Evolutionary biology

royalsocietypublishing.org/journal/rsbl

Evolution of moult-migration is directly
linked to aridity of the breeding grounds
in North American passerines

Research

Claudie Pageau1, Christopher M. Tonra3, Mateen Shaikh2, Nancy J. Flood1
and Matthew W. Reudink1

Cite this article: Pageau C, Tonra CM, Shaikh
M, Flood NJ, Reudink MW. 2020 Evolution of
moult-migration is directly linked to aridity of
the breeding grounds in North American
passerines. Biol. Lett. 16: 20200155.
http://dx.doi.org/10.1098/rsbl.2020.0155

Received: 11 March 2020
Accepted: 7 May 2020

Subject Areas:
evolution, ecology, behaviour
Keywords:
moult-migration, passeriformes, phylogenetic
analysis, evolution, North America, stopover

Author for correspondence:
Matthew W. Reudink
e-mail: mreudink@tru.ca

1

Department of Biological Sciences, and 2Department of Mathematics and Statistics, Thompson Rivers
University, Kamloops BC V2C 0C8, Canada
3
School of Environment and Natural Resources, The Ohio State University, Columbus, OH 43210, USA
CP, 0000-0003-0371-5602
To avoid energy allocation conflicts, birds generally separate breeding,
migration and moult during the annual cycle. North American passerines typically moult on the breeding grounds prior to autumn migration. However,
some have evolved a moult-migration strategy in which they delay moult
until stopping over during autumn migration. Rohwer et al. (2005) proposed
the ‘push–pull hypothesis’ as an explanation for the evolution of this moult
strategy, but it has not been empirically tested. Poor conditions on the breeding
grounds at the end of the summer would push birds to depart prior to moult,
while productive stopover locations would pull them. We tested for a relationship between moult-migration and breeding grounds aridity as measured
by the normalized difference vegetation index. Our results strongly support
the ‘push’ aspect of the push–pull hypothesis and indicate that arid breeding
grounds, primarily in western North America, would drive species to evolve
stopover moult-migration, although this relationship may depend upon
migration distance.

1. Background
Feathers are unique to birds and are critical to nearly every aspect of their
biology, including flight, thermoregulation and visual communication [1].
Each year, birds must exchange old feathers for fresh ones by moulting [2,3].
An energetically expensive stage of the annual cycle [4–8], moulting requires
specific resources to produce high-quality feathers. Hence, the timing and
location of moult are crucial for the production of feathers of sufficient quality
to maximize lifetime reproductive success [9]. To avoid energy allocation conflicts, birds generally separate the most energetically expensive life-history
events during the annual cycle: breeding, migration and moult [10]. Most
migratory passerines complete their moult on the breeding grounds prior to
autumn migration. However, some have evolved a moult-migration strategy,
the ‘temporal overlap in the moult and migration life-history stages’ [11].
Moving to a stopover location i.e. a rest/refueling site during migration [12],
to moult (stopover moult-migration [11]) or moulting during migration (continuous moult-migration [11]) must confer benefits that outweigh the costs of
overlapping these two energetically demanding life-history stages. One advantage to continuous moult-migration could be reducing the overall time used
for these events, thus advancing arrival date at stationary non-breeding grounds,
which might provide various advantages (see below). For stopover moultmigration, an advantage might be the ability to acquire high-quality resources
to support moulting. For example, various species breeding in western North
America moult in the monsoon region of northwestern Mexico and southwestern
© 2020 The Author(s) Published by the Royal Society. All rights reserved.

which differ in moult strategy, for a total of 208 taxa. We classified species with respect to where they can undergo prebasic
moult (i.e. complete moult resulting in the basic plumage [4]):
breeding, wintering grounds or during migration. We followed
Tonra and Reudink’s [11] classification to specify which category
of moult-migration (stopover, continuous or suspended) the
species conformed to. To determine prebasic moulting strategy,
we used descriptions of moulting from Birds of North America
species accounts [27], Pyle [28] and the literature [13,14,29–33].
When variation in moulting strategy among individuals of the
same species was encountered, we classified that species as a
moult–migrant. Altitudinal migrants (six species) were categorized as moulting on their breeding grounds.

Migration distance was approximated as the distance (Mm) from
the centroid of the breeding distribution to the centroid of the
non-breeding distribution. Distribution maps from Birdlife International [34] were used in the calculation of NDVI. NDVI is a
measure of live green vegetation and was used to indicate the
aridity of the breeding grounds during the post-breeding
period in North America (1 July–31 August). Winter territoriality
category (yes or no) was taken from Birds of North America [27].
Data on the number of broods were retrieved from Birds of
North America [27]. This predictor was categorical: one or multiple broods. More information on how the predictors were
retrieved is available in the electronic supplementary material.

(c) Phylogeny
Using BirdTree.org [35], we downloaded 1000 possible trees of a
phylogeny subset containing our 200 species of passerines from
‘Hackett All Species: a set of 10 000 trees with 9993 OTUs each’
[36]. Using TreeAnnotator v. 1.10.4 [37], we then created a maximum clade credibility tree with our 1000 trees using 1% burn-in
(as states) and mean heights for node heights. We added the
eight subspecies in R [38] to obtain a maximum clade credibility
tree of 208 species and subspecies, which we used for all our
analysis. The visual representation of our phylogeny (figure 1)
was created using the phytools package of R [39].

(d) Statistical analysis
We used phylogenetically controlled analysis to investigate factors
associated with the evolution of moult-migration (including stopover, suspended and continuous moult; 45 species) and specifically
the stopover moult-migration strategy (13 species) for which the
push–pull hypothesis was originally devised. Note, however, that
we could only test the ‘push’ aspect of the hypothesis because moulting distributions are unavailable. In both (stopover or moultmigration), the response variables were binary, with 1 indicating
presence of stopover or moult-migration and 0 indicating absence.
We then used phylogenetic logistic linear models to test the predictors by creating a full model that included all explanatory variables
(NDVI, migration distance, number of broods and winter territoriality) and sequentially eliminated non-significant variables ( p > 0.05)
to arrive at a final best fit model. No explanatory variables were
highly correlated (all r < 0.31). Analyses were conducted in R [38]
using the package phyloglm [40]. The ‘logistic_MPLE’ method was
applied with a btol of 10, a log.alpha.bound of 10, and no bootstrap.

2. Methods
(a) Data collection
We collected data for 200 species and five subspecies of
migratory passerines breeding in Canada and/or the USA (electronic supplementary material, table S1). Three species (Vireo
gilvus, Haemorhous purpureus and Passerina ciris) were divided
into their Eastern and Western subspecies or populations,

3. Results
When we examined the factors associated with moult-migration,
both NDVI (z = −2.72, p = 0.006) and migration distance (z = 2.68,
p = 0.007) were retained in the final model, indicating that moult–
migrants were more likely to migrate longer distances and have

Biol. Lett. 16: 20200155

(b) Potential drivers of moult-migration

2

royalsocietypublishing.org/journal/rsbl

USA [13,14]. Here, the late-summer monsoon rains result in an
explosion of productivity that may ‘pull’ species to this area to
take advantage of abundant resources [14,15].
While the richness of the Mexican monsoon region may
‘pull’ moult–migrants to stopover, other factors may also
‘push’ them to depart prior to moult, including aridity of
the breeding grounds at the end of the summer or a time constraint at high latitudes. The combination of good conditions
at stopover locations with unfavourable conditions on the
breeding grounds during the post-breeding period forms
the push–pull hypothesis. This hypothesis has often been
proposed as an explanation for the evolution of stopover
moult-migration in the Mexican monsoon region [14,16,17],
but has not yet been empirically tested.
Several drivers of moult-migration evolution have been
hypothesized, such as migration distance, aridity of the
breeding grounds during the post-breeding period, winter
territoriality, length of the breeding season and number of
broods produced during the breeding season. Long migration
distances and a long breeding season, especially if raising
multiple broods, may reduce the time available between the
end of breeding and start of migration and not allow for
the replacement of all feathers [18–21]. Arid breeding
grounds could select for moult-migration because the lack
of resources at the end of summer in western North America
limits the ability to grow feathers of sufficient quality [14,22].
Finally, moult-migration might be favoured in species that
defend non-breeding territories because it would allow earlier arrival at the non-breeding grounds, and thus the
acquisition of higher quality territories [23,24].
Revealing the mechanisms responsible for the evolution
of overlapping life-history stages is critical to understanding
the dynamics of migratory bird populations and how they
are limited. Migration places enormous phenological constraints on avian life histories, the organization of which is
shaped by both biotic and abiotic factors [10]. Environmental
changes, including land-use and global climate alterations,
are shifting the availability of resources both spatially and
temporally, potentially altering the playing field of selection
[e.g. 25]. Understanding the factors that drive the organization of annual cycles will help us predict the resilience of
species to environmental change, as well as identify species
in need of proactive management [26].
Using phylogenetic comparative analyses, we examined
the hypothesis that unfavourable (dry) breeding grounds conditions during the post-breeding period act as a ‘push’ for (i)
moult-migration in North American passerines, and (ii) explicitly the evolution of stopover moult-migration in the Mexican
monsoon region. We used the normalized difference vegetation index (NDVI) as a measure of the aridity of breeding
grounds. We also tested other factors that have been proposed
to influence the evolution of moult-migration: migration
distance, winter territoriality and number of broods.

3

royalsocietypublishing.org/journal/rsbl
Biol. Lett. 16: 20200155

0.222

trait value

0.823

length = 33.502

Figure 1. Phylogeny of the 208 species of North American migrant passerines. The colour of the branches represents the average NDVI of the breeding grounds in
July and August for each passerine: green indicates high NDVI values and tan low values. The black circles indicate species that are stopover moult–migrants; these
are labelled with the scientific name of these species.
Table 1. Predictors of stopover and moult-migration included in the best model of the phylogenetic logistic regression following a stepwise regression
(backward elimination). Stopover = stopover moult-migration strategy only (n = 13 species), moult-migration = stopover, continuous and suspended moultmigration combined (n = 45 species).
coefficient

estimate

s.e.

z-value

p

stopover

intercept

0.70

0.88

0.79

0.43

moult-migration

NDVI
intercept

−6.31
−0.17

1.81
0.60

−3.49
−0.29

0.0005
0.77

migration distance

0.26

0.096

2.68

0.007

NDVI

−2.93

1.08

−2.72

0.006

breeding ranges that are drier in the post-breeding period than
non-moult–migrants. Next, we specifically examined the stopover moult-migration strategy. In this case, only NDVI of the
breeding range was included in the final model and was strongly
negatively associated with stopover (z = −3.49, p = 0.0005),
indicating that the breeding areas of stopover moult–migrants
were drier during the post-breeding period than those of
non-moult–migrants (table 1).

4. Discussion
We tested four hypotheses (aridity of the breeding grounds
during the post-breeding period, migration distance, presence
or absence of winter territoriality and number of broods) that
have been proposed to explain the evolution of moult-migration
in North American passerines. Of the four factors, our analyses
suggested that NDVI was evolutionarily associated with moultmigration, in particular the strategy that involves a stopover in
the Mexican monsoon region during migration. Taxa with

breeding grounds that are dry and unproductive during the
post-breeding period (indicated by low NDVI values) showed
a stopover moult-migration strategy much more often than
expected by chance alone. In addition, longer migration
distances were also evolutionarily associated with moultmigration when suspended, continuous and stopover strategies
are combined in one category.
Aridity has long been proposed as a driver of moultmigration; however, explicit tests of this hypothesis have been
lacking. Rohwer et al. [14] and Young [22] raised the idea that
arid breeding grounds may offer insufficient resources at the
end of the summer to grow high-quality feathers, which are
essential for flight performance during autumn migration. Substantial energy is required to synthesize new feathers [6–8], thus
having an abundance of high-quality resources during moulting is critical. Limitation of resources at the end of the
summer would act as a ‘push’ towards moult-migration in
North America, particularly in the West, where lowlands
become dry and unproductive (figure 2; [22]) at this time. The
Mexican monsoon region in northwestern Mexico and

Bullock’s oriole

4

Baltimore oriole

50° N
NDVI
1.00
0.75
0.50
0.25
0

40° N

30° N

20° N

130° W 120° W 110° W 100° W 90° W 80° W 70° W 60° W 130° W 120° W 110° W 100° W 90° W 80° W 70° W 60° W

Figure 2. Distribution maps representing NDVI of the breeding grounds (1 July to 31 August) of two North American passerines. Higher values indicate a greater
abundance of live green vegetation. Icterus galbula (Baltimore orioles) moult on their breeding grounds and their NDVI average is 0.83. Icterus bullockii (Bullocks
orioles) are stopover moult–migrants and their NDVI average is 0.35.
southwestern USA is an important stopover location to undergo
moult for migrant passerines [41] such as Icterus bullockii [42],
Tyrannus verticalis [17] and Piranga ludoviciana [32]. Monsoon
rains in this region in July and August result in an explosion
of resources available for migrant passerines on their way to
the non-breeding grounds [43]. An attraction to the Mexican
monsoon region, combined with the aridity of the breeding
grounds at the end of the summer, likely drove the evolution
of some western North American migrant passerines toward
stopover moult-migration.
For the alternative factors tested, only migration distance
was associated with moult-migration; winter territoriality and
number of broods were not present in the best models. As
expected, longer migration distances were associated with
moult-migration: by imposing a time constraint, they would
force moult outside of the breeding grounds [18,19]. This
result is in accordance with previous European studies on Sylviidae [19] and Western Palearctic passerines [44] that
indicated longer migration distance as a driver of moulting
strategies differing from the ancestral state (moult on the breeding grounds) [3]. A time constraint was also the reason behind
number of broods as a predictor, but our results suggest this
variable was not important in the evolution of moult-migration.
Winter territoriality was hypothesized as driving moultmigration and winter moult by Pérez & Hobson [24] and Lindström et al. [23], but our results concur with Rohwer et al. [14],
who did not support the winter territoriality hypothesis [23,24].
Our research examined 208 North American migrant passerines and classified each species as moult–migrant or not.
While some species have extensive data and were easy to fit
into a category (e.g. I. bullockii [42]), those exhibiting intraspecific variation in moulting strategy were more challenging.
In addition, Pyle et al. [28] described a wide variety of
post-breeding dispersal movements for moulting in many
passerines. These dispersal movements might be a type of
moult-migration; however, we took a conservative approach
in our analysis and did not account for these movements since
they do not fit the definition of moult-migration provided by
Tonra and Reudink [11]. Intra-specific variation also exists for
the explanatory variables (migration distance, winter territoriality, number of broods): thus, these categorizations at the species

level are purposely rough in an effort to describe broadscale taxonomic and geographic patterns, and a more detailed
and nuanced study that accounted for that variation at the
individual level would be useful for future research.
Our results strongly support the ‘push’ aspect of the push–
pull hypothesis proposed by Rohwer et al. [14]; arid breeding
grounds during the post-breeding period ‘push’ some migrant
passerines towards a stopover moult-migration strategy that
capitalizes on the abundant resources available in the Mexican
monsoon region during late-summer and early autumn. However, to fully understand push–pull dynamics, future studies
should explore the ‘pull’ aspect of the hypothesis. Our results
also indicate that migration distance played a role in the evolution of moult-migration. Understanding such environmental
drivers in species’ ecology is critical at this time, particularly
for the chronically understudied portions of the annual cycle
outside of breeding [26]. Given contemporary conservation challenges, such as climate and land-use change, this study raises the
question: how plastic are species in their ability to adopt or cease
a moult-migration strategy should aridity increase or decrease in
their breeding range? [25]. In addition, how could changing climatic conditions alter the relative strength of the ‘push’ and/or
‘pull’ of breeding and moulting grounds, respectively? The
answer to these questions could reveal which species will be
most resilient to ongoing environmental change.
Data accessibility. Data and code are accessible from Dryad (https://doi.
org/10.5061/dryad.vhhmgqnqq) [45].

Authors’ contributions. C.P. collected the data, performed phylogenetic analyses, analysed output data and wrote the manuscript. M.W.R.
conceptualized and supervised the project, and edited the manuscript.
C.M.T. and N.J.F. helped with literature, classification of moultmigration, and editing. C.M.T. also contributed to writing. M.S. provided
statistical support for phylogenetic analyses. All authors contributed substantially to revisions. All authors approved the final version of the
manuscript and agree to be held accountable for the content.

Competing interests. We declare we have no competing interests.
Funding. This research was funded by a Natural Sciences and Engineering Research Council of Canada Discovery Grant (M.W.R.); a British
Columbia Graduate Scholarship (C.P.); a Master’s research scholarship from the Fond de Recherche Nature et technologies (C.P.).

Acknowledgements. We would like to thank B. Turner, S. LaZerte,
T. Murphy and J. Kusack.

Biol. Lett. 16: 20200155

10° N

royalsocietypublishing.org/journal/rsbl

60° N

References

2.

3.

4.

6.

7.

8.

9.

10.

11.

12.
13.

14.

15.

16. Bridge ES, Ross JD, Contina AJ, Kellya JF. 2016 Do
molt-migrant songbirds optimize migration routes
based on primary productivity? Behav. Ecol. 27,
784–792. (doi:10.1093/beheco/arv199)
17. Barry JH, Butler LK, Rohwer S, Rohwer VG. 2009
Documenting molt-migration in western kingbird
(Tyrannus verticalis) using two measures of
collecting. Auk. 126, 260–267. (doi:10.1525/auk.
2009.07137)
18. Benson A-M, Winker K. 2015 High-latitude
passerine migrants overlap energetically demanding
events in autumn. Wilson J. Ornithol. 127,
601–614. (doi:10.1676/14-034.1)
19. Hall KSS, Tullberg BS. 2004 Phylogenetic analyses of
the diversity of moult strategies in Sylviidae in
relation to migration. Evol. Ecol. 18, 85–105.
(doi:10.1023/B:EVEC.0000017848.20735.8b)
20. Lesley J, Ogden E, Stutchbury BJM. 1996 Constraints
on double brooding in a neotropical migrant, the
hooded warbler. Condor. 98, 736–744. (doi:10.
2307/1369855)
21. Kjellén N. 1994 Moult in relation to migration in
birds—a review. Ornis Svevica. 4, 1–24.
22. Young BE. 1991 Annual molts and interruption of
the fall migration for molting in lazuli Buntings.
Condor 93, 236–250. (doi:10.2307/1368939)
23. Lindström Â, Pearson DJ, Hasselquist D, Hedenström
A, Bensch S, Âkesson S. 1993 The moult of
Barred Warblers Sylvia nisoria in Kenya—evidence
for a split wing-moult pattern initiated during the
birds’ first winter. IBIS. 135, 403–409. (doi:10.
1111/j.1474-919X.1993.tb02112.x)
24. Pérez GE, Hobson KA. 2006 Isotopic evaluation of
interrupted molt in northern breeding populations
of the loggerhead shrike. Condor. 108, 877–886.
(doi:10.1650/0010-5422(2006)108[877:
IEOIMI]2.0.CO;2)
25. Siepielski AM et al. 2017 Precipitation drives global
variation in natural selection. Science. 355,
959–962. (doi:10.1126/science.aag2773)
26. Marra PP, Cohen EB, Loss SR, Rutter JE, Tonra CM.
2015 A call for full annual cycle research in animal
ecology. Biol. Lett 11, 20150552. (doi:10.1098/rsbl.
2015.0552)
27. Rodewald P. (ed.) 2019 The Birds of North America.
Ithaca: Cornell Laboratory of Ornithology. See
https://birdsna.org.
28. Pyle P. 1997 Identification guide to North American
birds part I. Bolinas, CA: Slate Creek Press.
29. Pyle P, Saracco JF, DeSante DF. 2018 Evidence of
widespread movements from breeding to molting
grounds by North American landbirds. Auk. 135,
506–520. (doi:10.1642/AUK-17-201.1)
30. Jahn AE et al. 2013 Migration timing and wintering
areas of three species of flycatchers (Tyrannus)
breeding in the Great Plains of North America. Auk.
130, 247–257. (doi:10.1525/auk.2013.13010)
31. Butler LK, Rohwer S, Rogers M. 2006 Prebasic molt
and molt-related movements in ash-throated

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

flycatchers. Condor 108, 647–660. (doi:10.1093/
condor/108.3.647)
Butler LK, Donahue MG, Rohwer S. 2002 Moltmigration in western tanagers (Piranga ludoviciana):
age effects, aerodynamics, and conservation
implications. Auk. 119, 1010–1023. (doi:10.1093/
auk/119.4.1010)
Voelker G, Rohwer S. 1998 Contrasts in scheduling
of molt and migration in eastern and western
warbling-vireos. Auk. 115, 142–155. (doi:10.2307/
4089119)
BirdLife International. 2018 Bird species distribution
maps of the world. Version 2018.1. See http://
datazone.birdlife.org/species/requestdis.
Jetz W, Thomas GH, Joy JB, Hartmann K, Mooers AO.
2012 The global diversity of birds in space and time.
Nature. 491, 444–448. (doi:10.1038/nature11631)
Hackett SJ et al. 2008 A phylogenomic study of
birds reveals their evolutionary history. Science. 320,
1763–1768. (doi:10.1126/science.1157704)
Rambaut A, Drummond AJ. 2018 TreeAnnotator
v1.10.4: MCMC Output analysis. See http://beast.
community/.
R Core Team. 2019 R: a language and environment
for statistical computing. Vienna, Austria: R
Foundation for Statistical Computing. See https://
www.R-project.org.
Revell LJ. 2012 phytools 0.6.99: an R package for
phylogenetic comparative biology (and other
things). Methods. Ecol. Evol. 3, 217–223. (doi:10.
1111/j.2041-210X.2011.00169.x)
Ho LST, Ane C. 2014 A linear-time algorithm for
Gaussian and non-Gaussian trait evolution models.
Syst. Biol. 63, 397–408. (doi:10.1093/sysbio/syu005)
Leu M, Thompson CW. 2002 The potential
importance of migratory stopover sites as flight
feather molt staging areas: a review for neotropical
migrants. Biol. Conserv. 106, 45–56. (doi:10.1016/
S0006-3207(01)00228-2)
Pillar AG, Marra PP, Flood NJ, Reudink MW. 2016
Moult-migration in Bullock’s orioles (Icterus
bullockii) confirmed by geolocators and stable
isotope analysis. J. Ornithol. 157, 265–275. (doi:10.
1007/s10336-015-1275-5)
Comrie AC, Glenn EG. 1998 Principal components-based
regionalization of precipitation regimes across the
southwest United States and northern Mexico, with an
application to monsoon precipitation variability. Clim.
Res. 10, 201–215. (doi:10.3354/cr010201)
Kiat Y, Izhaki I, Sapir N. 2019 The effects of longdistance migration on the evolution of moult
strategies in Western-Palearctic passerines. Biol. Rev.
94, 700–720. (doi:10.1111/brv.12474)
Pageau C, Tonra CM, Shaikh M, Flood NJ, Reudink
MW. 2020 Data from: Evolution of moult–migration
is directly linked to aridity of the breeding grounds
in North American passerines. Dryad Digital
Repository. (https://doi.org/10.5061/dryad.
vhhmgqnqq)

Biol. Lett. 16: 20200155

5.

Gill FB. 1994 Ornithology, 2nd edn. New York, NY:
W.H. Freeman and Company.
Humphrey PS, Parkes KC. 1959 An approach to the
study of molts and plumages. Auk. 76, 1. (doi:10.
2307/4081839)
Svensson E, Hedenström A. 1998 A phylogenetic
analysis of the evolution of moult strategies in
western Palearctic warblers (Aves: Sylviidae).
Biol. J. Linn. Soc. 67, 3–276.
Hoye BJ, Buttemer WA. 2011 Inexplicable
inefficiency of avian molt? Insights from an
opportunistically breeding arid-zone species,
Lichenostomus penicillatus. PLoS ONE. 6, e16230.
(doi:10.1371/journal.pone.0016230)
Howell SNG. 2010 Molt in North American birds.
New York, NY: Houghton Mifflin Harcourt.
Lindström Â, Visser GH, Daan S. 1993 The energetic
cost of feather synthesis is proportional to basal
metabolic rate. Physiol. Zool. 66, 490–510. (doi:10.
1086/515898)
Dietz MW, Daan S, Masman D. 1992 Energy
requirements for molt in the kestrel Falco
tinnunculus. Physiol Zool. 65, 1217–1235. (doi:10.
1086/physzool.65.6.30158276)
Murphy ME, King JR. 1992 Energy and nutrient use
during moult by white-crowned sparrows
Zonotrichia leucophrys gambelii. Ornis Scand. 23,
304–313. (doi:10.2307/3676654)
Echeverry-Galvis MA, Hau M. 2013 Flight
performance and feather quality: paying the price
of overlapping moult and breeding in a tropical
highland bird. PLoS ONE. 8, e61106. (doi:10.1371/
journal.pone.0061106)
Wingfield JC. 2008 Organization of vertebrate
annual cycles: implications for control mechanisms.
Phil. Trans. R. Soc. B 363, 425–441. (doi:10.1098/
rstb.2007.2149)
Tonra C, Reudink MW. 2018 Expanding the
traditional definition of molt-migration. Auk 135,
1123–1132. (doi:10.1642/AUK-17-187.1)
Newton I. 2008 The migration ecology of birds,
1st edn. London, UK: Elsevier.
Pyle P, Leitner WA, Lozano-Angulo L, Avilez-Teran F,
Swanson H, Limón EG, Chambers MK. 2009
Temporal, spatial, and annual variation in the
occurrence of molt-migrant passerines in the
Mexican monsoon region. Condor. 111, 583–590.
(doi: 10.1525/cond.2009.090085)
Rohwer S, Butler LK, Froehlich D. 2005 Ecology and
demography of east–west differences in molt
scheduling of Neotropical migrant passerines. In
Birds of two worlds: the ecology and evolution of
migration (eds R Greenberg, PP Marra), pp. 87–105.
Baltimore, MD: Johns Hopkins University Press.
Rohwer S, Manning J. 1990 Differences in timing
and number of molts for Baltimore and Bullock’s
orioles: implications to hybrid fitness and theories
of delayed plumage maturation. Condor. 92,
125–140. (doi:10.2307/1368391)

royalsocietypublishing.org/journal/rsbl

1.

5