 bp)
Five passerine fossils and one vicariance event, split between New Zealand and Antarctica (85–52 mya),
Five passerine fossils and essentially no time-constraint for New Zealand and Antarctica vicariance event (4,000–20 Mya)
Thirteen fossils, of which five avian (no passerine)
One vicariance event, split between New Zealand and Antarctica (82 mya)
One vicariance event, split between New Zealand and Antarctica (82 mya)
One vicariance event, split between New Zealand and Antarctica (82 mya)
Dating method (program)
IGR relaxed clock, Bayesian inference (MrBayes 3.2)
IGR relaxed clock, Bayesian inference (MrBayes 3.2)
Bayesian rate autocorrelation (Multidivtime)
Penalized likelihood (r8s)
Nonparametric rate smoothing (r8s)
Local rates method (Rhino)
Within the suboscines, the Old World and New World taxa form two monophyletic clades, with the exception of the Colombian species Sapayoa aenigma (Broad-billed Sapayoa) which has been shown to belong to the Old World clade . It has been speculated that the ancestors of the New World suboscines became isolated in South America when it split from Antarctica around the end of the Cretaceous . The ancestor of the Old World suboscines may have colonized the African and Indian land masses (at this time already separated from Antarctica) from the West Antarctic Peninsula via island chains on the Kerguelen and Crozet Plateaus (along the Mascarene, Maldive and Ninety East Ridges) . The mean date for the split between the suboscines in the Old World and New World is estimated at 63.6 Mya (77–50 Mya) (Table 1, column A). This is 5–7 My younger than previous estimates  (Table 1, columns D and E), which raises doubts about the biogeographic scenario suggested for the dispersal of the Old World suboscines from Antarctica to Africa and India.
Some biogeographic analyses have suggested that the oscines evolved in Australia after its split from Antarctica [1, 2]. In our study, the mean date estimate for the earliest split in extant oscines is 59.4 Mya (71–46 Mya). Previous estimates by Barker et al.  are 3–5 My older (Table 1, columns D and E). The core Corvoidea clade began to radiate at 37.4 Mya (47–28 Mya) according to our estimates, compared to 40.4 and 38.6 Mya, as estimated by Barker et al.  using two different algorithms.
Eupasseri consists of all species of Passerida except Petroicidae, Picathartidae, Eupetidae and Callaeidae, all of which systematists at some point have included in Passerida (e.g.,  and papers cited therein). We estimate the age of Eupasseri at 43.5 Mya (53–33 Mya), which is only 1–3 My younger than the estimates by Barker et al.  (Table 1, columns D and E).
The hypothesis of a Cretaceous origin for the passerine radiation is controversial, as this date is much older than would be expected from the fossil record [83–85]. The main criticism is the use of the New Zealand–Antarctica vicariance event for calibrating the tree. The rationale for using this calibration is the distinct Gondwanan signature of the passerine phylogeny [80, 86], with the basal clades distributed in different southern continents. However, both overall phylogenetic patterns and molecular clock estimates support an old age of the passerine radiation. By applying the passerine evolutionary rate for mitochondrial genes , Barker et al.  estimated the age of the split between the New Zealand wrens and all other passerines at 87 Mya. Our data support a Cretaceous origin of Passeriformes when we apply recent estimates of gene-specific evolutionary rates by Lerner et al.  to nuclear genes analyzed in our study. Lerner et al.  used island ages to obtain a dated phylogeny and rates of evolution for a number of genes in a study of Hawaiian honeycreepers. Three of the genes examined in their study were also included in our analyses. For each gene, we estimated the root age using these rates and using the respective tree heights in substitutions from our main analysis (using the rate multiplier in MrBayes). Even though this yielded rather crude estimates, the overall patterns are consistent with those from our calibrated analysis (estimate from RAG1: 115.8 Mya, 137–97 Mya; GAPDH: 99.2 Mya, 119–81 Mya; ODC1: 56.2 Mya, 67–47 Mya).
Divergence-time estimates suitable for use as secondary calibrations for studies of passerine evolution
Most phylogenetic analyses of passerines do not include Acanthisitta, which precludes the use of the split between New Zealand and Antarctica to calibrate the tree. Owing to the paucity of the passerine fossil record and the difficulty with which they can be taxonomically identified below suborder level, there are few fossil passerines that can provide useful calibrations for molecular studies of divergence times. Until our understanding of geological vicariance events improves and more fossils are discovered and described, we are confined to using secondary calibration points to estimate divergence times within the passerine radiation. The extent of overlap in 95% credibility intervals of age estimates from different genes gives an idea of the suitability of an individual node as a secondary calibration point. Despite the large variations in 95% credibility intervals between genes, there is a fairly good overlap between individual gene estimates for most nodes (Additional file 4: Figure S2). We find that several nodes receiving a high posterior probability (1.0) in the phylogeny presented here provide good candidates for secondary calibration points, particularly the suboscines, oscines, core Corvoidea and Eupasseri (Figure 2, Table 1). Divergence-time estimates from our combined dataset can be used in further molecular analyses of the passerine phylogeny.
Implications of a Cretaceous age for the passerine radiation
Molecular data suggest that the earliest diversification of passerines took place in the late Cretaceous. In studies that use a narrow 85–82 Mya span for the New Zealand–Antarctica vicariance event to calibrate the passerine phylogeny [1, 2], this conclusion is inevitably reached by default. However, we also inferred a late Cretaceous age for the earliest evolution of passerines even when widening the age range for the vicariance event to the more realistic 85–52 Mya span and including passerine fossils for calibration (Table 1, column A). Furthermore, when we removed the influence of the New Zealand–Antarctica vicariance event and relied solely on the fossils, the median age of the passerine root fell in the Cretaceous (Table 1, column B), albeit with a wide 95% credibility interval reaching well into the Cenozoic. Cretaceous ages for the earliest passerine diversifications are also implied when using independent estimates of substitution rates for individual genes. Barker et al.  showed this for the mitochondrial cytochrome b gene, and our analyses yield similar results for two out of the three nuclear genes that were calibrated in this manner.
Another observation consistent with a Cretaceous origin of the passerines is that their phylogenetic relationships reveal a biogeographic pattern that has a clear Gondwanan signature. The basal oscines, the New World suboscines and the Old World suboscines, are all confined to continents that were once part of Gondwana. Although passerines could easily have dispersed, obscuring their biogeographic patterns, birds surprisingly often exhibit strong biogeographic patterns that are closely linked to their evolutionary history. This is true regardless of whether we study phylogeographic patterns within a species or geographic distributions of families and other higher-level taxa. The current distributions of the basal oscines, New World suboscines and Old World suboscines may be important in revealing the earliest history of the passerines, and this may also be true for the current distribution of the New Zealand wrens.
A Cretaceous origin for the passerine radiation has far-reaching implications, as it suggests the fossil record is severely incomplete [83, 84]. The oldest passerine fossils are currently known from the early Eocene (ca 55 Mya) of Australia , which may represent stem passerines . The oldest fossils in the Northern Hemisphere date to the early Oligocene (34–32 Mya) of Europe (see  and references therein). These fossils include representatives thought to be outside crown group Eupasseres [90, 91], a suboscine-like passerine , and nearly complete skeletons of passerines of unknown affinities [93–95]. The earliest fossil passerines that can be confidently assigned to an extant family are from the early Miocene, including a New Zealand wren  and the fossils used to calibrate the nodes in our analyses [51–53]. It is probable that much of the early diversification took place in the southern continents from where comparatively few fossil sites from that time period are known. For example, there is an absence of passerine fossils in avifaunas from late Cretaceous and Paleogene sites in South America, and there is a significant gap in the Australian fossil record between the early Eocene and late Oligocene.
Mayr  has remarked that a late Cretaceous age of the passerines makes this radiation unique among endothermic vertebrates in having remained morphologically “virtually unchanged for 80 million years”. This is a major reason for why he is skeptical of the suggested age for the group, in addition to the lack of Cretaceous fossils. Passerines are indeed a morphologically uniform group, but this is mostly true for body parts that do not directly relate to locomotion and feeding. In passerines, the morphology associated with locomotion and feeding has shown substantial evolutionary plasticity. Following the advent of DNA-based taxonomy, the literature has been filled with examples of groups that have evolved similar adaptations by convergence (e.g., ), which has misled avian systematists for hundreds of years. Unfortunately the fossil record of passerines is silent on this point because very little is known about the morphology associated with feeding and locomotion in the few Paleogene passerines found.
In this study, we used a Bayesian relaxed-clock approach to estimate the evolutionary time-scale of the major clades of passerines based on seven nuclear genes, five passerine fossils, and an updated interpretation of the New Zealand split from Antarctica. We found no support from molecular data or from the overall biogeographic patterns of the passerines to refute the possibility of a Gondwanan origin, and thus a Cretaceous age, of the group. Our results represent the best estimate of the passerine evolutionary time-scale currently available and add to the growing support for the early Cenozoic diversification of this order. This temporal framework can be used in further biogeographical, ecological, and co-evolutionary studies of the Passeriformes. Furthermore, our analysis provides estimates of divergence times for major groups of passerines, which can be used as secondary calibration points in future molecular studies.
The data sets supporting the results of this article are available in the GenBank (DNA sequences, accession nos. KF905607-KF905630) and TreeBASE (TreeBase link Study Accession URL: http://purl.org/phylo/treebase/phylows/study/TB2:S15185). Also see Additional file 6: text file S1 for MrBayes command files, including clock model settings and calibration information.
This work was supported by the Swedish Research Council (grants no. 621-2007-5280 and 621-2010-5321 to P.E.).
- Barker FK, Barrowclough GF, Groth JG: A phylogenetic hypothesis for passerine birds: taxonomic and biogeographic implications of an analysis of nuclear DNA sequence data. Proc R Soc Lond B. 2002, 269: 295-308. 10.1098/rspb.2001.1883.View ArticleGoogle Scholar
- Ericson PGP, Christidis L, Cooper A, Irestedt M, Jackson J, Johansson US, Norman JA: A Gondwanan origin of passerine birds supported by DNA sequences of the endemic New Zealand wrens. Proc R Soc Lond B. 2002, 269: 235-241. 10.1098/rspb.2001.1877.View ArticleGoogle Scholar
- Sutherland R: Basement geology and tectonic development of the greater New Zealand region: an interpretation from regional magnetic data. Tectonophysics. 1999, 308: 341-362. 10.1016/S0040-1951(99)00108-0.View ArticleGoogle Scholar
- Cooper RA, Millener PR: The New Zealand biota: historical background and new research. Trends Ecol Evol. 1993, 8: 429-433. 10.1016/0169-5347(93)90004-9.PubMedView ArticleGoogle Scholar
- Gaina C, Müller DR, Royer JY, Stock J, Hardebeck J, Symonds P: The tectonic history of the Tasman Sea: a puzzle with 13 pieces. J Geophys Res. 1998, 103: 12413-12433. 10.1029/98JB00386.View ArticleGoogle Scholar
- Gaina C, Müller RD, Roest WR, Symonds P: The opening of the Tasman Sea: a gravity anomaly animation. Earth interact. 1998, 2: 1-23.View ArticleGoogle Scholar
- Schellart WP, Lister GS, Toy VG: A Late Cretaceous and Cenozoic reconstruction of the Southwest Pacific region: tectonics controlled by subduction and slab rollback processes. Earth-Sci Rev. 2006, 76: 191-233. 10.1016/j.earscirev.2006.01.002.View ArticleGoogle Scholar
- Ladiges PY, Cantrill D: New Caledonia-Australian connection: biogeographic patterns and geology. Aust Syst Bot. 2007, 20: 383-389. 10.1071/SB07018.View ArticleGoogle Scholar
- Landis CA, Campbell HJ, Begg JG, Mildenhall DC, Paterson AM, Trewick SA: The Waipounamu Erosion Surface: questioning the antiquity of the New Zealand land surface and terrestrial fauna and flora. Geol Mag. 2008, 145: 173-197.View ArticleGoogle Scholar
- Waters JM, Craw D: Goodbye Gondwana? New Zealand biogeography, geology, and the problem of circularity. Syst Biol. 2006, 55: 351-356. 10.1080/10635150600681659.PubMedView ArticleGoogle Scholar
- Bunce M, Worthy TH, Phillips MJ, Holdaway RN, Willerslev E, Haile J, Shapiro B, Scofield RP, Drummond A, Kamp PJJ, Cooper A: The evolutionary history of the extinct ratite moa and New Zealand Neogene paleogeography. Proc Natl Acad Sci. 2009, 106: 20646-20651. 10.1073/pnas.0906660106.PubMedPubMed CentralView ArticleGoogle Scholar
- Worthy TH, Hand SJ, Nguyen JMT, Tennyson AJD, Worthy JP, Scofield RP, Boles WE, Archer M: Biogeographical and phylogenetic implications of an Early Miocene wren (Aves: Passeriformes: Acanthisittidae) from New Zealand. J Vert Paleontol. 2010, 30: 479-498. 10.1080/02724631003618033.View ArticleGoogle Scholar
- Barker FK, Cibois A, Schikler P, Feinstein J, Cracraft J: Phylogeny and diversification of the largest avian radiation. Proc Natl Acad Sci USA. 2004, 101: 11040-11045. 10.1073/pnas.0401892101.PubMedPubMed CentralView ArticleGoogle Scholar
- Alström P, Ericson PGP, Olsson U, Sundberg P: Phylogeny and classification of the avian superfamily Sylvioidea. Mol Phylogenet Evol. 2006, 38: 381-397. 10.1016/j.ympev.2005.05.015.PubMedView ArticleGoogle Scholar
- Ericson PGP, Christidis L, Irestedt M, Norman JA: Systematic affinities of the lyrebirds (Passeriformes: Menura), with a novel classification of the major groups of passerine birds. Mol Phylogenet Evol. 2002, 25: 53-62. 10.1016/S1055-7903(02)00215-4.PubMedView ArticleGoogle Scholar
- Ericson PGP, Olson SL, Irestedt M, Alvarenga H, Fjeldså J: Circumscription of a monophyletic family for the tapaculos (Aves: Rhinocryptidae): Psiloramphus in and Melanopareia out. J Ornithol. 2010, 151: 337-345. 10.1007/s10336-009-0460-9.View ArticleGoogle Scholar
- Ericson PGP, Johansson US: Phylogeny of Passerida (Aves: Passeriformes) based on nuclear and mitochondrial sequence data. Mol Phylogenet Evol. 2003, 29: 126-138. 10.1016/S1055-7903(03)00067-8.PubMedView ArticleGoogle Scholar
- Fjeldså J, Zuccon D, Irestedt M, Johansson US, Ericson PGP: Sapayoa aenigma: a New World representative of 'Old World suboscines'. Proc R Soc Lond B (Suppl.). 2003, 270: S238-S241. 10.1098/rsbl.2003.0075.View ArticleGoogle Scholar
- Fjeldså J, Irestedt M, Ericson PGP, Zuccon D: The Cinnamon Ibon Hypocryptadius cinnamomeus is a forest canopy sparrow. Ibis. 2010, 152: 747-760. 10.1111/j.1474-919X.2010.01053.x.View ArticleGoogle Scholar
- Fuchs J, Fjeldså J, Pasquet E: An ancient African radiation of corvoid birds (Aves: Passeriformes) detected by mitochondrial and nuclear sequence data. Zool Scr. 2006, 35: 375-385. 10.1111/j.1463-6409.2006.00238.x.View ArticleGoogle Scholar
- Fuchs J, Irestedt M, Fjeldså J, Couloux A, Pasquet E, Bowie RC: Molecular phylogeny of African bush-shrikes and allies: tracing the biogeographic history of an explosive radiation of corvoid birds. Mol Phylogenet Evol. 2012, 64: 93-105. 10.1016/j.ympev.2012.03.007.PubMedView ArticleGoogle Scholar
- Gelang M, Cibois A, Pasquet E, Olsson U, Alstrom P, Ericson PGP: Phylogeny of babblers (Aves, Passeriformes): major lineages, family limits and classification. Zool Scr. 2009, 38: 225-236. 10.1111/j.1463-6409.2008.00374.x.View ArticleGoogle Scholar
- Irestedt M, Johansson US, Parsons TJ, Ericson PGP: Phylogeny of major lineages of suboscines (Passeriformes) analysed by nuclear DNA sequence data. J Avian Biol. 2001, 32: 15-25. 10.1034/j.1600-048X.2001.320103.x.View ArticleGoogle Scholar
- Irestedt M, Ohlson JI, Zuccon D, Källersjö M, Ericson PGP: Nuclear DNA from old collections of avian study skins reveals the evolutionary history of the Old World suboscines (Aves, Passeriformes). Zool Scr. 2006, 35: 567-580. 10.1111/j.1463-6409.2006.00249.x.View ArticleGoogle Scholar
- Irestedt M, Fuchs J, Jønsson KA, Ohlson JI, Pasquet E, Ericson PGP: The systematic affinity of the enigmatic Lamprolia victoriae (Aves: Passeriformes) - an example of avian dispersal between New Guinea and Fiji over Miocene intermittent land bridges?. Mol Phylogenet Evol. 2008, 48: 1218-1222. 10.1016/j.ympev.2008.05.038.PubMedView ArticleGoogle Scholar
- Irestedt M, Ohlson JI: The division of the major songbird radiation into Passerida and “core Corvoidea“ (Aves: Passeriformes) - the species tree versus gene trees. Zool Scr. 2008, 37: 305-313. 10.1111/j.1463-6409.2007.00321.x.View ArticleGoogle Scholar
- Johansson US, Bowie RCK, Fjeldså J: Phylogenetic relationships within Passerida (Aves: Passeriformes): a review and a new molecular phylogeny based on three nuclear intron markers. Mol Phylogenet Evol. 2008, 48: 858-876. 10.1016/j.ympev.2008.05.029.PubMedView ArticleGoogle Scholar
- Jønsson KA, Fjeldså J, Ericson PGP, Irestedt M: Systematic placement of an enigmatic Southeast Asian taxon Eupetes macrocerus and implications for the biogeography of a main songbird radiation, the Passerida. Biol Lett. 2007, 3: 323-326. 10.1098/rsbl.2007.0054.PubMedPubMed CentralView ArticleGoogle Scholar
- Ohlson JI, Fjeldså J, Ericson PGP: Tyrant flycatchers coming out in the open: phylogeny and ecological radiation of Tyrannidae (Aves: Passeriformes). Zool Scr. 2008, 37: 315-335. 10.1111/j.1463-6409.2008.00325.x.View ArticleGoogle Scholar
- Zuccon D, Ericson PGP: Molecular and morphological evidences place the extinct New Zealand endemic Turnagra capensis in the Oriolidae. Mol Phylogenet Evol. 2012, 62: 414-426. 10.1016/j.ympev.2011.10.013.PubMedView ArticleGoogle Scholar
- Zuccon D, Prys-Jones R, Rasmussen PC, Ericson PGP: The phylogenetic relationships and generic limits of finches (Fringillidae). Mol Phylogenet Evol. 2012, 62: 581-596. 10.1016/j.ympev.2011.10.002.PubMedView ArticleGoogle Scholar
- Cooper A, Penny D: Mass survival of birds across the Cretaceous–Tertiary boundary: molecular evidence. Science. 1997, 275: 1109-1113. 10.1126/science.275.5303.1109.PubMedView ArticleGoogle Scholar
- Irestedt M, Fjeldså J, Johansson US, Ericson PGP: Systematic relationships and biogeography of the tracheophone suboscines (Aves: Passeriformes). Mol Phylogenet Evol. 2002, 23: 499-512. 10.1016/S1055-7903(02)00034-9.PubMedView ArticleGoogle Scholar
- Allen ES, Omland KE: Novel intron phylogeny (ODC) supports plumage convergence in orioles (Icterus). Auk. 2003, 120: 961-969.View ArticleGoogle Scholar
- Ronquist F, Teslenko M, Van der Mark P, Ayres DL, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP: MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model space. Syst Biol. 2012, 61: 539-542. 10.1093/sysbio/sys029.PubMedPubMed CentralView ArticleGoogle Scholar
- Huelsenbeck JP, Larget B, Alfaro ME: Bayesian phylogenetic model selection using reversible-jump Markov chain Monte Carlo. Mol Biol Evol. 2004, 21: 1123-1133. 10.1093/molbev/msh123.PubMedView ArticleGoogle Scholar
- Yang Z: Among-site rate variation and its impact on phylogenetic analyses. Trends Ecol Evol. 1996, 11: 367-372. 10.1016/0169-5347(96)10041-0.PubMedView ArticleGoogle Scholar
- Rambaut A, Drummond AJ: Tracer v1.4. 2007, Available from http://beast.bio.ed.ac.uk/TracerGoogle Scholar
- Nylander JAA, Wilgenbus JC, Warren DL, Swofford DL: AWTY (Are We There Yet?): A system for graphical exploration of MCMC convergence in Bayesian phylogenetics. Bioinformatics. 2008, 24: 581-583. 10.1093/bioinformatics/btm388.PubMedView ArticleGoogle Scholar
- R Core Team: R: A language and environment for statistical computing. 2012, Vienna, Austria: R Foundation for Statistical Computing, http://www.R-project.org/,Google Scholar
- Paradis E, Claude J, Strimmer K: APE: analyses of phylogenetics and evolution in R language. Bioinformatics. 2004, 20: 289-290. 10.1093/bioinformatics/btg412.PubMedView ArticleGoogle Scholar
- Schliep KP: PHANGORN: Phylogenetic analysis using R. Bioinformatics. 2011, 27: 592-593. 10.1093/bioinformatics/btq706.PubMedPubMed CentralView ArticleGoogle Scholar
- Heibl C: PHYLOCH: R language tree plotting tools and interfaces to diverse phylogenetic software packages. 2008, Available from: http://www.christophheibl.de/Rpackages.htmlGoogle Scholar
- Xie W, Lewis PO, Fan Y, Kuo L, Chen M-H: Improving marginal likelihood estimation for Bayesian phylogenetic model selection. Syst Biol. 2011, 60: 150-160. 10.1093/sysbio/syq085.PubMedPubMed CentralView ArticleGoogle Scholar
- Thorne JL, Kishino H: Divergence time and evolutionary rate estimation with multilocus data. Syst Biol. 2002, 51: 689-702. 10.1080/10635150290102456.PubMedView ArticleGoogle Scholar
- Lepage T, Bryant D, Philippe H, Lartillot N: A general comparison of relaxed molecular clock models. Mol Biol Evol. 2007, 24: 2669-2680. 10.1093/molbev/msm193.PubMedView ArticleGoogle Scholar
- Kass RE, Raftery AE: Bayes Factors. J Am Statist Ass. 1995, 90: 773-795. 10.1080/01621459.1995.10476572.View ArticleGoogle Scholar
- Ronquist F, Klopfstein S, Vilhelmsen L, Schulmeister S, Murray DL, Rasnitsyn AP: A total-evidence approach to dating with fossils, applied to the early radiation of the Hymenoptera. Syst Biol. 2012, 61: 973-999. 10.1093/sysbio/sys058.PubMedPubMed CentralView ArticleGoogle Scholar
- McLoughlin S: The breakup history of Gondwana and its impact on pre-Cenozoic floristic provincialism. Austral Syst Bot. 2001, 49: 271-300. 10.1071/BT00023.View ArticleGoogle Scholar
- Zachos J, Pagani M, Sloan L, Thomas E, Billups K: Trends, rhythms, and aberrations in global climate change 65 Ma to present. Science. 2001, 292: 686-693. 10.1126/science.1059412.PubMedView ArticleGoogle Scholar
- Nguyen JMT, Worthy TH, Boles WE, Hand SJ, Archer M: A new cracticid (Passeriformes: Cracticidae) from the Early Miocene of Australia. Emu. 2013, 113: 374-382. 10.1071/MU13017.View ArticleGoogle Scholar
- Boles WE: A songbird (Aves: Passeriformes: Oriolidae) from the Miocene of Riversleigh, northwestern Queensland, Australia. Alcheringa. 1999, 23: 51-56. 10.1080/03115519908619338.View ArticleGoogle Scholar
- Manegold A: Earliest fossil record of the Certhioidea (treecreepers and allies) from the early Miocene of Germany. J Ornithol. 2008, 149: 223-228. 10.1007/s10336-007-0263-9.View ArticleGoogle Scholar
- Boles WE: Fossil honeyeaters (Meliphagidae) from the Tertiary of Riversleigh, northwestern Queensland. Emu. 2005, 105: 21-26. 10.1071/MU03024.View ArticleGoogle Scholar
- Ericson PGP, Anderson CL, Britton T, Elzanowski A, Johansson US, Källersjö M, Ohlson JI, Parsons TJ, Zuccon D, Mayr G: Diversification of Neoaves: integration of molecular sequence data and fossils. Biol Lett. 2006, 2: 543-547. 10.1098/rsbl.2006.0523.PubMedPubMed CentralView ArticleGoogle Scholar
- Heled J, Drummond AJ: Calibrated tree priors for relaxed phylogenetics and divergence time estimation. Syst Biol. 2012, 61: 138-149. 10.1093/sysbio/syr087.PubMedPubMed CentralView ArticleGoogle Scholar
- Huelsenbeck JP, Bollback JP, Levine AM: Inferring the root of a phylogenetic tree. Syst Biol. 2002, 51: 32-43. 10.1080/106351502753475862.PubMedView ArticleGoogle Scholar
- Jønsson KA, Fjeldså J: Determining biogeographic patterns of dispersal and diversification in oscine passerine birds in Australia, Southeast Asia and Africa. J Biogeogr. 2006, 33: 1155-1165. 10.1111/j.1365-2699.2006.01507.x.View ArticleGoogle Scholar
- Irestedt M, Fjeldså J, Dalén L, Ericson PGP: Convergent evolution, habitat shifts and variable diversification rates in the ovenbird-woodcreeper family (Furnariidae). BMC Evol. Biol. 2009, 9: 268-10.1186/1471-2148-9-268.PubMedPubMed CentralView ArticleGoogle Scholar
- Moyle RG, Chesser RT, Brumfield RT, Tello JG, Marchese DJ, Cracraft J: Phylogeny and phylogenetic classification of the antbirds, ovenbirds, woodcreepers and allies (Aves: Passeriformes: infraorder Furnariides). Cladistics. 2009, 25: 386-405. 10.1111/j.1096-0031.2009.00259.x.View ArticleGoogle Scholar
- Tello JG, Moyle RG, Marchese DJ, Cracraft J: Phylogeny and phylogenetic classification of the tyrant flycatchers, cotingas, manakins and their allies (Aves: Tyrannides). Cladistics. 2009, 25: 429-467. 10.1111/j.1096-0031.2009.00254.x.View ArticleGoogle Scholar
- Ohlson JI, Irestedt M, Ericson PGP, Fjeldså J: Phylogeny and classification of the New World suboscines (Aves, Passeriformes). Zootaxa. 2013, 3613: 1-35.PubMedView ArticleGoogle Scholar
- Ericson PGP, Johansson US, Parsons TJ: Major divisions of oscines revealed by insertions in the nuclear gene c-myc: a novel gene in avian phylogenetics. Auk. 2000, 117: 1077-1086.View ArticleGoogle Scholar
- Ericson PGP, Zuccon D, Ohlson JI, Johansson US, Alvarenga H, Prum RO: Higher level phylogeny and morphological evolution of tyrant flycatchers, cotingas, manakins and their allies (Aves: Tyrannida). Mol Phylogenet Evol. 2006, 40: 471-483. 10.1016/j.ympev.2006.03.031.PubMedView ArticleGoogle Scholar
- Jønsson KA, Irestedt M, Fuchs J, Ericson PGP, Christidis L, Bowie RCK, Norman JA, Pasquet E, Fjeldså J: Explosive avian radiations and multi-directional dispersal across Wallacea: Evidence from the Campephagidae and other Crown Corvida (Aves). Mol Phylogenet Evol. 2008, 47: 221-236. 10.1016/j.ympev.2008.01.017.PubMedView ArticleGoogle Scholar
- Spellman GA, Cibois A, Moyle RG, Winker K, Barker FK: Clarifying the systematics of an enigmatic avian lineage: What is a bombycillid?. Mol Phylogenet Evol. 2008, 49: 691-1044. 10.1016/j.ympev.2008.09.002.View ArticleGoogle Scholar
- Graybeal A: Evaluating the phylogenetic utility of genes: a search for genes informative about deep divergences among vertebrates. Syst Biol. 1994, 43: 174-193.View ArticleGoogle Scholar
- Yang Z: On the best evolutionary rate for phylogenetic analysis. Syst Biol. 1998, 47: 125-133. 10.1080/106351598261067.PubMedView ArticleGoogle Scholar
- Collins TM, Fedrigo O, Naylor GJP: Choosing the best genes for the job: the case for stationary genes in genome-scale phylogenetics. Syst Biol. 2005, 54: 493-500. 10.1080/10635150590947339.PubMedView ArticleGoogle Scholar
- Townsend JP: Profiling phylogenetic informativeness. Syst Biol. 2007, 56: 222-231. 10.1080/10635150701311362.PubMedView ArticleGoogle Scholar
- Fischer M, Steel MA: Sequence length bounds for resolving a deep phylogenetic divergence. J Theoret Biol. 2009, 256: 247-252. 10.1016/j.jtbi.2008.09.031.View ArticleGoogle Scholar
- Klopfstein S, Kropf C, Quicke DLJ: An evaluation of phylogenetic informativeness profiles and the molecular phylogeny of Diplazontinae (Hymenoptera, Ichneumonidae). Syst Biol. 2010, 9: 226-241.View ArticleGoogle Scholar
- Susko E, Roger AJ: The probability of correctly resolving a split as an experimental design criterion in phylogenetics. Syst Biol. 2012, 61: 811-821. 10.1093/sysbio/sys033.PubMedView ArticleGoogle Scholar
- Goldman N: Phylogenetic information and experimental design in molecular systematics. Proc R Soc Lond B. 1998, 265: 1779-1786. 10.1098/rspb.1998.0502.View ArticleGoogle Scholar
- Dos Reis M, Yang Z: The unbearable uncertainty of Bayesian divergence time estimation. J Syst Evol. 2013, 51: 30-43. 10.1111/j.1759-6831.2012.00236.x.View ArticleGoogle Scholar
- Inoue J, Donoghue PCJ, Yang Z: The impact of the representation of fossil calibrations on Bayesian estimation of species divergence times. Syst Biol. 2010, 59: 74-89. 10.1093/sysbio/syp078.PubMedView ArticleGoogle Scholar
- Warnock RCM, Yang Z, Donoghue PCJ: Exploring uncertainty in the calibration of the molecular clock. Biol Lett. 2012, 8: 156-159. 10.1098/rsbl.2011.0710.PubMedPubMed CentralView ArticleGoogle Scholar
- Pyron RA: Divergence time estimation using fossils as terminal taxa and the origins of Lissamphibia. Syst Biol. 2011, 60: 466-481. 10.1093/sysbio/syr047.PubMedView ArticleGoogle Scholar
- Parham JF, Donoghue PCJ, Bell CJ, Calway TD, Head JJ, Holroyd PA, Inoue J, Irmis RB, Joyce WG, Ksepka DT, Patané JSL, Smith ND, Tarver JE, van Tuinen M, Yang Z, Angielczyk KD, Greenwood JM, Hipsley CA, Jacobs L, Makovicky PJ, Müller J, Smith KT, Theodor JM, Warnock RCM, Benton MJ: Best practices for justifying fossil calibrations. Syst Biol. 2012, 61: 346-359. 10.1093/sysbio/syr107.PubMedPubMed CentralView ArticleGoogle Scholar
- Cracraft J, Barker FK: Passeriformes. The Timetree of Life. Edited by: Hedges SB, Kumar S. 2009, New York: Oxford University Press, 423-431.Google Scholar
- Pereira SL, Baker AJ: A mitogenomic timescale for birds detects variable phylogenetic rates of molecular evolution and refutes the standard molecular clock. Mol Biol Evol. 2006, 23: 1731-1740. 10.1093/molbev/msl038.PubMedView ArticleGoogle Scholar
- Harshman J: Passerida. Version 23 June 2008 (under construction). 2008, http://tolweb.org/Passerida/29223/2008.06.23 in The Tree of Life Web Project, http://tolweb.org/Google Scholar
- Feduccia A: ‘Big bang’ for Tertiary birds?. Trends Ecol Evol. 2003, 18: 172-176. 10.1016/S0169-5347(03)00017-X.View ArticleGoogle Scholar
- Mayr G: Paleogene Fossil Birds. 2009, Heidelberg: SpringerView ArticleGoogle Scholar
- Mayr G: The age of the crown group of passerine birds and its evolutionary significance – molecular calibrations versus the fossil record. Syst Biodiv. 2013, 11: 7-13. 10.1080/14772000.2013.765521.View ArticleGoogle Scholar
- Cracraft J: Avian evolution, Gondwana biogeography and the Cretaceous-Tertiary mass extinction event. Proc R Soc Lond B. 2001, 268: 459-469. 10.1098/rspb.2000.1368.View ArticleGoogle Scholar
- Fleischer RC, McIntosh CE, Tarr CL: Evolution on a volcanic conveyor belt: using phylogeographic reconstructions and K–Ar based ages of the Hawaiian Islands to estimate molecular evolutionary rates. Mol. Ecol. 1998, 7: 533-545. 10.1046/j.1365-294x.1998.00364.x.PubMedView ArticleGoogle Scholar
- Lerner HRL, Meyer M, James HF, Hofreiter M, Fleischer RC: Multilocus resolution of phylogeny and timescale in the extant adaptive radiation of Hawaiian honeycreepers. Curr Biol. 2011, 21: 1838-1844. 10.1016/j.cub.2011.09.039.PubMedView ArticleGoogle Scholar
- Boles WE: The world’s oldest songbird. Nature. 1995, 374: 21-22.View ArticleGoogle Scholar
- Mayr G, Manegold A: The oldest European fossil songbird from the early Oligocene of Germany. Naturwissenschaften. 2004, 91: 173-177. 10.1007/s00114-004-0509-9.PubMedView ArticleGoogle Scholar
- Mayr G, Manegold A: New specimens of the earliest European passeriform bird. Acta Palaeont Polonica. 2006, 51: 315-323.Google Scholar
- Mayr G, Manegold A: A small suboscine-like passeriform bird from the early Oligocene of France. Condor. 2006, 108: 717-720. 10.1650/0010-5422(2006)108[717:ASSPBF]2.0.CO;2.View ArticleGoogle Scholar
- Bochenski ZM, Tomek T, Bujoczek M, Wertz K: A new passerine bird from the early Oligocene of Poland. J Ornithol. 2011, 152: 1045-1053. 10.1007/s10336-011-0693-2.View ArticleGoogle Scholar
- Bochenski ZM, Tomek T, Wertz K, Swidnicka E: The third nearly complete passerine bird from the early Oligocene of Europe. J Ornithol. 2013, 154: 923-931. 10.1007/s10336-013-0958-z.View ArticleGoogle Scholar
- Bochenski ZM, Tomek T, Swidnicka E: The first complete leg of a passerine bird from the early Oligocene of Poland. Acta Palaeontol Polonica. in pressGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.