Supporting Information
Species interactions constrain geographic range expansion through
evolutionary time
Alex L. Pigot* and Joseph A. Tobias
*To whom correspondence should be addressed. E-mail
alex.pigot@zoo.ox.ac.uk
Contents
Expanded Materials and Methods (Appendix S1)
Supplementary References
Supplementary Figures (S1–S4)
Data Tables (S1)
Statistical Tables (S2–S8)
1
Appendix S1 Expanded Materials and Methods
Sister species ages
We obtained sister species pairs and estimated divergence times from a recently published
molecular phylogeny of Furnariidae (Derryberry et al. 2011). This is unusually well sampled
for a diverse radiation, containing 97% of currently described species (N = 285), and highly
resolved, with more than 80% of nodes having >0.95 posterior probability (Table S1).
Derryberry et al. (2011) generated a maximum clade credibility (MCC) tree based on
multiple mitochondrial and nuclear genes and dated nodes using a relaxed clock Bayesian
approach in BEAST v. 1.5.2. Substitution model, rate heterogeneity, and base frequencies
were unlinked across partitions. No restrictions were placed on the topology so that
topological uncertainty was factored into the divergence date estimates.
In the absence of a detailed fossil record, Derryberry et al. (2011) used biogeographic
events to place priors on divergence times, including the closure of the Panamanian Isthmus
(3 million years ago [Ma]) and uplift of the Andes Cordillera (3.6 Ma). Priors were applied to
nodes separating sister species with current distributions either side of these barriers. While
absolute rates of secondary sympatry will be sensitive to inaccuracies in divergence date
estimates, this will apply equally across all lineages and thus our assessment of how rates of
secondary sympatry vary across lineages in response to divergence in the α-niche are unlikely
to be biased. Systematic biases in the estimation of divergence dates are also unlikely to
explain our key findings regarding the temporal dynamics of secondary sympatry. If the
timing of recent nodes were systematically underestimated then this could potentially lead to
an apparent speed-up in the rate of secondary sympatry with time since divergence. However,
under this scenario rates of secondary sympatry would be expected to have accelerated with
time since speciation across all species pairs regardless of habitat affiliations. Instead, our
2
analysis shows that rates of secondary sympatry have only accelerated with time since
divergence in sister species with conserved β-niches, a pattern only expected if biotic
interactions limit sympatry. In fact, it is generally expected that divergence times for recent
nodes will tend to be overestimated compared to older nodes because of the comparatively
large effect of ancestral polymorphism for young species (Edwards & Beerli 2000; Weir
2006). Inaccuracies in divergence estimates are therefore likely to make our analysis
conservative with respect to detecting speed-ups in the rate of secondary sympatry with time
since speciation.
Geographic data
Sister species were assigned as either sympatric (with overlapping breeding distributions) or
allopatric (with non-overlapping breeding distributions) (Mayr 1942) according to vector
polygon range maps kindly provided by C.D.L Orme and I.P.F Owens (Orme et al. 2005).
Avian distribution maps were based on the avian taxonomy of Sibley and Monroe (1990) and
so we updated species ranges to reflect recent taxonomic revisions, matching the distribution
data to the lineages represented in the Furnaridae tree. Range polygons were edited in
ArcMap v10.
We used a binary classification of sympatry, rather than the magnitude of overlap,
because we were interested simply in whether species are able to co-exist or not (Weir &
Price 2011). We did not distinguish between cases of parapatry (abutting geographic ranges)
or allopatry (geographically isolated) because for the purpose of our analysis both have the
same pattern of non-overlapping ranges. While the range maps we use provide the most
detailed information on species distributions available at these scales, they tend to
overestimate the extents of species occurrence, particularly in regions with complex
topography (Hurlbert & Jetz 2007). We therefore followed previous studies (e.g. Weir &
3
Price 2011) in using additional literature searches to further refine estimates of range overlap.
This removed spurious cases of sympatry (N = 32) caused by mapping inaccuracies, or low
mapping resolution. In most of these cases, the degree of sympatry predicted by range
polygons was limited to only very localised co-occurrence (< 10% of the smaller species
range) (Fig S1). In a few cases, however, range overlap estimated from polygon data was
high while secondary sources revealed that the species were not in sympatry (e.g. Cinclodes
palliatus and C. atacamensis). This was usually associated with occurrence in the Andes,
where turnover in species distributions occurs over finer spatial scales than the resolution of
polygon maps. Sister species range sizes and their areas of overlap (km2) were calculated
using the mapTools and PBSmapping libraries in R version 2.10.1 (Team, 2010).
Species Β-niches
We quantified β-niches using information on species macro-habitat affinities and elevational
range limits (Table S1) taken from Stotz et al. (1996) and supplemented with additional
sources for species representing recent taxonomic revisions. To maximise sample size, we
used three relatively coarse habitat categories (forest [F1–F15], scrub/grassland [N1–N14]
and aquatic habitats, including lakes/rivers/coasts [A1–A12]) when assigning β-niches (codes
refer to habitat types given in Stotz et al. [1996]). We used the first two categories listed by
Stotz et al. (1996) for each species, and treated these as primary and secondary habitat types.
We used additional sources from primary literature to update data for a small number of
poorly known species, or for recent taxonomic revisions. We note that our results remained
qualitatively unchanged when using the finer habitat classifications within the broad
categories outlined above (Table S2, S3).
4
Estimating rates of secondary sympatry
We modelled the process of secondary sympatry using continuous time multi-state Markov
models. This class of models is appropriate for processes that can be described as a set of
transition events between states but where the exact times of these events are not actually
observed. The approach has been widely used in the phylogenetic literature to model the
evolution of discrete character traits (e.g. Pagel 1994; Maddison & Maddison 2011). Here we
use it to model sympatry as a combined trait of two sister species rather than a property of an
individual lineage. Each sister pair contributes two observations: the geographic state at the
time of population divergence and that of the present day. We assume all species pairs were
initially allopatric. This assumption is valid for birds because there is substantial evidence
that allopatry (or parapatry) is the predominant mode of speciation, with sympatric speciation
being very rare (Coyne & Price 2000; Phillimore et al. 2008; Price 2008). We considered a
model with two states, allopatry and sympatry, and use maximum likelihood to estimate the
rate of transition from allopatry to sympatry. All models were implemented in R using the
packages msm (Jackson 2011) and nhm (Titman 2011).
An advantage of using multi-state Markov models to estimate rates of sympatry
amongst sister species is that it is robust to differences in the distribution of species ages
arising from variation in speciation and/or extinction rates. Weir and Price (2011) used an
alternative method based on the ages of the youngest and oldest sympatric and allopatric
nodes in the phylogenetic tree to provide upper and lower bounds on the time required to
attain sympatry. However, this method may bias estimated times to sympatry because (1) the
observed age of sympatric nodes is constrained by the overall distribution of divergence
times, and (2) sympatry need only be attained by any two member species of sister clades
potentially confounding rate estimates with clade richness. Our method avoids these
limitations because we estimate transition rates across the age distribution of both sympatric
5
and allopatric lineages, and because we focus only on sister species pairs. Importantly, this
has the further advantage of allowing us to examine the temporal patterns of secondary
sympatry.
Estimating the return to allopatry
The instantaneous probabilities of moving between states are governed by a set of transition
intensities (qallo-sym and qsym-allo), where the order of subscripted states denotes the direction of
movement (allopatry to sympatry, and vice versa). In our standard models presented in the
main text (Constant-Rate, Rate-Switch and Time-Variable models) we treat the movement
from allopatry to sympatry as an irreversible process by fixing qsym-allo = 0. Following the
attainment of sympatry, however, some species potentially undergo a contraction or shift in
their geographic ranges leading to a new phase of allopatry. We explored this possibility
using a State-Reversible model in which we allowed qsym-allo to vary and thus estimated the
rate at which species pairs return to allopatry following a period in sympatry. Estimated
return rates to allopatry were extremely slow and did not differ from zero at the 95%
confidence level (Table S5). We compared the fit of the State-Reversible model to our
Constant-Rate model using a likelihood ratio test and failed to reject our initial assumption
that the transition rate from sympatry to allopatry is zero. This was true regardless of whether
we included all species pairs or only those with conserved β-niches, and was robust to
variations in the threshold defining distances between species ranges (Table S5). Across
Furnariid sister pairs, secondary sympatry therefore appears to be a one-way process.
Quantifying the effects of species interactions and dispersal constraints
We examined the role of biotic interactions on the rate of secondary sympatry using two
complementary methods. First, we tested whether rates of secondary sympatry depend on the
morphological distance separating species. We note that although present day species
6
contrasts will tend to exceed those prevailing during the course of divergence, here we are
simply interested in the shape of the relationship between sympatry rates and morphological
distances. Our models therefore assume that while rates of morphological divergence may
vary across lineages, species pairs share a common model of trait divergence. Second, we
tested whether rates of secondary sympatry vary with time since speciation. We focus on a
simple ‘Rate-Switch model’ in which an initial constant rate (S1), switches to a different rate
(S2) at time ts (Figure S2). We estimated the maximum likelihood break point by fitting our
models using different values of ts in increments of 100 kyr from 0.1 Myr to a maximum
value (tmax) determined by the distribution of sister species ages (Figure S3). Specifically, we
only considered break point ages where at least 10 sister pairs occurred after the break point
because smaller sample sizes prevented model convergence and reliable parameter estimates
(Table S2). We highlight that the timing of the shift in the rate of secondary sympatry is
relative to the timing of speciation and not an absolute date in the past. In a selection of sister
species of varying age these rate shifts will therefore have occurred at various times in the
past (Figure S2).
Assuming that the strength of ecological interactions between species is inversely
related to the time since their divergence (Figure S4a), it may be possible to use the rates
estimated under the Rate-Switch model to partition the delay in the timing of secondary
sympatry arising due to biotic interactions and dispersal constraints (Figure S4). The rates
(S1) and waiting times to secondary sympatry (W1) amongst recently diverged species prior
to the break point will be influenced by the combined effects of both dispersal constraints and
strong biotic interactions (Figure S4a–c). In contrast, the rates (S2) and waiting times to
secondary sympatry (W2) amongst long diverged species after the break point will be
influenced primarily by dispersal constraints because biotic interactions will presumably be
weaker (Figure S4a–c). The difference between the observed waiting time to sympatry (W)
7
and that expected due to dispersal constraints (W2) thus provides an estimate for the delay in
secondary sympatry arising due to biotic interactions (ΔW = W – W2) (Figure S4d). We note
that estimates of ΔW were robust to data selection procedures, including the geographic
distance separating sister species (Table S2).
Clade-wide shifts in rates of sympatry
It is plausible that a single clade-wide and contemporaneous shift in the rate of secondary
sympatry could generate a pattern analogous to that expected from species interactions. To
test this idea we compared the fit of the Constant-Rate model to a Clade-Rate-Switch model
in which an initial rate S1 can switch to a higher or lower rate S2 at time ta in the past (Figure
S2b). We highlight that in contrast to the Rate-Switch model (Figure S2a), the timing of the
shift in the rate of secondary sympatry occurs at an absolute date in the past. In a selection of
sister species of varying age these rate shifts will therefore have occurred at various times
relative to the timing of speciation (Figure S2b). We denote the timing of the rate shift as ta
(rather than ts in the Rate-Switch model [Figure S2a]) to highlight that this represents an
absolute date in the past where all sister pairs in existence experienced a shift in rate. We
explored different values of ta from 10 Ma to 1 Ma in increments of 1 Myr, selecting the
switch point with the highest likelihood as the best model. In addition to considering a single
shift in rates, we also fit a Clade-Time-Variable model in which rates are allowed to change
continuously through time.
Models in which the rate of secondary sympatry has undergone a clade-wide decline
provide a better fit to the date that assuming a constant arte of secondary sympatry (Table
S8). However, we view biogeographical explanations for this finding as highly unlikely for
two main reasons. First, the furnariids are spread across the Neotropics, from Tierra del
Fuego to Mesoamerica and from sea level to over 5000m in elevation (Derryberry et al.
8
2011). Species pairs are thus separated by a multitude of different environmental gradients
and/or biogeographic barriers with contrasting histories (Table S1). Second, bird species are
known to respond to these barriers individualistically, with speciation and recolonisation
occurring at different rates according to variation in ecology, and in particular dispersal
limitation (Burney & Brumfield 2009). For example, avian lineages dispersing both
southwards and northwards after the closure of the Isthmus of Panama have tended to
produce speciation and recolonisation events at seemingly random intervals (Dacosta &
Klicka 2008; Miller et al. 2008). This lack of synchronicity is unlikely to generate bursts of
secondary sympatry amongst sister species at a given point in time, and instead is expected to
generate heterogeneity in rates of secondary sympatry over time (Figure 1b). A more
plausible explanation for any possible clade-wide slowdown in the rate of sympatry could be
related to niche filling in highly diverse and ecologically saturated communities (Rundell &
Price 2009). Although to our knowledge this pattern has never been detected, our finding that
rates of secondary sympatry are reduced amongst ecologically similar species is consistent
with the idea that range expansions are likely to become increasingly constrained as niche
space is saturated.
9
Supplemental References
Burney C.W. & Brumfield R.T. (2009). Ecology predicts levels of genetic differentiation in
Neotropical birds. Am. Nat., 174, 358-368.
Coyne J.A. & Price T.D. (2000). Little evidence for sympatric speciation in island birds.
Evolution, 54, 2166-2171.
Dacosta J.M. & Klicka J. (2008). The Great American Interchange in birds: a phylogenetic
perspective with the genus Trogon. Mol. Ecol., 17, 1328-1343.
Derryberry E.P., Claramunt S., Derryberry G., Chesser R.T., Cracraft J., Aleixo A. et al.
(2011). Lineage diversification and morphological evolution in a large-scale
continental radiation: the Neotropical ovenbirds and woodcreepers (Aves:
Furnariidae). Evolution, 65, 2973-2986.
Edwards S.V. & Beerli P. (2000). Perspective: gene divergence, population divergence, and
the variance in coalescence time in phylogeographic studies. Evolution, 54, 18391854.
Hurlbert A.H. & Jetz W. (2007). Species richness, hotspots, and the scale dependence of
range maps in ecology and conservation. P. Natl. Acad. Sci. USA, 104, 13384-13389.
Jackson C.H. (2011). Multi-state models for panel data: The msm package for R. J. Stat.
Softw., 38, 1-28.
Maddison W.P. & Maddison D.R. (2011). Mesquite: a modular system for evolutionary
analysis. v2.75. http://mesquiteproject.org.
Mayr E. (1942). Systematics and the origin of species. Columbia University Press, New
York.
Miller M.J., Bermingham E., Klicka J., Escalante P., Do Amaral F.S.R., Weir J.T. et al.
(2008). Out of Amazonia again and again: episodic crossing of the Andes promotes
diversification in a lowland forest flycatcher. Proc. R. Soc. B, 275, 1133-1142.
Orme C.D.L., Davies R.G., Burgess M., Eigenbrod F., Pickup N., Olson V.A. et al. (2005).
Global hotspots of species richness are not congruent with endemism or threat.
Nature, 436, 1016-1019.
Pagel M. (1994). Detecting correlated evolution on phylogenies: a general method for the
comparative analysis of discrete characters. Proc. R. Soc. B, 255, 37-45.
Phillimore A.B., Orme C.D.L., Thomas G.H., Blackburn T.M., Bennett P.M., Gaston K.J. et
al. (2008). Sympatric speciation in birds is rare: insights from range data and
simulations. Am. Nat., 171, 646-657.
10
Price T. (2008). Speciation in birds. Roberts and Company, Colorado.
Revell L.J. (2009). Size-correction and prinicpal components for interspecific comparative
studies. Evolution, 63, 3258-3268.
Rundell R.J. & Price T.D. (2009). Adaptive radiation, nonadaptive radiation, ecological
speciation and nonecological speciation. Trends Ecol. Evol., 24, 394-399.
Sibley C.G. & Monroe B.L. (1990). Distribution and taxonomy of the birds of the world.
Yale University Press, New Haven.
Stotz D.F., Fitzpatrick J.W., Parker T.A. & Moskovits D.K. (1996). Neotropical birds:
ecology and conservation. The University of Chicago Press Chicago.
R Development Core Team (2010). R: a language and environment for statistical computing.
Vienne, Austria: R Foundation for Statistical Computing. Retrieved from
http://www.R-project.org.
Titman A.C. (2011). Flexible nonhomogeneous markov models for panel observed data.
Biometrics, 67, 780-787.
Weir J.T. (2006). Divergent timing and patterns of species accumulation in lowland and
highland Neotropical birds. Evolution, 60, 842-855.
Weir J.T. & Price T.D. (2011). Limits to speciation inferred from times to secondary
sympatry and ages of hybridizing species along a latitudinal gradient. Am. Nat., 177,
462-469.
11
Figure S1 Range overlap (%) calculated from polygon range maps across a) all sister species pairs, b)
confirmed sympatric species and c) cases of spurious or localised sympatry.
12
Figure S2 Modelling of changes in the rate of secondary sympatry through time (S) under the a) RateSwitch model and b) Clade-Rate-Switch model. Species divergence times and present day geographic
states (green rectangles) are shown for four sister species pairs. Red bars on accompanying graphs
show the rate of secondary sympatry prior to (S1) and after (S2) the break point. In a), ts is time
following speciation when there is a significant shift in S. The relative timing of the shift after
speciation is identical across lineages, but the absolute date of the shift is variable because sister pairs
vary in age. As the age of the top pair is < ts it has yet to undergo a shift in S. In b), ta represents the
time before the present when there was a shift in S. This shift occurs at the same absolute time across
all species pairs that were present at the time of the shift, but the timing of the shift relative to the
timing of speciation varies, again because sister pairs vary in age. Pairs arising after the rate shift (i.e.
closer to the present, so that age < ta) are characterised purely by S2. In a), the increase in rates after
the putative break point corresponds to that estimated across furnariids. In b) a decrease in rates is
shown because this leads to a similar pattern of sympatry across present day sister species (Table S8).
13
Figure S3 Likelihood of different break points (ts) in the Rate-Switch model for a) all sister pairs
sisters with conserved β-niches, and sisters with conserved β-niches and that were separated by b) <
1000 km, c) < 500 km, d) < 250 km and e) < 125 km. Break points not significantly different from
the maximum log-likelihood value (black triangle) at the 95% confidence level (i.e. within 1.92 loglikelihood units) are highlighted in red. N = sample size.
14
Figure S4 Conceptual model for estimating the delay in secondary sympatry arising due to dispersal
limitation (W2) and species interactions (ΔW) between sister lineages (see ‘An evolutionary model of
species co-occurrence’ in the main text for justifications). We assume (a) that the strength of species
interactions (indicated by the width of the dark green triangle) declines with time since speciation,
while the influence of dispersal limitation is constant or increasing; and (b) that weakening
competition as species diverge ecologically may lead to an upward shift in the rate of secondary
sympatry (S) from S1 to S2 at time ts indicated by the vertical dashed line. These different rate
estimates generate (c) variation in the expected distribution (coloured curves, with colours
corresponding to those in [b]) and average waiting times (vertical lines) to sympatry. This allows us to
partition the overall waiting time to sympatry (W) into components due to dispersal limitation and
species interactions (d). While the overall waiting time to sympatry (W) will be driven by both
dispersal limitation and species interactions, the waiting time following the rate shift (W) will
primarily be due to dispersal limitation.
15
Table S1 Furnariid sister species pairs included in the analysis, with details on their estimated times of divergence, geographic states (allopatry/sympatry),
habitats and elevation ranges.
Sister
Species
Age
Sympatric
(Myr)
1
Anabacerthia amaurotis
1
Philydor lichtensteini
2
Anabazenops fuscus
2
A. dorsalis
3
Asthenes anthoides
3
A. hudsoni
4
Asthenes dorbignyi
4
A. baeri
5
Asthenes humilis
5
A. modesta
6
Asthenes maculicauda
6
A. virgata
7
Asthenes ottonis
7
Schizoeaca palpebralis
8
Asthenes pudibunda
8
Schizoeaca vilcabambae
9
Asthenes pyrrholeuca
9
Schizoeaca harterti
10
Asthenes sclateri
10
A. wyatti
11
Automolus leucophthalmus
11
A. lammi
12
Automolus rubiginosus
Distance
between
ranges (km)
conserved
Primary
F4
β-niche
5.43
1
0
1
6.12
0
1575
1
4.45
0
718
1
1.49
0
0
0
Habitat
Elevation (m)
Secondary
3
3
F1
F3
0
1100
4
N7
N1
0
1500
5
0
0
3
F6
1800
4300
11
N1
0
0
4
3500
4800
4
N2
1.49
0
10
1
N10
N9
61
1
N7
0
1634
1
1.23
0
0
0
0
217
1
1.54
0
0
1
3
4300
3
3300
4300
5
N2
N3
2800
3700
3
3000
3500
3
N2
F6
2500
4000
5
N10
F5
2800
3400
4
N1
N2
0
3000
3
N10
F5
2900
3500
3
2000
2900
2
3000
4300
8
0
1000
3
0
1000
2
0
2400
9
N7
N9
F1
F1
16
4600
N9
N10
0.19
0
2900
N10
N10
3.63
3
950
N9
0
1500
1250
1
1.83
600
0
0
1
data (n)
350
0
87
Max
F4
2.19
0
Min
F1
N7
1.33
Morphological
F4
F1
12
Hylocryptus erythrocephalus
13
Automolus rufipileatus
13
A. melanopezus
14
Campylorhamphus procurvoides
14
C. trochilirostris
15
Certhiaxis mustelinus
15
C. cinnamomeus
16
Cinclodes albiventris
16
C. olrogi
17
Cinclodes antarcticus
17
C. fuscus
18
Cinclodes aricomae
18
C. excelsior
19
Cinclodes nigrofumosus
19
C. taczanowskii
20
Cinclodes palliatus
20
C. atacamensis
21
Coryphistera alaudina
21
Anumbius annumbi
22
Cranioleuca albiceps
22
C. marcapatae
23
Cranioleuca antisiensis
23
C. curtata
24
Cranioleuca muelleri
24
C. vulpina
25
Cranioleuca pallida
25
C. pyrrhophia
26
Cranioleuca subcristata
26
C. hellmayri
27
Cranioleuca sulphurifera
F7
3.36
3.57
1
0
0
0
1
1
0.31
1.03
2.19
0.12
2.70
10.49
1.74
1
0
1
0
0
0
1
0
0
989
0
1255
104
0
0
240
1
1
0
1
1
1
1
1
1.67
0
1
0
0
1
1
F2
0
850
6
F1
0
0
4
0
500
7
F4
0
1200
21
A1
A8
0
0
3
A1
F14
0
1000
4
N10
0
5000
3
N2
1600
2800
3
F1
0
0
5
N10
A4
N9
0
4900
3
F6
N3
3500
4600
2
N10
N3
3200
5200
3
A4
0
0
3
A4
0
0
4
A10
N9
4400
5000
5
N9
A9
2800
4900
4
N1
0
700
4
N6
0
1000
7
F4
2200
3300
4
2400
3300
3
F4
1200
3100
6
F4
800
2500
5
F2
0
0
3
0
800
5
F8
0.20
0.54
2.24
0
0
1
847
62
0
1
1
1
17
5
F2
F4
0.30
1800
F3
F2
6.11
600
F5
N12
F4
800
2200
3
F1
F7
0
2500
7
F4
F1
300
2300
3
F4
F15
1500
3000
2
0
0
3
A1
27
Limnoctites rectirostris
28
Dendrocincla fuliginosa
28
D. anabatina
29
Dendrocincla merula
29
D. tyrannina
30
Dendrocolaptes certhia
30
D. sanctithomae
31
Dendrocolaptes picumnus
31
D. hoffmannsi
32
Dendroplex kienerii
32
D. picus
33
Drymornis bridgesii
33
Drymotoxeres pucherani
34
Furnarius cristatus
34
F. rufus
35
Furnarius figulus
35
F. leucopus tricolor
36
Geositta crassirostris
36
G. rufipennis
37
Geositta isabellina
37
G. saxicolina
38
Geositta punensis
38
G. cunicularia
39
Hylexetastes perrotii
39
H. stresemanni
40
Hylocryptus rectirostris
40
Clibanornis dendrocolaptoides
41
Lepidocolaptes angustirostris
41
L. albolineatus
42
Lepidocolaptes leucogaster
4.57
10.50
4.29
1.17
0
0
0
0
0
0
162
0
1
0
1
1
A1
0
1000
3
F1
0
1200
12
F1
0
1250
6
F1
0
600
4
F4
1300
2500
4
F1
0
1300
13
F1
0
1000
3
0
2800
5
0
0
3
F1
F4
F1
7.33
8.54
1
0
0
1117
1
0
F2
0
0
5
F3
F7
0
900
4
F7
N1
0
1000
3
2100
2950
4
F4
2.91
3.54
10.18
1
1
0
0
0
351
1
1
0
0
1000
3
N14
N1
N13
0
3500
4
F3
F15
0
900
3
F2
F3
0
800
10
600
2500
4
N2
3100
4400
8
N9
1800
3100
3
N9
3700
4900
8
3200
4600
9
0
4800
18
F1
0
0
11
F1
0
0
3
F8
600
1200
3
F1
0
1000
3
0
1200
4
0
1100
13
950
4000
4
N2
N9
5.67
4.00
0
1
1500
0
0
1
N9
N7
2.92
3.76
3.69
0
0
0
0
0
0
1
1
1
F7
N9
N4
F1
2.32
0
0
1
18
F11
F10
42
L. affinis
43
Leptasthenura andicola
43
L. aegithaloides
44
Leptasthenura striata
44
L. pileata
45
Leptasthenura striolata
45
L. platensis
46
Margarornis bellulus
46
M. stellatus
47
Nasica longirostris
47
Dendrexetastes rufigula
48
Ochetorhynchus phoenicurus
48
O. ruficaudus
49
Phacellodomus dorsalis
49
P. maculipectus
50
Phacellodomus sibilatrix
50
P. striaticeps
51
Phacellodomus striaticollis
51
P. ruber
52
Philydor erythrocercum
52
P. fuscipenne
53
Philydor pyrrhodes
53
Heliobletus contaminatus
54
Philydor ruficaudatum
54
Anabacerthia variegaticeps
55
Philydor rufum
55
P. erythropterum
56
Phleocryptes melanops
56
Limnornis curvirostris
57
Premnoplex tatei
2.76
3.47
0.27
0.36
9.39
3.81
0
1
0
0
1
0
38
0
374
198
0
0
1
1
1
1
1
0
F4
F11
1000
3100
3
N2
N3
3200
4700
3
N1
N2
0
4300
7
N2
F6
1550
3800
5
N2
F6
2800
4400
4
F9
N11
750
1200
2
F8
N11
0
800
3
F4
F5
1350
1600
2
F4
F5
900
2200
3
F2
F3
0
500
5
F1
F2
0
950
5
N1
N7
0
1200
5
1800
4200
4
N2
1.48
1.67
1.58
2.63
7.70
4.08
4.32
0
0
0
0
0
0
1
1701
0
0
90
1130
101
0
1
0
1
1
1
1
1
N2
2050
2800
4
N2
F7
1000
2500
4
F7
N1
0
0
3
N2
F6
2600
4000
3
N11
F8
0
800
4
N11
N14
0
700
4
F1
F4
0
700
5
F1
F4
0
1400
3
F1
F2
0
750
4
F4
F1
750
1800
3
0
1050
3
F1
F4
F1
700
2100
7
F4
F1
0
2200
4
0
900
3
F1
10.86
7.62
1
0
0
261
1
1
19
A1
A1
0
4300
5
A1
A3
0
0
3
800
2400
3
F4
57
P. brunnescens
58
Pseudasthenes steinbachi
58
P. cactorum
59
Pseudocolaptes lawrencii
59
P. boissonneautii
60
Pseudoseisura gutturalis
60
P. lophotes
61
Schizoeaca griseomurina
61
S. fuliginosa
62
Schizoeaca perijana
62
S. coryi
63
Sclerurus caudacutus
63
S. guatemalensis
64
Sclerurus rufigularis
64
S. mexicanus
65
Sclerurus scansor
65
S. albigularis
66
Simoxenops ucayalae
66
S. striatus
67
Sittasomus griseicapillus
67
Deconychura longicauda
68
Sylviorthorhynchus desmursii
68
Leptasthenura yanacensis
69
Synallaxis albilora
69
S. maranonica
70
Synallaxis brachyura
70
S. subpudica
71
Synallaxis cabanisi
71
S. moesta
72
Synallaxis candei
3.38
4.06
0
0
1103
0
1
1
F4
950
2500
11
N2
800
3000
3
N2
0
2400
5
F4
1800
3100
4
1400
3400
7
0
1000
3
0
800
4
2750
3200
4
3000
4300
8
N10
3000
3400
1
N10
2800
4100
3
F1
0
1100
8
0
1250
5
F4
1.75
0
0
0
N1
F7
0.16
0
104
1
5.18
0
0
192
152
1
1
F1
11.68
2.90
2.10
1
0
0
0
760
0
1
1
1
F8
N10
N10
3.11
F5
F5
F4
F1
0
900
4
F1
F4
0
1800
7
F1
F4
0
1250
4
F4
F1
700
2100
5
F1
F3
F1
11.64
1
0
1
F1
F2
F1
10.50
3.76
3.94
3.08
2.27
0
0
0
0
0
1158
2454
0
342
1414
0
1
0
1
1
20
0
1000
4
650
800
4
0
1550
19
0
1400
6
F9
F15
0
1000
3
F6
N3
3700
4500
3
F8
F15
0
600
5
F8
F15
0
1500
3
N14
N11
0
2000
6
F4
F15
1200
3200
1
F1
F15
0
1400
4
F3
F1
0
1200
4
0
1000
4
F7
72
S. erythrothorax
73
Synallaxis castanea
73
S. unirufa
74
Synallaxis cherriei
74
S. rutilans
75
Synallaxis courseni
75
S. azarae
76
Synallaxis hypospodia
76
S. spixi
77
Synallaxis scutata
77
S. cinerascens
78
Synallaxis stictothorax
78
S. zimmeri
79
Syndactyla dimidiatum
79
S. rufosuperciliata
80
Syndactyla ruficollis
80
S. subalaris
81
Tarphonomus certhioides
81
T. harterti
82
Thripadectes melanorhynchus
82
T. rufobrunneus
83
Thripadectes scrutator
83
T. flammulatus
84
Thripophaga fusciceps
84
T. cherriei
85
Upucerthia albigula
85
U. dumetaria
86
Upucerthia validirostris
86
U. jelskii
87
Xenerpestes singularis
4.06
3.10
0
1
318
0
1
1
F15
N14
0
1000
4
F4
F15
1300
2900
3
F4
F5
1700
3300
8
F1
F15
0
1050
3
0
900
10
F1
1.66
2.85
2.17
2.27
1.74
2.30
1.18
3.53
1.81
3.56
0
0
0
0
0
0
0
0
0
0
7
0
0
62
0
0
61
882
91
987
1
1
1
0
1
1
0
1
1
1
F4
F15
2700
3500
5
F4
F15
1250
3100
13
N6
N14
0
700
6
N14
N11
0
2200
5
F7
F8
0
1500
4
F1
F4
0
2000
3
N1
0
400
8
N2
1900
3000
3
F8
F1
0
1200
3
F1
F4
0
2600
7
F4
400
2900
5
F4
1000
2300
7
N1
0
1300
3
N2
1400
2900
3
F4
900
1750
3
F4
1200
2500
3
F4
F5
2100
3500
5
F4
F5
1400
3250
4
F2
F3
0
0
8
0
0
3
3050
3700
4
0
4000
4
2700
5000
4
3250
4600
3
1050
1700
3
F3
2.89
0
30
1
N2
N2
0.45
0
0
1
N9
N9
2.92
0
117
0
21
N1
F4
N2
87
X. minlosi
88
Xenops rutilans
88
X. tenuirostris
89
Xiphocolaptes major
89
X. promeropirhynchus
90
Xiphorhynchus eytoni
90
X. guttatus
91
Xiphorhynchus lachrymosus
91
X. flavigaster
92
Xiphorhynchus pardalotus
92
X. ocellatus
93
Xiphorhynchus spixii
93
X. elegans
94
Xiphorhynchus triangularis
94
X. erythropygius
5.87
1.22
3.02
2.66
2.54
4.61
4.15
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
F4
F1
0
1000
2
F4
F1
0
2400
8
F3
F2
0
0
4
F7
F8
0
1800
4
F4
F11
700
3100
8
F1
0
0
2
F1
F2
0
1100
5
F1
F14
0
1200
4
F1
F4
0
1500
5
F1
F4
0
1800
6
F1
F4
0
1600
5
F1
0
600
3
F1
0
500
4
F4
1100
2400
7
900
2200
4
F4
F1
Species habitat assignments and elevation range limits are from Stotz et al. (1996), supplemented with additional sources for species
representing recent taxonomic revisions. Habitat codes are: forest [F1–F15], scrub/grassland [N1–N14] and aquatic (including lakes/rivers/coasts
[A1–A12]).
22
Table S2 Rates of secondary sympatry across Furnariidae under alternative β-niche classifications and distance cut-offs.
Constant-Rate
Data
N
All pairs
β-niche
Conserved
β-niche
Conserved2
94
<1000km
71
<500km
65
<250km
61
<125km
55
79
61
S
0.07
(0.04,0.11)
0.09
(0.05,0.14)
0.07
(0.04,0.13)
0.1
(0.06,0.16)
0.11
(0.07,0.18)
0.12
(0.08,0.2)
0.14
(0.09,0.22)
W
14.52
(9.23,22.85)
11.59
(7.25,18.51)
13.83
(7.81,24.51)
9.84
(6.14,15.75)
8.97
(5.59,14.39)
8.09
(5.03,13.01)
7.21
(4.47,11.64)
Rate-Switch
AIC
83.5
68.2
49.6
62.5
59.3
55.8
51.9
S1
0.06
(0.03,0.11)
0.06
(0.03,0.12)
0
(0,62182.17)
0.07
(0.04,0.13)
0.08
(0.04,0.15)
0.08
(0.04,0.15)
0.09
(0.05,0.18)
W1
16.16
(8.92,29.29)
16.35
(8.58,31.17)
10.16x10-2
(0,6.42x10-10)
14.43
(7.54,27.64)
12.8
(6.69,24.49)
12.54
(6.52,24.14)
10.99
(5.71,21.17)
S2
0.1
(0.03,0.36)
0.33
(0.12,0.9)
0.14
(0.08,0.26)
0.54
(0.18,1.61)
0.52
(0.17,1.6)
4.57
(0.55,37.97)
4.56
(0.14,151.75)
W2
9.59
(2.75,33.39)
3.03
(1.11,8.26)
7.05
(3.89,12.76)
1.87
(0.62,5.6)
1.93
(0.63,5.96)
0.22
(0.03,1.82)
0.22
(0.01,7.29)
Time-Variable
ΔW
TS
Tmax
AIC
P
7.89
4.6
7.7
85.1
0.55
8.56
4.6
6.1
66.1
0.04
6.79
1.7
4.6
48.3
0.07
7.97
4.6
5.4
58.9
0.02
7.04
4.6
5.4
56.7
0.03
7.87
5.1
5.2
48.9
<0.01
6.99
4.6
4.6
45.2
<0.01
S0
0.05
(0.02,0.13)
0.04
(0.01,0.09)
0.03
(0.01,0.09)
0.04
(0.01,0.11)
0.05
(0.02,0.1)
0.03
(0.01,0.14)
0.03
(0,0.18)
b
0.08
(-0.16,0.32)
0.31
(0.06,0.55)
0.31
(0.03,0.58)
0.35
(0.04,0.65)
0.33
(0.1,0.56)
0.58
(0.05,1.11)
0.67
(0,1.34)
AIC
P
85.1
0.54
65.4
0.03
47.8
0.05
59.1
0.02
56.6
0.03
50.8
0.01
46.5
0.01
Rates and waiting times (95% CI) were estimated across all furnariid sister pairs (‘All pairs’), across those sister pairs occupying the same β-niche defined
using either coarse (β-niche conserved: forest, scrub/grassland, aquatic) or fine (β-niche conserved2: F1–F15, N1–N14, A1–A12) habitat classifications given
in Stotz et al. (1996), and for sisters species separated by different distance cut-offs. N = number of sister pairs; S = rate of secondary sympatry per sister pair
per million years (Myr); W = expected waiting time to secondary sympatry (Myr); S1, S2, W1 and W2 = the rates and waiting times to secondary sympatry
estimated under the RateSwitch model before and after the break point (TS), respectively and ΔW is the estimated waiting time to sympatry arising from biotic
interactions; Tmax = max break point age considered in the RateSwitch model, b = rate of change in transition rate with time since divergence estimated under
the Time-Variable model; P = significance calculated from likelihood ratio tests comparing the Rate-Switch and Time-Variable models to a Constant-Rate
model, respectively.
23
Table S3 Trait-dependent models of secondary sympatry fit to observed α-niche divergence
(z) across species pairs (N = 48), showing the effects of more stringent β-niche classification.
Variable
Bill.PC1
Bill.PC2
Bill.PC3
Bill
Euclidian
Morph.PC1
Morph.PC2
Morph.PC3
Morph.PC4
Morph.PC5
Morphology
Euclidian
Proportional
Change
2.87
(0.08,5.75)
6.51
(-1.58,15.27)
0.08
(-2.22,2.44)
4.25
(-0.13,8.83)
2.6
(-0.26,5.53)
5.16
(-1.39,12.15)
2.41
(0.18,4.68)
-0.53
(-2.75,1.74)
3.36
(-2.38,9.44)
4.1
(-0.29,8.69)
S (z=min(z))
W (z=min(z))
S (z=max(z))
W (z=max(z))
0.05
(0.13,0.02)
0.05
(0.15,0.02)
0.1
(0.28,0.04)
0.04
(0.13,0.01)
0.05
(0.14,0.02)
0.06
(0.15,0.02)
0.05
(0.13,0.02)
0.12
(0.3,0.05)
0.06
(0.19,0.02)
0.04
(0.13,0.01)
21.25
(7.65,59.03)
18.73
(6.51,53.83)
9.98
(3.62,27.51)
23.47
(7.75,71.11)
19.31
(7.06,52.78)
17.2
(6.53,45.31)
20.01
(7.55,53.03)
8.21
(3.34,20.18)
17.86
(5.29,60.28)
23.18
(7.55,71.13)
0.8
(6.44,0.1)
29.36
(34589.41,0.02)
0.11
(0.54,0.02)
2.74
(88.86,0.08)
0.67
(5.79,0.08)
8.94
(2722.91,0.03)
0.54
(2.54,0.11)
0.07
(0.38,0.01)
1.53
(166.98,0.01)
2.41
(77.5,0.07)
1.25
(0.16,10.04)
0.03
(0,40.12)
9.22
(1.86,45.77)
0.36
(0.01,11.8)
1.49
(0.17,12.78)
0.11
(0,34.06)
1.86
(0.39,8.77)
13.91
(2.62,73.99)
0.65
(0.01,71.59)
0.42
(0.01,13.37)
P
0.02
0.07
0.95
0.02
0.04
0.08
0.03
0.64
0.13
0.02
Species β-niches were based on habitat classifications given in Stotz et al. (1996) and elevational
range overlap (see Appendix 1). To aid comparison, values of z across species were standardised from
0 (minimum divergence) to 100 (maximum divergence). Proportional Change indicates the % change
in S per 1% increase in z. S and W = expected rates (per sister pair per million years [Myr]) and
waiting times to secondary sympatry (Myr), respectively, corresponding to the minimum (z = 0) and
maximum (z = 100) observed levels of α-niche divergence; P = significance calculated from a
likelihood ratio test comparing the Trait-Dependent model to a Constant-Rate model. Models were
only fitted to those sister pairs with geographical ranges less than 250 km apart.
24
Table S4 Trait-dependent models of secondary sympatry fit to observed α-niche divergence
(z) across species pairs, showing the effects of varying distance cut-offs.
Distance
cut-off
(km)
125
125
125
125
125
125
125
125
125
125
250
250
250
250
250
250
250
250
250
250
500
500
500
500
500
500
500
500
500
Variable
Bill.
PC1
Bill.
PC2
Bill.
PC3
Bill
Euclidian
Morph.
PC1
Morph.
PC2
Morph.
PC3
Morph.
PC4
Morph.
PC5
Morph
Euclidian
Bill.
PC1
Bill.
PC2
Bill.
PC3
Bill
Euclidian
Morph.
PC1
Morph.
PC2
Morph.
PC3
Morph.
PC4
Morph.
PC5
Morph
Euclidian
Bill.
PC1
Bill.
PC2
Bill.
PC3
Bill
Euclidian
Morph.
PC1
Morph.
PC2
Morph.
PC3
Morph.
PC4
Morph.
Proportional
Change
S (z=min(z))
W (z=min(z))
S (z=max(z))
W (z=max(z))
2.63
(0.17,5.15)
2.9
(-3.58,9.82)
1.07
(-0.96,3.15)
3.91
(0.02,7.95)
2.32
(-0.14,4.85)
2.55
(-3.29,8.74)
0.88
(-0.82,2.6)
0.18
(-1.49,1.88)
1.3
(-1.45,4.12)
3.64
(-0.13,7.55)
2.82
(0.38,5.32)
3.37
(-3.16,10.35)
0.49
(-1.4,2.42)
4.15
(0.28,8.18)
2.51
(0.06,5.01)
2.93
(-2.94,9.16)
1.06
(-0.63,2.79)
-0.17
(-1.92,1.62)
1.56
(-1.29,4.48)
3.91
(0.15,7.81)
3.08
(0.61,5.61)
4.49
(-1.94,11.33)
-0.33
(-2.24,1.62)
4.46
(0.52,8.54)
2.78
(0.3,5.32)
3.9
(-1.82,9.96)
1.16
(-0.56,2.92)
-0.81
(-2.67,1.08)
0.07
(0.16,0.03)
0.1
(0.23,0.05)
0.1
(0.23,0.04)
0.07
(0.16,0.03)
0.08
(0.18,0.04)
0.11
(0.23,0.05)
0.11
(0.22,0.05)
0.13
(0.28,0.06)
0.11
(0.23,0.05)
0.07
(0.17,0.03)
0.06
(0.14,0.03)
0.09
(0.2,0.04)
0.11
(0.23,0.05)
0.06
(0.14,0.02)
0.07
(0.15,0.03)
0.09
(0.2,0.04)
0.09
(0.19,0.04)
0.13
(0.27,0.06)
0.09
(0.19,0.04)
0.06
(0.14,0.02)
0.05
(0.12,0.02)
0.07
(0.16,0.03)
0.12
(0.25,0.06)
0.05
(0.12,0.02)
0.06
(0.13,0.03)
0.08
(0.16,0.04)
0.08
(0.17,0.04)
0.14
(0.29,0.07)
13.53
(6.07,30.17)
9.7
(4.29,21.89)
10
(4.42,22.64)
14.91
(6.12,36.35)
12.43
(5.66,27.3)
9.46
(4.35,20.57)
9.29
(4.52,19.1)
7.66
(3.62,16.19)
9.42
(4.39,20.23)
14.75
(5.95,36.55)
15.72
(7.1,34.81)
11.35
(5.02,25.68)
9.49
(4.27,21.08)
17.43
(7.17,42.34)
14.48
(6.61,31.7)
10.99
(5.06,23.89)
10.93
(5.35,22.32)
7.66
(3.65,16.08)
11.11
(5.13,24.07)
17.34
(7.02,42.8)
18.2
(8.24,40.23)
13.72
(6.18,30.45)
8.13
(3.95,16.75)
20.3
(8.32,49.53)
16.84
(7.69,36.89)
13.18
(6.2,28.02)
12.44
(6.05,25.56)
7.04
(3.5,14.15)
0.99
(6.6,0.15)
1.8
(643.55,0.01)
0.29
(1.24,0.07)
3.11
(72.07,0.13)
0.8
(5.45,0.12)
1.31
(258.71,0.01)
0.26
(0.91,0.07)
0.16
(0.52,0.05)
0.39
(3.53,0.04)
2.41
(48.92,0.12)
1.03
(6.74,0.16)
2.42
(884.12,0.01)
0.17
(0.66,0.04)
3.36
(76.17,0.15)
0.82
(5.44,0.12)
1.64
(326.42,0.01)
0.26
(0.92,0.08)
0.11
(0.4,0.03)
0.42
(4.17,0.04)
2.66
(53.15,0.13)
1.14
(7.69,0.17)
5.87
(1822.32,0.02)
0.09
(0.38,0.02)
3.85
(90.67,0.16)
0.92
(6.31,0.13)
3.5
(583.37,0.02)
0.26
(0.91,0.07)
0.06
(0.27,0.01)
1.01
(0.15,6.72)
0.56
(0,198.57)
3.44
(0.8,14.7)
0.32
(0.01,7.46)
1.25
(0.18,8.5)
0.76
(0,150.23)
3.89
(1.1,13.68)
6.38
(1.91,21.3)
2.6
(0.28,23.82)
0.41
(0.02,8.41)
0.97
(0.15,6.35)
0.41
(0,150.48)
5.79
(1.51,22.27)
0.3
(0.01,6.74)
1.22
(0.18,8.09)
0.61
(0,121.74)
3.79
(1.08,13.28)
9.05
(2.47,33.13)
2.37
(0.24,23.41)
0.38
(0.02,7.5)
0.87
(0.13,5.87)
0.17
(0,52.84)
11.35
(2.61,49.27)
0.26
(0.01,6.11)
1.09
(0.16,7.45)
0.29
(0,47.75)
3.91
(1.1,13.94)
15.88
(3.7,68.17)
1.44
0.08
12.1
0.35
2.89
25
P
0.02
0.24
0.31
0.02
0.04
0.25
0.33
0.83
0.34
0.03
0.01
0.19
0.62
0.01
0.02
0.2
0.24
0.85
0.26
0.02
0.01
0.11
0.73
0.01
0.01
0.11
0.21
0.38
0.28
500
1000
1000
1000
1000
1000
1000
1000
1000
1000
1000
None
None
None
None
None
None
None
None
None
None
PC5
(-1.27,4.23)
(0.18,0.04)
(5.65,25.93)
(3.03,0.04)
(0.33,25.28)
Morph
Euclidian
Bill.
PC1
Bill.
PC2
Bill.
PC3
Bill
Euclidian
Morph.
PC1
Morph.
PC2
Morph.
PC3
Morph.
PC4
Morph.
PC5
Morph
Euclidian
Bill.
PC1
Bill.
PC2
Bill
.PC3
Bill
Euclidian
Morph
.PC1
Morph.
PC2
Morph.
PC3
Morph.
PC4
Morph.
PC5
Morph
Euclidian
4.26
(0.46,8.21)
2.83
(0.49,5.22)
3.58
(-2.7,10.26)
-0.71
(-2.68,1.3)
3.99
(0.25,7.88)
2.46
(0.14,4.84)
3.21
(-2.34,9.08)
1.2
(-0.56,2.98)
-1.11
(-3.01,0.83)
1.57
(-1.13,4.34)
3.75
(0.09,7.53)
1.88
(0.17,3.62)
2.54
(-3.13,8.53)
-0.64
(-2.62,1.38)
2.57
(-0.15,5.35)
1.66
(-0.05,3.41)
2.35
(-2.4,7.33)
1.27
(-0.5,3.07)
-1.2
(-3.12,0.76)
1.81
(-0.81,4.49)
2.43
(-0.12,5.04)
0.05
(0.12,0.02)
0.05
(0.11,0.02)
0.07
(0.16,0.03)
0.13
(0.26,0.06)
0.05
(0.11,0.02)
0.06
(0.12,0.03)
0.07
(0.15,0.03)
0.07
(0.15,0.04)
0.14
(0.29,0.07)
0.07
(0.16,0.03)
0.05
(0.12,0.02)
0.05
(0.11,0.02)
0.07
(0.14,0.03)
0.1
(0.22,0.05)
0.05
(0.11,0.02)
0.06
(0.11,0.03)
0.07
(0.13,0.03)
0.06
(0.13,0.03)
0.13
(0.25,0.06)
0.06
(0.13,0.03)
0.05
(0.11,0.02)
20.35
(8.24,50.29)
19.56
(8.85,43.23)
14.07
(6.35,31.19)
7.92
(3.79,16.54)
21.36
(8.72,52.3)
17.81
(8.17,38.84)
13.72
(6.47,29.09)
13.79
(6.66,28.54)
6.98
(3.44,14.18)
13.6
(6.37,29)
21.12
(8.52,52.36)
19.46
(9.35,40.47)
15.3
(7.14,32.81)
9.56
(4.58,19.95)
20.33
(9.13,45.27)
18.1
(8.86,36.98)
15.12
(7.44,30.73)
16.56
(7.97,34.41)
7.98
(3.92,16.25)
16.64
(7.95,34.85)
20.2
(9.14,44.68)
3.19
(65.61,0.16)
0.83
(4.89,0.14)
2.39
(673.75,0.01)
0.06
(0.28,0.01)
2.35
(46.21,0.12)
0.64
(3.77,0.11)
1.72
(251.7,0.01)
0.24
(0.86,0.07)
0.05
(0.21,0.01)
0.35
(3.03,0.04)
1.87
(33.36,0.11)
0.33
(1.15,0.1)
0.8
(133.46,0)
0.06
(0.25,0.01)
0.62
(5.1,0.08)
0.29
(1.03,0.08)
0.67
(47.49,0.01)
0.21
(0.77,0.06)
0.04
(0.17,0.01)
0.36
(2.92,0.04)
0.54
(3.86,0.08)
0.31
(0.02,6.43)
1.2
(0.2,7.06)
0.42
(0,117.8)
16.14
(3.57,72.97)
0.43
(0.02,8.36)
1.56
(0.27,9.19)
0.58
(0,84.98)
4.19
(1.16,15.19)
21.32
(4.81,94.47)
2.86
(0.33,24.7)
0.53
(0.03,9.52)
3.02
(0.87,10.51)
1.25
(0.01,209.09)
18.18
(3.99,82.87)
1.61
(0.2,13.25)
3.47
(0.97,12.37)
1.49
(0.02,105.25)
4.7
(1.29,17.11)
26.7
(5.93,120.25)
2.77
(0.34,22.43)
1.84
(0.26,13.02)
0.01
0.01
0.14
0.47
0.01
0.02
0.14
0.21
0.24
0.24
0.02
0.03
0.18
0.52
0.04
0.06
0.18
0.18
0.21
0.16
0.04
To aid comparison, values of z across species were standardised from 0 (minimum divergence) to 100
(maximum divergence). Proportional Change indicates the % change in S per 1% increase in z. S and
W = expected rates (per sister pair per million years (Myr)) and waiting times to secondary sympatry
(Myr), respectively, corresponding to the minimum (z = 0) and maximum (z = 100) observed levels of
α-niche divergence; P = significance calculated from a likelihood ratio test comparing the TraitDependent model to a Constant-Rate model.
26
Table S5 Results of the State-Reversible model used to estimate the rate at which species
pairs return to allopatry following the attainment of sympatry.
Constant-Rate
Data
N
All pairs
94
β-nicheConserved1
79
β-nicheConserved2
61
s
0.07
(0.04,0.11)
0.09
(0.05,0.14)
0.07
(0.04,0.13)
AIC
83.47
68.23
49.61
State-Reversible
s
0.07
(0.04,0.11)
0.09
(0.05,0.14)
0.07
(0.04,0.13)
a
0
(0,4.56x10-4)
0
(0,5.17x10-5)
0
(0,3.54x10-4)
0.1
(0.06,0.16)
0.11
(0.07,0.18)
0.12
(0.08,0.2)
0.14
(0.09,0.22)
0
(0,2.30x10-4)
0
(0,2.01x10-13)
0
(0,1.27x10-4)
0
(0,1.03x10-5)
AIC
P
85.49
1
70.25
1
51.63
1
64.55
1
61.34
1
57.84
1
53.94
1
Distance cut-off
<1000km
71
<500km
65
<250km
61
<125km
55
0.1
(0.06,0.16)
0.11
(0.07,0.18)
0.12
(0.08,0.2)
0.14
(0.09,0.22)
62.52
59.34
55.81
51.92
Rates and waiting times (95% CI) were estimated across all furnariid sister pairs (‘All pairs’), across
those sister pairs occupying the same β-niche defined using either coarse (β-niche conserved1: forest,
scrub/grassland, aquatic) or fine (β-niche conserved2: F1–F15, N1–N14, A1–A12) habitat
classifications given in Stotz et al. (1996), and for sisters species separated by different distance cutoffs. N = number of sister pairs; S = rate of secondary sympatry per sister pair per million years
(Myr); a = rate of return to allopatry per sister pair per million years (Myr); P = significance
calculated from likelihood ratio tests comparing the State-Reversible model to the Constant-Rate
model.
27
Table S6 Morphological trait loadings for principle components of bill dimensions.
Trait
PC1
PC2
PC3
Bill length
0.908 0.419 -0.014
Bill width
0.743 -0.546 -0.387
Bill depth
0.808 -0.499
0.313
Results are from a phylogenetic Principal Components (PC) analysis (Revell et al. 2009) based on the
covariance matrix following log-transformation of species trait values.
Table S7 Morphological trait loadings for principle components of body dimensions.
Trait
PC1
PC2
PC3
PC4
PC5
Bill length
-0.865
0.498
0.058 -0.014 -0.025
Bill width
-0.773 -0.466
0.135 -0.407 -0.042
Bill depth
-0.830 -0.410
0.232
0.260
0.147
Tarsus
-0.769 -0.106 -0.544 -0.041
0.317
Wing
-0.758 -0.335 -0.285
0.171 -0.451
Results are from a phylogenetic Principal Components (PC) analysis (Revell et al. 2009) based on the
covariance matrix following log-transformation of species trait values.
28
Table S8 Clade-wide shifts in rates of secondary sympatry.
Constant-Rate
Data
N
All pairs
94
β-niche
Conserved
79
S
W
0.07
14.52
(0.04,0.11)
(9.23,22.85)
0.09
11.59
(0.05,0.14)
(7.25,18.51)
Clade-Rate-Switch
AIC
83.47
68.23
Clade-Time-Variable
S1
W1
S2
W2
0.78
1.27
0.06
16.16
(0.11,5.5)
(0.18,8.93)
(0.04,0.1)
(9.75,26.78)
3.96
0.25
0.07
14.3
(0,10992.07)
(0,701.79)
(0.04,0.12)
(8.29,24.67)
Ta
10
8
AIC
83.99
64.59
P
0.22
0.02
S0
b
0.14
-0.08
(0.02,1.3)
(-0.32,0.16)
1.29
-0.29
(0.14,11.62)
(-0.54,-0.05)
AIC
P
85.1
0.54
65.44
0.03
Rates and waiting times (95% CI) were estimated across all furnariid sister pairs (‘All pairs’) and across those sister pairs occupying the same β-niche (βniche conserved: forest, scrub/grassland, aquatic) using habitat classifications given in Stotz et al. (1996). N = number of sister pairs; S = rate of secondary
sympatry per sister pair per million years (Myr); W = expected waiting time to secondary sympatry (Myr); S1, S2, W1 and W2 = the rates and waiting times to
secondary sympatry estimated under the Clade-Rate-Switch model before and after the breakpoint (Ta), respectively; b = rate of change in transition rate
through time estimated under the CladeTimeVariable model; P = significance calculated from likelihood ratio tests comparing the Clade-Rate-Switch and
Clade-Time-Variable models to a Constant-Rate model, respectively.
29