Access to this full-text is provided by Wiley.
Content available from Journal of Avian Biology
This content is subject to copyright. Terms and conditions apply.
Page 1 of 14
© 2023 e Authors. Journal of Avian Biology published by John Wiley & Sons Ltd on behalf of
Nordic Society Oikos
is is an open access article under the terms of the Creative Commons
Attribution License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited.
Subject Editor: Suvi Ruuskanen
Editor-in-Chief: Jan-Åke Nilsson
Accepted 24 November 2022
doi: 10.1111/jav.03006
00
1–14
2023: e03006
JOURNAL OF
AVIAN BIOLOGY
www.avianbiology.org
Journal of Avian Biology e evolution of egg shape across birds has been associated with breeding ecology,
body shape constraints and nest microclimate, among other factors. We model the
effects of migratory status, climate, clutch size and egg volume on egg shape variation
over the distribution of fork-tailed flycatchers Tyrannus savana. Although migratory
status and climatic variables appear to be influencing intraspecific egg shape, these
effects are not significant when accounting for nest identity as a random factor (i.e.
eggs from the same clutch are more similar than to other clutches). Moreover, the
differences that we observe in egg shape are not explained by variation in egg size.
Finally, within a breeding population of migratory fork-tailed flycatchers, egg shape
does not vary with respect to egg-laying order and/or female wing length (standardized
by weight). Egg shape is highly variable within populations of fork-tailed flycatchers
but not within clutches, suggesting that female traits, apart from migratory status and
wing morphology, constrain egg shape variation.
Keywords: behavioral strategies, clutch size, egg shape, latitudinal gradient, maternal
effects, morphological evolution
Introduction
Avian eggs have evolved in a variety of shapes and colors (Hauber 2014), despite the
highly conserved functional nature of their structure which is key to embryonic devel-
opment. A hypothesis proposed to explain the evolution of egg shape in birds states
that a reduced abdominal space, resulting from adaptations for flight, constrains the
Egg shape variation across the distribution of the partially
migratory fork-tailed flycatcher Tyrannus savana
Valentina Gómez-Bahamón ✉1,2,3, Elizabeth R. Chen3, Diego T. Tuero 4, María de las Nieves Sabio4,
Kevin Tkach 4, Marcelo Assis5, Neander M. Heming 6, Miguel Â. Marini7 and John M. Bates 3
1Dept of Biological Sciences, Univ. of Illinois at Chicago, Chicago, IL, USA
2Dept of Biological Sciences, Pennsylvania State Univ., Muller Laboratory, University Park, PA, USA
3Negaunee Integrative Research Center, Field Museum of Natural History, Chicago, IL, USA
4Depto de Ecología, Genética y Evolución, IEGEBA (CONICET-UBA), Facultad de Ciencias Exactas y Naturales, Univ. de Buenos Aires, Intendente
Güiraldes 2160, Ciudad Universitaria, Buenos Aires, Argentina
5Programa de Pós-Graduação em Ecologia, Inst. de Ciências Biológicas, Univ. de Brasília, Brasília, Distrito Federal, Brazil
6Laboratório de Ecologia Aplicada à Conservação, DCB, Univ. Estadual de Santa Cruz, Ilhéus, BA, Brazil
7Laboratório de Ecologia e Conservação de Aves, Depto de Zoologia, Inst. de Ciências Biológicas, Univ. de Brasília, Brasília, Distrito Federal, Brazil
Correspondence: Valentina Gómez-Bahamón (vgbahamon@gmail.com)
Research article
14
Page 2 of 14
shape of eggs during egg formation (Inverson and Ewert
1991). e most widely studied aspect of egg shape varia-
tion is egg elongation (or ellipticity), but pointedness (polar
asymmetry) (Biggins et al. 2018) and concavity (Hays et al.
2020) are also important features of egg shape. Mixed evi-
dence has been found on whether elongation and asymmetry
are correlated (Stoddard et al. 2017), or if they are evolv-
ing independently (Montgomerie et al. 2021). Moreover, the
factors influencing egg shape have been suggested to differ
depending on the taxonomic scale at which comparisons are
addressed (Stoddard et al. 2019, Montgomerie et al. 2021).
Across the entire avian radiation, Stoddard et al. (2017)
found that species with more pointed wings tend to lay more
elliptical (elongated) and asymmetric (pointed) eggs than spe-
cies that have rounder wings. Rounded wings are also associated
with less dispersive, non-migratory and year-round territorial
defense (Sheard et al. 2020). Other anatomical features apart
from wing shape have also been found to correlate with egg
elongation. For instance, birds with a narrow pelvis and an
elongated post – acetabular region (the skeletal area behind
the insertion of the femur) tend to lay more elongated eggs
(Shatkovska et al. 2018, Montgomerie et al. 2021). In petrels
(family Procellariidae), those that have shorter pelvises tend
to produce rounder eggs, while females with longer pelvises
produce more elongated eggs (Warham 1990). Elongation was
also found to correlate with oviduct morphology in different
species of Galliformes (Montgomerie et al. 2021).
Although egg elongation is correlated with oviduct and
pelvis morphology, egg asymmetry (pointedness) appears
to be associated with adaptations to prevent breakage and
incubation efficiency, suggesting that different aspects of egg
shape may be evolving independently (Montgomerie et al.
2021). e substrate where birds lay their eggs also may be an
important factor contributing to variation in egg pointedness
(Montgomerie et al. 2021), as well as nest features. Studies
in common murres Uria aalge suggest that the pointed shape
of their eggs, which are laid on rocky surfaces of densely
populated cliffs, evolved to prevent the eggs from rolling off
(MacGillivray 1852, Gill 2007). is has been challenged by
studies showing that the pointed shape of eggs in this spe-
cies has evolved instead as an adaptation to maintain struc-
tural integrity by reducing pressure from the incubating
parent against rocky surfaces, and for protection from debris
excreted by parents and neighboring birds (Birkhead et al.
2017). Overall, all these studies point to multiple factors
associated with nesting ecology influencing pointedness in
common murres, as opposed to intrinsic anatomical con-
straints in adult females.
Egg shape has also been found to evolve in correlation with
climate and microhabitat conditions in Australian birds, where
more rounded eggs are found in open nests in hot and dry
environments, and more elongated eggs in humid environ-
ments and domed nests (Duursma et al. 2018). Moreover,
although in theory egg shape varies in association with clutch
size for optimizing incubation efficiency (Barta and Szekely
1997), clutch size was not found to be a factor that explains the
variation in egg shape across all birds (Stoddard et al. 2017).
However, Stoddard et al. (2017) did not assess the role of intra-
specific variation. It is well known that within and between
related avian species, clutch size varies in strong association
with latitude, with species that breed in the temperate regions
laying more eggs than those that breed in tropical regions (e.g.
in African birds (Moreau 1944), European birds (Lack 1947a,
b) and birds across the world (Jetz et al. 2008)). us, it may be
important to account for the effect of clutch size on egg shape
across the geographic distribution of species.
Different ways of quantifying egg shape have been used
in these studies (reviewed by Biggins et al. 2018), including
models based on geometric shapes and Fourier functions.
ese models differ in how well (or poorly) they fit different
shapes of eggs, the number of parameters required to describe
egg shape and the ease of implementation. Moreover, to our
knowledge, these models have not been used to calculate egg
volume. Developing methods to determine egg volume is
important for studies on eggs that lack data on their weight at
the time of laying, as is often the case for clutches in museum
collections and for eggs monitored in the wild which lose
weight as incubation progresses due to evapotranspiration.
An assumption of the hypothesis that the body cavity is
an important factor constraining egg elongation, is that egg
shape variation results from optimizing egg volume (Inverson
and Ewert 1991). Moreover, empirical evidence suggests that
egg volume is an important factor influencing egg shape
(Stoddard et al. 2017, Montgomerie et al. 2021). Here, we
develop a new geometric model based on conic sections and
tangent points that accurately fit passerine egg shapes with
three parameters (width, depth and height), and implement
this model to calculate both egg shape and volume variation
across the distribution of a Neotropical songbird with migra-
tory and non-migratory populations.
Fork-tailed flycatcher Tyrannus savana is a species composed
of a long distance austral migratory subspecies (T. s. savana)
and three non-migratory subspecies (T. s. monachus, T. s. cir-
cumdatus, T. s. sanctaemartae) (del Hoyo et al. 2004). e
non-migratory subspecies of fork-tailed flycatchers appear to
have evolved via an event of loss of migration on the wintering
grounds of the migratory subspecies, which led to reproductive
isolation and correlated evolution of wing shape, tail length
(Gómez-Bahamón et al. 2020a). and flight feather morphol-
ogy (Gómez-Bahamón et al. 2020b). Non-migratory fork-
tailed flycatchers have significantly more rounded wings, with
lower hand-wing index values (Gómez-Bahamón et al. 2020a)
than the migratory taxon. Both migratory and non-migratory
subspecies lay eggs in cupped shaped nests across their distribu-
tion (Marini et al. 2009, Salvador 2013, pers. obs.).
We studied the geographic variation of egg shape across
a latitudinal gradient in migratory and non-migratory fork-
tailed flycatchers. We ask whether the trends observed across
birds, where stronger flyers were found to lay more elon-
gated and pointed eggs hold at an intraspecific level (i.e. do
migratory fork-tailed flycatchers have more elongated and
asymmetric eggs than non-migratory fork-tailed flycatch-
ers). Using a model-based approach, we study the relation-
ship among egg shape and volume 1) between migratory
Page 3 of 14
and non-migratory fork-tailed flycatchers, 2) within clutches
across latitude and 3) within a breeding population. We also
account for climatic conditions, clutch size, egg volume and
breeding latitude as factors that may explain shape variation
of fork-tailed flycatcher eggs.
Methods
Data collection
We took horizontal photographs of eggs of fork-tailed fly-
catchers from museum collections and field studies in the
wild. We photographed eggs in a standardized horizontal
position to avoid errors in egg shape quantification caused
by the natural rest of eggs with the pointed end lower than
the blunt end (Biggins et al. 2018). In total, we took images
at 13 different institutions (including eggs from a total of
31 localities across 10 countries from Mexico to Argentina,
Supporting information) and additionally from three field
sites. Individual eggs were photographed at these field sites:
1) El Destino Nature Reserve in the Buenos Aires Province in
Argentina, 2) Tomo Grande Nature Reserve in the Vichada
Department of Colombia and 3) in Monteria, in the Córdoba
Department of Colombia (Supporting information). Eggs
were returned to the nests following photography and after
taking egg weight with a digital balance (weight was taken only
for eggs that were laid the same day of data collection), and
egg length and width. Measurements were made with an SPI
metric dial caliper with 0–150 mm range. We photographed
a total of 378 eggs corresponding to sampling localities across
the breeding distribution of this species (Fig. 1). Of these, 274
eggs belong to the migratory subspecies T. s. savana, 100 to
the non-migratory subspecies T. s. monachus and 3 to the also
non-migratory T. s. sanctaemartae. Among all eggs, 303 had
information about nest identity (ID), corresponding to a total
of 108 nests; 71 nests of T. s. savana, 35 to T. s. monachus and
2 to T. s. sanctaemartae (Supporting information).
Measuring egg shape
We implemented a novel mathematical model to describe
egg shape based on conic sections and tangent points. is
method (model 3, hereafter) requires three parameters (sharp
end, blunt end and the widest point of the egg). We used
these three control points to draw four ellipses that visually fit
the model to the shape of the egg and obtained shape param-
eters (these ellipses account for the sharp end, blunt end and
Figure 1. Sampling distribution. e map shows localities where the eggs were collected, or where pictures of eggs were taken in the wild.
e blue polygon represents the breeding distribution of the migratory subspecies T. s. savana, yellow represents the breeding distribution
of the non-migratory T. s. monachus and red of the non-migratory T. s. sanctaemartae.
Page 4 of 14
side ellipses shown in Fig. 2). e positions of the three con-
trol points are independent. We manually adjusted the three
control points to the contour of the egg (at the upper point
of the sharp end, lower point of the blunt end and the wid-
est point of the egg; Fig. 2A) until the model approximated
the shape. We also automated this process using an image
processing algorithm that identifies foreground from back-
ground. is automation was only possible in images with
a good distinction between egg contour versus background.
Additionally, we visually checked that the control points
from the automated processing were placed correctly on the
sharp end, blunt end and the widest point of the egg, respec-
tively. Graphically, four ellipses describe the contour of these
eggs (in Fig. 2A and B they correspond to the cyan, magenta
and yellow ellipses). e cyan ellipse is constrained by the
magenta ellipse; they have the same aspect ratio and the same
curvature where the two ellipses are tangent. Given these two
constraints, the matching curvatures and the matching aspect
ratios, three conditions are enough to solve for the cyan
ellipse (i.e. which are the lengths of A, B and ι + b, in Fig. 2B).
When the three control points in the image are chosen, a
geometric transformation is done so that these control points
correspond to the following coordinates in a plane: ↑ (0,+1),
↓ (0,−1) and ↙ (A,F). e other points are determined
following:
-á+ñ
á- ñ
á+ ñ
á- ñ
á+ ñ
¯á-ñ
01
01
,
,
,
,
,
,
af
af
AF
AF
áñ==+
=+
-
æ
è
çö
ø
÷
áñ=á
ñ=
-
AF BF
F
F
ab AB
fb
,
,,
input1
1
1
1
3
c
cc
where χ is the scaling factor of the cyan and magenta ellipses.
e implicit equations for the ellipses are calculated with the
following equations:
Ýæ
è
çö
ø
÷+-
æ
è
çö
ø
÷<
Ý+
æ
è
çö
ø
÷+-
--
æ
è
çö
ø
÷-+
x
a
yf
b
x
AQyF
FQx
A
22
22
1
12kQQ yF
FQQ
x
A
yF
F
æ
è
çö
ø
÷-
--
æ
è
çö
ø
÷
+<
Þ-
æ
è
çö
ø
÷+-
--
æ
è
çö
ø
÷-
1
21
12
22
QQkxx
A
yF
F
x
A
yF
B
-
æ
è
çö
ø
÷-
--
æ
è
çö
ø
÷
+<
ßæ
è
çö
ø
÷+-
æ
è
çö
ø
÷<
QQ
Q
1
21
1
22
kk
k
=- +
-
æ
è
çö
ø
÷=+
11
11
2
F
FQ
where the broad arrows represent the position of the ellipses
on Fig. 2A and B. e standard equation of an ellipse is equal
to 1, but because we were interested in the inside and not the
outside, the right-hand side of the equation is < 1. Moreover,
κ and ϙ are intermediate constants to simplify the equations.
e equivalent equations in parametric form to draw them
in a plane are:
Figure 2. Egg shape measurements. (A) Example of an egg photograph demonstrating the newly described geometric (here model 3) that
we used to fit egg shape. e arrows point to the three control points that correspond to the upper point of the sharp end, lower point of
the blunt end and the widest point of the egg. (B) Graphic representation of newly described model 3, which is based on conic sections and
tangent points that we used to fit egg shape. e equations correspond to the shape indices that describe elongation and two types of asym-
metry (center asymmetry and polar asymmetry).
Page 5 of 14
Ý á
Ý
Ý
ñ=áñ+á±ñ +á ±ñ
áñ=á ñ+ á- -ñ+
xy fa b
xy FAF
,, ,cos ,sin
,, ,
000
01
1
qq
Q//( )
,c
os
/( ),sin
,,
21
1
1
121
1
1
0
-
+á- -ñ
++
+á+ -ñ
áñ=á
k
kJ
k
kJ
AF
AF
xy Fññ+ á+ -ñ+-
+á+ -ñ
++
+á- -ñ
QAF AF
KAF
,/( ),cos
/( ),si
1121
1
1
121
1
1
k
kJ
knn
,, ,cos ,sin
J
qqßá ñ=áñ+á±ñ +á ±ñ
xy FA B000
-+££
++
--££+
--
--
sin/() sin/
()
cos/() cos/(
11
11
12
11
21
12
11
2
kJ k
kJ 11 -k)
We implemented this model using a newly developed graphi-
cal interface which we will publish elsewhere. is app was
written using the MacOS Xcode App, the programing lan-
guage Swift and by leveraging the computing power using the
Metal graphics platform. e app allows the fitting of model
3 to the photos of the eggs manually and automatically for
photos in which the foreground and background can be dis-
tinguished. With the app, all eggs were rescaled to have a
height of 2 units, which allowed all measurements to be com-
parable, independent of egg size. However, we also included a
scale (a ruler) to compute size (volume and surface area). For
those pictures that did not include a ruler, we measured egg
height using a caliper, and incorporated these measurements
into model 3 for scaling.
Egg volume and surface area
We quantified relative volume based on the ellipses’ param-
eters as follows:
vol=+ -
-+
()
-+
()
--
-+
æ
è
ç
ç
ç
AFF
FF
FF BF
FF
2
22
22
2
11
16
56 531
16
()
()
cos--
-
æ
è
ç
ç
ö
ø
÷
÷
ö
ø
÷
÷
1
2
4
1
F
F
()
And we quantified relative surface area with the following
formulas:
area
if area
=¶
¶
¶
æ
è
çö
ø
÷+¶
¶
æ
è
çö
ø
÷
ß<
ß
ß
=+
-
+
ò
1
2
1
4
22
J
JJJJ
xxy
AB AA[] BB
A
B
A
B
AB
sin
[]
--æ
è
çö
ø
÷
-æ
è
çö
ø
÷
æ
è
ç
ç
ç
ç
ç
ç
ö
ø
÷
÷
÷
÷
÷
÷
ß= =
1
2
2
1
1
if area 11
4
1
4
1
1
1
2
2
AA B
AB AA B
A
B
A
B
()
[]
sinh
+
ß> =+
æ
è
çö
ø
÷-
æ
è
çö
ø
÷-
æ
è
-
if area
çç
ç
ç
ç
ç
ç
ö
ø
÷
÷
÷
÷
÷
÷
area
if area
=¶
¶
¶
æ
è
çö
ø
÷+¶
¶
æ
è
çö
ø
÷
ß<
ß
ß
=+
-
+
ò
1
2
1
4
22
J
JJJJ
xxy
AB AA[] BB
A
B
A
B
AB
sin
[]
--æ
è
çö
ø
÷
-æ
è
çö
ø
÷
æ
è
ç
ç
ç
ç
ç
ç
ö
ø
÷
÷
÷
÷
÷
÷
ß= =
1
2
2
1
1
if area 11
4
1
4
1
1
1
2
2
AA B
AB AA B
A
B
A
B
()
[]
sinh
+
ß> =+
æ
è
çö
ø
÷-
æ
è
çö
ø
÷-
æ
è
-
if area
çç
ç
ç
ç
ç
ç
ö
ø
÷
÷
÷
÷
÷
÷
To find scaled values, we multiplied the resulting relative vol-
ume and surface area by the volume or area (respectively) of a
sphere that has a radius equal to ½ of the egg’s length.
Egg shape indices
We calculated three shape indices: elongation, center asym-
metry and polar asymmetry with the following equations
(Fig. 2B for reference):
Elongation
bB
A
Center asymmetryb
B
PolarasymmetryA
=++
=+
=
()
()
i
i
2
aa
Because both asymmetries are strongly correlated, we focused
most of the downstream ecological and morphological analy-
ses described below on center asymmetry.
We fitted single predictor linear regressions to all the inde-
pendent variables listed below, with elongation and center
asymmetry as dependent variables. To account for nest ID,
we also fitted linear mixed models including nest ID as a ran-
dom variable. We fitted mixed models using the lmer func-
tion in the R package lme4 (Bates et al. 2015). To test for
significance, we ran a Wald test using the R package car (Fox
and Weisberg 2019), which assesses the constraints of sta-
tistical parameters against a null hypothesis with a χ2 distri-
bution. After comparing Akaike information criterion (AIC)
values of all single predictor linear and mixed models, we ran
multiple predictor regression models with the variables that
had the lowest AIC values and those with 2 units of differ-
ence among them. e only variables fitting these conditions
were: nest ID as a random variable, with center asymmetry
as the dependent variable, and migratory status and clutch
size as the independent variables. We then ran our multiple
Page 6 of 14
predictor models on those variables. Finally, we also ran all
models for eggs laid only by migratory birds and only by non-
migratory birds. All analyses were done using R ver. 4.0.2
(www.r-project.org).
Migratory status
We assigned migratory status to eggs laid by birds of the T.
s. savana subspecies (which nest in southern South America)
and non-migratory status (year-round resident) to eggs laid
by birds of T. s. monachus and T. s. sanctaemartae (which
nest from Northern South America to Mexico). As part of
our fieldwork, we have documented the presence of T. s.
monachus in eastern Colombia in the breeding and non-
breeding seasons confirming its non-migratory status.
Moreover, we confirmed that there are observations of year-
round fork-tailed flycatchers in northern South America
from data in the eBird platform (Sullivan et al. 2009). We
include the three eggs from sanctaemartae with non-migra-
tory monachus because based on population genomic stud-
ies, sanctaemartae is clustered genetically within monachus
(Gómez-Bahamón et al. 2020), and both populations are
non-migratory. Moreover, sanctaemartae eggs are not outli-
ers in shape when compared to monachus (the range in elon-
gation measured for monachus eggs is: 1.198–1.560. For
center asymmetry, the range in monachus is: 1.098–1.539.
e values for elongation for the three sanctaemartae eggs
are: 1.347, 1.395 and 1.442, and for center asymmetry:
1.456, 1.341 and 1.334).
Latitude
We determined egg-laying latitude based on georeferenc-
ing locality data where the eggs were collected or studied in
the wild using Google Earth. We wrote a search term on the
Google Earth search option with the name of the locality
written on the specimen’s label. We then allowed the Google
Earth search software to zoom in to the locality. We ran-
domly chose a point from this zoomed image. We assume
this approximation is appropriate given the resolution of the
downstream ecological raster analyses we performed, which
were at a resolution of 10 min of a degree.
We separated eggs laid by migratory and non-migratory
females and fitted single predictor linear models with latitude
as an independent variable and shape indices as dependent
variables. Because our dataset included eggs laid north and
south of the equator, we fitted a model including both eggs
laid by migratory and non-migratory birds with absolute val-
ues of latitude as an independent variable and shape indices as
dependent variables, and additionally fitted latitude as a poly-
nomial by adding latitude2 as independent variable. ese
different ways of describing latitude were treated as indepen-
dent single predictor variables in linear models. Latitude is a
surrogate for climatic and ecological processes that might not
be represented in the other variables that we tested. us, we
included latitude to understand the role of geographic pat-
terns, but did not consider it as a direct biological factor.
Clutch size
We refered to clutch size as the number of eggs collected from
a single nest or the total number of eggs that were found in a
nest in the field. One caveat is that because there is a chance
that a nest was collected before the female had ended laying
the totality of a clutch, or that we photographed eggs in the
wild before the female was done laying, minimum clutch size
is an underestimate.
We also checked that none of the clutches analyzed
contained eggs from the parasitic shiny cowbird Molothrus
bonariensis. Shiny cowbird eggs significantly differ in shape
and color to those of fork-tailed flycatchers. Moreover, we
assumed that eggs found in a nest were from the same female
because throughout our field studies, we have never found
more than one egg laid on the same day (females of fork-
tailed flycatchers lay one egg every day and begin incubation
when they reach their maximum cultch size). Also, we have
no knowledge of reports of multi-female egg-laying in the
same nest within the genus Tyrannus.
Climatic variables
We extracted bioclimatic variables from the WorldClim data-
base (Fick and Hijmans 2017) using the R package raster
(Hijmans and van Etten 2012) to combine with our georefer-
enced locality data. We used a download resolution of 10 min
of a degree and extracted variables using the bilinear method
option. We focused on isothermality and temperature sea-
sonality (standard deviation × 100). Isothermality is equal
to the mean of monthly maximum temperature minus the
minimal temperature, divided by annual temperature range
(maximum temperature of the warmest month minus mini-
mal temperature of the coldest month), all multiplied by 100.
The effect of Nest ID
Because females lay more than one egg in a nest, those from
the same nest might be more similar to each other compared
to those from different nests. For instance, in other species of
birds, eggs from the same clutch are more similar in size to
each other than to eggs of other clutches (Christians 2002),
highlighting the importance of maternal effects for egg traits.
Nest ID was used in mixed linear models as a random variable.
Egg-laying order, female wing length and egg
elongation within a breeding population
To test whether there is a relationship between female flight
morphology and egg shape within a breeding population in
which all individuals are migratory, we studied egg shape in
the wild at El Destino Nature Reserve in Argentina. We com-
piled data from nest monitoring across 7 breeding seasons: in
2011–2015 and 2018–2019. We measured egg length and
width using a dial caliper. Because we did not have photo-
graphs of all these eggs, we estimated egg elongation by divid-
ing egg length over width for this analysis. We also weighed
Page 7 of 14
eggs with a digital balance and marked them with a num-
ber indicating egg-laying order with a non-toxic waterproof
marker. Because females lay one egg every day until the clutch
is complete, we had information for some clutches about the
order in which the eggs were laid. If the nest was found when
all, or some, of the eggs had already been laid, we did not
include those clutches for laying order analyses.
We captured the breeding females by using mist nets
placed at ~4 m from the nest. We banded them with a unique
metal ring and color bands with unique combinations for
later resighting, and we measured their wing length and body
weight. Because weight in birds fluctuates across the yearly
cycle (especially in migratory birds) and throughout the
day, females were only captured during the breeding season
at dawn. One caveat is that females may be carrying an egg
in their oviduct when captured, so there might be variation
in adult weight due to this. We fitted a linear model with
elongation (egg length/width) as the dependent variable and
female wing length standardized by body weight as the inde-
pendent variable. Finally, to determine if the eggs laid first
or last were more spherical or elongated, we ran an ANOVA
with egg order as the treatment variable.
Results
Migratory status
We found that eggs from migratory fork-tailed flycatch-
ers tend to be more elongated than eggs of non-migratory
fork-tailed flycatchers (Fig. 3A). Although the difference was
significant (Table 1; F = 4.56, p = 0.03) and it represented
the model with the lowest AIC value of all the models tested
(Table 1), the effect size was small (Table 1; R2 = 0.01) and
non-significant when accounting nest ID as a random factor
(Table 1). With respect to center asymmetry and polar asym-
metry, eggs of migratory birds tended to be less asymmetric
than those of non-migrants, but the difference was not sig-
nificant (Fig. 3B, C and Table 1; F = 2.97, p = 0.09). In terms
of egg size, there was no difference between eggs of migratory
and non-migratory fork-tailed flycatchers for either mean vol-
ume or surface area (Fig. 3D and E, respectively). However,
egg sizes in migratory birds occupied a larger morphological
space than eggs of non-migratory birds, with migratory birds
laying the largest and smallest eggs in our dataset (Fig. 3D).
Nevertheless, there was a weak and non-significant correla-
tion among elongation and volume (Fig. 3F and Table 1;
F = 3.70, p = 0.06), with a small effect size (R2 = 0.01).
Latitude
We found a non-linear trend in egg shape for both elonga-
tion (Fig. 4A) and center asymmetry (Fig. 4B), in which
eggs towards equatorial latitudes were more elongated, and
none approximated a sphere (Fig. 4A and B). is trend was
significant for elongation when treating latitude as an abso-
lute value (Table 1; F = 5.84, p = 0.02) and when adding a
polynomial term of latitude2 to the model (Table 1; F = 4.49,
p = 0.03). Eggs towards temperate regions (both to the north
and south) tended to occupy a larger shape space (with elon-
gated, asymmetric and more spherical eggs (Fig. 4A and B)).
Center asymmetry was not found to be associated with abso-
lute latitude (Fig. 4B and Table 1; F = 0.41, p = 0.52), but
it did weakly when adding latitude2 to the model (Table 1;
F = 3.81, p = 0.05). Volume did not show the latitudinal
trend of eggs from tropical latitudes occupying a different
size space (Fig. 4C), as was found for egg shape.
Clutch size
Our data reported here, and our field observations in the two
study areas in the wild (one located in Colombia and one
in Argentina), further documented that fork-tailed flycatch-
ers at tropical latitudes do not lay more than 3 eggs, while
in temperate latitudes they lay up to 5 (Jahn et al. 2014).
We found that clutch size was smaller towards the equa-
tor (Fig. 4D, when considering absolute values of latitude:
F = 8.621, p = 0.004). However, clutch size was not found to
have a linear association with elongation (Table 1; F = 0.86,
p = 0.36) nor asymmetry (Table 1; F = 0.1, p = 0.76).
Climatic variables
Neither isothermality nor temperature seasonality were found
to be significant factors explaining the trends observed in egg
elongation (Fig. 4E, F and Table 1; isothermality F = 0.31,
p = 0.58; temperature seasonality F = 0.12, p = 0.73). However,
center asymmetry did seem to be associated with changes in
both these variables, with more asymmetric eggs occurring
at sites with increasing isothermality (Fig. 4G and Table 1;
F = 8.09, p = 0.005) and less asymmetric eggs with increasing
temperature seasonality (Fig. 4H and Table 1; F = 9.05, p <
0.001). However, these patterns were non-significant when
including nest ID as a random variable (below).
The effect of Nest ID
Eggs from the same nest were more similar to each other in
shape indices than to eggs of other clutches, in both migra-
tory and non-migratory fork-tailed flycatchers (Fig. 5). When
we included nest ID as a random variable to our models, the
effect of migration on elongation was not significant (Table 1;
χ2 = 1.17, p = 0.28), nor were the effects of any of the other
variables tested (Table 1). e effect of migratory status on cen-
ter asymmetry was also not significant when accounting for
nest ID as a random variable (Table 1; χ2 = 0.65, p = 0.42),
nor was the effect of any of the other variables tested (Table 1).
Model comparison
Overall, the model with lowest AIC value was a single pre-
dictor linear model in which migratory status was the inde-
pendent variable and elongation was the dependent variable
(Table 1). For center asymmetry, the model with lowest
AIC value included migration as the independent variable.
However, both models had a difference in AIC with a null
Page 8 of 14
model of 2.6 units for elongation and 0.97 units for center
asymmetry (Table 1).
Eggs of migratory and non-migratory populations
considered independently
We found that temperature seasonality had a significant
effect on egg elongation and center asymmetry of migratory
birds, but not for those that are non-migratory (Supporting
information). Isothermality had also a significant effect on
egg elongation in eggs laid by migratory females, and latitude
on center asymmetry (Supporting information). However,
these effects were insignificant when accounting for nest ID.
Models without nest ID were found significant because of
the lack of independence between same clutch eggs, i.e. same
clutch eggs are not true replicates, and not climatic variables.
Egg elongation within a migratory breeding
population
Egg shape did not vary with egg-laying order in a migratory
population in Argentina (Supporting information; F = 2.76,
p = 0.10, n = 89 nests). However, based on a Tukey–Kramer
post-hoc analysis, we found that the first egg laid tended to
be significantly lighter than the eggs that follow (Supporting
information, F = 6.21, p = 0.01). Egg elongation was not
correlated with female wing length/body weight within this
breeding population of migratory fork-tailed flycatchers in
Argentina (Supporting information; F = 1.70, p = 0.19). Egg
elongation was also not correlated with egg weight in this
population (Supporting information; F = 0.74, p = 0.39).
Finally, egg elongation did not vary with sampling year
(Supporting information).
In summary, we found that although eggs of migratory
and non-migratory fork-tailed flycatchers differed in shape
(Fig. 3), migratory status was not a significant predictor of
egg shape when accounting for nest ID (Fig. 5 and Table 1).
We also found that eggs from the same clutch were more
similar to each other than to those from other clutches
(Fig. 5). Moreover, eggs laid by non-migratory birds did not
occupy the more spherical morphospace (Fig. 4A). Likewise,
although egg asymmetry was correlated with temperature
seasonality and isothermality in single factor models (Table 1
and Fig. 4), these effects were non-significant when account-
ing for nest ID. Finally, clutch size and egg volume also were
not significant predictors of egg shape.
Figure 3. Migratory status and egg shape. (A) Elongation of eggs laid by migratory (1-blue) fork-tailed flycatchers is significantly greater
than that of eggs laid by non-migratory (2-yellow) fork-tailed flycatchers. (B) e center asymmetry and (C) polar asymmetry do not sig-
nificantly differ between eggs from migratory (1-blue) and non-migratory (2-yellow) fork-tailed flycatchers. Egg size does not differ between
migratory types as observed by (D) volume measurements and (E) surface area, although variation is larger in migratory birds. (F) Egg
elongation and volume are not correlated.
Page 9 of 14
Table 1. Models with intrinsic and extrinsic factors influencing egg shape in fork-tailed flycatchers.
Single variable models Elongation Elongation with nest ID as random variable
Predictor AIC Coeff. F t p R2AIC Coeff. fixed t p X2
Null −1012.6 1.350623 412.2 < 2e-16 −955.91 1.351142 243.3
Migration status −1015.2 −0.01547 4.563 −2.136 0.0333 0.01187 −948.02 −0.012616 −1.082 0.2794 1.1701
Isothermality −902.36 0.000182 0.314 0.56 0.576 −0.002 −856.969 0.0003482 0.678 0.4977 0.4599
Temperature seasonality −902.16 −8.37E-07 0.116 −0.341 0.733 −0.0026 −846.977 −1.48E-06 −0.36 0.7192 0.1293
|Latitude| −907.87 −0.018085 5.842 −2.417 0.0162 0.01408 −864.51 −0.0171 −1.463 0.1436 2.1394
Latitude + latitude2−906.52 −0.0003138 4.485 −2.118 0.0349 0.01017 −856.65 −0.0003167 −1.245 0.2132 1.5496
Volume −992 1.76E-05 3.695 1.922 0.0553 0.00719 −903.57 1.23E-05 1.055 0.2914 1.1129
Clutch size −803.1 −0.003976 0.856 −0.925 0.356 −0.0005 −945.453 4.31E-05 0.007 0.9941 1.00E-04
Single variable models Center asymmetry Center asymmetry with nest ID as random variable
Predictor AIC Coeff. F t p R2AIC Coeff. fixed t p X2
Null −846.22 1.295522 318 < 2e-16 −707.282 1.299424 199.5
Migration status −847.19 0.015544 2.968 1.723 0.0857 0.00514 −699.193 0.011079 0.808 0.419 0.6531
Isothermality −745.91 0.0011625 8.087 2.844 0.00473 0.02048 −617.217 0.0009912 1.578 0.1146 2.4897
Temperature seasonality −746.86 −0.013105 9.049 −3.008 0.00282 0.02319 −622.574 −0.012203 −1.741 0.08168 3.0312
|Latitude| −738.28 −0.006126 0.407 −0.638 0.524 −0.0018 −621.166 −0.006056 −0.424 0.6715 0.1798
Latitude + latitude2−741.68 0.0003684 3.805 1.951 0.0519 0.00821 −614.348 0.0003129 1.012 0.3113 1.025
Volume −820.39 1.70E-05 2.188 1.479 0.14 0.00318 −660.795 2.41E-05 1.442 0.1494 2.0781
Clutch size −654.36 0.001694 0.095 0.308 0.758 −0.003 −697.504 0.003876 0.552 0.5809 0.3048
Multi variable model Elongation with Nest ID as random variable
Fixed effects AIC Coeff. t Corr. p X2
Migration + clutch size −937.57 −0.085
Migration 0.0127036 −1.08 0.28 1.1672
Clutch size 0.0005756 0.099 0.9214 0.0097
Coeff. = coefficient.
Bold text indicate p values <0.05.
Page 10 of 14
Discussion
Intraspecific variation in egg shape has not been studied in
detail in many avian species. We studied the association of
egg shape with climatic factors, clutch size, egg volume and
migratory status across the distribution of fork-tailed fly-
catchers Tyrannus savana. We approached this question from
different levels of variation including within clutches, within
a breeding population and among migratory and non-migra-
tory populations. We found that when we accounted for nest
identity as a random factor, there was no significant effect of
the variables tested on egg shape variation. Our study cau-
tions that if nest identity is not considered in studies on egg
trait ecology and evolution, erroneous conclusions may result
from factors that appear to be significant without including
nest identity in the models.
Figure 4. Egg morphology across latitude and climatic variables. (A) Egg elongation across latitude, (B) center asymmetry across latitude,
(C) volume across latitude and (D) clutch size across latitude. (E) Elongation as a function of isothermality, and (F) as a function of tem-
perature seasonality. (G) Center asymmetry as a function of isothermality and (H) as a function of temperature seasonality. Eggs from
migratory fork-tailed flycatchers are in blue and those of non-migratory birds are in yellow. Colored lines represent respective linear regres-
sions and gray lines represent polynomial regressions with an added term of latitude2.
Figure 5. Within nest egg shape variation. (A) Egg elongation within nests of migratory fork-tailed flycatchers ordered by latitude, (B)
elongation within nests of non-migratory fork-tailed flycatchers ordered by latitude, (C) center asymmetry within nests of migratory fork-
tailed flycatchers ordered by latitude and (D) center asymmetry within nests of non-migratory fork-tailed flycatchers ordered by latitude.
Blue colored boxes represent nests from migratory birds and yellow boxes those from non-migratory birds.
Page 11 of 14
Egg morphology in migratory and non-migratory
fork-tailed flycatchers
It is well known that anatomical differences exist between
migratory and non-migratory birds. Migratory birds tend to
have more pointed wings (Lockwood et al. 1998, Phillips et al.
2018), and species that colonize small islands and are year-
round residents have often evolved flightless morphologies,
with a reduction in muscle fiber and the evolution of longer
hindlimbs (Wright et al. 2016). Although not yet quantified,
anatomical differences among fork-tailed flycatcher females
from migratory and non-migratory populations may not be
large enough to impose a significant pressure affecting egg
shape.
On the other hand, several ecological traits tend to be phy-
logenetically conserved due to niche conservatism, as are sim-
ilarities of closely related species in life history traits, and the
factors that influence them (Wiens and Graham 2005). us,
morphological traits within species and among closely related
species that are associated with a species’ niche may not affect
the correlated evolution of egg shape. However, observed dif-
ferences at higher taxonomic levels (Stoddard et al. 2017)
may be reflecting body plans that are ecologically very differ-
ent to the extent that egg shape is correlated with ecology at
the macroevolutionary scale (i.e. eggs of stronger flyers tend
to be elongated). Within fork-tailed flycatchers, egg shape did
not significantly differ between migratory and non-migratory
populations in our analyses when including nest ID as a ran-
dom factor, suggesting that instead, maternal effects, repro-
ductive traits or genetic factors associated with egg shape are
constraining variation.
We also found that fork-tailed flycatcher eggs in equato-
rial latitudes did not tend to be more spherical (Fig. 4). We
found that morphospace occupied by eggs increases with lati-
tude (Fig. 4A-D), suggesting that some factor in the tropics
is constraining egg shape as well as female traits, although we
caution that the lower sample size in these latitudes could be
skewing our results.
Another possibility is that ecological pressures associated
with life history traits may be influencing egg shape evolu-
tion. For example, incubation periods in tropical areas have
been found to be approximately 10% longer, and chick
growth rates about 23% slower (Austin et al. 2020). It has
been hypothesized that these slower developmental rates
occur because parents incubate less frequently to avoid pre-
dation, in turn reducing nest temperature and resulting in
slower developmental rates (Austin et al. 2020). Egg shape
could be responding to optimizing egg temperature under
these ecological conditions. However, there are discrepancies
across bird studies on the assumption that predation rates are
higher towards the tropics suggesting that instead the trend is
that there are higher predation rates in transformed environ-
ments irrespective of latitude (Martin 1996). Longer devel-
opmental times still suggest that towards the tropics birds
incubate for shorter amounts of time per day (Austin et al.
2020).
For embryonic development to be successfully accom-
plished, the temperature at which an egg is maintained dur-
ing incubation must remain within a narrow temperature
range (Turner 2002), and gas exchange must be balanced
with dehydration (Duursma et al. 2018). In theory, diffusiv-
ity of water vapor increases at higher elevations due to lower
barometric pressures, and potential adaptations to prevent
dehydration could require a combination of adaptations such
as: reduction of egg pore area, increase in initial water con-
tent, increase in egg shell thickness and alteration of adult
nest attendance (Carey 1980). Similar factors may be at play
in warmer and drier areas in which alterations in eggshell
structure can influence shape. Moreover, more spherical eggs
are expected to be more favorable in areas where water vapor
diffusion is higher, because more spherical eggs have less sur-
face area for a given volume than an elongated egg (Hoyt
1976). In fact, eggs laid in hot and dry areas in Australia were
found to be more rounded than those in regions that are wet
and cooler (Duursma et al. 2018).
Our data for fork-tailed flycatchers did not follow this pat-
tern, because eggs laid in equatorial, warmer and less seasonal
environments, tended to be more elongated relative to the
more spherical eggs from temperate latitudes (Fig. 4). We
also found no evidence of higher initial water retention via
increase in egg volume, because egg size and surface area did
not vary with latitude or migratory status (Fig. 3D and E),
although it is possible that shell volume is not a precise indi-
cator of water content.
Egg morphology within a breeding population of
migratory fork-tailed flycatchers
Macroevolutionary research suggests that female anatomical
features associated with flight likely influence egg elongation
(Stoddard et al. 2017). At the intraspecific level, we find that
egg elongation is not correlated with female wing length/
female weight (Supporting information). In other intraspe-
cific studies, great tits Parus major exhibit significantly differ-
ent egg morphologies with egg-laying order (You et al. 2009,
Hargitai et al. 2013), with later eggs being more spherical,
having thinner shells and being larger (Hargitai et al. 2013).
We found that the fork-tailed flycatcher eggs did not differ
in shape with egg-laying order (Supporting information). An
important life history difference between these two species is
that hatching in great tits is asynchronous (You et al. 2009),
while hatching in fork-tailed flycatchers occurs in synchrony
(i.e. females do not begin incubating until they have laid
all their eggs in a clutch). is raises the question as to how
important is breeding synchrony to egg traits. Although
egg volume has been found to decrease with laying order
in spotless starlings Sturnus unicolor, another species with
synchronous hatching (Monclús et al. 2017), in fork-tailed
flycatchers, the first egg was significantly lighter than subse-
quent eggs (Supporting information). ese data sets suggest
that there is much to be learned about the drivers of mor-
phological differences of eggs associated with laying order.
Page 12 of 14
Our finding that egg shape variation in populations of
migratory and non-migratory fork-tailed flycatchers is high,
but constrained at the nest level, suggests that egg shape or
the anatomical features associated with it, might be under a
strong genetic control. In a review of different avian stud-
ies, Christians (2002) found that egg size at the population
level is widely variable, but not within a female nest. Neither
female weight, nor experimental manipulation of food avail-
ability or ambient temperature, are factors that influence egg
size variation in different species of birds (Christians 2002).
e heritability of egg size was found to be high (Christians
2002), suggesting that a genetic component associated with
egg traits may be stronger than environmental influences.
The possible effects of female age on egg shape
and size
In some species of birds, egg-laying date and egg size are
known to change with female age; older females tend to lay
larger eggs and lay earlier in the nesting season compared
to younger females (Hipfner et al. 1997). Moreover, eggs of
black-legged kittiwakes Rissa tridactyla increase in volume
and shape index (breadth × 100/length) with female age
(Coulson 1963), while in spotless starlings, eggs increase in
yolk volume with female age but not in size (Muriel et al.
2019). Age is not a factor included in our models, future
studies on known-aged fork-tailed flycatcher and other spe-
cies with migratory and non-migratory populations could
shed light on the underlying mechanisms of observed varia-
tion among life history traits, environmental factors and egg
morphology across distributions.
Egg morphology across years
Eggs shape among migratory and non-migratory fork-tailed
flycatchers did not change consistently with sampling year in
a span of 140 years (from 1880 to 2020) (Supporting infor-
mation). Non-migratory females appeared to lay more elon-
gated eggs towards the present (Supporting information),
while eggs laid by migratory females were not significantly
changing over time. We found the same trend with asym-
metry; non-migratory females appeared to be laying more
asymmetric eggs towards the present (Supporting informa-
tion). However, we did find a consistent and significant trend
for both eggs laid by migratory and non-migratory females
becoming smaller towards the present (Supporting informa-
tion). is trend of eggs becoming smaller over time is con-
sistent with studies showing that migratory birds themselves
are evolving to be smaller (Weeks et al. 2020), as are resident
Amazonian birds (Jirinec et al. 2021), which appears to be
linked to climate change (Weeks et al. 2020, Jirinec et al.
2021). Future studies should focus on documenting egg
volume across species of birds and the relationship with
chick and adult development. However, within a breeding
population of migratory fork-tailed flycatchers, over 7 years
sampling we find there is no evidence of shape (Supporting
information) or size change. is may be an insufficient time
period to find changes, and therefore highlight the need for
more datasets sampling across multiple decades (and longer)
to better understand these patterns and processes.
e patterns that we document here would not have been
possible to study if we had not measured eggs both in the
field and in museum collections across different localities. As
distributions of species become broader, accounting for varia-
tion in egg shape across geography, as opposed to including
measurements from eggs representing a few sites of a species’
distribution, is also important. Marini et al. (2020) docu-
mented that there are ~5 million eggs of ~1.97 million egg
sets in museum collections worldwide which is well of tempo-
ral data for some species in some regions; however, they also
note that there are still areas of the world and species that are
poorly sampled such as tropical South America. Measuring
and photographing eggs in the wild across the Neotropics,
and creating regional collections, would help document,
monitor and understand change over time in a critical life
history phase of birds.
Conclusions
Our results highlight that although there was a wide varia-
tion in egg shape across the broad distribution of fork-tailed
flycatchers, there was little variation at the nest level. is fact
about intra-nest variation eliminates significant patterns of
factors that appeared to be significant when nest identity was
not included as a random factor. We argue that accounting
for nest identity is imperative in studies on the ecology and
evolution of egg traits, as eggs from the same nest tend to be
more alike to each other in multiple morphological traits.
Acknowledgements – We thank the following museum collections
where we photographed eggs: Collecção José Caetano Sobrinho,
Naturhistorisches Museum Wien, American Museum of Natural
History, Western Foundation of Vertebrate Zoology, Museum
für Naturkunde Berlin, Museum für Naturkunde Berlin, Max
Schönwetter Collection at Universität Halle-Wittenberg, Field
Museum of Natural History, Universidade de Brasília, Museum of
Comparative Zoology, Max Schönwetter Collection at Universität
Halle-Wittenberg, Instituto Alexander von Humboldt, Museo
de la Plata, Museo Argentino de Ciencias Naturales and the staff
who assisted us during our visits. We thank Reserva Natural Tomo
Grande and Reserva Natural el Destino and SELVA: Investigación
para la conservación en el Neotrópico for supporting our research.
We are grateful to multiple field assistants including Luz Dary Rivas,
John Fredy Barrera, Steven Barrera Rivas, María Alejandra Meneses,
María Isabel Castaño, Belén Fuentes and Jesús A. Rodriguez. We
thank Ben Marks for orientation at the egg collection at the Field
Museum. We thank Nicholas Bayly for sharing a picture of two eggs
of a nest monitored in Córdoba, Colombia. We thank Henry F.
Howe, Marcella Baiz, Felix Grewe, Richard Ree, Trevor Price, Chris
Clark and David Toews for insightful discussions and comments on
manuscripts, as well as members of the Toews and Price labs. We
thank Trevor Price, Nick Bayly and Roberto Márquez for advice
on statistical analyses. We thank the anonymous reviewers for their
careful reading of our manuscript and their insightful comments
and suggestions.
Page 13 of 14
Funding – is project was funded by the Grainer Bioinformatic
Center at the Field Museum of Natural History, the Elmer Hadley
Graduate Research Award from the Department of Biological
Sciences at the University of Illinois at Chicago, the Field Museum
Armour Gradute Student Fellowship, the Consejo Nacional de
Investigaciones Científicas y Técnicas (CONICET-Argentina:
11220170100592CO) and the University of Buenos Aires,
Argentina: 21020180700302BA.
Ethics statement – e eggs measure and monitored in the wild are
under a research permit in Argentina and Colombia. No eggs were
collected during this study.
Author contributions
Valentina Gómez-Bahamón: Conceptualization (lead);
Data curation (supporting); Formal analysis (lead); Funding
acquisition (lead); Investigation (lead); Methodology (lead);
Project administration (equal); Software (supporting);
Supervision (supporting); Validation (lead); Visualization
(lead); Writing – original draft (lead); Writing – review
and editing (lead). Elizabeth Chen: Data curation (lead);
Formal analysis (supporting); Software (lead). Diego Tuero:
Funding acquisition (supporting); Investigation (support-
ing); Resources (supporting); Supervision (supporting);
Writing – review and editing (supporting). María Sabio:
Data curation (supporting). Kevin Tkach: Data curation
(supporting). Marcelo Assis: Data curation (supporting).
Neander Heming: Data curation (supporting); Formal anal-
ysis (supporting); Writing – review and editing (supporting).
Miguel Marini: Data curation (supporting); Investigation
(supporting). John Bates: Conceptualization (lead); Funding
acquisition (lead); Investigation (supporting); Project
administration (equal); Resources (lead); Supervision (lead);
Writing – original draft (supporting); Writing – review and
editing (supporting).
Transparent peer review
e peer review history for this article is available at https://
publons.com/publon/10.1111/jav.03006.
Data availability statement
Data available from the figshare digital repository: https://
figshare.com/articles/dataset/Egg_shape_and_size_
data/17102876 (Gómez-Bahamón et al. 2022).
Supporting information
e Supporting information associated with this article is
available with the online version.
References
Austin, S. H., Douglas Robinson, W., Robinson, T. R., Ellis, V. A.
and Ricklefs, R. E. 2020. Development syndromes in New
World temperate and tropical songbirds. – PLoS One 15:
e0233627.
Barta, Z. and Szekely, T. 1997. e optimal shape of avian eggs.
– Funct. Ecol. 11: 656–662.
Bates, D., Mächler, M., Bolker, B. M. and Walker, S. C. 2015.
Fitting linear mixed-effects models using lme4. – J. Stat. Softw.
67: 1–48.
Biggins, J. D., ompson, J. E. and Birkhead, T. R. 2018. Accu-
rately quantifying the shape of birds’ eggs. – Ecol. Evol. 8:
9728–9738.
Birkhead, T. R., ompson, J. E., Jackson, D. and Biggins, J. D.
2017. e point of a Guillemot’s egg. – Ibis 159: 255–265.
Carey, C. 1980. Adaptation of the avian egg to high altitude. – Am.
Zool. 20: 449–459.
Christians, J. K. 2002. Avian egg size: variation within species and
inflexibility within individuals. – Biol. Rev. 77: 1–26.
Coulson, J. 1963. Egg size and shape in the kittiwake Rissa tridac-
tyla and their use in estimating age composition of populations.
– Proc. Zool. Soc. Lond. 140: 221–226.
del Hoyo, J., Elliott A. and Christie D. A. (eds) 2004. Handbook
of the birds of the world: Volume 9: Cotingas to pipits and
wagtails. – Lynx Editions.
Duursma, D. E., Gallagher, R. V., Price, J. J. and Griffith, S. C.
2018. Variation in avian egg shape and nest structure is
explained by climatic conditions. – Sci. Rep. 8: 4141.
Fick, S. E. and Hijmans, R. J. 2017. WorldClim 2: new 1-km
spatial resolution climate surfaces for global land areas. – Int.
J. Climatol. 37: 4302–4315.
Fox, J. and Weisberg, S. 2019. An R companion to applied regres-
sion, 3rd edn. – Sage.
Gill, F. B. 2007. Ornithology, 3rd edn. – W. H. Freeman.
Gómez-Bahamón, V., Chen, E. R., Tuero, D. T., Sabio, M. N.,
Tkach, K., Assis, M., Heming, N. M., Marini, M. Â. and Bates,
J. M. 2022. Data from: Egg shape variation across the distribu-
tion of the partially migratory fork-tailed flycatcher Tyrannus
savana. – Figshare Digital Repository, https://figshare.
com/articles/dataset/Egg_shape_and_size_
data/17102876.
Gómez-Bahamón, V., Márquez, R., Jahn, A. E., Miyaki, C. Y.,
Tuero, D. T., Laverde-R, O., Restrepo, S. and Cadena, C. D.
2020a. Speciation Associated with Shifts in Migratory Behavior
in an Avian Radiation. – Curr Biol. 30: 1312–1321.e6.
Gómez-Bahamón, V., Tuero, D. T., Castaño, M. I., Jahn, A. E.,
Bates, J. M. and Clark, C. J. 2020b. Sonations in migratory
and non-migratory fork-tailed flycatchers Tyrannus savana. –
Integr. Comp. Biol. 60: 1147–1159.
Hargitai, R., Nagy, G., Márton, H. and Török, J. 2013. Effects of
experimental calcium availability, egg parameters and laying
order on great tit Parus major eggshell pigmentation patterns.
– Ibis 155: 561–570.
Hauber, M. E. 2014. e book of eggs: a life-size guide to the eggs
of six hundred of the world’s bird species (Bates, J. M. and
Becker, B. eds). – Ivy Press.
Hays, I. R., Ljubicic, I. and Hauber, M. E. 2020. e shape of
avian eggs: assessment of a novel metric for quantifying eggshell
conicality. – Auk 137: ukaa021.
Hijmans, R. J. and van Etten, J. 2012. raster: geographic analysis
and modeling with raster data. – R package, http://CRAN.R-
project.org/package=raster.
Hipfner, J. M., Gaston, A. J. and de Forest, L. N. 1997. e role
of female age in determining egg size and laying date of thick-
billed murres. – J. Avian Biol. 28: 271–278.
Page 14 of 14
Hoyt, D. F. 1976. e effect of shape on the surface–volume rela-
tionships of birds’ eggs. – Condor 78: 343–349.
Inverson, J. B. and Ewert, M. A. 1991. Physical characteristics of
reptilian eggs and a comparison with avian eggs. – In: Deeming,
D. C. and Ferguson, M. W. J. (eds), Incubation: its effect on
embryonic development in birds and reptiles. Cambridge Univ.
Press, pp. 87–100.
Jahn, A. E., Tuero, D. T., Mamani, A. M., Bejarano, V., Masson,
D. A. and Aguilar, E. 2014. Drivers of clutch-size in fork-tailed
flycatchers Tyrannus savana at temperate and tropical latitudes
in South America. – Emu 114: 337–342.
Jetz, W., Sekercioglu, C. H. and Bohning-Gaese, K. 2008. e
worldwide variation in avian clutch size across species and
space. – PLoS Biol. 6: 2650–2657.
Jirinec, V., Burner, R. C., Amaral, B. R., Bierregaard Jr., R. O.,
Fernández-Arellano, G., Hernández-Palma, A., Johnson, E. I.,
Lovejoy, T. E., Powell, L. L., Rutt, C. L., Wolfe, J. D. and
Stouffer, P. C. 2021. Morphological consequences of climate
change for resident birds in intact Amazonian rainforest. – Sci.
Adv. 7: eabk1743.
Lack, D. 1947a. e significance of clutch-size, part I. – Ibis 89:
302–352.
Lack, D. 1947b. e significance of clutch-size, part III. – Ibis 90:
25–45.
Lockwood, R., Swaddle, J. P. and Rayner, J. M. V. 1998. Avian
wingtip shape reconsidered: wingtip shape indices and morpho-
logical adaptations to migration. – J. Avian Biol. 29: 273–292.
MacGillivray, W. 1852. A history of british birds. – Scott, Webster
& Geary.
Marini, M. Â., Hall, L., Bates, J., Steinheimer, F. D., McGowan, R.,
Silveira, L. F., Lijtmaer, D. A., Tubaro, P. L., Córdoba-Córdoba,
S., Gamauf, A., Greeney, H. F., Schweizer, M., Kamminga, P.,
Cibois, A., Vallotton, L., Russell, D., Robinson, S. K., Sweet, P.
R., Frahnert, S., Corado, R. and Heming, N. M. 2020. e five
million bird eggs in the world’s museum collections are an inval-
uable and underused resource. – Auk 137: ukaa036.
Marini, M. A., Lobo, Y., Esteves Lopes, L., Fernandes França, L.
and Vieira de Paiva, L. 2009. Biologia reprodutiva de Tyrannus
savana (Aves, Tyrannidae) em cerrado do Brasil Central. – Biota
Neotrop. 9: 55–63.
Martin, T. E. 1996. Life history evolution in tropical and south
temperate birds: what do we really know? – J. Avian Biol. 27:
263–272.
Monclús, R., Muriel, J., Pérez-Rodríguez, L., Møller, A. P. and Gil,
D. 2017. e role of the mating system and intraspecific brood
parasitism in the costs of reproduction in a passerine bird. –
Oecologia 185: 629–639.
Montgomerie, R., Hemmings, N., ompson, J. E. and Birkhead,
T. R. 2021. e shapes of birds’ eggs: evolutionary constraints
and adaptations. – Am. Nat. 198: E215–E230.
Moreau, R. E. 1944. Clutch-size: a comparative study, with special
reference to African birds. – Ibis 86: 286–347.
Muriel, J., Pérez-Rodríguez, L. and Gil, D. 2019. Age-related patterns
of yolk androgen deposition are consistent with adaptive brood
reduction in spotless starlings. – Behav. Ecol. Sociobiol. 73: 160.
Phillips, A. G., Töpfer, T., Böhning-Gaese, K. and Fritz, S. A. 2018.
Evidence for distinct evolutionary optima in the morphology
of migratory and resident birds. – J. Avian Biol. 49: e01807.
Salvador, S. A. 2013. Biología de la Tijereta Tyrannus savana en el
Departamento General San Martín, Córdoba, Argentina. –
Hitoria Nat. 3: 47–59.
Shatkovska, O. V., Ghazali, M., Mytiai, I. S. and Druz, N. 2018.
Size and shape correlation of birds’ pelvis and egg: impact of
developmental mode, habitat and phylogeny. – J. Morphol.
279: 1590–1602.
Sheard, C., Neate-Clegg, M. H. C., Alioravainen, N., Jones, S. E.
I., Vincent, C., MacGregor, H. E. A., Bregman, T. P., Clara-
munt, S. and Tobias, J. A. 2020. Ecological drivers of global
gradients in avian dispersal inferred from wing morphology. –
Nat. Commun. 11: 2463.
Stoddard, M. C., Sheard, C., Akkaynak, D., Yong, E. H., Mahade-
van, L. and Tobias, J. A. 2019. Evolution of avian egg shape:
underlying mechanisms and the importance of taxonomic scale.
– Ibis 161: 922–925.
Stoddard, M. C., Yong, E. H., Akkaynak, D., Sheard, C., Tobias,
J. A. and Mahadevan, L. 2017. Avian egg shape: form, function
and evolution. – Science 356: 1249–1254.
Sullivan, B. L., Wood, C. L., Iliff, M. J., Bonney, R. E., Fink, D.
and Kelling, S. 2009. eBird: a citizen-based bird observation
network in the biological sciences. – Biol. Conserv. 142:
2282–2292.
Turner, J. S. 2002. Maintenance of egg temperature. – In: Deem-
ing, D. C. (ed.), Avian incubation: behavior, environment and
evolution. Oxford Univ. Press, pp. 119–142.
Warham, J. 1990. e petrels: their ecology and breeding systems.
– Academic Press.
Weeks, B. C., Willard, D. E., Zimova, M., Ellis, A. A., Witynski,
M. L., Hennen, M. and Winger, B. M. 2020. Shared morpho-
logical consequences of global warming in North American
migratory birds. – Ecol. Lett. 23: 316–325.
Wiens, J. J. and Graham, C. H. 2005. Niche conservatism: inte-
grating evolution, ecology and conservation biology. – Annu.
Rev. Ecol. Evol. Syst. 36: 519–539.
Wright, N. A., Steadman, D. W., Witt, C. C. and Estes, J. A. 2016.
Predictable evolution toward flightlessness in volant island
birds. – Proc. Natl Acad. Sci. USA 113: 4765–4770.
You, Y., Feng, J., Wang, H., Wang, J., Dong, C., Su, X., Sun, H.
and Gao, W. 2009. Variation in egg size and nestling growth
rate in relation to clutch size and laying sequence in great tits
Parus major. – Prog. Nat. Sci. 19: 427–433.
