Developmental, ecological and biological drivers of macroevolutionary trajectories of postcranial morphological diversification in

Camilo López-Aguirre

A thesis in fulfillment of the requirements for the degree of Doctor of Philosophy

School of Biological, Earth and Environmental Sciences

Faculty of Science

August 2020

3

4

5

6

7

Table of Contents List of abbreviations...... 12 Acknowledgments ...... 13 List of figures ...... 16 List of tables ...... 20 Chapter 1: General introduction……………………………………………………………………………. 21 References ...... 24 Chapter 2: Postcranial heterochrony, , integration and disparity in the prenatal ossification in bats (Chiroptera)………………………………………………………………..44 Abstract ...... 44 Introduction ...... 45 Materials and methods ...... 52 Data collection ...... 52 Sequence heterochrony ...... 55 Metric growth ...... 56 Developmental modularity ...... 57 Developmental disparity and integration ...... 58 Results ...... 60 Sequence heterochrony ...... 60 Developmental growth ...... 62 Developmental modularity ...... 63 Disparity and integration ...... 64 Discussion ...... 67 Sequence heterochrony and developmental growth ...... 67 Developmental modularity ...... 72 Disparity and integration ...... 74 Conclusions ...... 77 References ...... 78 Chapter 3: Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera) ...... 89 Abstract ...... 89 Introduction ...... 90 Materials and Methods ...... 96 Data collection ...... 96 Ontogenetic allometric trajectories ...... 97 of ontogenetic allometric trajectories ...... 99 Results ...... 100 Ontogenetic allometry ...... 100 Allometric space analysis...... 105 Evolution of ontogenetic allometric trajectories ...... 106 Discussion ...... 109 Conclusions ...... 115

8

References ...... 116 Chapter 4: Prenatal developmental trajectories of fluctuating asymmetry in humeri ...... 132 Abstract ...... 133 Introduction ...... 134 Methods ...... 140 Sampling ...... 140 Data collection ...... 141 Estimation of asymmetry ...... 142 Data analysis ...... 144 Results ...... 145 Symmetric and asymmetric shape variation ...... 145 Longitudinal and cross-sectional asymmetry ...... 146 Discussion ...... 149 Conclusions ...... 153 References ...... 154 Chapter 5: Variation in cross-sectional shape and biomechanical properties of the bat humerus under Wolff’s law ...... 160 Abstract ...... 160 Introduction ...... 161 Materials and methods ...... 165 Sample description and digitisation...... 165 Extraction of landmark and biomechanical data ...... 167 Ecological and phylogenetic characterisation of species ...... 169 Statistical analyses ...... 169 Results ...... 171 Scaling of biomechanical and shape data ...... 171 Ecological differences in morpho- and mechanospaces ...... 172 Discussion ...... 176 Scaling of biomechanical and shape data ...... 177 Ecological differences in shape and biomechanical properties...... 178 Conclusions ...... 183 References ...... 183 Chapter 6: Phylogeny and foraging behaviour shape the modular morphological variation in bat humeri ...... 189 Abstract ...... 189 Introduction ...... 190 Methods ...... 196 Sample specimens...... 196 Morphometric data ...... 198 Phylogenetic, ecological and biological traits ...... 199 9

Statistical analysis ...... 200 Morphological disparity, integration and modularity ...... 203 Results ...... 204 Landmarking accuracy ...... 204 Humeral morphological variation ...... 204 Morphospace reconstruction ...... 208 Morphological modularity and integration ...... 212 Discussion ...... 213 Drivers of humeral morphological variation ...... 214 Morphological modularity and integration ...... 219 Conclusions ...... 221 References ...... 222 Chapter 7: General conclusions ...... 231 Main findings ...... 231 Future studies ...... 237 Summary ...... 239 References ...... 240 Appendices ...... 265 Appendix Figure 1 ...... 265 Appendix Figure 2 ...... 265 Appendix Figure 3 ...... 266 Appendix Figure 4 ...... 267 Appendix Figure 5 ...... 267 Appendix Figure 6 ...... 268 Appendix Figure 7 ...... 268 Appendix Figure 8 ...... 269 Appendix Figure 9 ...... 269 Appendix Figure 10 ...... 270 Appendix Table 1 ...... 272 Appendix Table 2 ...... 274 Appendix Table 3 ...... 275 Appendix Table 4 ...... 276 Appendix Table 5 ...... 278 Appendix Table 6 ...... 279 Appendix Table 7 ...... 282 Appendix Table 8 ...... 285 Appendix Table 9 ...... 287

10

Appendix Table 10 ...... 288 Appendix Table 11 ...... 289 Appendix Table 12 ...... 291 Appendix Table 13 ...... 293 Appendix peer-reviewed publication 1 ...... 294 Appendix peer-reviewed publication 2 ...... 294

11

List of abbreviations

Antisymmetry (AS); (AER); Carnivore (C); Carpal Length (CaL);

Caudal Vertebral Width (CavL); centroid size (CS); Cervical Vertebral Width (CvL); circularity (CircMaxR); Clavicular Length (CL); Crown to Rump Length (CRL); Directional

Asymmetry (DA); Developmental Instability (DI); evolutionary developmental biology

(EvoDevo); Femoral Length (FL); Fibular Length (FiL); Fluctuating asymmetry (FA);

Frugivore (F); Generalised Linear Model (GLM); Gleaning (G); Hawking (H); hedgehog signalling pathway (SHH); homeobox proteins (HOX); Homogeneity of Slopes test

(HOS); Humeral Length (HL); Ilium Length (IlL); Ischium Length (IsL); Lateral Plate

Mesoderm (LPM); Leading Edge Vortex (LEV); Least Square (LS); Linear Discriminant

Function Analysis (LDA); Lumbar Vertebral Width (LvL); Manual Phalanges Length

(MPL); maximum second moment of area (Imax); Measurement Error (ME).; Metacarpal

Length (McL); Metatarsal Length (MtL); minimum second moment of area (Imin); morphogenetic protein 2 (Bmp2); Particle Image Velocimetry (PIV); Pedal Phalanges

Length (PPL); phylogenetic ANOVAs (PGLS); polar moment of inertia of an area (J);

Principal Component Analysis (PCA); Pubis bone Length (PuL); Radial Length (RL); Rib

Length (RL); Sacral Vertebral Width (SvL); Scapular Length (SL); second moment of area about the x axis (Ix); second moment of area about the y axis (Iy); Sternum Length (StL);

Tarsal Length (TaL); Terrestrial Locomoting (TL); Tibial Length (TiL); Thoracic Vertebral

Width (TvL); Trawling (T); Ulnar Length (UL); Upstand Roosting (UR); zone of polarising activity (ZPA).

12

Acknowledgments

Parece desproporcionado en un documento tan extenso dedicado a exponer mi trabajo, dedicar tan sólo unos párrafos reconociendo el aporte, nada corto de vital, de todos lo que me han acompañado en este reto desde que me fui de Colombia.

Entiendan lo escueto de este reconocimiento como evidencia de lo difícil que me es agradecerles en proporción a lo que me han dado. Esta es sin duda la sección que más me ha costado, pero más me ha gratificado escribir. No sólo este documento, sino todo trabajo que realice de aquí en adelante, va dedicado a mi padre, mi viejo. Aunque no lograste acompañarme en vida hasta este último tramo del doctorado, el desinterés y la valentía que mostraste luchando un cáncer incurable lejos de tus hijos, no es nada menos que la muestra más desgarradora de la humildad de tu amor desinteresado.

Seré débil, y a veces fallaré, pero prometo que mi vida y cada pedazo de ella, será un tributo a ti. Este es sólo el comienzo.

A mi familia quiero agradecerles la paciencia, incondicionalidad y desinterés con la que siempre me han apoyado. Su complicidad en todas mis decisiones, a pesar de las diferencias y la distancia que a veces puedo establecer, son tan responsables de este trabajo como yo. A mi madre, agradecerte tu capacidad de asombro y de escucha.

Espero con los años aprender a escuchar con tanta humildad y respeto como tu lo has hecho todos los días de mi vida. A mi hermana, agradecerte tu disposición inquebrantable a guiarme en las aguas, a veces innavegables, de ser migrante, de cuestionarme y desafiarme a encontrar el mínimo básico que me define. A Laura, por enseñarme a encontrar fortaleza en exponer mi vulnerabilidad. Tu presencia en mi vida es sin duda mi más grande debilidad, de la que más he aprendido y con la que

13 más me he conocido. Es en esa debilidad que he encontrado mis alegrías más duraderas. Agradecimientos también van a los Lopera y Jaime Roa, su apoyo y fortaleza durante la enfermedad de mi papá hicieron posible que terminara este trabajo. A mis mejores amigos, José, Juan y Valeria, mi compromiso a seguir creciendo como persona nutriéndome de su compañía. Estar lejos de ustedes ha sido duro, pero prometo seguir buscando formas de igualar su incondicionalidad. A los Castañeda

Gómez, gracias por invitarme a ser una pequeña parte de su familia. Ha sido un privilegio tratar de entenderlos en sus individualidades y colectividades. Special thanks go to Chris and the Stavrou family for showing me the best face of Australia in all its complexity. Ευχαριστώ πολύ.

I would like to thank my main supervisors Drs. Suzanne Hand, Laura A. B. Wilson and external supervisor Dr. Daisuke Koyabu for their trust and support, especially during the last year. Thanks for letting me explore my own ideas and work independently under your guidance. I feel privileged for having the opportunity to work and learn from professionals that I respect and admire profoundly. Finally, I would like to thank the members of the palaeolab at UNSW, Naomi, Antonia, Troy, Chris, Roy, Ana,

Michael, Mike, Corey and Tim, for their company and advice during my years as a member of the lab, and in particular to Troy Myers for the free coffee, albeit insufficient.

14

List of publications and presentations Peer-reviewed publications 1. López-Aguirre, C. Hand, S. J. Koyabu, D. Nguyen, T. S. Wilson, L. A. B. (2019) Postcranial modularity, integration and disparity in the prenatal development of bats (Chiroptera). BMC . 19:75. 2. López-Aguirre, C. Hand, S. J. Koyabu, D. Nguyen, T. S. Wilson, L. A. B. (2019) Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera). Journal of Experimental Zoology B. 332:36-49. In review López-Aguirre, C. Hand, S. J. Koyabu, D. Nguyen, T. S. Wilson, L. A. B. Phylogeny and foraging behaviour shape modular morphological variation in bat humeri. Submitted to Journal of Anatomy López-Aguirre, C. Hand, S. J. Koyabu, D. Nguyen, T. S. Wilson, L. A. B. Prenatal developmental trajectories of fluctuating asymmetry in bat humeri. Submitted to Evolution & Development Conference presentations 1. Oral presentation: Biomechanics and not ecology describe variation in cross- sectional shape of the humerus in bats. 18th International Bat Research Conference. 28 July-1 August 2019. 2. Oral presentation: The morphogenetic basis of mammalian : Allometric trajectories and ossification heterochronies in prenatal skeletogenesis of bats. 12th International congress of vertebrate morphology. Prague, Czechia. 21-25 July 2019. 3. Oral presentation: born to fly: Postcranial development in bats. 64th Australian Society Meeting. Brisbane, Australia. 1-5 July 2018. 4. Oral presentation: Postcranial modularity, integration and disparity in the prenatal development of bats (Chiroptera). 18th Australasian Bat Society Conference. Richmond, Australia. 3-6 April 2018.

15

List of figures

Figure 1.1. Evolutionary history of the polyphyletic origin of vertebrate flight, and of extinct (red) and extant (black) species of all three groups of flying vertebrates. Silhouettes represent wingspan ranges per group...... 23 Figure 1.2. Homological anatomical for flight in (top), birds (middle) and bats (bottom). Colours group the bones that form each of sections of the forelimb: pectoral girdle (brown), stylopod (red), zeugopod (green) and autopod (blue)...... 24 Figure 2.1. 3D virtual models of ontogenetic series of H. blanfordi, representing the samples from which raw measurements were taken from postcranial elements...... 53 Figure 2.2. Relative timing of onset of ossification (ranks) of 24 postcranial bones in Aves (red), Chiroptera (green), and non-volant mammals (blue). Standardised ranking of bone ossification onset ranges from 0 (first to start ossification) to 1 (last to start ossification)...... 61 Figure 2.3. Principal Component Analysis of the ossification sequences of 24 postcranial bones (relative ranks from 0 to 1) in 39 vertebrate species analysed in this study. Species are plotted across PC1 (49.47%) and PC2 (17.86%), and are grouped as Aves (red), Chiroptera (green), and non-volant mammals (blue)...... 62 Figure 2.4. Neighbour-joining clustering analysis of ossification sequence of 24 postcranial bones (relative ranks from 0 to 1) in bats. Colours represent the two best- supported groups. Numbers represent level of support at each node after 10,000 randomisations...... 63 Figure 2.5. Developmental integration (eigenvalue dispersion) and disparity (bone size variance) across 24 postcranial bones in bats. Dotted and dashed lines mark average integration and disparity respectively...... 65 Figure 2.6. Average developmental integration (eigenvalue dispersion) and disparity (bone size variance) of elements of the appendicular and axial developmental modules in bats. Dotted and dashed lines mark intermodule mean integration and disparity respectively...... 66 Figure 2.7. Developmental integration (eigenvalue dispersion) and disparity (bone size variance) across prenatal stages for bats examined in this study. Values at each stage are an average across all bones...... 68 Figure 2.8. 95% confidence intervals of values of disparity (A,C) and integration (B,D) across bat developmental stages (A,B) and bones (C,D) based on 10,000 bootstrap replicates...... 70 Figure 3.1. Three-dimensional virtual reconstruction of skeleton of Aselliscus dongbacana (stage 21), showing 24 postcranial bones measured. From left to right, skeleton viewed on ventral, lateral and dorsal view...... 97 Figure 3.2. Boxplots of the distribution of inter-trajectory angle values of pairwise comparisons between species. Whiskers represent minimum and maximum values for each set of comparisons. From left to right, pairwise comparisons are summarised as comparisons between species of different suborders (intersubordinal). Numbers and the postion of “x” within each boxplot represent the number of pairwise comparisons and the average, respectively...... 102

16

Figure 3.3. Ontogenetic allometric trajectories of Chiroptera at species (A) and subordinal (B) levels. Trajectories are derived from an homogeneity of slope test, plotting log transformed geometric means in the x axis (i.e. log(Size)) and the PC1 of the predicted values of multivariate regression of shape ratios on size in the y axis (Shape(Predicted))...... 103 Figure 3.4. Allometric space (A) and phyloallomspace (B) of Chiroptera. Each point represents the allometric trajectory of a species, convex hulls represent the allometric space occupied by each species (B), and the different colours in the phylogenetic relationships projected in the phyloallomspace represent each suborder (B)...... 106 Figure 3.5. Ancestral state reconstruction of the magnitude of shape change with size, based on the ontogenetic allometric slope vector length. Colours in each branch represent the magnitude of shape change with size for each species and each reconstructed ancestor, red-to-pink branches representing higher levels of change, and blue-to-green branches lower levels of change...... 108 Figure 3.6. Levels of allometry-corrected disparity (Procrustes variance) for seven bat species analysed. Numbers on top of each bar represent significant differences across species based on pairwise permutations test: 1 signifies significant differences with A. dongbacana, 2 with A. stoliczkanus, 3 with C. sphinx, 4 with H. blanfordi, 5 with H. larvatus, 6 with K. hardwickii and 7 with R. thomasi...... 109 Figure 4.1. PCAs of symmetric (A) and asymmetric (B) components of humeral cross- sectional shape variation across prenatal development. Colours represent developmental stages. Stages 1 to 10 represent early to late prenatal development...... 145 Figure 4.2. Probability density functions of distribution of values of longitudinal (signed FA6, A) and cross-sectional FA (PC1-2 scores, B-C). Barplots show individual asymmetry values from which density functions were estimated...... 147 Figure 4.3. Boxplots of longitudinal (unsigned FA6, left) and cross-sectional (CFA, right) FA values across development. Stages 1 to 10 represent early to late prenatal development...... 148 Figure 4.4. Scatterplot of association between longitudinal (unsigned FA6) and cross- sectional (CFA) fluctuating asymmetry across prenatal development. Dot colours represent developmental stages, stages 1 to 10 representing early to late prenatal development...... 149 Figure 5.1. Phylogenetic reconstruction of the evolutionary relationships between the sampled species (top), based on Shi and Rabosky (2015). Colours of branches represent different foraging and roosting categories. Species classified based on body size, represented as discretised centroid size (CS) categories: small-sized (triangle), medium-sized (square) and large-sized species (diamond). Subordinal clades are marked in the phylogeny. Lineage through time plot (bottom) showing temporal accumulation of lineages in our sample...... 166 Figure 5.2. Three-dimensional virtual reconstruction of the humerus of Austronomus australis, showing a schematic representation of the cross-sectioning and landmarking protocol used in this study. Landmarks were collected at the intersection of equiangular radii and the periosteal contour of the cross section...... 168 Figure 5.3. Allometric trajectories of scaling of biomechanical traits with size (humeral length, HL) in the bat humeri, depending on foraging guild category (FG): carnivore (C),

17 frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown colour represents species with upstand roosting (UR)...... 172 Figure 5.4. Boxplots of values of biomechanical traits of the bat humeri, depending on foraging guild category (FG): carnivore (C), frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown data points represent species with upstand roosting (UR)...... 173 Figure 5.5. Morphospace of humerus allometry-corrected cross-sectional shape (A) and allometry-phylogeny-corrected (B) shape data. Data points are colour-coded by foraging guild category: carnivore (C), frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown data points represent species with upstand roosting (UR). Deformation grids show cross-sectional shape of species at opposite extremes of variation along PC1 and PC2...... 175 Figure 5.6. Phenotypic disparity based on shape variance for humeral cross-sectional shape. Shape disparity was deconstructed based on foraging guild categories: carnivore (C), frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown data points represent species with upstand roosting (UR). .... 176 Figure 6.1. Phylogenetic relationships between sampled taxa based on Shi and Rabosky (2015)’s phylogeny. Branch colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= ). 3D models of humeri illustrate humeral diversity in sampled taxa. Represented taxa clockwise from bottem left to bottom right: Desmodus rotundus, Furipterus horrens, grandis, Macroglossus minimus, Myotis daubentoni, Molossus molossus...... 198 Figure 6.2. Landmarking protocol used to quantify humeral morphology. From left to right humeri are presented in anterior (far left), medial (centre left), posterior (centre right) and lateral (far right) views. Proximal (top right) and distal (bottom right) epiphyses are also presented. Homologous landmarks are represented by numbers 0- 30 and curves used to place semi-landmarks are represented by C0-C5...... 199 Figure 6.3. Humeral shape disparity of whole-bone (left), diaphyseal (centre) and epiphyseal (right) morphology. Shape disparity was decomposed based on foraging guild categories: C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= terrestrial locomotion...... 208 Figure 6.4. Morphospace (PCA, A and C) and phylogenetically-corrected morphospace (pPCA, B and D) based on whole-bone shape data. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H= hawkers, T= trawlers and TL= terrestrial locomotion), and polygon colours suborder (purple= , yellow= Yinpterochiroptera). Landmark heatmaps of shape change represent magnitude of shape variation across each PC by the comparing the minimum and maximum of each component. Humeri 3D models represent position of landmark heatmaps; red colours representing greater variation and yellow colours lower variation...... 209 Figure 6.5. Diaphyseal (left) and epiphyseal (right) morphospaces of humeral morphology based on PCAs of shape data. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H= hawkers, T= trawlers and TL= terrestrial locomotion), and polygon colours suborder (purple= Yangochiroptera, yellow= Yinpterochiroptera)...... 210

18

Figure 6.6. Diaphyseal (A and C) and epiphyseal (B and D) phylogenetically-corrected morphospaces of humeral morphology based on pPCAs of shape data. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H= hawkers, T= trawlers and TL= terrestrial locomotion), and polygon colours suborder (purple= Yangochiroptera, yellow= Yinpterochiroptera)...... 212 Figure 6.7. PLS biplot of first two axes of diaphyseal and epiphyseal shape covariation. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H= hawkers, T= trawlers and TL= terrestrial locomotion)...... 213

19

List of tables

Table 2.1. List of mammal and bird species analysed in this study...... 54 Table 2.2. Description of 25 linear measurements of the postcranial skeleton of bat fetuses used in this study. Numbers indicate the modularity hypotheses where each bone was included (see table 3 for number coding)...... 57 Table 2.3. Kendall’s tau results testing nine different modularity hypotheses of metric growth. P values shown in parentheses. Values in bold were statistically significant before Bonferroni correction, and asterisks indicate significance after correction (P<0.05/29=0.0017)...... 64 Table 3.1. Individual multivariate linear regressions for the test of allometric growth, based on an isometric growth null hypothesis...... 102 Table 3.2. Pairwise comparisons of statistical differences in vector length (magnitude of shape change with size; top) and angle (direction of allometric vector; bottom) of the ontogenetic allometric trajectory between species...... 104 Table 3.3. Pairwise comparisons of statistical differences in the slope intercept of the allometric trajectories (top) and p values of the difference between each pair of species based on 10,000 iterations (bottom) of the ontogenetic allometric trajectory between species...... 107 Table 4.1. Procrustes ANOVA (Shape, CS) and ANOVA (HL) statistical tests of significance of fluctuating asymmetry (FA), directional asymmetry (DA) and measurement error (ME) in cross-sectional shape, centroid size (CS) and humeral length (HL). Side factor tests for DA, Individual:Side interaction tests for FA and replicate tests for ME...... 146 Table 4.2. ANOVA test results for statistically significant differences in longitudinal (FA6) and cross-sectional (CFA) humeral FA across developmental stages...... 148 Table 5.1. Individual Procrustes linear regressions for the test of scaling in shape and biomechanical data with humeral length (HL), based on an isometric null hypothesis. Biomechanical properties and HL were log-transformed...... 171 Table 5.2. Individual Procrustes ANOVA for the test of differences in shape and biomechanical data across foraging and roosting guilds...... 173 Table 6.1. Procrustes ANOVA (PLM) for different hypotheses of shape variation. Significance test was based on 1000 iterations...... 205 Table 6.2. Kmult- statistic test of phylogenetic signal on shape data. Significance test was based on 1000 iterations...... 205 Table 6.3. Phylogenetic procrustes ANOVA for different hypotheses of shape variation. Significance test was based on 1000 iterations...... 206 Table 6.4. Procrustes ANOVA for hypotheses of covariation between humeral shape and aerodynamic properties (WL and AR). Significance test was based on 1000 iterations...... 207

20

Chapter 1 General introduction Evolution of vertebrate flight

Pterosaurs, birds and bats are the only vertebrate capable of self-powered flight (Rayner 1988). However, the phylogenetic relationships between these groups and their position in the evolutionary history of vertebrates have shown that this feature evolved in each group independently, as a result of

(Rayner 1988). Self-powered flight was a key innovation that provided a major ecological opportunity for these groups, allowing them to diversify to a vast range of previously empty niches (Stroud and Losos 2016). As a result, both birds and bats are some of the most speciose groups of living vertebrates. It is estimated that pterosaurs were also abundant and diverse before their , with over 160 fossil species now described (Prentice et al. 2011; Butler et al. 2013).

Historically, the convergent evolution of vertebrate flight was asynchronous, with pterosaurs evolving flight first (≈240 million years ago (Mya)), followed by birds (≈140

Mya), and bats (≈60 Mya) ( and Shedlock 2009). Throughout their evolution, flying vertebrates have been extremely successful at colonising a variety of previously unexploited niches, reaching remarkably high levels of biodiversity today (McGowan and Dyke 2007) (Fig. 1.1).

The relatively complete fossils of pterosaurs and birds have facilitated the study of the evolution of flight in these groups. Morphofunctional analysis of early paravians

21 suggests that self-powered flight did not appear early in the evolutionary history of

Paraves, even though many of these species already shared several traits that are common in modern birds (Witton and Habib 2010; Dececchi et al. 2016). flight evolution has been an area of increasing interest. Traditionally, several studies suggested that due to their massive body mass, large pterosaurs were flightless

(Henderson 2010). This assumption was based on extrapolations of avian biomechanical models used to interpret the pterosaur fossil record (Sato et al. 2009).

Recently, this approach has been debated, proposing that this model underestimates the aerial capabilities of large pterosaurs, and that in fact even the largest species flew

(Witton and Habib 2010).

Bat flight evolution, on the other hand, has been comparably more difficult to study due to the incompleteness of the fossil record (Eiting and Gunnell 2009; Giannini

2012). The main evolutionary question that researchers have been seeking to answer is the order in which flight and echolocation evolved in bats, with no clear consensus being reached (Arita and Fenton 1997; Adams 2008; Simmons, et al. 2008; Veselka, et al. 2010; Cooper, et al. 2012; Giannini 2012; Adams and Shaw 2013). Onychonycteris finneyi, the oldest bat fossil, displays all the morphological characteristics needed to generate self-powered flight, indicating that flight evolved early in the history of this group (Simmons et al. 2008). Moreover, no transitional form between true flying bats and their putatively non-volant ancestor has been found, leaving a significant gap in the evolutionary history of bat flight. Nonetheless, a transition from an ancestral shrew-like arboreal insectivore to an Onychonycteris-like bat ancestor has been

22 proposed as the most plausible scenario (Simmons et al. 2008; Amador et al. 2019a;

Amador et al. 2019b).

Figure 1.1. Evolutionary history of the polyphyletic origin of vertebrate flight, and biodiversity of extinct (red) and extant (black) species of all three groups of flying vertebrates. Silhouettes represent wingspan ranges per group. Although the current understanding of the evolution of vertebrate flight remains debated, it has been suggested that all three groups share some morphological and physiological adaptations in body form vital for the kinematics of flight (Dumont 2010).

Since the of all flying vertebrates evolved to become , forelimb morphological modifications are of interest (Fig. 2). Reduction of cortical bone thickness (Cubo and Casinos 1998), increased bone density (Dumont 2010), structural changes in the pectoral girdle (Currey and Alexander 1985), and elongation of forelimb

23 bones are some of the adaptations that these groups share (Lee and Simons 2015). It has also been found that all three groups have relatively small genomes when compared to their respective close relatives (Organ and Shedlock 2009), suggesting that constricted genome size has been correlated with the evolution of flight in vertebrates (Hafner et al. 1984; Organ and Shedlock 2009; Kapusta et al. 2017).

Figure 1.2. Homological forelimb anatomical adaptations for flight in pterosaurs (top), birds (middle) and bats (bottom). Colours group the bones that form each of sections of the forelimb: pectoral girdle (brown), stylopod (red), zeugopod (green) and autopod (blue). Despite these similarities, bats, birds and pterosaurs also show clear morphological differences that make the wings of each group unique. Based on linear measurements of several wing bones, McGowan and Dyke (2007) suggested that the morphospace 24 occupied by the wings of each group is completely different from the rest, and that competitive displacement did not limit the of flight and wing morphology in birds and bats. Moreover, this study also showed that each evolutionary event produced very different wing morphologies, suggesting that vertebrate flight evolved each time from completely different strategies (McGowan and Dyke 2007).

Not only has vertebrate flight evolved multiple times, but also there has been a diversification of flight patterns within each group. Flight styles have co-evolved with different ecologies enabling species to occupy novel ecomorphological spaces (Witton

2008; Wang et al. 2011; Arita et al. 2014). The of flight and diet is one of the most important factors in the diversification of these groups. In pterosaurs, different types of wing morphology have been associated with specific trophic guilds

(Witton 2008). Flight-diet coevolution has also been reported in birds and bats

(Marchetti et al. 1995; Denzinger and Schnitzler 2013) and flight-echolocation coevolution in bats (Aldridge 1987).

Vertebrate flight kinematics

Flight kinematics also show that flying strategies are strikingly different across all three groups (Witton and Habib 2010; Muijres et al. 2012a). Particle Image Velocimetry (PIV) is one approach that has been particularly effective at recreating flight kinematics

(Johansson et al. 2010; Muijres et al. 2011; Muijres et al. 2012a; Von Busse et al. 2012;

Muijres et al. 2014; Hakansson et al. 2015; Johansson et al. 2016). PIV is a flow visualisation method that describes the vorticity and wake structure formed in a fluid as a result of an object exerting a specific force to that fluid. A wide range of studies

25 that applied PIV to the study of flight kinematics in bats and birds have shown critical differences between these groups (Hedenstrom et al. 2009; Johansson et al. 2010;

Wolf et al. 2010; Muijres et al. 2012a; Johansson et al. 2016). As a result of the upstroke and downstroke movement during flapping flight, both birds and bats create two characteristic vortices that provide both positive lift and increased acceleration

(Muijres et al. 2012a).

Most of the vorticity is generated during the downstroke, when both positive lift and acceleration are at their highest during the flapping cycles (Lindhe Norberg and Winter

2006; Iriarte-Diaz et al. 2011; MacAyeal et al. 2011; Riskin et al. 2012; Cheney et al.

2014). Bats, however, also generate an additional vortex during the upstroke (i.e. the

‘’reversed vortex loop”) that provides additional thrust at the expense of lift (Muijres et al. 2012a). As a result, upstroke in bats is kinematically more active than in birds, providing both thrust and negative lift (Muijres et al. 2012a).

Bats also create a Leading Edge Vortex (LEV) during the downstroke, as a way to enhance their kinematic performance (Muijres et al. 2008). By developing LEVs, bats generate positive lift during the downstroke (Muijres et al. 2008). LEVs are span-wise vortices generated on the upper surface of the leading edge of a wing during the downstroke of the wingbeat (Ellington et al. 1996). However, LEVs have only been detected for slow-flying bats (Muijres et al. 2008; Muijres et al. 2014). LEVs are a common mechanism to produce lift in flying organisms, also being reported in insects

(Ellington et al. 1996; Johansson et al. 2013), birds (Warrick et al. 2009; Muijres et al.

2012b) and even seeds (Lentink et al. 2009).

26

On the other hand, pterosaur flight kinematics have been strikingly more difficult to study. Until recently, most studies have used kinematic models of to describe pterosaur flight (Sato et al. 2009). However, this approach has been challenged because the anatomy of each group is considerably different, possibly leading to considerable misinterpretations of pterosaur ecology (Witton and Habib

2010).

The ontogenetic basis of vertebrate flight

The genetics of wing development in vertebrates has also been a field of increasing interest (Swartz and Konow 2015; Petit et al. 2017). Understanding the genetic signalling pathways and regulatory mechanisms that control morphological development has widened our understanding of the evolutionary processes that led to the evolution of vertebrate wings. Despite the morphological differences between bat and bird wings, both groups share similar genetic mechanisms of forelimb development (Shubin et al. 1997; Kapusta et al. 2017).

Vertebrates’ happens along three different axes: dorsal-ventral, anterior-posterior and proximal-distal. In the early stages, limbs form from rudimentary buds of undifferentiated mesenchymal cells that are coated with an ectoderm layer (Petit et al. 2017). Growth in each of these axes is controlled by a different signalling pathway. Dorsal-ventral growth (from the back of the hand to the palm) is governed by the WNT gene family, whereas anterior-posterior growth (from I to V) depends on the zone of polarising activity (ZPA) that is regulated by the (SHH) signalling pathway (Petit et al. 2017). Finally, proximal-distal

27 growth (from shoulder to fingers) is controlled by the apical ectodermal ridge (AER) that influences the proliferation of the mesenchyme (Petit et al. 2017).

Another key event during the development of the wing is the forelimb-hindlimb differentiation. Vertebrate limb ontogeny undergoes differential genetic signalling that triggers different morphogenetic processes (Pieretti et al. 2015). This is particularly important for birds and bats as they show the most divergent forelimb-hindlimb differentiation among vertebrates (Pieretti et al. 2015). This process is determined by the paired-like homeodomain transcription factor 1 (Pitx1), responsible for hindlimb- type morphological development (Petit et al. 2017). Tbx4 and Tbx5 transcription factors also play a key role in the morphological specification of the limb. Tbx4 triggers a hindlimb-like morphological specialisation, whereas Tbx5 triggers a forelimb-like morphological specialisation (Petit et al. 2017). Tbx4 and Tbx5 expression is induced by different homeobox (HOX) proteins present in the caudal and rostral regions of the lateral plate mesoderm (Petit et al. 2017). Malfunction of these signalling pathways can lead to wing-to-leg or leg-to-wing transformations in birds and bats (Spielmann et al. 2012).

Forelimb elongation is another key aspect of wing development. In mammals, Prrx1 gene encodes the paired related homeobox 1 protein that enhances cellular proliferation in the forelimb, promoting its elongation (Cretekos et al. 2008).

Differences in gene evolution associated with flight have also been found between birds and mammals (Machado et al. 2016). When comparing rates of positive selection in a pool of flight-associated genes, birds have shown positive selection in almost twice the number of genes than mammals (Machado et al. 2016). This indicates that bird

28 wing development involves a greater number of genes and signalling pathways than bat wing development (Machado et al. 2016).

Flight biomechanics

Wing bone biomechanics is another crucial component in the evolution of flight in vertebrates. The evolutionary transition between terrestrial and aerial locomotion created novel biomechanical models of stress and strain under which wing bones had to perform optimally in order to allow flying vertebrates to truly colonise aerial niches

(Swartz et al. 1992; Kirkpatrick 1994).

Compared to terrestrial locomotion, flight requires a completely different arrangement of forces to be applied to the forelimb (Kirkpatrick 1994). During flight, shearing stress has been found to be higher than in terrestrial locomotion, and bending stress to be lower (Kirkpatrick 1994). Bone laminarity, the circumferential vascularisation of the bone, is a conspicuous trait in bird humeri, but absent in bats, that has been associated with stress and strain resistance (Simons 2010; Simons et al. 2011; Simons and

O'Connor P 2012; Marelli and Simons 2014). During the wingbeat cycle, bat humeri experience critical levels of both stress and strain during the downstroke, the most kinematically active stage of the wingbeat (Swartz et al. 1992; Watts et al. 2001).

During the downstroke, torsion and shearing stress are crucial traits that determine flying performance (Swartz et al. 1992).

Breaking stress resistance is another difference in wing biomechanics between birds and bats. Comparing values of maximum breaking stress and experienced stress in the humerus, bird humeri have a greater safety factor than bat humeri, meaning that birds could potentially resist larger stresses and support larger body sizes (Kirkpatrick 1994). 29

Bat ecomorphological evolution

Compared to birds and pterosaurs, the behavioural and ecological diversity of bats exert extra functional pressures on forelimb morphology (Swartz and Konow 2015).

With over 1400 species, bats are the second most speciose group of mammals after rodents. With an evolutionary history spanning more than 60 Myr, bats evolved several extreme specialisations (e.g. flight, echolocation, upside down roosting) that allowed them to fill niches that were previously inaccessible for any other mammal group (Teeling et al. 2000; Jones and Teeling 2006; Bergou et al. 2015). The morphological disparity of bats is an extraordinary example of adaptive radiation among mammals, reflected in considerable body size variation and cranial phenotypic diversity of forms specialised for a wide variety of niches (i.e. nectarivory, insectivory, carnivory, frugivory, sanguivory) (Santana and Cheung 2016). Although the adaptive radiation of cranial morphology in bats has been explored in some detail (Santana et al. 2011; Dumont et al. 2012; Santana et al. 2012), the concordance between cranial and postcranial morphological integration remains unexamined, and evolutionary trends in the extent of diversity in postcranial shape and size have not been tested in an explicit ecological and functional framework (Arita and Fenton 1997).

Coupled with the evolutionary trend towards a wide ecomorphological radiation of bat cranial morphology, ecological demands could have promoted phenotypic diversification and morphological disparity across the evolution of the forelimb

(Hedenstrom and Johansson 2015). A clear example of this may be the variety of wing shapes among bats (Norberg and Rayner 1987). Studies of the camber, wing loading and aspect ratio of bat wings have been able to recognise ecomorphological aerial

30 guilds. These guilds generally classify bats according to two gradients: slow to fast flight, and low to high manoeuvrability (Canals et al. 2011). The interaction of these two gradients reflects the correlation between wing morphology and the ecology of species (Marinello and Bernard 2014). Different intersecting points between the two gradients reflect specific trophic guilds. Slow, highly manoeuvrable flight usually indicates understory gleaners, whereas fast, poorly manoeuvrable flight usually indicates open-air hawkers (Gardiner et al. 2011).

Bats perform a wide variety of tasks using their forelimbs aside from flying, such that that forelimb ecomorphology corresponds not only to functional needs related to different flight behaviours (e.g. hovering, gleaning), but also climbing, grooming, pup carrying, feeding, roosting, walking and even wing-clicking echolocation (Dietz 1973;

Swartz et al. 1992; Riskin et al. 2005; Schliemann and Goodman 2011; Boonman et al.

2014). Previous studies suggest that differential functional demands, resulting from specific ecological and behavioural needs, are related to morphological differences found in the wings of bats (de Camargo and de Oliveira 2012; Schmieder et al. 2015).

Studying the evolution of some of these adaptations has contributed to the understanding of bat evolution, resulting in evolutionary hypotheses such as the convergent evolution of terrestrial locomotion and upstand roosting, and the polyphyletic origin of nectarivory (Fenton 2010). Given that, novel forelimb functional strategies would have had to originate to successfully exploit these specialised niches

(e.g. walking for efficient terrestrial locomotion and hovering flight for nectarivory), it could be inferred that morphological disparity in forelimb morphology

31 could coevolve with other non-locomotory ecological adaptations of other kinds

(Adams 2008). However, this hypothesis remains untested.

To test this hypothesis, one could focus on analysing the instances of phenotypic radiation in the forelimb morphology of bats associated with different ecomorphological specialisations, as a proxy to understand how morphological disparity is generated over time and across taxa (Oyston et al. 2015). A promising approach to test this hypothesis is evolutionary developmental biology (EvoDevo).

Examining the ontogenetic mechanisms and processes behind the generation of phenotypic novelties has been successful in reconstructing aspects of the evolutionary history of species (Hall 2000; Hallgrímsson et al. 2012). In bats, EvoDevo studies have shown how morphogenesis responds to newborn functional needs (Koyabu and Son

2014), the ontogenetic basis of cranial ecomorphological diversity in modern taxa

(Camacho et al. 2019), and have helped reconstruct evolutionary trajectories of adaptive innovations (Wang et al. 2017; Nojiri et al. 2018). Another developmental approach to understanding phenotypic evolution is to study the homeostatic processes that organisms experience to favour the appearance of functionally optimal phenotypes (Klingenberg and Nijhout 1999; Leamy and Klingenberg 2005). A way to explore this is through fluctuating asymmetry (i.e. random deviations from the ideally symmetric body plan of an organism), which can be interpreted as an index for the levels of instability animals experience during development and how they buffer such stress (Klingenberg and McIntyre 1998). Studying fluctuating asymmetry from an evolutionary perspective can provide novel insights into the interplay between the

32 mechanisms through which phenotypes are either optimised (to improve fitness), or changed (to facilitate diversification) (Dongen 2006; Neubauer et al. 2020).

This project will assess the developmental, ecological and evolutionary aspects of bat wing evolution in the form of five general objectives: 1) compare the prenatal ossification of the postcranium in bats, birds and non-volant mammals; 2) quantify ontogenetic allometric disparity in Chiroptera and assess the presence of a phylogenetic signal; 3) assess the presence and magnitude of fluctuating asymmetry across humeral development in bats; 4) reconstruct the cross-sectional morpho- biomechanical evolution of the humerus in bats; and 5) investigate ecological and phylogenetic drivers of phenotypic disparity in the bat humerus.

Traditional methods to study long bone morphology have been limited in their capabilities and usually restricted to the description of length and thickness ratios; in turn they have limited the capacity to make inferences about evolutionary trends and phylogenetic signal. This project will apply novel virtual bone modelling methods that help illuminate the evolutionary drivers of phenotypic radiation of forelimb morphology, and to also trace potential macroevolutionary phylogenetic signals of disparity and integration (Frelat et al. 2012; Wilson and Humphrey 2015).

Alcohol-preserved, whole bat bodies were micro-CT scanned to build 3D virtual models to describe overall external variation in shape and size (Frelat, Katina et al. 2012), as well as cross-sectional internal shape variation implementing three-dimensional geometric morphometric analysis (Wilson and Humphrey 2015). For the first objective, ontogenetic series of ecomorphologically diverse species were used to explore the heterochrony of postcranial development (Koyabu and Son 2014). For the second

33 objective, whole-bone linear measurements from the ontogenetic series were used to describe allometric trajectories and test for differences across lineages. For the third objective, humeral whole-bone and cross-sectional asymmetry was investigated across prenatal development. For the fourth objective, to reconstruct bat wing evolution, cross-sectional forelimb shape of all extant lineages was incorporated into a molecular, fossil-calibrated phylogeny to reconstruct the evolutionary history of forelimb phenotypic radiation in bats. Lastly, for the fifth objective, the influence of ecology, biology and phylogeny on humeral whole-bone shape variation in modern bat lineages was evaluated.

References

Adams DC. 2014. A method for assessing phylogenetic least squares models for shape and other high-dimensional multivariate data. Evolution 68:2675-2688. Adams DC, Collyer ML. 2015. Permutation tests for phylogenetic comparative analyses of high-dimensional shape data: what you shuffle matters. Evolution 69:823-829. Adams RA. 2008. Morphogenesis in Bat Wings: Linking Development, Evolution and Ecology. Cells Tissues Organs 187:13-23. Adams RA, Shaw JB. 2013. Time’s Arrow in the Evolutionary Development of Bat Flight. Bat evolution, ecology and conservation: 21-46. Aldridge HDJN. 1987. Turning Flight of Bats. Journal of Experimental Biology 128:419- 425. Anderson JF, Hall-Martin A, Russell DA. 1985. Long-bone circumference and weight in mammals, birds and . Journal of Zoology 207:53-61. Arita HT, Fenton MB. 1997. Flight and echlocation in the ecology and evolution of bats. Trends in Ecology & Evolution 12:53-58. Arita HT, Vargas-Barón J, Villalobos F. 2014. Latitudinal gradients of genus richness and endemism and the diversification of New World bats. Ecography 37: 1024-1033. Attard MR, Parr WC, Wilson LA, Archer M, Hand SJ, Rogers TL, Wroe S. 2014. Virtual reconstruction and prey size preference in the mid Cenozoic thylacinid, Nimbacinus dicksoni (Thylacinidae, Marsupialia). PLOS ONE 9:e93088. Attard MR, Wilson LA, Worthy TH, Scofield P, Johnston P, Parr WC, Wroe S. 2016. Moa diet fits the bill: virtual reconstruction incorporating mummified remains and prediction of biomechanical performance in avian giants. Proc Biol Sci 283. Bahlman JW, Price‐Waldman RM, Lippe HW, Breuer KS, Swartz SM. 2016. Simplifying a wing: diversity and functional consequences of digital joint reduction in bat wings. Journal of Anatomy 229:114-127.

34

Bell E, Andres B, Goswami A. 2011. Integration and dissociation of limb elements in flying vertebrates: a comparison of pterosaurs, birds and bats. J Evol Biol 24:2586- 2599. Bergou AJ, Swartz SM, Vejdani H, Riskin DK, Reimnitz L, Taubin G, Breuer KS. 2015. Falling with Style: Bats Perform Complex Aerial Rotations by Adjusting Wing Inertia. PLOS Biology 13: e1002297. Bininda-Emonds OR, Jeffery JE, Sánchez-Villagra MR, Hanken J, Colbert M, Pieau C, Selwood L, Ten Cate C, Raynaud A, Osabutey CK, et al. 2007. Forelimb-hindlimb developmental timing changes across phylogeny. BMC Evol Biol 7:182. Boonman A, Bumrungsri S, Yovel Y. 2014. Nonecholocating fruit bats produce biosonar clicks with their wings. Current biology : CB 24:2962-2967. Brassey CA, Sellers WI. 2014. Scaling of Convex Hull Volume to Body Mass in Modern Primates, Non-Primate Mammals and Birds. PLOS ONE 9: e91691. Butler RJ, Benson RBJ, Barrett PM. 2013. Pterosaur diversity: Untangling the influence of sampling biases, Lagerstätten, and genuine biodiversity signals. Palaeogeography, Palaeoclimatology, Palaeoecology 372:78-87. Canals M, Iriarte-Diaz J, Grossi B. 2011. Biomechanical, Respiratory and Cardiovascular Adaptations of Bats and the Case of the Small Community of Bats in Chile. In: Klika V, editor. Biomechanics in Applications: InTech. p. 299-322. Carter RT, Adams RA. 2016. Integrating Ontogeny of Echolocation and Locomotion Gives Unique Insights into the Origin of Bats. Journal of Mammalian Evolution 23:413– 421. Cheney JA, Allen JJ, Swartz SM. 2017. Diversity in the organization of elastin bundles and intramembranous muscles in bat wings. J Anat 230:510-523. Cheney JA, Ton D, Konow N, Riskin DK, Breuer KS, Swartz SM. 2014. Hindlimb motion during steady flight of the lesser dog-faced fruit bat, Cynopterus brachyotis. PLOS ONE 9:e98093. Collyer ML, Adams DC. 2013. Phenotypic trajectory analysis: comparison of shape change patterns in evolution and ecology. Hystrix, the Italian Journal of Mammalogy 24:75-83. Cooper KL, Tabin CJ. 2008. Understanding of bat wing evolution takes flight. Genes Dev 22:121-124. Cooper L, Cretekos CJ, Sears KE. 2012. The evolution and development of mammalian flight. Wiley Interdisciplinary Reviews: Developmental Biology 1:773-779. Cretekos CJ, Wang Y, Green ED, Martin JF, Rasweiler JJt, Behringer RR. 2008. Regulatory divergence modifies limb length between mammals. Genes Dev 22:141- 151. Cretekos CJ, Weatherbee SD, Chen CH, Badwaik NK, Niswander L, Behringer RR, Rasweiler JJt. 2005. Embryonic staging system for the short-tailed fruit bat, Carollia perspicillata, a model organism for the mammalian order Chiroptera, based upon timed pregnancies in captive-bred animals. Dev Dyn 233:721-738. Cubo J, Casinos A. 1998. The variation of the cross-sectional shape in the long bones of birds and mammals. Annales des Sciences Naturelles - Zoologie et Biologie Animale 19:51-62. Currey JD, Alexander MR. 1985. The thickness of the walls of tubular bones. Journal of Zoology 206:453-468.

35

Dahan G, Trabelsi N, Safran O, Yosibash Z. 2016. Verified and validated finite element analyses of humeri. J Biomech 49:1094-1102. Datzmann T, von Helversen O, Mayer F. 2010. Evolution of nectarivory in phyllostomid bats (Phyllostomidae Gray, 1825, Chiroptera: Mammalia). BMC Evol Biol 10: 165. de Camargo NF, de Oliveira HFM. 2012. Sexual Dimorphism in Sturnira lilium (Chiroptera, Phyllostomidae): Can Pregnancy and Pup Carrying Be Responsible for Differences in Wing Shape? PLOS ONE 7: e49734. Dececchi TA, Larsson HC, Habib MB. 2016. The wings before the bird: an evaluation of flapping-based locomotory hypotheses in bird antecedents. PeerJ 4:e2159. Denzinger A, Schnitzler HU. 2013. Bat guilds, a concept to classify the highly diverse foraging and echolocation behaviors of microchiropteran bats. Front Physiol 4:164. Dickinson M. 2008. locomotion: a new spin on bat flight. Curr Biol 18:R468-470. Dietz CL. 1973. Bat walking behavior. Journal of Mammalogy 54: 790-792. Dumont ER. 2010. Bone density and the lightweight skeletons of birds. Proceedings of the Royal Society B: Biological Sciences 277:2193-2198. Dumont ER, Davalos LM, Goldberg A, Santana SE, Rex K, Voigt CC. 2012. Morphological innovation, diversification and invasion of a new adaptive zone. Proc Biol Sci 279:1797- 1805. Dumont ER, Grosse IR, Slater GJ. 2009. Requirements for comparing the performance of finite element models of biological structures. J Theor Biol 256:96-103. Dumont ER, Piccirillo J, Grosse IR. 2005. Finite-element analysis of biting behavior and bone stress in the facial skeletons of bats. Anat Rec A Discov Mol Cell Evol Biol 283:319-330. Dumont ER, Samadevam K, Grosse I, Warsi OM, Baird B, Davalos LM. 2014. Selection for mechanical advantage underlies multiple cranial optima in new world leaf-nosed bats. Evolution 68:1436-1449. Eiting TP, Gunnell GF. 2009. Global Completeness of the Bat Fossil Record. Journal of Mammalian Evolution 16:151-173. Elangovan V, Yuvana Satya Priya E, Raghuram H, Marimuthu G. 2007. Wing morphology and flight development in the short-nosed fruit bat Cynopterus sphinx. Zoology (Jena) 110:189-196. Ellington CP, vandenBerg C, Willmott AP, Thomas ALR. 1996. Leading-edge vortices in insect flight. Nature 384:626-630. Fenton MB. 2010. Convergences in the diversification of bats. Current Zoology (formerly Acta Zoologica Sinica) 56: 454-468. Ferry-Graham L, Bolnick D, Wainwright P. 2002. Using Functional Morphology to Examine the Ecology and Evolution of Specialization. Integrative and Comparative Biology 42:265–277. Foley NM, Thong VD, Soisook P, Goodman SM, Armstrong KN, Jacobs DS, Puechmaille SJ, Teeling EC. 2015. How and why overcome the impediments to resolution: lessons from rhinolophid and hipposiderid bats. Mol Biol Evol 32:313-333. Frelat MA, Katina S, Weber GW, Bookstein FL. 2012. Technical note: A novel geometric morphometric approach to the study of long bone shape variation. American Journal of Physical Anthropology 149:628-638. Gardiner JD, Codd JR, Nudds RL. 2011. An association between ear and tail morphologies of bats and their foraging style. Canadian Journal of Zoology 89:90-99.

36

Giannini NP. 2012. Toward an integrative theory on the origin of bat flight. In: Gunnell GF, Simmons NB, editors. Evolutionary History of Bats: Fossils, Molecules and Morphology. Cambridge: Cambridge University Press. p. 353-384. Gill PG, Purnell MA, Crumpton N, Brown KR, Gostling NJ, Stampanoni M, Rayfield EJ. 2014. Dietary specializations and diversity in feeding ecology of the earliest stem mammals. Nature 512:303-305. Goswami A, Binder WJ, Meachen J, O'Keefe FR. 2015. The fossil record of phenotypic integration and modularity: A deep-time perspective on developmental and evolutionary dynamics. Proc Natl Acad Sci U S A 112:4891-4896. Goswami A, Milne N, Wroe S. 2011. Biting through constraints: cranial morphology, disparity and convergence across living and fossil carnivorous mammals. Proc Biol Sci 278:1831-1839. Goswami A, Polly PD. 2010. The influence of modularity on cranial morphological disparity in Carnivora and Primates (Mammalia). PLOS ONE 5:e9517. Haba C, Oshida T, Sasaki M, Endo H, Ichikawa H, Masuda Y. 2008. Morphological variation of the Japanese raccoon dog: implications for geographical isolation and environmental . Journal of Zoology 274:239-247. Hafner M, Remsen Jr J, Lanyon S. 1984. Bird Versus Mammal Morphological Diversity. Evolution 38: 1154-1156. Hakansson J, Hedenstrom A, Winter Y, Johansson LC. 2015. The wake of hovering flight in bats. J R Soc Interface 12:20150357. Hallgrimsson B, Percival CJ, Green R, Young NM, Mio W, Marcucio R. 2015. Morphometrics, 3D Imaging, and Craniofacial Development. Curr Top Dev Biol 115:561-597. Hand SJ, Archer M. 2005. A new hipposiderid genus (Microchiroptera) from an early Miocene bat community in Australia. Palaeontology 48. Hand SJ, Lee DE, Worthy TH, Archer M, Worthy JP, Tennyson AJ, Salisbury SW, Scofield RP, Mildenhall DC, Kennedy EM, et al. 2015. Miocene Fossils Reveal Ancient Roots for New Zealand's Endemic Mystacina (Chiroptera) and Its Rainforest Habitat. PLOS ONE 10:e0128871. Hand SJ, Sigé B, Archer M, Gunnell GF, Simmons NB. 2015. A New Early (Ypresian) Bat from Pourcy, Paris Basin, France, with Comments on Patterns of Diversity in the Earliest Chiropterans. Journal of Mammalian Evolution 22:343-354. Hand SJ, Weisbecker V, Beck RM, Archer M, Godthelp H, Tennyson AJ, Worthy TH. 2009. Bats that walk: a new evolutionary hypothesis for the terrestrial behaviour of New Zealand's endemic mystacinids. BMC Evol Biol 9:169. Hautier L, Bennett NC, Viljoen H, Howard L, Milinkovitch MC, Tzika AC, Goswami A, Asher RJ. 2013. Patterns of ossification in southern versus northern placental mammals. Evolution 67:1994-2010. Hedenstrom A, Johansson LC. 2015. Bat flight: aerodynamics, kinematics and flight morphology. Journal of Experimental Biology 218:653-663. Hedenstrom A, Johansson LC, Spedding GR. 2009. Bird or bat: comparing airframe design and flight performance. Bioinspir Biomim 4:015001. Henderson DM. 2010. Pterosaur body mass estimates from three-dimensional mathematical slicing. Journal of Vertebrate Paleontology 30:768-785. Iriarte-Diaz J, Riskin DK, Breuer KS, Swartz SM. 2012. Kinematic plasticity during flight in fruit bats: individual variability in response to loading. PLOS ONE 7:e36665. 37

Iriarte-Diaz J, Riskin DK, Willis DJ, Breuer KS, Swartz SM. 2011. Whole-body kinematics of a fruit bat reveal the influence of wing inertia on body accelerations. J Exp Biol 214:1546-1553. Johansson LC, Engel S, Kelber A, Heerenbrink MK, Hedenstrom A. 2013. Multiple leading edge vortices of unexpected strength in freely flying hawkmoth. Sci Rep 3:3264. Johansson LC, Hakansson J, Jakobsen L, Hedenstrom A. 2016. Ear-body lift and a novel thrust generating mechanism revealed by the complex wake of brown long-eared bats (Plecotus auritus). Sci Rep 6:24886. Johansson LC, Wolf M, Hedenstrom A. 2010. A quantitative comparison of bird and bat wakes. J R Soc Interface 7:61-66. Jones G, Teeling EC. 2006. The evolution of echolocation in bats. Trends Ecol Evol 21:149-156. Jones KE, Safi K. 2011. Ecology and evolution of mammalian biodiversity Introduction. Philosophical Transactions of the Royal Society B-Biological Sciences 366:2451-2461. Kapusta A, Suh A, Freschotte C. 2017. Dynamics of genome size evolution in birds and mammals. Proceedings of the National Academy of Science 114:E1460–E1469. Kirkpatrick S. 1994. Scale Effects on the Stresses and Safety Factors in the Wing Bones of Birds and Bats. Journal of Experimental Biology 190:195-215. Klingenberg CP. 2013. Cranial integration and modularity: insights into evolution and development from morphometric data. Hystrix, the Italian Journal of Mammalogy 24:43-58. Klingenberg CP. 2009. Morphometric integration and modularity in configurations of landmarks: tools for evaluating a priori hypotheses. Evol Dev 11:405-421. Koyabu D, Endo H, Mitgutsch C, Suwa G, Catania KC. 2011. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. BMC EvoDevo 2: 21. Koyabu D, Son N. 2014. Patterns of postcranial ossification and sequence heterochrony in bats: Life histories and developmental trade‐offs. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 322:607-618. Koyabu D, Werneburg I, Morimoto N. 2014. Mammalian skull heterochrony reveals modular evolution and a link between cranial development and brain size. Nature Communications 5: 3625. Lee AH, Simons E. 2015. Wing bone laminarity is not an adaptation for torsional resistance in bats. PeerJ 3: e823. Lentink D, Dickson WB, van Leeuwen JL, Dickinson MH. 2009. Leading-Edge Vortices Elevate Lift of Autorotating Plant Seeds. Science 324:1438-1440. Lin AQ, Jin LR, Shi LM, Sun KP, Berquist SW, Liu Y, Feng J. 2011. Postnatal development in Andersen's leaf-nosed bat Hipposideros pomona: flight, wing shape, and wing bone lengths. Zoology (Jena) 114:69-77. Lindhe Norberg UM, Winter Y. 2006. Wing beat kinematics of a nectar-feeding bat, Glossophaga soricina, flying at different flight speeds and Strouhal numbers. J Exp Biol 209:3887-3897. MacAyeal LC, Riskin DK, Swartz SM, Breuer KS. 2011. Climbing flight performance and load carrying in lesser dog-faced fruit bats (Cynopterus brachyotis). J Exp Biol 214:786- 793. 38

Machado JP, Johnson WE, Gilbert MT, Zhang G, Jarvis ED, O'Brien SJ, Antunes A. 2016. Bone-associated gene evolution and the origin of flight in birds. BMC Genomics 17:371. Maier JA, Rivas-Astroza M, Deng J, Dowling A, Oboikovitz P, Cao X, Behringer RR, Cretekos CJ, Rasweiler JJt, Zhong S, et al. 2017. Transcriptomic insights into the genetic basis of mammalian limb diversity. BMC Evol Biol 17:86. Marchetti K, Price T, Richman A. 1995. Correlates of Wing Morphology with Foraging Behaviour and Migration Distance in the Genus Phylloscopus. Journal of Avian Biology 26:177-181. Marelli CA, Simons EL. 2014. Microstructure and cross-sectional shape of limb bones in Great Horned Owls and Red-tailed Hawks: how do these features relate to differences in flight and hunting behavior? PLOS ONE 9:e106094. Marinello MM, Bernard E. 2014. Wing morphology of Neotropical bats: a quantitative and qualitative analysis with implications for habitat use. Canadian Journal of Zoology 92:141-147. Marshall KL, Chadha M, deSouza LA, Sterbing-D'Angelo SJ, Moss CF, Lumpkin EA. 2015. Somatosensory substrates of flight control in bats. Cell Rep 11:851-858. McCabe K, Henderson K, Pantinople J, Richards HL, Milne N. 2017. Curvature reduces bending strains in the quokka femur. PeerJ 5:e3100. McCurry MR, Evans AR, McHenry CR. 2015. The sensitivity of biological finite element models to the resolution of surface geometry: a case study of crocodilian crania. PeerJ 3:e988. McGowan AJ, Dyke GJ. 2007. A morphospace-based test for competitive exclusion among flying vertebrates: did birds, bats and pterosaurs get in each other's space? J Evol Biol 20:1230-1236. McNamara KJ. 2012. Heterochrony: the Evolution of Development. Evolution: Education and Outreach 5:203-218. Monteiro LR, Nogueira MR. 2011. Evolutionary patterns and processes in the radiation of phyllostomid bats. BMC Evol Biol 11:137. Muijres FT, Christoffer Johansson L, Winter Y, Hedenstrom A. 2014. Leading edge vortices in lesser long-nosed bats occurring at slow but not fast flight speeds. Bioinspir Biomim 9:025006. Muijres FT, Johansson LC, Barfield R, Wolf M, Spedding GR, Hedenstrom A. 2008. Leading-Edge Vortex Improves Lift in Slow-Flying Bats. Science 319: 1250-1253. Muijres FT, Johansson LC, Bowlin MS, Winter Y, Hedenstrom A. 2012. Comparing aerodynamic efficiency in birds and bats suggests better flight performance in birds. PLOS ONE 7:e37335. Muijres FT, Johansson LC, Hedenstrom A. 2012. Leading edge vortex in a slow-flying passerine. Biol Lett 8:554-557. Muijres FT, Johansson LC, Winter Y, Hedenstrom A. 2011. Comparative aerodynamic performance of flapping flight in two bat species using time-resolved wake visualization. J R Soc Interface 8:1418-1428. Norberg RA, Rayner JMV. 1987. Ecological Morphology and Flight in Bats (Mammalia; Chiroptera): Wing Adaptations, Flight Performance, Foraging Strategy and Echolocation. Philos Trans R Soc Lond B Biol Sci 316:335-427.

39

O'higgins P, Milne N. 2013. Applying geometric morphometrics to compare changes in size and shape arising from finite elements analyses. Hystrix, the Italian Journal of Mammalogy 24:126-132. Organ CL, Shedlock AM. 2009. Palaeogenomics of pterosaurs and the evolution of small genome size in flying vertebrates. Biol Lett 5:47-50. Oyston JW, Hughes M, Wagner PJ. 2015. What limits the morphological disparity of clades? Interface Focus 5: 20150042. Panyutina A, Korzun LP, Kuznetsov AN. 2015. Flight of Mammals: From Terrestrial Limbs to Wings: Springer, New York. Parr WCH, Wroe S, Chamoli U, Richards HS, McCurry MR, Clausen PD, McHenry C. 2012. Toward integration of geometric morphometrics and computational biomechanics: New methods for 3D virtual reconstruction and quantitative analysis of Finite Element Models. Journal of Theoretical Biology 301:1-14. Petit F, Sears KE, Ahituv N. 2017. Limb development: a paradigm of gene regulation. Nature Reviews Genetics 18: 245-258. Pieretti J, Gehrke AR, Schneider I, Adachi N, Nakamura T, Shubin NH. 2015. Organogenesis in deep time: A problem in genomics, development, and paleontology. Proc Natl Acad Sci U S A 112:4871-4876. Polly PD, Stayton CT, Dumont ER, Pierce SE, Rayfield EJ, Angielczyk KD. 2016. Combining geometric morphometrics and finite element analysis with evolutionary modeling: towards a synthesis. Journal of Vertebrate Paleontology 36:e1111225. Prentice KC, Ruta M, Benton MJ. 2011. Evolution of morphological disparity in pterosaurs. Journal of Systematic Palaeontology 9:337-353. Rayner JM. 1988. The evolution of vertebrate flight. Biological Journal of the Linnean Society 34:269-287. Rayner JMV. 2011. Mechanics and physiology of flight in fossil vertebrates. Transactions of the Royal Society of Edinburgh: Earth Sciences 80:311-320. Riskin DK, Bergou A, Breuer KS, Swartz SM. 2012. Upstroke wing flexion and the inertial cost of bat flight. Proc Biol Sci 279:2945-2950. Riskin DK, Bertram JEA, Hermanson JW. 2005. Testing the hindlimb-strength hypothesis: non-aerial locomotion by Chiroptera is not constrained by the dimensions of the femur or tibia. Journal of Experimental Biology 208:1309-1319. Riskin DK, Iriarte-Diaz J, Middleton KM, Breuer KS, Swartz SM. 2010. The effect of body size on the wing movements of pteropodid bats, with insights into thrust and lift production. J Exp Biol 213:4110-4122. Riskin DK, Parsons S, Schutt WA, Jr., Carter GG, Hermanson JW. 2006. Terrestrial locomotion of the New Zealand short-tailed bat Mystacina tuberculata and the common vampire bat Desmodus rotundus. J Exp Biol 209:1725-1736. Riskin DK, Willis DJ, Iriarte-Diaz J, Hedrick TL, Kostandov M, Chen J, Laidlaw DH, Breuer KS, Swartz SM. 2008. Quantifying the complexity of bat wing kinematics. J Theor Biol 254:604-615. Rojas D, Vale Á, Ferrero V, Navarro L. 2012. The role of frugivory in the diversification of bats in the Neotropics. Journal of 39:1948-1960. Ross D, Marcot JD, Betteridge KJ, Nascone-Yoder N, Bailey CS, Sears KE. 2013. Constraints on Mammalian forelimb development: insights from developmental disparity. Evolution 67:3645-3652.

40

Ruff C, Holt B, Trinkaus E. 2006. Who's afraid of the big bad Wolff?: "Wolff's law" and bone functional adaptation. Am J Phys Anthropol 129:484-498. Santana SE, Cheung E. 2016. Go big or go fish: morphological specializations in carnivorous bats. Proceedings of the Royal Society B: Biological Sciences 283:20160615. Santana SE, Dumont ER, Davis JL. 2010. Mechanics of bite force production and its relationship to diet in bats. Functional Ecology 24:776-784. Santana SE, Grosse IR, Dumont ER. 2012. Dietary hardness, loading behavior, and the evolution of skull form in bats. Evolution 66:2587-2598. Santana SE, Strait S, Dumont ER. 2011. The better to eat you with: functional correlates of tooth structure in bats. The better to eat you with: functional correlates of tooth structure in bats. Sato K, Sakamoto KQ, Watanuki Y, Takahashi A, Katsumata N, Bost CA, Weimerskirch H. 2009. Scaling of soaring seabirds and implications for flight abilities of giant pterosaurs. PLOS ONE 4:e5400. Schileo E, Taddei F, Malandrino A, Cristofolini L, Viceconti M. 2007. Subject-specific finite element models can accurately predict strain levels in long bones. J Biomech 40:2982-2989. Schliemann H, Goodman SM. 2011. A new study on the structure and function of the adhesive organs of the Old World sucker-footed bat (Myzopoda: Myzopodidae) of Madagascar. Verh. Naturwiss. Ver. Hamburg. Schmieder DA, Benitez HA, Borissov IM, Fruciano C. 2015. Bat Species Comparisons Based on External Morphology: A Test of Traditional versus Geometric Morphometric Approaches. PLOS ONE 10:e0127043. Shubin NH, Tabin CJ, Carroll S. 1997. Fossils, genes and the evolution of animal limbs. Nature 388:639–648. Simmons NB, Seymour KL, Habersetzer J, Gunnell GF. 2010. Inferring echolocation in ancient bats. Nature 466:E8; discussion E9. Simmons NB, Seymour KL, Habersetzer J, Gunnell GF. 2008. Primitive Early Eocene bat from Wyoming and the evolution of flight and echolocation. Nature 451:818-821. Simons EL. 2010. Forelimb skeletal morphology and flight mode evolution in pelecaniform birds. Zoology (Jena) 113:39-46. Simons EL, Hieronymus TL, O'Connor PM. 2011. Cross sectional geometry of the forelimb skeleton and flight mode in pelecaniform birds. J Morphol 272:958-971. Simons EL, O'Connor P M. 2012. Bone laminarity in the avian forelimb skeleton and its relationship to flight mode: testing functional interpretations. Anat Rec (Hoboken) 295:386-396. Smith KK. 2001. Heterochrony revisited: the evolution of developmental sequences. Biological Journal of the Linnean Society 73:169-186. Smith KK. 2002. Sequence heterochrony and the evolution of development. J Morphol 252:82-97. Smith KK. 2003. Time's arrow: heterochrony and the evolution of development. International Journal of Developmental Biology 47:613-621. Spielmann M, Brancati F, Krawitz PM, Robinson PN, Ibrahim DM, Franke M, Hecht J, Lohan S, Dathe K, Nardone AM, et al. 2012. Homeotic Arm-to-Leg Transformation Associated with Genomic Rearrangements at the PITX1 Locus. American Journal of Human Genetics 91:629-635. 41

Sterbing-D'Angelo S, Chadha M, Chiu C, Falk B, Xian W, Barcelo J, Zook JM, Moss CF. 2011. Bat wing sensors support flight control. Proc Natl Acad Sci U S A 108:11291- 11296. Stroud JT, Losos JB. 2016. Ecological Opportunity and Adaptive Radiation. Annual Review of Ecology, Evolution, and 47:507-532. Swartz SM, Bennett MB, Carrier DR. 1992. Wing bone stresses in free flying bats and the evolution of skeletal design for flight. Nature 359: 726–729. Swartz SM, Konow N. 2015. Advances in the study of bat flight: the wing and the wind 1. Canadian Journal of Zoology 93: 977-990. Swartz SM, Middleton KM. 2008. Biomechanics of the bat limb skeleton: Scaling, material properties and mechanics. Cells Tissues Organs 187:59-84. Teeling EC, Scally M, Kao D, Romagnoli M, Springer MS, Stanhope M. 2000. Molecular evidence regarding the origin of echolocation and flight in bats. Nature 403:188–192. Tokita M. 2015. How the pterosaur got its wings. Biol Rev Camb Philos Soc 90:1163- 1178. Velazco P, Gregorin R, Voss RS, Simmons NB. 2014. Extraordinary Local Diversity of Disk-winged Bats (Thyropteridae: Thyroptera) in Northeastern Peru, with the Description of a New Species and Comments on Roosting Behavior. American Museum Novitates 3795. Veselka N, McErlain DD, Holdsworth DW, Eger JL, Chhem RK, Mason MJ, Brain KL, Faure PA, Fenton MB. 2010. A bony connection signals laryngeal echolocation in bats. Nature 463:939-942. Von Busse R, Hedenstrom A, Winter Y, Johansson LC. 2012. Kinematics and wing shape across flight speed in the bat, Leptonycteris yerbabuenae. Biol Open 1:1226-1238. Wang X, McGowan AJ, Dyke GJ. 2011. Avian wing proportions and flight styles: first step towards predicting the flight modes of mesozoic birds. PLOS ONE 6:e28672. Wang Z, Zhu T, Xue H, Fang N, Zhang J, Zhang L, Pang J, Teeling EC, Zhang S. 2017. Prenatal development supports a single origin of laryngeal echolocation in bats. Nature Ecology & Evolution 1: 0021. Warrick DR, Tobalske BW, Powers DR. 2009. Lift production in the hovering hummingbird. Proc Biol Sci 276:3747-3752. Watts P, Mitchell E, Swartz S. 2001. A computational model for estimating the mechanics of horizontal flapping flight in bats: Model description and validation. Journal of Experimental Biology 204:2873-2898. Weisbecker V. 2011. Monotreme ossification sequences and the riddle of mammalian skeletal development. Evolution 65:1323-1335. Weisbecker V, Goswami A, Wroe S, Sánchez-Villagra MR. 2008. Ossification heterochrony in the therian postcranial skeleton and the marsupial-placental dichotomy. Evolution 62:2027-2041. Werneburg I, Tzika AC, Hautier L, Asher RJ, Milinkovitch MC, Sánchez‐Villagra MR. 2013. Development and embryonic staging in non‐model organisms: the case of an afrotherian mammal. Journal of Anatomy 222:2-18. Wilson L, Humphrey LT. 2015. A Virtual geometric morphometric approach to the quantification of long bone bilateral asymmetry and cross‐sectional shape. American Journal of Physical Anthropology 4: 541-556. Wilson LAB, Schradin C, Mitgutsch C, Galliari FC, Mess A, Sánchez-Villagra MR. 2010. Skeletogenesis and sequence heterochrony in rodent evolution, with particular 42 emphasis on the African striped mouse, Rhabdomys pumilio (Mammalia). Organisms Diversity & Evolution 10:243-258. Witton MP. 2008. A new approach to determining pterosaur body mass and its implications for pterosaur flight. Zitteliana 28:143-158. Witton MP, Habib MB. 2010. On the size and flight diversity of giant pterosaurs, the use of birds as pterosaur analogues and comments on pterosaur flightlessness. PLOS ONE 5:e13982. Wolf M, Johansson LC, von Busse R, Winter Y, Hedenstrom A. 2010. Kinematics of flight and the relationship to the vortex wake of a Pallas' long tongued bat (Glossophaga soricina). J Exp Biol 213:2142-2153. Wroe S, Chamoli U, Parr WC, Clausen P, Ridgely R, Witmer L. 2013. Comparative Biomechanical Modeling of Metatherian and Placental Saber-Tooths: A Different Kind of Bite for an Extreme Pouched Predator. PLOS ONE 8:e66888.

43

Chapter 2 Postcranial heterochrony, modularity, integration and disparity in the prenatal ossification in bats (Chiroptera) 1

Abstract

Self-powered flight is one of the most energy-intensive types of locomotion found in vertebrates. It is also associated with a range of extreme morpho-physiological adaptations that evolved independently in three different vertebrate groups.

Considering that development acts as a bridge between the genotype and phenotype on which selection acts, studying the ossification of the postcranium can potentially illuminate our understanding of bat flight evolution. However, the ontogenetic basis of vertebrate flight remains largely understudied. Advances in quantitative analysis of sequence heterochrony and morphogenetic growth have created novel approaches to study the developmental basis of diversification and the evolvability of skeletal morphogenesis. Assessing the presence of ontogenetic disparity, integration and

1 A version of this has been accepted for publication: López-Aguirre, C., Hand, S.J., Koyabu, D. et al. Postcranial heterochrony, modularity, integration and disparity in the prenatal ossification in bats (Chiroptera). BMC Evol Biol 19, 75 (2019). A copy is included in Appendix peer-reviewed publication 1. * I certify that this publication was a direct result of my research towards this PhD, including the design, analysis and interpretation of results, and that reproduction in this thesis does not breach copyright regulations. I acknowledge my co-authors for their supervisory guidance and their contributions to the publication.

44 modularity from an evolutionary approach allows assessing whether flight may have resulted in evolutionary differences in the magnitude and mode of development in bats. We quantitatively compared the prenatal ossification of the postcranium (24 bones) between bats (14 species), non-volant mammals (11 species) and birds (14 species), combining for the first time prenatal sequence heterochrony and developmental growth data. Sequence heterochrony was found across groups, showing that bat postcranial development shares patterns found in other flying vertebrates but also those in non-volant mammals. In bats, modularity was found as an axial-appendicular partition, resembling a mammalian pattern of developmental modularity and suggesting flight did not repattern prenatal postcranial covariance in bats. Combining prenatal data from 14 bat species, this study represents the most comprehensive quantitative analysis of chiropteran ossification to date. Heterochrony between the wing and leg in bats could reflect functional needs of the newborn, rather than ecological aspects of the adult. Bats share similarities with birds in the development of structures involved in flight (i.e. handwing and sternum), suggesting that flight altriciality and early ossification of pedal phalanges and sternum are common across flying vertebrates. These results indicate that the developmental modularity found in bats facilitates intramodular phenotypic diversification of the skeleton. Integration and disparity increased across developmental time in bats. We also found a delay in the ossification of highly adaptable and evolvable regions (e.g. handwing and sternum) that are directly associated with flight performance.

Introduction

Pterosaurs, birds and bats are the only vertebrates capable of self-powered flight

(herein refered to as flight, (Rayner 1988). However, the phylogenetic relationships 45 between these groups and their position in the evolutionary history of vertebrates have shown that this feature evolved in each group independently, as a result of convergent evolution (Rayner 1988; Tokita 2015). Historically, the convergent evolution of vertebrate flight was asynchronous, with pterosaurs evolving flight first

(≈240 Mya), followed by birds (≈140 Mya), and bats (≈60 Mya) (Organ and Shedlock

2009; Fong, et al. 2012; Amador, Arevalo, et al. 2018). Flight was a key innovation that provided a major ecological opportunity for these groups, allowing them to diversify into a vast range of previously empty niches (Stroud and Losos 2016). As a result, both birds and bats are some of the most speciose groups of living vertebrates. It is estimated that pterosaurs were also abundant and diverse before their extinction, with over 160 fossil species described (Butler, et al. 2013).

It has been suggested that all three groups share some morphological and physiological adaptations in body regions that are vital for the kinematics of flight

(Dumont 2010). Reduction of cortical bone thickness (Cubo and Casinos 1998), increased bone density (Dumont 2010), morphological changes in the pectoral girdle

(Currey and Alexander 1985), and elongation of forelimb bones are some of the adaptations that these groups share (Lee and Simons 2015). It has also been found that all three groups have relatively small genomes when compared to their respective close relatives (Organ and Shedlock 2009), suggesting that constricted genome sizes has been evolutionarily correlated with the evolution of self-powered flight in vertebrates (Organ and Shedlock 2009; Kapusta, et al. 2017).

The relatively abundant fossil record of pterosaurs and birds has facilitated the study of the evolution of flight in these groups. Bat flight evolution, on the other hand, has been comparably more difficult to study due to the incompleteness of the fossil 46 record, limiting the scope for research (Eiting and Gunnell 2009). Studying how the development of organisms reflects their evolutionary history (Hall 2000; Hendrikse, et al. 2007; Weisbecker, et al. 2008; Maxwell and Harrison 2009; Koyabu, et al. 2011;

Organ, et al. 2015), represents an alternative to assess the extent of interspecific variation in postcranial morphology over the course of ontogeny, data that could provide a baseline for further study into the evolution of mammalian flight (Wang,

Zhu, et al. 2017; Nojiri, et al. 2018).

Based on previous studies that used ossification to study morphogenesis, we now know that the development of specific traits is not homogenous across taxa that share those features (Bininda-Emonds, Jeffery, et al. 2007). This has been demonstrated for both analogous and homologous traits (Cretekos, et al. 2008). Comparisons of ossification sequences have shown that the interaction between ontogeny and ecology has a domino effect on the modification and specialisation of the mammalian skeleton

(Wilson, et al. 2010; Ross, et al. 2013). Nevertheless, previous studies have suggested that it may not always be possible to differentiate between functional and phylogenetic developmental fluctuations (Koyabu and Son 2014). The delimitation between ontogenetic changes reflecting a phylogenetic signal and changes reflecting ecomorphological adaptations is not clear when analysing developmental sequences without testing a specific hypothesis (Koyabu and Son 2014).

There are two general approaches to study changes in morphology during development: sequence heterochrony and developmental growth analyses (Smith

2002). Sequence heterochrony analyses document variations in the timing and order in which a group of traits starts developing, permitting the study of developmental

47 events that are not explicitly characterized by size or shape (Smith 2002).

Developmental growth refers to differences in the timing of growth onset, and growth rate that different traits experience during the development of an individual (Smith

2002). Studying sequence heterochrony has been particularly useful to elucidate developmental differences in patterns of ossification in a wide range of vertebrate taxa

(Smith 2001, 2002, 2003; Ross, et al. 2013; Werneburg, et al. 2013; Koyabu,

Werneburg and Morimoto 2014). Among vertebrates, mammals exhibit the widest range of skeletal specialisations for different locomotor and feeding strategies (Hafner, et al. 1984; Weisbecker, et al. 2008; Maxwell and Harrison 2009; Koyabu, et al. 2011).

As a result, mammals have successfully adapted to more ecosystems than any other vertebrate group (Jones and Safi 2011). Sequence heterochrony studies of skeletal ossification in mammals have suggested an ontogenetic basis for the functional adaptability of the postcranium (Weisbecker, et al. 2008; Koyabu, et al. 2011; Hautier, et al. 2013; Koyabu 2017). Sequence heterochrony in mammals has also been linked to both phylogeny and life history (Koyabu, Werneburg and Morimoto 2014; Koyabu

2017).

Developmental growth studies have helped to elucidate evolutionary changes in ontogenetic trajectories across lineages (Klingenberg 2010). By focusing on the morphological changes across the development of a structure, it is possible to recreate the relationship between shape and size and how it responds to ecological, genetic and ontogenetic constrains (Wilson and Sánchez-Villagra 2010b; Wilson 2013a; Randau and Goswami 2017; Randau and Goswami 2018). This principle has been applied to the evolutionary history of different taxa, reconstructing ancestral states and historical

48 divergences and diversifications, providing a developmental perspective to the study of evolution (Gerber, et al. 2008; Adams and Nistri 2010; Ramirez-Chaves, et al. 2016;

Esquerré, et al. 2017).

Prenatal and postnatal development have been found to be drastically dissimilar processes for some species (Wilson 2011b; Halley 2016; Zelditch, et al. 2016;

Werneburg and Geiger 2017), experiencing different selective pressures (Zelditch, et al. 2016; Werneburg and Geiger 2017), and resulting in differential effects on morphological disparity at adulthood (Wilson, et al. 2010; Wilson 2013a; Zelditch, et al.

2016; Werneburg and Geiger 2017). A prenatal developmental basis has been suggested for the phenotypic diversity in some mammal species, proposing that prenatal development shows more variability as it does not experience strong environmental changes (Wilson, et al. 2010; Wilson 2013a; Zelditch, et al. 2016;

Werneburg and Geiger 2017). Moreover, a link between altriciality and the capacity of a system to evolve (i.e. evolvability; (Hendrikse, et al. 2007)) has been proposed, in which incomplete development at birth and extended developmental times have been hypothesised to facilitate adaptability (Zelditch, et al. 2016; Werneburg and Geiger

2017).

Another aspect of mammalian evolution that has been increasingly studied is how the interaction between morphological disparity and integration shape the phenotypic evolution of a clade (Wilson and Sánchez-Villagra 2010b; Gerber 2013; Goswami, et al.

2014a; Swartz and Konow 2015). Morphological disparity refers to the phenotypic variability within a set of individuals, whereas morphological integration indicates a correlation in the morphological variation of a set of traits (Klingenberg 2014). Recent

49 quantitative studies have examined the role that the interaction between disparity and integration could play in the evolutionary history of a group, leading to two, mutually- exclusive hypotheses tractable to testing: 1) high levels of integration restrains disparity, canalising all phenotypes to a similar state (Goswami, et al. 2014b), or 2) high levels of integration create an “evolutionary buffer” that facilitates disparity, leading to phenotypic diversification (Goswami, et al. 2014b). These two hypotheses help to define the range of possible forms that the disparity-integration interaction can take in biological systems. Furthermore, integration can create groups of traits that show high within-group correlations, but that are weakly connected and therefore relatively independent from other groups (i.e. modularity; (Wagner and Altenberg 1996;

Wagner, et al. 2007)). Consequently, one could expect that modularity would vary in its configuration as integration of the entire system is higher, as all traits are strongly correlated.

Despite increased interest in the morphological development and ontogeny of mammals (Sánchez-Villagra 2002; Weisbecker, et al. 2008; Wilson and Sánchez-Villagra

2010b; Werneburg, et al. 2013; Thean, et al. 2017; Wang, Zhu, et al. 2017), many questions regarding the ontogenetic basis of modern phenotypic diversity remain unanswered. Bats are an excellent example of ecomorphological diversity within

Mammalia, having a wide range of dietary specialisations (Santana, et al. 2012b), and exhibiting a range of locomotor strategies (Dickinson 2008). Bats show a unique, highly derived postcranial body plan adapted to flight (e.g. elongated forelimbs, reduced bone cortical thickness, specialised pectoral girdle morphology), but different bat species are also capable of walking and in some extreme cases swimming (Riskin, et al.

50

2005; Riskin, et al. 2006; Riskin, et al. 2008). To date, few studies have focused on quantifying and interpreting heterochrony of the postcranium in bats (Adams 2000;

Koyabu, Werneburg, Morimoto, et al. 2014). Koyabu and Son (Koyabu and Son 2014) showed that bats have accelerated ossification of the phalanges of the hindlimbs and thumb during prenatal development, and suggested that this pattern could be a functional response to the need to continuously attach to their mothers and to the substrate of their roosts. This is particularly relevant in bats because newborns are incapable of flying for several weeks or up to several months after birth (Koyabu and

Son 2014). Given that vertebrate flight is polyphyletic, it is possible that the processes underlying postcranial development in bats converged with those of other flying vertebrates. Consequently, bat postcranial ossification could diverge from a more phylogenetically constrained mammalian development to a more functionally convergent flying vertebrate development. However, our understanding of the developmental basis of vertebrate flight remains limited.

Combining sequence heterochrony and metric growth data, this study compared the prenatal ossification of the postcranium in bats, birds and non-volant mammals, evaluating whether flight may have shaped the magnitude and mode of postcranial development in bats compared to non-volant mammals and birds. We used the most comprehensive sampling of prenatal bats to date. We quantify the levels of disparity and integration across bones during ossification, to assess whether integration restrains or promotes disparity during the development of bats. Considering that flight and the associated morphological specialisations are exclusive to bats among mammals, we expect to find a shift in the postcranial development of bats compared

51 to non-volant mammals, showing some similarities with other flying vertebrates. Also, based on the embryological origin and functional differences of different regions of the postcranium, we expect to find evidence of ontogenetic modularity. As a result, we expect a positive correspondence between magnitudes of disparity and integration within each module, as well as differences in levels of disparity and integration between modules.

Materials and methods

Data collection

To study prenatal postcranial ossification in bats, 66 specimens representing developmental series of 11 bat species (Aselliscus dongbacana, A. stoliczkanus,

Cynopterus sphinx, Hesperoptenus blanfordi, Hipposideros larvatus, Kerivoula hardwickii, Miniopterus schreibersii, Myotis sp., Rhinolophus pearsoni, R. pusillus, R. thomasi) were sampled, comprising five families and members from both chiropteran suborders Yinpterochiroptera and Yangochiroptera (Table 2.1). Specimens were collected and prepared in Vietnam for a study of sequence heterochrony in bats

(Koyabu and Son 2014), and stored in 70% ethanol. Grey scale images of specimens were acquired using a microfocal X-ray CT system at the University Museum,

University of Tokyo (TXS225-ACTIS, TESCO, Tokyo) with 70 kV source voltage and µ114

A source currents at a resolution of 36 µm (Fig. 2.1). Segmentation of the skeleton from other tissues was performed using the thresholding tool in MIMICS v. 20 software

(Materialise NV), using the bone (CT) predefined set as a basis. Finer thresholding to separate CT-values of osseous and non-osseous structures followed the Half Maximum

Height method (HMH) (Spoor, et al. 1993). This technique did not allow us to

52 reconstruct cartilaginous tissue during the growth of the skeleton, limiting our capacity to describe the growth of non-osseous tissue in the bone.

Figure 2.1. 3D virtual models of ontogenetic series of H. blanfordi, representing the samples from which raw measurements were taken from postcranial elements.

Ossification sequence data of 24 postcranial elements were recorded by analysing the

3D virtual models of the postcranial skeleton using MIMICS v. 20 software (Materialise

NV). Raw models of scanned specimens were thresholded to generate the virtual models of the postcranial skeleton (Fig. 2.1). Ossification sequences for three additional species were compiled from previous studies (Appendix Figure 1): Rousettus amplexicaudatus (Weisbecker, et al. 2008), Myotis ater (Koyabu and Son 2014) and M. lucifugus (Adams 1992). Prenatal ossification sequences of 11 non-volant mammal species and 14 bird species were consolidated based on previously published studies

(Koyabu and Son 2014). Our sample included three orders, eight families of non-volant mammals, and six orders and seven families of birds.

53

Table 2.1. List of mammal and bird species analysed in this study.

Class Order Family Species Source Mammalia Chiroptera Hipposideridae Aselliscus dongbacana This study (n= 14)

Aselliscus stoliczkanus This study (n= 16) Hipposideros larvatus This study (n= 4) Pteropodidae Cynopterus sphinx This study (n= 4) Rousettus amplexicaudatus (Weisbecker, et al. 2008) Rhinolophidae Rhinolophus pearsoni This study (n= 1) Rhinolophus pusillus This study (n= 1) Rhinolophus thomasi This study (n= 10) Vespertilionidae Hesperoptenus blanfordi This study (n= 5) Kerivoula hardwickii This study (n= 7) Myotis sp. This study (n= 2) Myotis ater (Koyabu and Son 2014) Myotis lucifugus (Koyabu and Son 2014) Miniopteridae Miniopterus schreibersii This study (n= 2) (Koyabu and Son 2014) Rodentia Muridae Rattus norvegicus (Koyabu and Son 2014) Meriones unguiculatus (Koyabu and Son 2014) Mus musculus (Koyabu and Son 2014) Rhabdomys pumilio (Koyabu and Son 2014) Cricetidae Peromyscus melanophrys (Koyabu and Son 2014) Mesocricetus auratus (Koyabu and Son 2014) Octodontidae Octodon degus (Koyabu and Son 2014) Eulipotyphla Soricidae Cryptotis parva (Koyabu and Son 2014) Talpidae Talpa europaea (Koyabu and Son 2014) Cetartiodactyla Bovidae Bos taurus (Koyabu and Son 2014) Suidae Sus scrofa Aves Galliformes Phasianidae Meleagris gallopavo (Maxwell, et al. 2010) (Maxwell, et al. 2010) Coturnix coturnix (Maxwell, et al. 2010) Gallus gallus (Maxwell, et al. 2010) Struthioniformes Struthionidae Struthio sp. (Maxwell, et al. 2010) Casuariiformes Casuariidae Dromaius novaehollandiae (Maxwell, et al. 2010) Anseriformes Anatidae Somateria mollissima (Maxwell, et al. 2010) Anas platyrhynchos (Maxwell, et al. 2010) Cairina moschata (Maxwell, et al. 2010) Pelecaniformes Phalacrocoracidae Phalacrocorax auritus (Maxwell, et al. 2010) Charadriiformes Laridae Sterna hirundo (Maxwell, et al. 2010) Larus ridibundus (Maxwell, et al. 2010) Larus argentatus (Maxwell, et al. 2010) Larus canus (Maxwell, et al. 2010) Stercorariidae Stercorarius skua

54

Sequence heterochrony

For each specimen, the ossification level of each bone was codified as one of three categories: unossified, ossification onset, or partly ossified. To consolidate the ossification sequence of each species, we followed a modified absolute rank r standardisation implemented in previous studies (Goswami 2007; Sánchez-Villagra, et al. 2008; Weisbecker, et al. 2008; Wilson, et al. 2010; Koyabu, et al. 2011). The traditional approach standardises the absolute rank r by the maximum number of ranks, which has been shown to result in high variability in the relative rank of the first bones to ossify (Koyabu, Werneburg and Morimoto 2014). To address this, the relative rank of each species was scaled from 0 (i.e. earliest ossification event) to 1 (i.e. latest ossification event), removing possible interspecific differences in the maximum rank number (Koyabu, Werneburg, Morimoto, et al. 2014). Only species with well-resolved developmental series (i.e. ≥ 4 ranks) were included in further analyses.

To compare the prenatal ossification sequences of bats with both non-volant mammals and flying vertebrates, we controlled for anatomical differences between birds and mammals, standardising a unique anatomical nomenclature of homologous structures. Consequently, the avian furcula was matched with the mammalian clavicle.

Given the complex structure of the avian synsacrum, the ossification of the avian sacral vertebrae was paired with the lumbar and sacral mammalian vertebrae. Finally, in order to visualise developmental differences across groups, we performed a Principal

Component Analysis (PCA) of the ossification sequences of all 39 species pooled as bats, birds and non-volant mammals. All analyses were performed with PAST 3.18

(Hammer, et al. 2001).

55

Metric growth

3D virtual reconstructions of the postcranial skeleton of all 66 specimens were used to collect a total of 25 linear measurements (Table 2.2). Measurements were obtained from 3D renderings of the virtual models using the measure distance tool in MIMICS v.

20 software (Materialise NV). Measurements were logarithmically transformed prior to analysis. Missing values were replaced using linear interpolation with the na.approx function of the R package zoo version 1.8 (Zeileis, et al. 2018). Mean values of each measurement were used to create a single developmental growth sequence for each species.

For bone categories representing more than a single bone (e.g. ribs and vertebrae) we always measured the same bone, usually being the one first to ossify within the category (see Table 2.2 for specifications). To control possible bilateral asymmetry due to developmental instability in our data, we measured bones only on the left side for all specimens. Since all analyses were performed at ordinal level, all species were included in further analysis, including R. pearsoni and R. pusillus that were represented only by one individual each.

56

Table 2.2. Description of 25 linear measurements of the postcranial skeleton of bat fetuses used in this study. Numbers indicate the modularity hypotheses where each bone was included (see table 3 for number coding).

Measurement Acronym Description Module Crown-Rump Length CRL Length from the top of the head to bottom of torso -

Humeral Length HL Left humerus length 3, 4, 6, 8 Clavicular length CL Left clavicle length 2, 6, 8 Scapular length SL Left scapula length 2, 6, 8 Femoral length FL Left femur length 3, 7, 8 Rib length RiL First left rib length 2, 9 Tibial length TL Left tibia length 3, 7, 8 Fibular length FiL Left fibula length 3, 7, 8 Radial length RL Left radius length 3, 4, 6, 8 Ulnar length UL Left ulna length 3, 4, 6, 8 Sternum length StL Sternum length 2, 8 Manual phalange length MPL Proximal phalange of second digit of left forelimb 3, 5, 6, 8 Pedal phalange length PPL Proximal phalange of first digit of left hindlimb 3, 7, 8 Metacarpal length McL Metacarpus of second digit of left forelimb 3, 5, 6, 8 Metatarsal length MtL Metatarsal of first digit of left hindlimb 3, 7, 8 Tarsal length TaL Length of calcaneus of left hindlimb 3, 7, 8 Carpal length CaL Length of scaphoid of left forelimb 3, 5, 6, 8 Cervical vertebral width CvL C1 vertebral width 1, 2, 9 Thoracic vertebral width TvL T1 vertebral width 1, 2, 9 Lumbar vertebral width LvL L1 vertebral width 1, 2, 9 Sacral vertebral width SvL S1 vertebral width 1, 2, 9 Caudal vertebral width CavL vertebral width of first caudal vertebra 1, 2, 9 Length from top of iliac crest to the edge of Ilium length IlL triradiate joint 2, 7, 8 Length from bottom of ramus to edge of triradiate Ischium length IsL joint 2, 7, 8 Length from bottom of ramus to edge of triradiate Pubis bone length PuL joint 2, 7, 8

Developmental modularity

To assess the presence of correlation between prenatal ossification ranks we conducted a neighbour-joining clustering analysis based on the scaled ossification ranks of each bone, bootstrapping each node with 10,000 permutations (Koyabu,

Werneburg and Morimoto 2014). Kendall’s  was used to test modularity in prenatal ossification ranks for bats, birds and non-volant mammals separately, performing pairwise comparisons between all 24 bones based on their relative ossification ranks.

57

Following previous studies (Goswami, et al. 2014b), seven different functional and anatomical modularity hypotheses were tested: 1) vertebral column, 2) axial skeleton,

3) appendicular skeleton, 4) armwing, 5) handwing, 6) forelimb and 7) hindlimb.

Additionally, to explore the ontogenetic basis of modularity, we developed two new hypotheses based on the embryological origin of the bone: 8) abaxial skeleton or lateral plate mesoderm (LPM)-derived, and 9) primaxial skeleton or somite-derived

(see Table 2.2 for specifications, (Burke and Nowicki 2003; Buchholtz and Stepien

2009; Hautier, et al. 2010)). All hypotheses were evaluated for Chiroptera, non-volant mammals and birds.

Finally, to visualise the similarities between bone growth patterns informed by the modules obtained in the two previous analyses, we performed a PCA grouping all 24 bones into the best supported modularity model. All analyses were performed with

PAST 3.18 (Hammer, et al. 2001).

Developmental disparity and integration

Morphological integration was interpreted as the relative eigenvalue standard deviation (i.e. eigenvalue dispersion) of a PCA of all 24 bones for all 66 bat specimens.

With this approach, high levels of dispersion mean that few eigenvectors account for a high proportion of variance, indicating strong integration (Pavlicev, et al. 2009).

Morphological disparity was measured as the statistical variance of the linear measurements of each bone. To test differences in disparity and integration between hypothetical developmental modules we averaged the values obtained for all bones within each module. Finally, to describe temporal shifts in disparity and integration across the prenatal development of Chiroptera, we staged all bat specimens based on

58 criteria previously published for different bat species (Cretekos, et al. 2005; Wyant and

Adams 2008; Hockman, et al. 2009; Wang, et al. 2010; Tokita, et al. 2012). Since CS staging systems have not been developed for nine of the 11 species in the prenatal sample, specimens were classified into one of 10 developmental stages (1-10), based on the development of external features, bone ossification sequence and crown to rump length (CRL, Fig. 2.1). Linear Discriminant Function Analysis (LDA) was performed to statistically assess the accuracy of our staging system and visualise the spatial separation between the stages (Appendix Figure 2). By combining both datasets we circumvent the issue of establishing discrete categories along a continuous morphometric dimension, informing our staging system with a discrete component of development. Generalised Linear Model (GLM) was used to test the correlation between integration and disparity across bones and developmental time and PCA to visualise the distance between species in the developmental space in PAST 3.18

(Hammer, et al. 2001).

The effect of uneven bat species representation across developmental stages on values of disparity and integration was assessed with Pearson’s correlations. Also,

Pearson’s correlations were used to test whether the number of specimens per developmental stage influences values of disparity and integration. Uncertainty in the values of disparity and integration was assessed by estimating 95% confidence intervals with 10,000 bootstrap replicates using the boot package version 1.3 in R

(Canty and Ripley 2017).

59

Results

Sequence heterochrony

General patterns of sequence heterochrony were evident across the chiropteran skeleton. Bones of the stylopod (proximal section) and zeugopod of both limbs, and the pectoral girdle are the first ones to ossify, followed by bones of the autopod (distal section) of both limbs and all but the caudal vertebrae of the spinal column (Fig. 2.2).

Carpals, caudal vertebrae, ischium and sternum are the last bones to ossify.

Metatarsals and pedal phalanges showed the highest variance in their relative ranking, whereas the clavicle was invariably the first bone to ossify. Compared to the other groups, postcranial ossification sequence in bats have similarities with both non-volant mammals and birds (Fig. 2.2). Vertebral column ossification in bats is delayed compared to non-volant mammals and birds, showing a distinctive developmental process in this region unique to bats, a pattern also found for the ischium. In contrast, bats and non-volant mammals show an early ossification of the stylopod and zeugopod of both limbs, compared to birds. Bats and birds shared a delay in the ossification of the sternum, compared to non-volant mammals. All groups had a late ossification of the carpals.

60

Figure 2.2. Relative timing of onset of ossification (ranks) of 24 postcranial bones in Aves (red), Chiroptera (green), and non-volant mammals (blue). Standardised ranking of bone ossification onset ranges from 0 (first to start ossification) to 1 (last to start ossification).

PCA analysis revealed a clear differentiation between the prenatal ossification of the postcranium of mammals and birds, with bats and non-volant mammals occupying a single developmental space (Fig. 2.3). PC1 and PC2 combined explained 67.33% of the variation. Chiroptera’s development showed the highest variability in our sample, occupying the largest space across PC1 and PC2. Negative values along PC1 (49.47%) indicate delayed ossification of the caudal vertebrae, manual phalanges, pubis and ischium, whereas positive values indicate early ossification of clavicle, humerus and radius. PC2 (17.86%) is associated with early ossification of the presacral vertebrae.

61

Figure 2.3. Principal Component Analysis of the ossification sequences of 24 postcranial bones (relative ranks from 0 to 1) in 39 vertebrate species analysed in this study. Species are plotted across PC1 (49.47%) and PC2 (17.86%), and are grouped as Aves (red), Chiroptera (green), and non-volant mammals (blue).

Developmental growth

In contrast to our results of sequence heterochrony in early-ossifying bones, the last bones to ossify were the ones with the shortest lengths across bat species (Appendix

Table 1). Such is the case for the sternum and the carpals, where for most species the adjusted ranks were frequently zero. Overall, CRL had the highest variance of all measurements, followed by UL and RL (see Table 2.2 for abbreviations). CaL and CavL showed the lowest variance, followed by TaL and StL. Regarding the limbs, measurements of the zeugopod showed the highest variance, whereas measurements of the autopod showed the lowest. With the exception of the cervical vertebrae, most of the sections of the spinal column had similar magnitudes of variance.

62

Developmental modularity

Permutated neighbour-joining cluster analysis revealed two well-defined clusters of bones in the postcranium, one comprising the stylopod and zeugopod of both limbs as well as both girdles and the ribcage, and the other comprising the autopod of both limbs and most of the spinal column (Fig. 2.4).

Figure 2.4. Neighbour-joining clustering analysis of ossification sequence of 24 postcranial bones (relative ranks from 0 to 1) in bats. Colours represent the two best-supported groups. Numbers represent level of support at each node after 10,000 randomisations.

Of all nine modularity hypotheses tested with Kendall’s  across all birds, bats and non- volant mammals, six were statistically significant before Bonferroni correction (Table

2.3). Axial and appendicular modules were significant for all taxonomic groups, whereas the forelimb and hindlimb functional modules were significant only for

63

Chiroptera and non-volant mammals. After Bonferroni correction, however, the appendicular and axial hypotheses were both significant for Chiroptera and non-volant mammals, whereas only the appendicular was significant for birds. The abaxial hypothesis was significant for both mammal groups only.

Table 2.3. Kendall’s tau results testing nine different modularity hypotheses of metric growth. P values shown in parentheses. Values in bold were statistically significant before Bonferroni correction, and asterisks indicate significance after correction (P<0.05/29=0.0017).

Chiroptera Aves Non-volant mammals Vertebral column (1) 0.2 (0.8065) -0.111 (1) 0.738 (0.129) Axial skeleton (2) 0.879 (<0.0001)* 0.576 (0.1662) 0.806 (<0.0001)* Appendicular skeleton (3) 0.818 (0.0002)* 0.795 (<0.0001)* 0.907 (<0.0001)* Armwing (4) 1 (1) 0.816 (1) 1 (1) Handwing (5) 1 (1) 0.333 (1) 1 (1) Forelimb functional (6) 1 (0.0085) 0.745 (0.0691) 1 (0.0085) Hindlimb functional (7) 0.714 (0.0187) 0.837 (0.0061) 0.837 (0.0061) Abaxial skeleton (8) 0.856 (<0.0001)* 0.495 (<0.0053) 0.775 (<0.0001)* Primaxial skeleton (9) 0.467 (0.2596) 0.286 (0.5593) 0.828 (0.0354) PCA results grouping linear measures of all 24 bones based on this axial-appendicular model showed a clear distinction between the two modules, with only four variables showing an overlap between both modules (Appendix Figure 3).

Disparity and integration

Values of integration ranged from 0.011 to 0.458 across the postcranium (mean

0.088). Seven bones showed integration values higher than average, with metacarpals and pedal phalanges with the highest values (0.458 and 0.338 respectively; Fig. 2.5).

Manual phalanges, tibia and fibula showed the lowest values (< 0.02; Fig. 2.5).

Disparity in bats ranged from 0.002 to 0.138 (mean 0.061), with the highest values found in the radius and fibula, and the lowest values in the carpals and the thoracic vertebrae. The appendicular module showed higher values of integration and disparity than the axial module, both higher than average (0.175 for disparity and 0.11 for

64 integration), whereas both integration and disparity in the axial module were below average (0.075 and 0.07) (Fig. 2.6). Disparity was significantly different between modules, whereas integration was not (ANOVA, disparity: P= <0.001, F(1,23)= 15.49; integration: 0.333, F(1,23)= 0.978). GLM did not reveal a significant association between integration and disparity across bones (P= 0.832, R² = 0.002; Appendix Figure 4). Our results of disparity and integration across developmental time show that both disparity and integration increase across time, but neither follows a clear temporal pattern (Fig.

2.7).

Figure 2.5. Developmental integration (eigenvalue dispersion) and disparity (bone size variance) across 24 postcranial bones in bats. Dotted and dashed lines mark average integration and disparity respectively.

LDA showed that the staging system implemented for the bat specimens successfully separated developmental stages (93.94% of specimens correctly classified, Appendix

Figure 2). Stages were characterised by a CRL range and ossification events specific to each stage (Appendix Table 2). Disparity across developmental stages had a tendency to increase but neither linearly nor exponentially (R2= 0.31 and 0.36). Integration, on the other hand, reached its highest values in stages 6 and 10, but did not show a clear pattern of increase that fitted either a linear model or exponential curve (R2= 0.26 and

65

0.06). Pearson’s correlations showed non-significant relationships between disparity, integration and the number of specimens per stage (disparity: ρ= 0.103, P= 0.775; integration: ρ= -0.19, P= 0.584; Appendix Figure 5). Pearson’s correlation also showed that differences in species sampling across stages had no effect in values of disparity and integration (disparity: ρ= 0.006, P= 0.986; integration: ρ= -0.228, P= 0.525). GLM showed that integration and disparity are not correlated across developmental time

(P= 0.199, R² = 0.196; Appendix Figure 6). Bootstrapping revealed that all disparity and integration values (both across bones and developmental stages) fall within the 95% confidence intervals, indicating statistical significance (Fig. 2.8).

Figure 2.6. Average developmental integration (eigenvalue dispersion) and disparity (bone size variance) of elements of the appendicular and axial developmental modules in bats. Dotted and dashed lines mark intermodule mean integration and disparity respectively.

66

Discussion

Sequence heterochrony and developmental growth

The ossification of the clavicle is consistently the first event across birds and mammals, a pattern well-known for vertebrates (Sánchez-Villagra 2002). Compared with birds, mammals (both bats and non-volant mammals) show a delayed ossification of the stylopod and zeugopod of both limbs, indicating that stylopod and zeugopod development in bats resembles the general pattern found in Mammalia (Koyabu and

Son 2014). Patterns in our results support heterochrony between the forelimb and hindlimb in bats and birds, compared to non-volant mammals (Fig. 2.2). Similarities between bats and birds in ossification sequences of pedal phalanges could indicate a developmental response to locomotory needs of newborns (Maxwell, et al. 2010;

Koyabu and Son 2014). Despite behavioural differences in the neonates, and because newborn bats and birds are incapable of self-powered flight, locomotion and roosting depend mostly on hindlimb functional performance (Maxwell, et al. 2010; Koyabu and

Son 2014). Based on recent palaeontological findings suggesting newborn pterosaurs were incapable of flight (Wang, Kellner, et al. 2017), we suggest that hindlimb heterochrony could be a trait shared across flying vertebrates (Maxwell, et al. 2010;

Koyabu and Son 2014). Our results support the hypothesis that accelerated development of the foot in bats correlates with the roosting ecology of newborn pups, which cannot fly and need to remain attached to either the mother or the roosting site at all times, until they achieve self-powered flight (Koyabu and Son 2014). In contrast, delayed development of the carpals could be related to the functional importance of the wrist for the folding of the wing during the upstroke, a key kinematic process

67 during flight (Riskin, et al. 2012). Bats also show accelerated ossification of the jaw for suckling and attachment to the mothers’ nipples during flight (Koyabu and Son 2014).

Figure 2.7. Developmental integration (eigenvalue dispersion) and disparity (bone size variance) across prenatal stages for bats examined in this study. Values at each stage are an average across all bones.

Marked heterochrony in the ossification of the sternum of bats and birds could indicate an ontogenetic basis for flight altriciality in flying vertebrates. The presence of a sternal ridge or in the sternum is a crucial adaptation for flight in vertebrates, as it provides additional surface for the attachment of muscles involved in wingbeat movements (Norberg 1990). Delayed ossification and longer developmental times have recently been proposed to be associated with evolvability and adaptability, although thus far only within the context of domestication (Werneburg and Geiger

2017). Our findings of delayed sternum ossification along with its importance for locomotion across flying vertebrates (Norberg and Rayner 1987; Norberg 1990), may also be fitting with that hypothesis. Finally, unsynchronised ossification of the pelvic bones reflects the different chondrogenous centres that lead to the formation of the hip as a single structure (Ruth 1932), a pattern previously seen in other mammalian

68 groups but not in birds (Maxwell, et al. 2010; Wilson, et al. 2010; Koyabu and Son

2014).

PCA of ossification sequences reveals a clear separation between mammals and birds along the first principal component (PC1), showing that phylogenetic relationships shaped our results (Fig. 2.3). Based on the apparent mammal (negative loadings) to bird (positive loadings) polarity across PC1, it can be argued that bones with positive loadings exemplify bird development, whereas bones with negative values exemplify mammal development. All sections of the vertebral column had negative loadings, whereas the stylopod and zeugopod of both limbs had positive loadings, indicating that axial skeleton development differentiates mammal development, whereas appendicular skeleton development differentiates bird development (Appendix Table

3).

69

Figure 2.8. 95% confidence intervals of values of disparity (A,C) and integration (B,D) across bat developmental stages (A,B) and bones (C,D) based on 10,000 bootstrap replicates.

70

Similarities between birds and bats in postcranial development suggest an ontogenetic basis to the convergent evolution of vertebrate flight. Our results show some similarities with recent palaeontological findings that informed the prenatal development of pterosaurs (Wang, Kellner, et al. 2017). It is hypothetised that newborn pterosaurs were incapable of self-powered flight due to an altricial wing, a trait shared with bats and birds (Maxwell, et al. 2010; Koyabu and Son 2014; Wang,

Kellner, et al. 2017). Moreover, based on our results, we can suggest a shared ontogenetic basis to flight altriciality, as a general delay in the ossification of the wing and sternum compared with the hindlimb is found across all flying vertebrate groups

(Maxwell, et al. 2010; Koyabu and Son 2014; Wang, Kellner, et al. 2017). Relative timing of morphogenesis and incomplete prenatal development could act as an evolutionary promoter of adaptability and evolvability in the forelimb of flying vertebrates, facilitating flight adaptations to evolve (Werneburg and Geiger 2017).

Previously, prenatal developmental timing has been associated with evolvability in cranial shape of carnivorans, also showing an ontogenetic source for variability associated with domestication in canids (Werneburg and Geiger 2017; Machado, et al.

2018; Wilson 2018).

Our results also showed that the development of the vertebral column of bats deviates from both the non-volant mammals and birds. In comparison to other mammals, unique morphological features found in the vertebral column of bats have been linked to the roosting and feeding ecology of the species (Gaudioso, et al. 2017). Our findings of sequence heterochrony in the vertebral column indicate that the morphological patterns found in this region in previous studies could have an ontogenetic basis,

71 originating during the prenatal development of the skeleton (Wilson and Sánchez-

Villagra 2010b; Gaudioso, et al. 2017).

Evolutionary changes in bone size have been correlated with heterochronic development of those structures (Sánchez-Villagra, et al. 2008). Our result that bones that grow to be relatively small in the adult have a late ossification onset does not reflect this phenomenon (Fig. 2.5). Instead, our results could reflect that at the stage sampled much of the growth for the elements examined here was yet to be completed

(i.e. postnatally), leading to a narrower depiction of its total growth path (Werneburg, et al. 2016). Future studies could combine pre- and postnatal developmental data to trace the complete development of the hip.

Developmental modularity

Our analysis did not reject the hypothesis that postcranial developmental modularity is present in bats, birds and non-volant mammals (Fig. 2.3-2.4, Table 2.3). The presence of an axial-appendicular partition in both mammal groups, not found in birds, could indicate that flight did not repattern ontogenetic modularity in flying vertebrates, as bats and non-volant mammals shared the same pattern for postcranial modularity.

Furthermore, our results supported the presence of an appendicular skeleton module, previously unknown in bats (Goswami, et al. 2009), and an axial module previously found in placental prenatal development (Goswami, et al. 2009). We suggest a developmental module hypothesis, where the shared LPM-derived origin of all bones of the appendicular skeleton develop as a unit (Wallin, et al. 1994).

Modularity in the appendicular skeleton of mammals has been reported to respond to functional pressures and selection on short times scales, as in domestication (Hanot, et

72 al. 2017, 2018). Presence of developmental modularity in mammals has revealed a clear ontogenetic division between marsupials and placentals, each group presenting different patterns of postcranial modularity (Goswami, et al. 2009). In particular, placental mammals showed strong evidence of an appendicular module and a module including both girdles (Goswami, et al. 2009). However, previous studies suggested that bats are an exception to this trend, showing low covariation between hind- and forelimb both during their development and adulthood (Young and HallgrÍmsson 2005;

Weisbecker, et al. 2008; Goswami, et al. 2009; Bell, et al. 2011), in contrast to our results. Each of these studies focused on modularity in mammals as a whole, including only one bat species in their sample. The contrasting results could be due to differences in sample composition, with low bat species representation obscuring patterns only discernible when analysing multiple species representing different lineages. With 14 bat species, our study represents the most comprehensive quantitative study of prenatal developmental biology in chiropterans to date. Also, our results of developmental modularity were informed by two different datasets that combined sequence heterochrony and metric growth data, contrary to previous studies that focused only on one.

Our two sets of results could indicate that the functional differences between the forelimb and hindlimb in adult bats are not reflected during prenatal development

(Adams 2008; Eghbali, et al. 2017). Moreover, integrated development of homologous structures could facilitate morphological diversity, enabling novel functional ecologies to evolve (Klingenberg 2009, 2010; Goswami, et al. 2014b; Klingenberg 2014). The latter may indicate that the correlated development of both limbs in bats could

73 facilitate morphological disparity in adult forms. In bats, neonate hindlimb size is similar to its adult size whereas neonate forelimb size is about one third of adult size

(Koyabu and Son 2014). Since the forelimb is not developed sufficiently for flight at birth and requires extended postnatal time to be large enough to be fully functional, it was suggested that prenatal bats invest in earlier development of the hindlimb

(Koyabu and Son 2014). Given the present results, we hypothesise that modular development in the appendicular skeleton could represent a trade-off between the accelerated development of the foot (and functionality in newborns), and the delayed development of the wing (and its functionality in later development; (Riskin, et al.

2012)). Delayed prenatal development has been linked to higher adaptability and evolvability to environmental pressures in domesticated carnivorans (Werneburg and

Geiger 2017).

Disparity and integration

High levels of disparity in the zeugopod of both limbs document the developmental basis of its functional variability (Fig. 2.4). Compared to the ancestral mammalian condition, zeugopod reduction is a convergent trait common across multiple clades

(Sears, et al. 2007). This reduction arguably enabled locomotory specialisations that led to the adaptive radiation of mammals (Sears, et al. 2007). Diminished growth rate of the zeugopod – in particular the ulna and fibula – is a developmental process that has been found in several different mammal groups, arguing for a shared ontogenetic mechanism for zeugopod reduction (Sears, et al. 2007). Our results indicate that zeugopod development is highly variable across bat species, possibly reflecting functional differences associated with ecological traits of the species (e.g. flying

74 behaviours, locomotory specialisations, and body sizes) (Giannini, et al. 2004; Bininda-

Emonds, Jeffery, et al. 2007; Wilson and Sánchez-Villagra 2010a). Forearm length is widely used to inform field identification of bat species, providing a good indicator of the interspecific morphological delimitation in many taxa (Jarrin, et al. 2010).

Additionally, relative forearm length is also a good estimate of several biomechanical properties of the wing that reflect ecological differences across species (Swartz and

Middleton 2008). We hypothesise that high prenatal disparity of the zeugopod could be indicative of ecomorphological diversity in adult bats.

Our PCA of species could reflect functional differences by showing that C. sphinx (the only plant-eater and non-echolocating species in our sample for this analysis) was the most dissimilar (Appendix Figure 7 and Appendix Table 4). Plant-visiting bats exhibit foraging behaviours uncommon in animalivorous species (e.g. prolonged hovering flight) that could represent novel functional needs for the postcranium. However, the

PCA also shows that the subordinal spaces did not overlap, raising the question of whether our results evidence a phylogenetic signal, rather than functional.

Descriptions of the broad foraging guilds for these bat families suggest that all animalivorous species in our sample are aerial hawkers (Jones, et al. 2016), indicating a phylogenetic signal in our results. Nevertheless, the ecology of some of the species is either poorly known (e.g. Aselliscus) or highly adaptable (e.g. Kerivoula and Myotis species are known to switch between hawking and gleaning) (Faure and Barclay 1994;

Schulz 2000). Expanding the sample to include frugivorous yangochiropteran species

(i.e. phyllostomids) could yield further insights.

75

Phylogenetic studies have reconstructed clade-specific evolutionary trajectories of joint and muscle reduction in the wings of bats, closely associated with flying behaviours in response to feeding strategies (Bahlman, et al. 2016). Such correspondence between ecological diversity and developmental differences has been previously reported in other mammals (Wilson and Sánchez-Villagra 2010b). Finally, low levels of integration in the zeugopod and a generalised increase in disparity indicate that integration could promote disparity during prenatal development

(Goswami, et al. 2014a).

Differences in disparity between modules also illustrate the ontogenetic basis of the phenotypic diversity and functional variability of the limbs in bats. Prenatal development has been suggested to respond to the functional needs of the neonates, rather than the ecological niche of the species (Koyabu and Son 2014). As such, lower than average disparity in the axial module suggests that this section of the skeleton does not experience high evolutionary pressure to diversify prenatally (Giannini, et al.

2006; Adams 2008; Cooper and Tabin 2008; Koyabu and Son 2014).

Our results provide strong evidence for increase in disparity over ontogeny (Fig. 2.7), a pattern rarely reported in mammals (Goswami, et al. 2014a; Zelditch, et al. 2016).

Moreover, ours represents the first report of this pattern both in bats and in prenatal development. Previous studies have shown that integration canalises phenotypic variation in deep-time (Marroig, et al. 2003; Porto, et al. 2008; Marroig, et al. 2009;

Goswami, et al. 2014a; Goswami, et al. 2015), and that changes in size can work as evolutionary buffers for adaptive radiation (Marroig and Cheverud 2005a). Differences between morphological over deep time and our results across

76 developmental time support the hypothesis that integration facilitates disparity from an ontogenetic perspective (Goswami, et al. 2014a; Goswami, et al. 2015). This congruence in temporal patterns differs from the mismatch in patterns of integration and disparity found across bones and modules, demonstrating that temporality rather than functionality shapes the interaction between developmental disparity and integration (Goswami, et al. 2014a; Randau and Goswami 2017).

Conclusions

Combining data of sequence heterochrony and developmental growth from 14 bat species, this study represents the most comprehensive quantitative analysis of chiropteran prenatal ossification to date. Sequence heterochrony between the autopod of fore- and hindlimbs could reflect functional needs of the newborn, rather than ecological aspects of the adult. Sequence heterochrony also showed that bats have similarities with birds in the ossification of structures involved in flight (i.e. handwing and sternum), suggesting that flight altriciality and early ossification of pedal phalanges and sternum are common across flying vertebrates. Developmental modularity was detected both in ossification sequence and metric growth of the postcranium, suggesting an axial-appendicular partition of the postcranium that deviates from the general pattern reported for mammals. This partition possibly corresponds to genetic and ontogenetic aspects of the development of the postcranium (e.g. the LPM-derived nature of all appendicular bones), rather than ecomorphological characteristics of bats. The marked difference in values of disparity between modules reflects the phenotypic diversity of the appendicular skeleton in response to interspecific functional differences of the wing. Our results reject the

77 hypothesis that morphological variation in the fore- and hindlimbs of bats is dissociated. Integration and disparity increased across prenatal stages, supporting the hypothesis that integration facilitates disparity. Moreover, we found that this interaction is only evident from a temporal perspective (i.e. across developmental time) rather than from a morphofunctional one (i.e. between functional modules of the skeleton). Finally, our results show not only an increase in disparity across developmental time, but also a delayed ossification in highly adaptable and evolvable regions, both patterns rarely reported in wild mammals.

References

Adams DC, Nistri A. 2010. Ontogenetic convergence and evolution of foot morphology in European cave salamanders (Family: Plethodontidae). BMC Evolutionary Biology 10. Adams RA. 1992. Comparative Skeletogenesis of the Forearm of the Little Brown Bat (Myotis-Lucifugus) and the Norway Rat (Rattus-Norvegicus). Journal of Morphology 214:251-260. Adams RA. 2008. Morphogenesis in Bat Wings: Linking Development, Evolution and Ecology. Cells Tissues Organs 187:13-23. Adams RA. 2000. Wing ontogeny, shifting niche dimensions, and adaptive landscapes. In: Adams RA, Pedersen SC, editors. Ontogeny, Functional Ecology and Evolution of Bats. Cambridge: Cambridge University Press. p. 275-316. Adams RA, Shaw JB. 2013. Time’s Arrow in the Evolutionary Development of Bat Flight. Bat evolution, ecology and conservation: 21-46. Aldridge HDJN. 1987. Turning Flight of Bats. Journal of Experimental Biology 128:419- 425. Amador LI, Arevalo RLM, Almeida FC, Catalano SA, Giannini NP. 2018. Bat Systematics in the Light of Unconstrained Analyses of a Comprehensive Molecular Supermatrix. Journal of Mammalian Evolution 25:37-70. Arita HT, Fenton MB. 1997. Flight and echlocation in the ecology and evolution of bats. Trends in Ecology & Evolution 12:53-58. Arita HT, Vargas-Barón J, Villalobos F. 2014. Latitudinal gradients of genus richness and endemism and the diversification of New World bats. Ecography 37: 1024-1033. Bahlman JW, Price‐Waldman RM, Lippe HW, Breuer KS, Swartz SM. 2016. Simplifying a wing: diversity and functional consequences of digital joint reduction in bat wings. Journal of Anatomy 229:114-127. Bell E, Andres B, Goswami A. 2011. Integration and dissociation of limb elements in flying vertebrates: a comparison of pterosaurs, birds and bats. J Evol Biol 24:2586- 2599.

78

Bergou AJ, Swartz SM, Vejdani H, Riskin DK, Reimnitz L, Taubin G, Breuer KS. 2015. Falling with Style: Bats Perform Complex Aerial Rotations by Adjusting Wing Inertia. PLOS Biology 13: e1002297. Bininda-Emonds OR, Jeffery JE, Sánchez-Villagra MR, Hanken J, Colbert M, Pieau C, Selwood L, Ten Cate C, Raynaud A, Osabutey CK, et al. 2007. Forelimb-hindlimb developmental timing changes across tetrapod phylogeny. BMC Evol Biol 7:182. Boonman A, Bumrungsri S, Yovel Y. 2014. Nonecholocating fruit bats produce biosonar clicks with their wings. Current biology : CB 24:2962-2967. Buchholtz EA, Stepien CC. 2009. Anatomical transformation in mammals: developmental origin of aberrant cervical anatomy in tree sloths. Evolution & Development 11:69-79. Burke AC, Nowicki JL. 2003. A new view of patterning domains in the vertebrate mesoderm. Developmental Cell 4:159-165. Butler RJ, Benson RBJ, Barrett PM. 2013. Pterosaur diversity: Untangling the influence of sampling biases, Lagerstätten, and genuine biodiversity signals. Palaeogeography, Palaeoclimatology, Palaeoecology 372:78-87. Canals M, Iriarte-Diaz J, Grossi B. 2011. Biomechanical, Respiratory and Cardiovascular Adaptations of Bats and the Case of the Small Community of Bats in Chile. In: Klika V, editor. Biomechanics in Applications: InTech. p. 299-322. Canty A, Ripley BS. 2017. boot: Bootstrap R (S-Plus) Functions. Version 1.3-20. Cheney JA, Ton D, Konow N, Riskin DK, Breuer KS, Swartz SM. 2014. Hindlimb motion during steady flight of the lesser dog-faced fruit bat, Cynopterus brachyotis. PLoS ONE 9:e98093. Cooper KL, Tabin CJ. 2008. Understanding of bat wing evolution takes flight. Genes Dev 22:121-124. Cooper L, Cretekos CJ, Sears KE. 2012. The evolution and development of mammalian flight. Wiley Interdisciplinary Reviews: Developmental Biology 1:773-779. Cretekos CJ, Wang Y, Green ED, Martin JF, Rasweiler JJt, Behringer RR. 2008. Regulatory divergence modifies limb length between mammals. Genes Dev 22:141- 151. Cretekos CJ, Weatherbee SD, Chen CH, Badwaik NK, Niswander L, Behringer RR, Rasweiler JJt. 2005. Embryonic staging system for the short-tailed fruit bat, Carollia perspicillata, a model organism for the mammalian order Chiroptera, based upon timed pregnancies in captive-bred animals. Dev Dyn 233:721-738. Cubo J, Casinos A. 1998. The variation of the cross-sectional shape in the long bones of birds and mammals. Annales des Sciences Naturelles - Zoologie et Biologie Animale 19:51-62. Currey JD, Alexander MR. 1985. The thickness of the walls of tubular bones. Journal of Zoology 206:453-468. de Camargo NF, de Oliveira HFM. 2012. Sexual Dimorphism in Sturnira lilium (Chiroptera, Phyllostomidae): Can Pregnancy and Pup Carrying Be Responsible for Differences in Wing Shape? PLoS ONE 7: e49734. Dececchi TA, Larsson HC, Habib MB. 2016. The wings before the bird: an evaluation of flapping-based locomotory hypotheses in bird antecedents. PeerJ 4:e2159. Denzinger A, Schnitzler HU. 2013. Bat guilds, a concept to classify the highly diverse foraging and echolocation behaviors of microchiropteran bats. Front Physiol 4:164. Dickinson M. 2008. : a new spin on bat flight. Curr Biol 18:R468-470. 79

Dietz CL. 1973. Bat walking behavior. Journal of mammalogy 54: 790-792. Dumont ER. 2010. Bone density and the lightweight skeletons of birds. Proceedings of the Royal Society B: Biological Sciences 277:2193-2198. Eghbali H, Shahabi S, Najafi N, Mehdizadeh R, Yousefi S, Sharifi M. 2017. Postnatal growth, wing development and age estimations in the Mediterranean horseshoe bat Rhinolophus euryale (Chiroptera: Rhinolophidae) in Kerend cave, western Iran. Mammalia 82: 276-287. Eiting TP, Gunnell GF. 2009. Global Completeness of the Bat Fossil Record. Journal of Mammalian Evolution 16:151-173. Ellington CP, vandenBerg C, Willmott AP, Thomas ALR. 1996. Leading-edge vortices in insect flight. Nature 384:626-630. Esquerré D, Sherratt E, Keogh JS. 2017. Evolution of extreme ontogenetic allometric diversity and heterochrony in pythons, a clade of giant and dwarf snakes. Evolution 71:2829-2844. Faure PA, Barclay RMR. 1994. Substrate-Gleaning Versus Aerial-Hawking - Plasticity in the Foraging and Echolocation Behavior of the Long-Eared Bat, Myotis-Evotis. Journal of Comparative Physiology a-Sensory Neural and Behavioral Physiology 174:651-660. Fenton MB. 2010. Convergences in the diversification of bats. Current Zoology (formerly Acta Zoologica Sinica) 56: 454-468. Fong JJ, Brown JM, Fujita MK, Boussau B. 2012. A Phylogenomic Approach to Vertebrate Phylogeny Supports a Turtle-Archosaur Affinity and a Possible Paraphyletic Lissamphibia. PLoS ONE 7. Frelat MA, Katina S, Weber GW, Bookstein FL. 2012. Technical note: A novel geometric morphometric approach to the study of long bone shape variation. American Journal of Physical Anthropology 149:628-638. Gardiner JD, Codd JR, Nudds RL. 2011. An association between ear and tail morphologies of bats and their foraging style. Canadian Journal of Zoology 89:90-99. Gaudioso PJ, Diaz MM, Barquez RM. 2017. Morphology of the axial skeleton of seven bat genera (Chiroptera: Phyllostomidae). An Acad Bras Cienc 89:2341-2358. Gerber S. 2013. On the Relationship between the Macroevolutionary Trajectories of Morphological Integration and Morphological Disparity. PLoS ONE 8. Gerber S, Eble GJ, Neige P. 2008. Allometric space and allometric disparity: a developmental perspective in the macroevolutionary analysis of morphological disparity. Evolution 62:1450-1457. Giannini N, Goswami A, Sánchez-Villagra MR. 2006. Development of integumentary structures in Rousettus amplexicaudatus (Mammalia: Chiroptera: Pteropodidae) during late-embryonic and fetal stages. Journal of mammalogy 95:993-1001. Giannini NP, Abdala F, Flores DA. 2004. Comparative postnatal ontogeny of the skull in Dromiciops gliroides (Marsupialia : Microbiotheriidae). American Museum Novitates:1- 17. Goswami A. 2007. Cranial modularity and sequence heterochrony in mammals. Evol Dev 9:290-298. Goswami A, Binder WJ, Meachen J, O'Keefe FR. 2015. The fossil record of phenotypic integration and modularity: A deep-time perspective on developmental and evolutionary dynamics. Proc Natl Acad Sci U S A 112:4891-4896.

80

Goswami A, Smaers JB, Soligo C, Polly PD. 2014. The macroevolutionary consequences of phenotypic integration: from development to deep time. Philos Trans R Soc Lond B Biol Sci 369:20130254. Goswami A, Weisbecker V, Sánchez-Villagra MR. 2009. Developmental modularity and the marsupial-placental dichotomy. J Exp Zool B Mol Dev Evol 312B:186-195. Hafner M, Remsen Jr J, Lanyon S. 1984. Bird Versus Mammal Morphological Diversity. Evolution 38: 1154-1156. Hakansson J, Hedenstrom A, Winter Y, Johansson LC. 2015. The wake of hovering flight in bats. J R Soc Interface 12:20150357. Hall BK. 2000. Evo-devo or devo-evo--does it matter. Evol Dev 2:177-178. Halley AC. 2016. Prenatal brain-body allometry in mammals. Brain Behavior Evolution 88:14-24. Hammer O, Harper DAT, Ryan PD. 2001. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica 4:9pp. Hanot P, Herrel A, Guintard C, Cornette R. 2018. The impact of artificial selection on morphological integration in the appendicular skeleton of domestic horses. J Anat. Hanot P, Herrel A, Guintard C, Cornette R. 2017. Morphological integration in the appendicular skeleton of two domestic taxa: the horse and donkey. Proc Biol Sci 284. Hautier L, Bennett NC, Viljoen H, Howard L, Milinkovitch MC, Tzika AC, Goswami A, Asher RJ. 2013. Patterns of ossification in southern versus northern placental mammals. Evolution 67:1994-2010. Hautier L, Weisbecker V, Sánchez-Villagra MR, Goswami A, Asher RJ. 2010. Skeletal development in sloths and the evolution of mammalian vertebral patterning. Proceedings of the National Academy of Sciences of the United States of America 107:18903-18908. Hedenstrom A, Johansson LC. 2015. Bat flight: aerodynamics, kinematics and flight morphology. Journal of Experimental Biology 218:653-663. Hedenstrom A, Johansson LC, Spedding GR. 2009. Bird or bat: comparing airframe design and flight performance. Bioinspir Biomim 4:015001. Henderson DM. 2010. Pterosaur body mass estimates from three-dimensional mathematical slicing. Journal of Vertebrate Paleontology 30:768-785. Hendrikse JL, Parsons TE, Hallgrimsson B. 2007. Evolvability as the proper focus of evolutionary developmental biology. Evol Dev 9:393-401. Hockman D, Mason MK, Jacobs DS, Illing N. 2009. The role of early development in mammalian limb diversification: a descriptive comparison of early limb development between the Natal long-fingered bat (Miniopterus natalensis) and the mouse (Mus musculus). Dev Dyn 238:965-979. Iriarte-Diaz J, Riskin DK, Willis DJ, Breuer KS, Swartz SM. 2011. Whole-body kinematics of a fruit bat reveal the influence of wing inertia on body accelerations. J Exp Biol 214:1546-1553. Jarrin VP, Flores C, Salcedo J. 2010. Morphological variation in the short-tailed fruit bat (Carollia) in Ecuador, with comments on the practical and philosophical aspects of boundaries among species. Integr Zool 5:226-240. Johansson LC, Engel S, Kelber A, Heerenbrink MK, Hedenstrom A. 2013. Multiple leading edge vortices of unexpected strength in freely flying hawkmoth. Sci Rep 3:3264.

81

Johansson LC, Hakansson J, Jakobsen L, Hedenstrom A. 2016. Ear-body lift and a novel thrust generating mechanism revealed by the complex wake of brown long-eared bats (Plecotus auritus). Sci Rep 6:24886. Johansson LC, Wolf M, Hedenstrom A. 2010. A quantitative comparison of bird and bat wakes. J R Soc Interface 7:61-66. Jones G, Teeling EC. 2006. The evolution of echolocation in bats. Trends Ecol Evol 21:149-156. Jones KE, Safi K. 2011. Ecology and evolution of mammalian biodiversity Introduction. Philosophical Transactions of the Royal Society B-Biological Sciences 366:2451-2461. Jones PL, Page RA, Ratcliffe JM. 2016. To Scream or to Listen? Prey Detection and Discrimination in Animal-Eating Bats. In: Springer, editor. Bat Bioacoustics. p. 93-116. Kapusta A, Suh A, Freschotte C. 2017. Dynamics of genome size evolution in birds and mammals. Proceedings of the National Academy of Science 114:E1460–E1469. Kirkpatrick S. 1994. Scale Effects on the Stresses and Safety Factors in the Wing Bones of Birds and Bats. Journal of Experimental Biology 190:195-215. Klingenberg CP. 2009. Morphometric integration and modularity in configurations of landmarks: tools for evaluating a priori hypotheses. Evol Dev 11:405-421. Klingenberg CP. 2014. Studying morphological integration and modularity at multiple levels: concepts and analysis. Philos Trans R Soc Lond B Biol Sci 369:20130249. Klingenberg CP. 2010. There's something afoot in the evolution of ontogenies. BMC Evol Biol 10:221. Koyabu D. 2017. Prenatal postcranial development in two species of sympatric Japanese wood mice (Apodemus argenteus and A. speciosus): a comparison of arboreal versus terrestrial congeners. Journal of Veterinary Medical Science 79:952- 956. Koyabu D, Endo H, Mitgutsch C, Suwa G, Catania KC. 2011. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. BMC EvoDevo 2: 21. Koyabu D, Son N. 2014. Patterns of postcranial ossification and sequence heterochrony in bats: Life histories and developmental trade‐offs. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 322:607-618. Koyabu D, Werneburg I, Morimoto N, Zollikofer CPE, Forasiepi A, Endo H, Kimura J, Ohdachi SD, Son NT, Sánchez-Villagra MR. 2014. Mammalian skull heterochrony reveals modular evolution and a link between cranial development and brain size. Nat Commun 5:3625. Lee AH, Simons E. 2015. Wing bone laminarity is not an adaptation for torsional resistance in bats. PeerJ 3: e823. Lentink D, Dickson WB, van Leeuwen JL, Dickinson MH. 2009. Leading-Edge Vortices Elevate Lift of Autorotating Plant Seeds. Science 324:1438-1440. Lindhe Norberg UM, Winter Y. 2006. Wing beat kinematics of a nectar-feeding bat, Glossophaga soricina, flying at different flight speeds and Strouhal numbers. J Exp Biol 209:3887-3897. MacAyeal LC, Riskin DK, Swartz SM, Breuer KS. 2011. Climbing flight performance and load carrying in lesser dog-faced fruit bats (Cynopterus brachyotis). J Exp Biol 214:786- 793.

82

Machado FA, Zahn TMG, Marroig G. 2018. Evolution of morphological integration in the skull of Carnivora (Mammalia): Changes in lead to increased evolutionary potential of facial traits. Evolution. Machado JP, Johnson WE, Gilbert MT, Zhang G, Jarvis ED, O'Brien SJ, Antunes A. 2016. Bone-associated gene evolution and the origin of flight in birds. BMC Genomics 17:371. Marchetti K, Price T, Richman A. 1995. Correlates of Wing Morphology with Foraging Behaviour and Migration Distance in the Genus Phylloscopus. Journal of Avian Biology 26:177-181. Marelli CA, Simons EL. 2014. Microstructure and cross-sectional shape of limb bones in Great Horned Owls and Red-tailed Hawks: how do these features relate to differences in flight and hunting behavior? PLoS ONE 9:e106094. Marinello MM, Bernard E. 2014. Wing morphology of Neotropical bats: a quantitative and qualitative analysis with implications for habitat use. Canadian Journal of Zoology 92:141-147. Marroig G, Cheverud JM. 2005. Size as a line of least evolutionary resistance: Diet and adaptive morphological radiation in New World monkeys. Evolution 59:2005. Marroig G, Shirai LT, Porto A, de Oliveira FB, De Conto V. 2009. The Evolution of Modularity in the Mammalian Skull II: Evolutionary Consequences. Evolutionary Biology 36:136-148. Marroig G, Vivo M, Cheverud JM. 2003. Cranial evolution in sakis (Pithecia, Platyrrhini) II: evolutionary processes and morphological integration. Journal of Evolutionary Biology 17:144-155. Maxwell EE, Harrison LB. 2009. Methods for the analysis of developmental sequence data. Evol Dev 11:109-119. Maxwell EE, Harrison LB, Larsson HCE. 2010. Assessing the phylogenetic utility of sequence heterochrony: evolution of avian ossification sequences as a case study. Zoology 113:57-66. McGowan AJ, Dyke GJ. 2007. A morphospace-based test for competitive exclusion among flying vertebrates: did birds, bats and pterosaurs get in each other's space? J Evol Biol 20:1230-1236. Muijres FT, Christoffer Johansson L, Winter Y, Hedenstrom A. 2014. Leading edge vortices in lesser long-nosed bats occurring at slow but not fast flight speeds. Bioinspir Biomim 9:025006. Muijres FT, Johansson LC, Barfield R, Wolf M, Spedding GR, Hedenstrom A. 2008. Leading-Edge Vortex Improves Lift in Slow-Flying Bats. Science 319: 1250-1253. Muijres FT, Johansson LC, Bowlin MS, Winter Y, Hedenstrom A. 2012. Comparing aerodynamic efficiency in birds and bats suggests better flight performance in birds. PLOS ONE 7:e37335. Muijres FT, Johansson LC, Hedenstrom A. 2012. Leading edge vortex in a slow-flying passerine. Biol Lett 8:554-557. Muijres FT, Johansson LC, Winter Y, Hedenstrom A. 2011. Comparative aerodynamic performance of flapping flight in two bat species using time-resolved wake visualization. J R Soc Interface 8:1418-1428. Nojiri T, Werneburg I, Son NT, Tu VT, Sasaki T, Maekawa Y, Koyabu D. 2018. Prenatal cranial bone development of Thomas's horseshoe bat (Rhinolophus thomasi): with special reference to petrosal morphology. J Morphol 279: 809-827. 83

Norberg RA, Rayner JMV. 1987. Ecological Morphology and Flight in Bats (Mammalia; Chiroptera): Wing Adaptations, Flight Performance, Foraging Strategy and Echolocation. Philos Trans R Soc Lond B Biol Sci 316:335-427. Norberg UM. 1990. Vertebrate Flight, Mechanics, Physiology, Morphology, Ecology and Evolution: Springer-Verlag. Organ CL, Cooper LN, Hieronymus TL. 2015. Macroevolutionary developmental biology: Embryos, fossils, and phylogenies. Dev Dyn 244:1184-1192. Organ CL, Shedlock AM. 2009. Palaeogenomics of pterosaurs and the evolution of small genome size in flying vertebrates. Biol Lett 5:47-50. Oyston JW, Hughes M, Wagner PJ. 2015. What limits the morphological disparity of clades? Interface Focus 5: 20150042. Pavlicev M, Cheverud JM, Wagner GP. 2009. Measuring Morphological Integration Using Eigenvalue Variance. Evolutionary Biology 36:157-170. Petit F, Sears KE, Ahituv N. 2017. Limb development: a paradigm of gene regulation. Nature Reviews Genetics 18: 245-258. Pieretti J, Gehrke AR, Schneider I, Adachi N, Nakamura T, Shubin NH. 2015. Organogenesis in deep time: A problem in genomics, development, and paleontology. Proc Natl Acad Sci U S A 112:4871-4876. Porto A, de Oliveira FB, Shirai LT, De Conto V, Marroig G. 2008. The Evolution of Modularity in the Mammalian Skull I: Morphological Integration Patterns and Magnitudes. Evolutionary Biology 36:118-135. Prentice KC, Ruta M, Benton MJ. 2011. Evolution of morphological disparity in pterosaurs. Journal of Systematic Palaeontology 9:337-353. Ramirez-Chaves HE, Wroe SW, Selwood L, Hinds LA, Leigh C, Koyabu D, Kardjilov N, Weisbecker V. 2016. Mammalian development does not recapitulate suspected key transformations in the evolutionary detachment of the mammalian middle ear. Proc Biol Sci 283. Randau M, Goswami A. 2018. Shape Covariation (or the Lack Thereof) Between Vertebrae and Other Skeletal Traits in Felids: The Whole is Not Always Greater than the Sum of Parts. Evolutionary Biology 45: 196-210. Randau M, Goswami A. 2017. Unravelling intravertebral integration, modularity and disparity in Felidae (Mammalia). Evol Dev 19:85-95. Rayner JM. 1988. The evolution of vertebrate flight. Biological Journal of the Linnean Society 34:269-287. Riskin DK, Bergou A, Breuer KS, Swartz SM. 2012. Upstroke wing flexion and the inertial cost of bat flight. Proc Biol Sci 279:2945-2950. Riskin DK, Bertram JEA, Hermanson JW. 2005. Testing the hindlimb-strength hypothesis: non-aerial locomotion by Chiroptera is not constrained by the dimensions of the femur or tibia. Journal of Experimental Biology 208:1309-1319. Riskin DK, Parsons S, Schutt WA, Jr., Carter GG, Hermanson JW. 2006. Terrestrial locomotion of the New Zealand short-tailed bat Mystacina tuberculata and the common vampire bat Desmodus rotundus. J Exp Biol 209:1725-1736. Riskin DK, Willis DJ, Iriarte-Diaz J, Hedrick TL, Kostandov M, Chen J, Laidlaw DH, Breuer KS, Swartz SM. 2008. Quantifying the complexity of bat wing kinematics. J Theor Biol 254:604-615.

84

Ross D, Marcot JD, Betteridge KJ, Nascone-Yoder N, Bailey CS, Sears KE. 2013. Constraints on Mammalian forelimb development: insights from developmental disparity. Evolution 67:3645-3652. Ruth EB. 1932. A study of the development of the mammalian pelvis. Anat Rec (Hoboken) 53:207-225. Sánchez-Villagra MR. 2002. Comparative patterns of postcranial ontogeny in therian mammals: an analysis of relative timing of ossification events. J Exp Zool 294:264-273. Sánchez-Villagra MR, Goswami A, Weisbecker V, Mock O, Kuratani S. 2008. Conserved relative timing of cranial ossification patterns in early mammalian evolution. Evol Dev 10:519-530. Sánchez-Villagra MR, Goswami A, Weisbecker V, Mock O, Kuratani S. 2008. Conserved relative timing of cranial ossification patterns in early mammalian evolution. Evol Dev 10:519-530. Santana SE, Cheung E. 2016. Go big or go fish: morphological specializations in carnivorous bats. Proceedings of the Royal Society B: Biological Sciences 283:20160615. Santana SE, Grosse IR, Dumont ER. 2012b. Dietary hardness, loading behavior, and the evolution of skull form in bats. Evolution 66:2587-2598. Santana SE, Strait S, Dumont ER. 2011. The better to eat you with: functional correlates of tooth structure in bats 25: 839-847. Sato K, Sakamoto KQ, Watanuki Y, Takahashi A, Katsumata N, Bost CA, Weimerskirch H. 2009. Scaling of soaring seabirds and implications for flight abilities of giant pterosaurs. PLOS ONE 4:e5400. Schliemann H, Goodman SM. 2011. A new study on the structure and function of the adhesive organs of the Old World sucker-footed bat (Myzopoda: Myzopodidae) of Madagascar. Verh. Naturwiss. Ver. Hamburg. Schmieder DA, Benitez HA, Borissov IM, Fruciano C. 2015. Bat Species Comparisons Based on External Morphology: A Test of Traditional versus Geometric Morphometric Approaches. PLOS ONE 10:e0127043. Schulz M. 2000. Diet and foraging behavior of the golden-tipped bat, Kerivoula papuensis: A spider specialist? Journal of mammalogy 81:948-957. Sears KE, Behringer RR, Rasweiler JJt, Niswander L. 2007. The Evolutionary and Developmental Basis of Parallel Reduction in Mammalian Zeugopod Elements. Am Nat 169:105-117. Shubin NH, Tabin CJ, Carroll S. 1997. Fossils, genes and the evolution of animal limbs. Nature 388:639–648. Simmons NB, Seymour KL, Habersetzer J, Gunnell GF. 2008. Primitive Early Eocene bat from Wyoming and the evolution of flight and echolocation. Nature 451:818-821. Simons EL. 2010. Forelimb skeletal morphology and flight mode evolution in pelecaniform birds. Zoology (Jena) 113:39-46. Simons EL, Hieronymus TL, O'Connor PM. 2011. Cross sectional geometry of the forelimb skeleton and flight mode in pelecaniform birds. J Morphol 272:958-971. Simons EL, O'Connor P M. 2012. Bone laminarity in the avian forelimb skeleton and its relationship to flight mode: testing functional interpretations. Anat Rec (Hoboken) 295:386-396. Smith KK. 2001. Heterochrony revisited: the evolution of developmental sequences. Biological Journal of the Linnean Society 73:169-186. 85

Smith KK. 2002. Sequence heterochrony and the evolution of development. J Morphol 252:82-97. Smith KK. 2003. Time's arrow: heterochrony and the evolution of development. International Journal of Developmental Biology 47:613-621. Spielmann M, Brancati F, Krawitz PM, Robinson PN, Ibrahim DM, Franke M, Hecht J, Lohan S, Dathe K, Nardone AM, et al. 2012. Homeotic Arm-to-Leg Transformation Associated with Genomic Rearrangements at the PITX1 Locus. American Journal of Human Genetics 91:629-635. Spoor CF, Zonneveld FW, Macho GA. 1993. Linear Measurements of Cortical Bone and Dental Enamel by Computed-Tomography - Applications and Problems. American Journal of Physical Anthropology 91:469-484. Stroud JT, Losos JB. 2016. Ecological Opportunity and Adaptive Radiation. Annual Review of Ecology, Evolution, and Systematics 47:507-532. Swartz SM, Bennett MB, Carrier DR. 1992. Wing bone stresses in free flying bats and the evolution of skeletal design for flight. Nature 359: 726-729. Swartz SM, Konow N. 2015. Advances in the study of bat flight: the wing and the wind 1. Canadian Journal of Zoology 93: 977-990. Swartz SM, Middleton KM. 2008. Biomechanics of the bat limb skeleton: Scaling, material properties and mechanics. Cells Tissues Organs 187:59-84. Teeling EC, Scally M, Kao D, Romagnoli M, Springer MS, Stanhope M. 2000. Molecular evidence regarding the origin of echolocation and flight in bats. Nature 403:188–192. Thean T, Kardjilov N, Asher RJ. 2017. Inner ear development in cetaceans. J Anat 230:249-261. Tokita M. 2015. How the pterosaur got its wings. Biol Rev Camb Philos Soc 90:1163- 1178. Tokita M, Abe T, Suzuki K. 2012. The developmental basis of bat wing muscle. Nature Communications 3: 1302. Veselka N, McErlain DD, Holdsworth DW, Eger JL, Chhem RK, Mason MJ, Brain KL, Faure PA, Fenton MB. 2010. A bony connection signals laryngeal echolocation in bats. Nature 463:939-942. Von Busse R, Hedenstrom A, Winter Y, Johansson LC. 2012. Kinematics and wing shape across flight speed in the bat, Leptonycteris yerbabuenae. Biol Open 1:1226-1238. Wagner GP, Altenberg L. 1996. Perspective: Complex adaptations and the evolution of evolvability. Evolution 50:967-976. Wagner GP, Pavlicev M, Cheverud JM. 2007. The road to modularity. Nature Reviews Genetics 8:921-931. Wallin J, Wilting J, Koseki H, Fritsch R, Christ B, Balling R. 1994. The role of Pax-1 in axial skeleton development. Development 120:1109-1121. Wang X, McGowan AJ, Dyke GJ. 2011. Avian wing proportions and flight styles: first step towards predicting the flight modes of mesozoic birds. PLOS ONE 6:e28672. Wang XL, Kellner AWA, Jiang SX, Cheng X, Wang Q, Ma YX, Paidoula Y, Rodrigues T, Chen H, Sayao JM, et al. 2017. Egg accumulation with 3D embryos provides insight into the life history of a pterosaur. Science 358:1197-1201. Wang Z, Han N, Racey PA, Ru BH, He GM. 2010. A comparative study of prenatal development in Miniopterus schreibersii fuliginosus, Hipposideros armiger and H. pratti. Bmc Developmental Biology 10: 10.

86

Wang Z, Zhu T, Xue H, Fang N, Zhang J, Zhang L, Pang J, Teeling EC, Zhang S. 2017. Prenatal development supports a single origin of laryngeal echolocation in bats. Nature Ecology & Evolution 1: 0021. Warrick DR, Tobalske BW, Powers DR. 2009. Lift production in the hovering hummingbird. Proc Biol Sci 276:3747-3752. Watts P, Mitchell E, Swartz S. 2001. A computational model for estimating the mechanics of horizontal flapping flight in bats: Model description and validation. Journal of Experimental Biology 204:2873-2898. Weisbecker V, Goswami A, Wroe S, Sánchez-Villagra MR. 2008. Ossification heterochrony in the therian postcranial skeleton and the marsupial-placental dichotomy. Evolution 62:2027-2041. Werneburg I, Geiger M. 2017. Ontogeny of domestic dogs and the developmental foundations of carnivoran domestication. Journal of Mammalian Evolution 24:323-343. Werneburg I, Laurin M, Koyabu D, Sánchez-Villagra MR. 2016. Evolution of organogenesis and the origin of altriciality in mammals. Evolution & Development 18:229-244. Werneburg I, Tzika AC, Hautier L, Asher RJ, Milinkovitch MC, Sánchez‐Villagra MR. 2013. Development and embryonic staging in non‐model organisms: the case of an afrotherian mammal. Journal of Anatomy 222:2-18. Wilson L, Humphrey LT. 2015. A Virtual geometric morphometric approach to the quantification of long bone bilateral asymmetry and cross‐sectional shape. American Journal of Physical Anthropology 4: 541-556. Wilson LA. 2013. Allometric disparity in rodent evolution. Ecol Evol 3:971-984. Wilson LA, Sánchez-Villagra MR. 2010a. Diversity trends and their ontogenetic basis: an exploration of allometric disparity in rodents. Proc Biol Sci 277:1227-1234. Wilson LAB. 2011. Comparison of prenatal and postnatal ontogeny: cranial allometry in the African striped mouse (Rhabdomys pumilio). Journal of mammalogy 92:407-420. Wilson LAB. 2018. The evolution of ontogenetic allometric trajectories in mammalian domestication. Evolution 72: 867-877. Wilson LAB, Sánchez-Villagra MR. 2010b. Diversity trends and their ontogenetic basis: an exploration of allometric disparity in rodents. Proceedings of the Royal Society B- Biological Sciences 277:1227-1234. Wilson LAB, Schradin C, Mitgutsch C, Galliari FC, Mess A, Sánchez-Villagra MR. 2010. Skeletogenesis and sequence heterochrony in rodent evolution, with particular emphasis on the African striped mouse, Rhabdomys pumilio (Mammalia). Organisms Diversity & Evolution 10:243-258. Witton MP. 2008. A new approach to determining pterosaur body mass and its implications for pterosaur flight. Zitteliana 28:143-158. Witton MP, Habib MB. 2010. On the size and flight diversity of giant pterosaurs, the use of birds as pterosaur analogues and comments on pterosaur flightlessness. PLoS ONE 5:e13982. Wolf M, Johansson LC, von Busse R, Winter Y, Hedenstrom A. 2010. Kinematics of flight and the relationship to the vortex wake of a Pallas' long tongued bat (Glossophaga soricina). J Exp Biol 213:2142-2153. Wyant KA, Adams RA. 2007. Prenatal growth and development in the Angolan free- tailed bat, Mops condylurus (Chiroptera : Molossidae). Journal of mammalogy 88:1248-1251. 87

Young NM, HallgrÍmsson B. 2005. Serial and the Evolution of Mammalian Limb Covariation Structure. Evolution 59: 2691-2704. Zeileis A, Grothendieck G, Ryan JA, Ulrich JM, Andrews F. 2018. zoo: S3 Infrastructure for Regular and Irregular Time Series (Z's Ordered Observations). Version 1.8-1: R Development Core Team. Zelditch ML, Calamari ZT, Swiderski DL. 2016. Disparate Postnatal Ontogenies Do Not Add to the Shape Disparity of Infants. Evolutionary Biology 43:188-207.

88

Chapter 3 Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera)2

Abstract

Most morphological and physiological adaptations associated with bat flight are concentrated in the postcranium, reflecting strong functional demands for flight performance. Despite an association between locomotory diversity and trophic differentiation, postcranial morphological diversity in bats remains largely unexplored.

Evolutionary developmental biology is a novel approach providing a link between the analysis of genotypic and phenotypic variation resulting from selective pressures. To quantify the morphological diversity of the postcranium in bats, and to explore its developmental basis, we reconstructed the postcranial allometric trajectories of nine bat species from different prenatal developmental series, representing five families and both suborders. We tested for allometric growth in Chiroptera and also quantified levels of allometric disparity and inter-trajectory distances. Using a phylogenetic scaffold, we assessed whether ontogenetic differences reflect evolutionary relationships. We found significant allometric growth trajectories in almost all species.

Interspecific trajectory distances showed lower variance within Yinpterochiroptera than within Yangochiroptera and between suborders. Each suborder occupied non-

2 A version of this has been accepted for publication: López‐Aguirre, C, Hand, SJ, Koyabu, D, Son, NT, Wilson, LAB. Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera). J Exp Zool (Mol Dev Evol). 2019; 332: 36– 49. A copy is included in Appendix peer-reviewed publication 1. * I certify that this publication was a direct result of my research towards this PhD, including the design, analysis and interpretation of results, and that reproduction in this thesis does not breach copyright regulations. I acknowledge my co-authors for their supervisory guidance and their contributions to the publication.

89 overlapping sections of allometric space, showing changes in the growth rates of specific bones for each suborder. Allometry-corrected disparity was significantly higher in larger species. Statistically significant phylogenetic signal in our results suggests that there is an ontogenetic basis for the postcranial morphological diversity in modern bats. Ancestral state reconstruction also showed an increase in the amount of change in shape with size in the larger species studied. We hypothesise that differences in allometric patterns among bat taxa may reflect a size-dependent evolutionary constraint, whereby variability in body size and allometric patterns are associated.

Introduction

Variability, coupled with its potential for adaptability, are the basic components of evolution (Hallgrímsson, et al. 2012). Variation can occur across different interacting dimensions (e.g. genetic, phenotypic, ecological, behavioural) that generate advantageous traits when responding to selective pressures, promoting evolvability

(Hallgrímsson, et al. 2012). Understanding the complex interaction between variability and selective pressures has been one of the main focuses of evolutionary biology, and has expanded the methodologies with which variation is quantified (Raff 1996;

McNamara 2012). One approach of increasing interest is evolutionary developmental biology (EvoDevo), the study of how development changes over evolutionary time

(Hall 2000; Love and Raff 2003).

EvoDevo has proven to be a particularly useful tool to study the evolution of speciose clades with broad ecological diversity (Adams 2008; Hockman, et al. 2009; Wilson and

Sánchez-Villagra 2010b; Wilson, et al. 2010; Cooper, et al. 2012; Wilson 2013a; Wilson

2013b; Koyabu and Son 2014; Eckalbar, et al. 2016). Previous findings have highlighted

90 a developmental component to the evolutionary divergence between placentals and marsupials (Weisbecker, et al. 2008; Goswami, et al. 2009), developmental differences between the placental clades Boreoeutheria (i.e. Laurasiatheria and Euarchontoglires) and Atlantogenata (i.e. Afrotheria and Xenarthra; Hautier et al., 2013), the ontogenetic basis of ecomorphological diversity of modern rodents (Zelditch, et al. 2004; Zelditch, et al. 2006; Wilson and Sánchez-Villagra 2010b; Wilson 2013a; Wilson 2013b) and the developmental basis of morpho-functional adaptations in cetaceans (Armfield, et al.

2013; Boessenecker and Fordyce 2015; Cooper, et al. 2017; Thean, et al. 2017).

An alternative approach for the analysis of morphogenesis is the study of allometry within a phenotypic space (Gerber, et al. 2008; Klingenberg 2010; Wilson and Sánchez-

Villagra 2010b; Wilson 2013a; Sánchez-Villagra, et al. 2017). Size-dependent growth is intrinsic to the development of any organism, and changes in size-dependent development can reflect ecomorphological diversity within a clade (Klingenberg 2010).

Recent studies have proposed novel multidimensional interpretations of allometry, compiling the developmental trajectories of multiple structures into a multivariate, allometric space (Wilson and Sánchez-Villagra 2010b; Wilson 2013a; Randau and

Goswami 2017; Randau and Goswami 2018). Allometric spaces facilitate the quantification of variation in the developmental trajectories of different taxa (i.e. allometric polarity), enabling the depiction of ontogenetic signal in evolutionary processes (Gerber, et al. 2008; Adams and Nistri 2010; Wilson and Sánchez-Villagra

2010b; Sánchez-Villagra, et al. 2017). The distribution of species within an allometric space can yield information on the evolutionary origin of apomorphies in a phylogeny

(Adams and Nistri 2010), ontogenetic signals in the ecologic partitioning of a group

91

(Wilson and Sánchez-Villagra 2010b), and the evolutionary changes in the development of different organisms in response to ecological pressures (Wilson and

Sánchez-Villagra 2010a; Wilson 2011c; Sánchez-Villagra, et al. 2017).

With over 1400 living species, bats (order Chiroptera) represent the second largest clade of mammals, and one of the best-known examples of ecomorphological radiation in Mammalia (Falkowski, et al. 2005; Bininda-Emonds, Cardillo, et al. 2007;

Figueirido, et al. 2012). Flight was one of the most important evolutionary novelties during this radiation (Arita and Fenton 1997; Adams 2008; Cooper and Tabin 2008;

Cooper, et al. 2012; Adams and Shaw 2013). Moreover, bats have evolved behaviours

(e.g. terrestrial locomotion) that represent novel functional needs for a flight-adapted postcranium (Hand, et al. 2009; Wang, Zhu, et al. 2017). Many studies have reviewed bat intraordinal phylogenetic relationships at higher taxonomic levels, leading to evolutionary hypotheses in which these adaptations are interpreted to have evolved multiple times within Chiroptera (Teeling, et al. 2005; Foley, et al. 2015). Based on morphological and ecological characteristics, the order was traditionally divided into

Microchiroptera (“”) and Megachiroptera (“”; Dobson, 1875).

Today two different suborders are recognised – Yangochiroptera (i.e.

Emballonuroidea, and ) and Yinpterochiroptera (i.e.

Rhinolophoidea and Pteropodidae). The Yangochiroptera-Yinpterochiroptera arrangement contradicted the monophyly of laryngeal echolocation, raising the unresolved question of whether echolocation evolved once in the common ancestor of all bats, or evolved independently in yangochiropterans and rhinolophoids (Teeling, et al. 2000; Jones and Teeling 2006; Wang, Zhu, et al. 2017).

92

Previous studies have examined the development of the cochlea to untangle the evolutionary history of laryngeal echolocation in bats, providing evidence for a monophyletic origin in the common ancestor of chiropterans, followed by a loss of echolocation in pteropodids (Wang, Zhu, et al. 2017). Based on this, it was argued that there was not a developmental signal in the subordinal partition of the order, and instead autapomorphic change in the development of the cochlea in Pteropodidae

(Wang, Zhu, et al. 2017). However, this conclusion has also been challenged on the basis of developmental data (Nojiri, et al. 2018), leaving the question of whether a developmental signal exists in the subordinal divergence in Chiroptera unresolved.

Developmental studies in bats have also shown heterochronic acceleration in the development of the foot and thumb compared to other mammals (Koyabu and Son

2014), the regulatory role that wing membrane has in the development of the wing

(Adams, 2008), postnatal compensatory growth between the autopod and zeugopod

(Adams 2008; Adams and Shaw 2013), and a developmental basis for the diversification of New World leaf-nosed bats (Sears 2014). However, most research in bat development is based on the study of few model species (e.g. Carollia perspicillata) to represent this ecologically diverse group (Cretekos, et al. 2005; Giannini, et al. 2006;

Wyant and Adams 2008; Hockman, et al. 2009; Tokita, et al. 2012; Carter and Adams

2015; Eghbali, et al. 2017; Reyes-Amaya, et al. 2017).

Many of these studies focus exclusively on postnatal ontogeny, due in part to the difficulty associated with obtaining samples of prenatal ontogenetic series for most mammals (Milinkovitch and Tzika 2007). Reported differences in the ontogenetic processes across developmental stages in different mammal species highlights the

93 importance of describing it in its entirety (Zelditch, et al. 2003; Frederich and

Vandewalle 2011; Wilson 2011b; Ponssa and Candioti 2012; Halley 2016; Werneburg and Geiger 2017). It has been argued that these differences have an evolutionary origin, as each developmental stage responds to different selective pressures and functional needs (Werneburg & Geiger 2017; Zelditch et al. 2016). Consequently, a link between variability in the morphogenetic processes and morphological disparity in adult forms has been theorised (Werneburg & Geiger, 2017; Wilson et al., 2010;

Wilson, 2013a, 2013b; Zelditch et al., 2016), whereby most of a species’ disparity is thought to arise prenatally, reflecting a stage in ontogeny that may be exposed to fewer selective pressures compared to postnatal and adult stages (Wilson, 2011;

Zelditch et al., 2016). Moreover, it has been postulated that highly adaptable and evolvable traits tend to appear earlier or to reach maturity late in development, as this allows longer developmental processes that enable traits to diversify (Werneburg &

Geiger, 2017; Zelditch et al., 2016).

A limiting factor to our understanding of the evolution of postcranial adaptations for flight in bats (e.g. forelimb bone elongation, reduced cortical thickness, interdigital membranes) is the patchiness of the fossil record, which lacks postcranial transitional forms between the non-volant bat ancestor and the oldest bat fossils known, and between modern bats with novel locomotory strategies and their most recent common ancestors (Eiting and Gunnell 2009; Hand, et al. 2009; Hand, Lee, et al. 2015).

Furthering understanding of the developmental basis of flight adaptations in bats could inform understanding about the correspondence between the adaptations to selective pressures that taxa experience and fluctuations in their developmental

94 trajectories (Geiger, et al. 2014; Koyabu, Werneburg, Morimoto, et al. 2014; Ramirez-

Chaves, et al. 2016).

Given the strong molecular support for the Yangochiroptera-Yinpterochiroptera subordinal arrangement of Chiroptera (Teeling, et al. 2002; Teeling, et al. 2005;

Springer 2013; Tsagkogeorga, et al. 2013; Lei and Dong 2016), and its implications for the interpretation of the evolution of bats (i.e. paraphyly of “microbats” and laryngeal echolocation), studying morphogenesis is a promising approach that could merge morphological and developmental data to inform evolutionary hypotheses

(Klingenberg 2010). Furthermore, evidencing differences in allometric trajectories across bat species could identify a developmental basis for the morphological diversity in modern bats (Sears, 2014). Considering the lack of knowledge about the role that development has in modern bat phenotypic diversity, the purpose of this study is to quantify ontogenetic allometric disparity in Chiroptera and assess the presence of a phylogenetic signal in the allometric trajectories of bats. We focused on prenatal postcranial development because this encompasses the highest number of morphological adaptations for flight. Moreover, it has been argued that prenatal development better reflects the developmental basis of evolutionary diversification

(Gould, 1998; Zelditch et al., 2003, 2016). Allometric trajectories of different species were reconstructed based on a multivariate interpretation of allometry, allowing us to assess variability in allometric trajectories across taxa, and to test for a phylogenetic signal in our results that would suggest an ontogenetic basis for modern bat diversity and phylogenetic relationships. Postcranial traits under selective pressure associated with diversification of bats would be expected to show morphogenetic variability with

95 a strong phylogenetic signal, indicating a developmental basis to the postcranial morphological variability in modern bats.

Materials and Methods

Data collection

Linear measurements of 24 postcranial bones (Fig. 3.1; see Table 2.2 for details on measuring protocol) were collected from 64 prenatal embryonic specimens representing nine species (Aselliscus dongbacana, A. stoliczkanus, Cynopterus sphinx,

Hesperoptenus blanfordi, Hipposideros larvatus, Kerivoula hardwickii, Miniopterus schreibersii, Myotis sp. and Rhinolophus thomasi), five families and both suborders.

The measurements were: Humeral length (HL), clavicular length (CL), scapular length

(SL), femoral length (FL), length of first rib (RiL), tibial length (TL), fibular length (FiL), radial length (RL), length of ulna (UL), length of sternum (StL), length of manual phalange (MPL), length of pedal phalange (PPL), metacarpal length (McL), metatarsal length (MtL), tarsal length (TaL), carpal length (CaL), length of cervical vertebrae (CvL), length of thoracic vertebrae (TvL), length of lumbar vertebrae (LvL), length of sacral vertebrae (SvL), length of caudal vertebrae (CavL), length of illium (IlL), length of ischium (IsL) and pubis bone length (PuL). Some bones with late ossification onset are absent in early prenatal stages, leading to measurements with zero values that do not represent absence in adults.

96

Figure 3.1. Three-dimensional virtual reconstruction of skeleton of Aselliscus dongbacana (stage 21), showing 24 postcranial bones measured. From left to right, skeleton viewed on ventral, lateral and dorsal view.

Specimens were collected and prepared in Vietnam by N.T.S. and D.K. for a study of sequence heterochrony in bats (Koyabu and Son 2014), and stored in 70% ethanol. 3D models of the skeletons of specimens were constructed using a microfocal X-ray CT system at the University Museum, University of Tokyo (TXS225-ACTIS, TESCO, Tokyo) with 70 kV source voltage and µ114 A source currents at a resolution of 36 µm.

Postcranial osseous elements were segmented using the thresholding tool in MIMICS v. 20 software (Materialise NV). Finer segmentation was performed, thresholding CT- values of osseous and non-osseous tissues using the Half Maximum Height method

(i.e. gradual change in CT numbers at the boundary of a structure) (Spoor, et al. 1993).

Measurements of postcranial elements were obtained directly from the 3D virtual models of the postcranial skeleton of the embryos using the measuring tool in MIMICS v. 20 software (Materialise NV).

Ontogenetic allometric trajectories

To reconstruct the allometric trajectories of the species, we used log-shape ratios to deconstruct morphological variability represented by linear measurements into shape and size (Mosimann and James 1979). Log-shape ratios calculate a single size variable

97 that represents the overall size of the complete object and a vector of shape variables.

For each specimen, size was calculated as the geometric mean of all 24 postcranial measurements. Given the nature of our sample (i.e. measurements in early prenatal stages with zero values), we calculated the geometric mean of each specimen after transforming our measurements by adding a value of one to all raw measurements and then subtracting one from the geometric mean (Hornung and Reed 1990). Each measurement was then divided by size to obtain a shape ratio for each measurement.

Shape ratios and geometric means were log-transformed and used as input data for all analyses.

To test the null hypothesis of isometry (i.e. no change in shape with size), individual linear regressions of size on shape ratios were performed both at species and subordinal level, and statistical significance was assessed with 10,000 iterations using the procD.lm function from the R package Geomorph version 3.0.4 (Adams, et al.

2013; Adams, et al. 2017). To analyse differences in the allometric trajectories between taxa, we first tested differences in slope angle and length with a

Homogeneity of Slopes test (HOS; “shape ~ size * species” and “shape ~ size * suborder”) (Collyer, et al. 2015). HOS estimate distances among specimens to test differences in the allometric trajectories between pairs of taxa based on a null hypothesis of common allometry with different intercepts (Esquerré, et al. 2017;

Wilson 2018). Pairwise comparisons of the slope angles (the direction of shape change with size) and lengths (magnitude of shape change with size) were extracted from the

HOS, and statistical significance was evaluated using 10,000 iterations. HOS pairwise comparisons that showed parallel allometric trajectories (i.e. did not reject the null

98 hypothesis of common slopes) were then tested for differences in the intercept along the shape axis, with a test for differences in Least Square (LS) means (Esquerré, et al.

2017; Wilson 2018). This test assesses whether different allometric trajectories share the same intercept (H0) or not (H1), and was performed with the advanced.procD.lm function of the R package Geomorph, with 10,000 iterations to assess statistical significance. All tests were performed at the species and subordinal levels.

Ontogenetic allometric trajectories were visualised by plotting the first principal component of the predicted values extracted from the shape-size regression against size (Adams and Nistri 2010; Esquerré, et al. 2017; Wilson 2018). All analyses were performed in R version 3.3.2.

Evolution of ontogenetic allometric trajectories

To visualize the evolution of the allometric trajectories across species and suborders, an allometric space was constructed with a principal component analysis of a matrix of allometric coefficients retrieved from a multivariate regression of shape on size, computed for each separately. With this approach, the allometric trajectory of each taxon is represented by a single point in the allometric space. Subsequently, a phylogenetic scaffold was projected into the allometric space to produce a phyloallomspace (Esquerré, et al. 2017) using the phylomorphospace function of the R package phytools version 06-20 (Revell 2012). Phylogenetic relationships and calibrated divergence estimates follow a recent, comprehensive extant bat phylogeny by Shi and Rabosky (2015).

To test whether allometric differences between taxa had a phylogenetic signal, Kmult

(K-) statistic was computed as implemented in the physignal function of Geomorph

99

(Adams 2014). This test evaluated whether closely related taxa were more similar in postcranial allometry than expected under a Brownian motion model of evolution.

Following Wilson (2018), we implemented a multivariate interpretation of K- by using the principal components that accounted for at least 95% of the variance in allometric space. Finally, to see how the magnitude of shape change with size has evolved across in bats, the phylogenetic framework was used to plot the evolutionary changes in slope vector lengths across species using maximum likelihood ancestral reconstruction, implementing the contMap function of the package phytools version 06-20 (Revell

2012). To compare variability in allometric trajectories across bats, allometric disparity was quantified as the Procrustes variance of the scores of the first five PCs, representing in total 95% of the variance in allometric space (Adams, et al. 2013;

Adams, et al. 2017). Additionally, in order to estimate the level of dispersion along the allometric trajectory of each taxon, allometry-corrected disparity was measured as the

Procrustes variance of the residuals of a linear model fit using the morphol.disparity function in Geomorph (Piras, et al. 2011; Adams, et al. 2013; Adams, et al. 2017).

Allometric and allometry-corrected disparity were quantified at species and subordinal levels. Statistical significance of pairwise comparisons was estimated with a permutation test (1000 iterations). Due to small sample size (N= 2), M. schreibersii and

Myotis sp. were excluded from this analysis.

Results

Ontogenetic allometry

The null hypothesis of isometric growth was rejected for most species based on individual multivariate regressions (Table 3.1), with the exception of M. schreibersii

100 and Myotis sp., where small sample size could have affected these results. At subordinal level, only Yinpterochiroptera showed significant allometric growth, whereas Yangochiroptera did not (Yinpterochiroptera: F(1,46)= 5.219, Z= 1.292, P=

0.027; Yangochiroptera: F(1,14)= 0.549, Z= 0.254, P= 0.477). HOS showed statistically significant differences in the slope of the allometric trajectories at species level but not at subordinal level (Species: F(54,46)= 2.459, Z= 2.114, P= 0.012; Subordinal: F(61,60)=

0.589, Z= 0.534, P= 0.331). Of the 36 species-level pairwise comparisons, six showed statistically significant differences in magnitude, whereas only two showed statistically significant differences in angle, meaning that most species share the same allometric slope (Table 3.2). Based on boxplots of the pairwise interspecific trajectory distances

(i.e. slope angle in degrees), allometric trajectories show a broader range of slopes in yinpterochiropteran species than in the yangochiropteran species we studied (Fig. 3.2).

Furthermore, yangochiropteran species showed similar degrees of separation from species of both suborders (Fig. 3.2).

101

Figure 3.2. Boxplots of the distribution of inter-trajectory angle values of pairwise comparisons between species. Whiskers represent minimum and maximum values for each set of comparisons. From left to right, pairwise comparisons are summarised as comparisons between species of different suborders (intersubordinal). Numbers and the position of “x” within each boxplot represent the number of pairwise comparisons and the average, respectively.

Table 3.1. Individual multivariate linear regressions for the test of allometric growth, based on an isometric growth null hypothesis.

Species SS MS Rsq F Z Pr(>F) A. dongbacana 77.635 77.635 0.98327 705.38 3.4945 0.0001 A. stoliczkanus 1097.3 1097.29 0.98457 893.05 3.6077 0.0001 C. sphinx 504.17 504.17 0.97283 71.604 1.3307 0.01945 H. blanfordi 340.4 340.4 0.98064 151.94 2.4961 0.0037 H. larvatus 8.5919 8.5919 0.99581 475.18 2.2477 0.01945 K. hardwickii 222.869 222.87 0.99332 743.8 3.2837 0.0002 M. schreibersii 0.020855 0 1 0 0.5 Myotis sp. 10.117 10 1 0 0.5 R. thomasi 653.25 653.25 0.9887 699.69 2.9461 0.0001 Intercepts of allometric trajectories were significantly different both at species and subordinal levels (Species: F(55,54)= 1739, Z= 4.089, P< 0.001; Subordinal: F(62,61)= 1783,

Z= 3.965, P< 0.001). Interspecific pairwise comparisons of differences in intercept showed that all species with a common allometric slope do not have significantly different intercepts, meaning that, in those cases, species have overlapping allometric

102 trajectories (Table 3.3). Visualisation of the predicted allometric slopes shows variability both in angle and magnitude of the allometric trajectories at species and subordinal levels (Fig. 3.3).

Figure 3.3. Ontogenetic allometric trajectories of Chiroptera at species (A) and subordinal (B) levels. Trajectories are derived from an homogeneity of slope test, plotting log transformed geometric means in the x axis (i.e. log(Size)) and the PC1 of the predicted values of multivariate regression of shape ratios on size in the y axis (Shape(Predicted)).

103

Table 3.2. Pairwise comparisons of statistical differences in vector length (magnitude of shape change with size; top) and angle (direction of allometric vector; bottom) of the ontogenetic allometric trajectory between species.

MAGNITUDE A. dongbacana A. stoliczkanus C. sphinx H. blanfordi H. larvatus K. hardwickii M. schreibersii Myotis sp. R. thomasi A. dongbacana 121.709 58.347 81.449 1.222 169.350 10.857 47.183 231.293 A. stoliczkanus 0.005 63.362 40.260 122.931 47.642 132.566 74.525 109.584 C. sphinx 0.126 0.122 23.102 59.569 111.004 69.204 11.164 172.946 H. blanfordi 0.110 0.433 0.615 82.671 87.902 92.306 34.266 149.844 H. larvatus 0.984 0.081 0.336 0.248 170.573 9.635 48.406 232.515 K. hardwickii 0.056 0.552 0.143 0.289 0.083 180.208 122.167 61.942 M. schreibersii 0.954 0.557 0.711 0.651 0.960 0.471 58.041 242.150 Myotis sp. 0.682 0.533 0.923 0.775 0.707 0.404 0.810 184.110 R. thomasi 0.001 0.062 0.012 0.028 0.009 0.468 0.398 0.253 ANGLE A. dongbacana A. stoliczkanus C. sphinx H. blanfordi H. larvatus K. hardwickii M. schreibersii Myotis sp. R. thomasi A. dongbacana 3.738 3.366 3.559 0.406 3.848 15.825 3.221 3.927 A. stoliczkanus 0.031 0.371 0.179 4.143 0.110 19.563 0.517 0.189 C. sphinx 0.023 0.287 0.193 3.772 0.482 19.191 0.145 0.561 H. blanfordi 0.067 0.653 0.581 3.965 0.289 19.384 0.338 0.368 H. larvatus 0.447 0.153 0.149 0.183 4.253 15.419 3.626 4.332 K. hardwickii 0.188 0.857 0.401 0.639 0.279 19.673 0.627 0.079 M. schreibersii 0.406 0.406 0.406 0.415 0.433 0.440 19.046 19.752 Myotis sp. 0.355 0.604 0.906 0.744 0.405 0.594 0.469 0.706 R. thomasi 0.081 0.672 0.272 0.499 0.192 0.897 0.419 0.562 0.000

104

Allometric space analysis

The first two components of allometric space accounted for 78.6% of the variation between the ontogenetic allometric trajectories of the species. Both suborders occupied a clearly delimited, non-overlapping region of allometric space, with Yinpterochiroptera spread across PC1 of allometric space and both suborders clearly separated across PC2 (Fig. 3.4A).

Similarly, by projecting the phylogenetic scaffold into the allometric space, it was possible to reconstruct a clear phylogenetic structuring between suborders (Fig. 3.4B). These results not only show highly variable slopes across all species examined, but also that the allometric spaces of both suborders are approximately of the same size (Yangochiroptera= 0.169 and

Yinpterochiroptera= 0.247, P=0.075). Since the principal components of allometric space represent the deviation from the mean allometric trajectory, negative or positive values in a given axis indicate lower or faster than average lengthening of a given structure, respectively. Faster growth of the carpals, the sternum and the caudal vertebrae and a slower lengthening of the radius, ulna and humerus are associated with PC1 (65.63%), whereas faster growth of the manual phalanges, and slower growth of the thoracic, cervical and lumbar vertebrae are associated with PC2 (12.96%) (Appendix Table 5). Subordinal differences in PC1 loadings were also found for the lumbar vertebrae (Appendix Table 5).

105

Figure 3.4. Allometric space (A) and phyloallomspace (B) of Chiroptera. Each point represents the allometric trajectory of a species, convex hulls represent the allometric space occupied by each species (B), and the different colours in the phylogenetic relationships projected in the phyloallomspace represent each suborder (B).

Evolution of ontogenetic allometric trajectories

Multivariate tests of phylogenetic signal in allometric space indicate a significant phylogenetic structuring in the variability in allometric trajectories across species (K- statistic= 1.098, P= 0.029). Ancestral state reconstruction showed that on average yangochiropteran species have lower magnitude shape change with size, compared with yinpterochiropteran species. C. sphinx and M. schreibersii had the highest magnitude of shape change with size, whereas Myotis sp. and K. hardwickii had the lowest magnitude of change (Fig. 3.5).

106

Table 3.3. Pairwise comparisons of statistical differences in the slope intercept of the allometric trajectories (top) and p values of the difference between each pair of species based on 10,000 iterations (bottom) of the ontogenetic allometric trajectory between species.

A. dongbacana A. stoliczkanus C. sphinx H. blanfordi H. larvatus K. hardwickii M. schreibersii Myotis sp. R. thomasi A. dongbacana -2.014 -1.705 -1.487 -0.902 -2.900 -0.928 -0.714 -1.266 A. stoliczkanus 0.982 -0.496 -1.243 -1.809 -1.672 -1.397 -1.063 -1.401 C. sphinx 0.979 0.612 -0.797 -1.251 -0.852 -1.738 -0.646 -0.786 H. blanfordi 0.962 0.934 0.759 -1.450 -1.558 -1.301 -0.890 -1.169 H. larvatus 0.815 1.000 0.891 0.970 -2.363 -0.699 -0.952 -1.488 K. hardwickii 1.000 0.976 0.764 0.950 0.996 -1.881 -1.737 -2.092 M. schreibersii 0.803 0.879 0.958 0.883 0.711 0.948 -0.768 -1.143 Myotis sp. 0.732 0.856 0.689 0.794 0.860 0.970 0.753 -0.782 R. thomasi 0.906 0.979 0.747 0.959 0.986 0.988 0.851 0.775

107

Figure 3.5. Ancestral state reconstruction of the magnitude of shape change with size, based on the ontogenetic allometric slope vector length. Colours in each branch represent the magnitude of shape change with size for each species and each reconstructed ancestor, red-to-pink branches representing higher levels of change, and blue-to- green branches lower levels of change.

Both allometric and allometry-corrected disparity were higher in Yinpterochiroptera

(0.247 and 9680.94, respectively) than in Yangochiroptera (0.169 and 3683.85, respectively) but neither showed significant differences (P=0.664 and 0.378, respectively). At species level, the largest species (C. sphinx) had the highest allometry- corrected disparity followed by R. thomasi, whereas H. larvatus had the lowest followed by A. dongbacana. Pairwise comparisons revealed that morphological disparity in C. sphinx had the highest number of significant differences (all species except R. thomasi), followed by A. dongbacana and R. thomasi with three (Fig. 3.6).

108

Figure 3.6. Levels of allometry-corrected disparity (Procrustes variance) for seven bat species analysed. Numbers on top of each bar represent significant differences across species based on pairwise permutations test: 1 signifies significant differences with A. dongbacana, 2 with A. stoliczkanus, 3 with C. sphinx, 4 with H. blanfordi, 5 with H. larvatus, 6 with K. hardwickii and 7 with R. thomasi. Discussion

Our results provide evidence of a developmental signal for both phenotypic diversity and possibly subordinal partitioning in modern bats. Our results also suggest that the magnitude of postcranial allometric disparity in each bat suborder is different, and that it can be traced back to prenatal morphogenesis (Sears 2014). Given the controlled environment of the foetus during prenatal development, it is reasonable to argue that prenatal morphogenesis is not conditioned by highly variable external or environmental conditions (Wilson 2011a; Zelditch, et al. 2016; Werneburg and Geiger

2017). Because morphogenetic variability at prenatal stages has been directly linked with morphological variability at later stages (Werneburg and Geiger 2017), early onset of morphogenetic changes provides extended developmental times that may in turn facilitate phenotypic variation, signalling an ontogenetic basis to diversification

(Werneburg and Geiger 2017). Congruently, short developmental times could

109 represent a limiting factor for diversification, suggesting that prenatal development has a major role in either constraining (i.e. canalisation) or facilitating (i.e. patterning) diversification of shape.

Here, postcranial allometric trajectories showed differences in the prenatal development of bats both at species and subordinal levels. Our results are the first to show evidence within Mammalia of a significant phylogenetic signal in the prenatal ontogenetic processes, supporting the hypothesis that changes in prenatal development could promote diversification (Sears 2014; Werneburg and Geiger 2017).

We used multivariate generalisation of allometry to provide a standardised and quantifiable approach to assess the developmental basis of phylogenetic and ecological divergences (Goswami, et al. 2009; Wilson and Sánchez-Villagra 2010b;

Hautier, et al. 2013; Wilson 2013a; Wilson, et al. 2015). Differences between suborders in PC1 loadings reveal bone-specific differences in the allometric trajectories of both suborders. These results indicate that subordinal differences in allometric trajectories appear early in development (i.e. prenatally), suggesting that the subordinal dichotomy in Chiroptera could involve a developmental basis.

At the species level, C. sphinx is clearly separated from all others along PC1, suggesting that frugivory - a trait exclusive to C. sphinx within our sample - could correlate with an accelerated growth of the manual phalanges. Our results could suggest changes in the signalling pathways of digit development in C. sphinx. Sears, et al. (2006) found that bat forelimb digits show higher rates of chondrocytes proliferation, associated with increased morphogenetic protein 2 (Bmp2) expression. Signalling pathways associated with wing membrane morphogenesis (i.e. Fgf10) can also act as regulators of wing

110 muscle and possibly bone growth (Tokita, et al. 2012). Our results for C. sphinx also echo previous findings of differences in the postnatal development of the wing between insectivorous and frugivorous species (Adams 2008). Moreover, intraguild differences in postnatal growth have been reported between tropical and temperate insectivorous bat species (Kunz, et al. 2009). However, because our sample includes only one frugivorous species, prenatal allometric trajectories of other frugivorous species need to be studied to better test a diet-related developmental shift. Similar results have been interpreted as evidence for an ontogenetic basis for shape disparity in the palate of phyllostomid bats, a family that exhibits the greatest cranial shape diversity within Chiroptera (Sears 2014). Contrary to most studies that have focused on the evolution of the cranium where there was no ontogenetic signal in phylogenetic relationships (Wang, Zhu, et al. 2017; Nojiri, et al. 2018), our results show a clear ontogenetic signal in the subordinal phylogenetic relationships in Chiroptera.

Nevertheless, this does not exclude that other aspects of their biology could have played a significant role during their divergence.

Reconstruction of each subordinal allometric space exposed allometric patterning in specific regions of the postcranium for each suborder. Yangochiroptera exhibited faster than average growth of the manual phalanges compared with

Yinpterochiroptera (Koyabu and Son 2014), suggesting a tendency towards precociality. However, considering the variability in the development between different regions of the body (i.e. altricial development of the autopod and precocial development of the foot compared to other mammals), bats are difficult to define as either precocial or altricial (Koyabu and Son 2014; Nojiri, et al. 2018). Slower growth of

111 the thoracic, cervical and lumbar vertebrae, on the other hand, indicates a shift in the prenatal ontogenesis of presacral vertebral development in the yinpterochiropterans examined here. Presacral developmental patterning in mammals has been studied based on deviations in sloths and manatees from the ‘rule of seven’ cervical vertebral count (Buchholtz and Stepien 2009; Hautier, et al. 2010; Bohmer, et al. 2018). Efforts have focused on testing two competing hypotheses, one arguing for changes in Hox genes expression (Bohmer, et al. 2018), and the other for a change in the tissues of origin for some vertebrae (i.e. primaxial-abaxial repatterning; (Hautier, et al. 2010).

Our results raise interesting questions about the morphogenesis of the vertebral column in mammals, beyond changes in vertebral count.

The subordinal spatial separation in allometric space also suggests developmental differences between suborders. Ecomorphological diversity in bats is arguably one of the clearest examples of adaptive radiation in mammals, a radiation that has led to high within group variability (e.g. dietary breadth) that often obscures differences (e.g. evolution of laryngeal echolocation) (Teeling, et al. 2005; Springer 2013). Moreover, the ecomorphological diversity in our yangochiropteran sample is narrow, limiting our capacity to extrapolate our results to the whole order. Previous studies have described prenatal developmental differences between ecologically distinct rodent groups that point to an ontogenetic basis for this differentiation (Wilson and Sánchez-Villagra

2010a; Wilson 2013a). It is more difficult to attribute these differences to ecological characteristics specific to each bat suborder. Further studies could focus on testing the developmental differences found here across a broader ecomorphological range of bat species.

112

Alternatively, we can suggest that the developmental differences that characterise the suborders do not reflect ecological adaptations that evolved in each group. This is underscored by the values of allometric disparity in each suborder. Differences in allometric disparity have helped to describe evolutionary differences between closely related taxa in mammals (Wilson and Sánchez-Villagra 2010a; Wilson 2013a; Sánchez-

Villagra, et al. 2017). Greater disparity in our yinpterochiropteran sample indicates that the developmental trajectories in this group may not be evolutionarily constrained, leading to a diversification in the morphogenetic processes of the species (Wilson and

Sánchez-Villagra 2010b; Wilson 2013a). Similar results have been reported in the three major clades of rodents (Wilson 2013a). Sciurid rodents occupy a unique developmental space within Rodentia, characterised by diminished allometric disparity, and lower variability in intragroup intertrajectory distances (Wilson 2013a).

Molecular, morphological and fossil data places the divergence between crown group bat suborders between around 70 and 55 Ma (Springer 2013; Tsagkogeorga, et al.

2013; Lei and Dong 2016). Developmental trajectories are generally thought of as being highly conserved between sister taxa (Hautier, et al. 2010). However, deep- rooted divergences allow for different developmental mechanisms to evolve between closely-related taxa (Wilson 2013a; Sánchez-Villagra, et al. 2017). The early divergence of bat suborders could have allowed the evolution of differences in the ontogenetic allometric trajectories between suborders. Previous studies suggesting an ontogenetic component to evolutionary splits within Mammalia are mostly limited to the marsupial-placental dichotomy (Weisbecker, et al. 2008; Goswami, et al. 2009), and some Miocene diversification events in hominids and rodents (Cobb and O’Higgins

113

2004; Cardini and O’Higgins 2005). Following the placental-marsupial split approximately 160 Mya (Luo, et al. 2011), the divergence between bat suborders represents the second oldest split in Mammalia to be supported by developmental data. Moreover, phenotypic variability does not necessarily correlate with diversification rates in globally distributed mammal taxa (e.g. Sciuridae), but rather seems to be a geographic and ecologically mediated process associated with the age of each clade (Zelditch, et al. 2015). Finally, which subordinal pattern diverged from the ancestral condition is unclear. Future studies could study bat allometric disparity in deep time by analysing the fossil record of individual taxa across time (Goswami, et al.

2014a) in order to reconstruct the ancestral condition in Chiroptera.

Our ancestral state reconstruction analysis showed a general decrease in the amount of shape change with size at the tips of the phylogeny. This suggests that, in our sample, size variability does constrain evolvability of allometric trajectories in the postcranium, and is consistent with previous findings of size-dependent heterochrony and shape change in the mammalian skull (Cardini and Polly 2013). Morphogenesis is inherently size-dependent, indicating that size plays an important role in evolvability

(Marroig and Cheverud 2005a; Cardini and Polly 2013; Porto, et al. 2013). Size- dependent ontogenesis (both pre- and postnatal) has shown to be allometric for some mammal groups, tracing a direct link between phenotypic diversity and size-correlated shape change (Marroig and Cheverud 2005b; Cardini and Polly 2013; Porto, et al.

2013). It is hypothesised that size can act as an evolutionary mechanism favouring most parsimonious pathways as a line of least evolutionary resistance (Marroig and

Cheverud 2005b). Because changes in size are inherent in developmental processes,

114 one could argue that allometric disparity in ontogenetic trajectories is positively correlated with size variability, promoting phenotypic heterogeneity (Zelditch, et al.

2004; Marroig and Cheverud 2005b; Cardini and Polly 2013; Porto, et al. 2013).

Increase in the amount of shape change with size in C. sphinx and M. schreibersii

(bigger species in our study within each suborder) could indicate that larger size promotes evolvability within the phenotypic space of each suborder (Marroig and

Cheverud 2005b). However, our sample did not include the largest species within each suborder and our results could be tested by including additional bat species to more comprehensively sample the trophic and phenotypic diversity found in Chiroptera.

Significantly high allometric disparity in C. sphinx may support our size-dependent interpretation, or reflect a distinctive pattern based on trophic differences (i.e. C. sphinx is the only frugivorous species in our sample).

Conclusions

This study represents a novel approach to study the ontogenetic basis of phenotypic variability in bats, focusing on prenatal postcranial development, a difficult and understudied approach for non-model species. Based on our results we conclude: 1) differences in the prenatal morphogenesis of the postcranium can be found across bats; 2) these differences are evident as taxon-specific patterns of acceleration in the growth of specific bones ; 3) there is a significant phylogenetic signal in the variability of allometric trajectories of both suborders, showing a clear differentiation in the allometric subspace that each occupies; 4) size might have shaped the evolvability of developmental processes in bats; and 5) our study is the first to propose a measurable role of ontogeny in postcranial morphological differences among bats.

115

References

Adams DC. 2014. A generalized K statistic for estimating phylogenetic signal from shape and other high-dimensional multivariate data. Syst Biol 63:685-697. Adams DC, Collyer ML. 2018. Phylogenetic ANOVA: Group-clade aggregation, biological challenges, and a refined permutation procedure. Evolution 72:1204-1215. Adams DC, Collyer ML, Kaliontzopoulou A, Sherratt E. 2017. Geomorph: software for geometric morphometric analyses. R package. Version 3.0.5: https://cran.r- project.org/package=geomorph. Adams DC, Nistri A. 2010. Ontogenetic convergence and evolution of foot morphology in European cave salamanders (Family: Plethodontidae). BMC Evolutionary Biology 10. Adams DC, Otárola-Castillo E, Paradis E. 2013. Geomorph: an R package for the collection and analysis of geometric morphometric shape data. Methods in Ecology and Evolution 4:393-399. Adams DC, Rohlf FJ. 2000. Ecological character displacement in Plethodon: Biomechanical differences found from a geometric morphometric study. Proceedings of the National Academy of Sciences 97:4106-4111. Adams RA. 1992. Comparative Skeletogenesis of the Forearm of the Little Brown Bat (Myotis-Lucifugus) and the Norway Rat (Rattus-Norvegicus). Journal of Morphology 214:251-260. Adams RA. 2008. Morphogenesis in Bat Wings: Linking Development, Evolution and Ecology. Cells Tissues Organs 187:13-23. Adams RA. 2000. Wing ontogeny, shifting niche dimensions, and adaptive landscapes. In: Adams RA, Pedersen SC, editors. Ontogeny, Functional Ecology and Evolution of Bats. Cambridge: Cambridge University Press. p. 275-316. Adams RA, Shaw JB. 2013. Time’s Arrow in the Evolutionary Development of Bat Flight. Bat evolution, ecology and conservation: 21-46. Aldridge HDJN. 1987. Turning Flight of Bats. Journal of Experimental Biology 128:419- 425. Amador LI, Almeida FC, Giannini NP. 2019. Evolution of Traditional Aerodynamic Variables in Bats (Mammalia: Chiroptera) within a Comprehensive Phylogenetic Framework. Journal of Mammalian Evolution 27: 549-561.. Amador LI, Arevalo RLM, Almeida FC, Catalano SA, Giannini NP. 2018. Bat Systematics in the Light of Unconstrained Analyses of a Comprehensive Molecular Supermatrix. Journal of Mammalian Evolution 25:37-70. Amador LI, Giannini NP, Simmons NB, Abdala V. 2018. Morphology and Evolution of Sesamoid Elements in Bats (Mammalia: Chiroptera). American Museum Novitates 2018:1-40, 40. Amador LI, Simmons NB, Giannini NP. 2019. Aerodynamic reconstruction of the primitive fossil bat Onychonycteris finneyi (Mammalia: Chiroptera). Biology Letters 15:20180857. Arbour JH, Curtis AA, Santana SE. 2019. Signatures of echolocation and dietary ecology in the adaptive evolution of skull shape in bats. Nat Commun 10:2036. Arita HT, Fenton MB. 1997. Flight and echlocation in the ecology and evolution of bats. Trends in Ecology & Evolution 12:53-58.

116

Arita HT, Vargas-Barón J, Villalobos F. 2014. Latitudinal gradients of genus richness and endemism and the diversification of New World bats. Ecography 37: 1024-1033. Armfield BA, Zheng Z, Bajpai S, Vinyard CJ, Thewissen J. 2013. Development and evolution of the unique cetacean dentition. PeerJ 1:e24. Bahlman JW, Price‐Waldman RM, Lippe HW, Breuer KS, Swartz SM. 2016. Simplifying a wing: diversity and functional consequences of digital joint reduction in bat wings. Journal of Anatomy 229:114-127. Bastide P, Ane C, Robin S, Mariadassou M. 2018. Inference of Adaptive Shifts for Multivariate Correlated Traits. Syst Biol 67:662-680. Bell E, Andres B, Goswami A. 2011. Integration and dissociation of limb elements in flying vertebrates: a comparison of pterosaurs, birds and bats. J Evol Biol 24:2586- 2599. Bergou AJ, Swartz SM, Vejdani H, Riskin DK, Reimnitz L, Taubin G, Breuer KS. 2015. Falling with Style: Bats Perform Complex Aerial Rotations by Adjusting Wing Inertia. PLOS Biology 13: e1002297. Bininda-Emonds OR, Cardillo M, Jones KE, MacPhee RD, Beck RM, Grenyer R, Price SA, Vos RA, Gittleman JL, Purvis A. 2007. The delayed rise of present-day mammals. Nature 446:507-512. Bininda-Emonds OR, Jeffery JE, Sánchez-Villagra MR, Hanken J, Colbert M, Pieau C, Selwood L, Ten Cate C, Raynaud A, Osabutey CK, et al. 2007. Forelimb-hindlimb developmental timing changes across tetrapod phylogeny. BMC Evol Biol 7:182. Boerma DB, Barrantes JP, Chung C, Chaverri G, Swartz SM. 2019. Specialized landing maneuvers in Spix's disk-winged bats (Thyroptera tricolor) reveal linkage between roosting ecology and landing biomechanics. The Journal of Experimental Biology:jeb.204024. Boerma DB, Breuer KS, Treskatis TL, Swartz SM. 2019. Wings as inertial appendages: how bats recover from aerial stumbles. The Journal of Experimental Biology:jeb.204255. Boessenecker RW, Fordyce RE. 2015. Anatomy, feeding ecology, and ontogeny of a transitional baleen whale: a new genus and species of Eomysticetidae (Mammalia: Cetacea) from the Oligocene of New Zealand. PeerJ 3:e1129. Bohmer C, Amson E, Arnold P, van Heteren AH, Nyakatura JA. 2018. Homeotic transformations reflect departure from the mammalian 'rule of seven' cervical vertebrae in sloths: inferences on the Hox code and morphological modularity of the mammalian neck. BMC Evol Biol 18:84. Boonman A, Bumrungsri S, Yovel Y. 2014. Nonecholocating fruit bats produce biosonar clicks with their wings. Current biology : CB 24:2962-2967. Brown EE, Cashmore DD, Simmons NB, Butler RJ. 2019. Quantifying the completeness of the bat fossil record. Palaeontology 62: 757-776. Buchholtz EA, Stepien CC. 2009. Anatomical transformation in mammals: developmental origin of aberrant cervical anatomy in tree sloths. Evolution & Development 11:69-79. Burke AC, Nowicki JL. 2003. A new view of patterning domains in the vertebrate mesoderm. Developmental Cell 4:159-165. Butler RJ, Benson RBJ, Barrett PM. 2013. Pterosaur diversity: Untangling the influence of sampling biases, Lagerstätten, and genuine biodiversity signals. Palaeogeography, Palaeoclimatology, Palaeoecology 372:78-87. 117

Canals M, Iriarte-Diaz J, Grossi B. 2011. Biomechanical, Respiratory and Cardiovascular Adaptations of Bats and the Case of the Small Community of Bats in Chile. In: Klika V, editor. Biomechanics in Applications: InTech. p. 299-322. Canty A, Ripley BS. 2017. boot: Bootstrap R (S-Plus) Functions. Version 1.3-20. Cardini A, O’Higgins P. 2005. Post-natal ontogeny of the mandible and ventral cranium in Marmota species (Rodentia, Sciuridae): allometry and phylogeny. Zoomorphology 124:189-203. Cardini A, Polly PD. 2013. Larger mammals have longer faces because of size-related constraints on skull form. Nature Communications 4. Carter RT, Adams RA. 2015. Postnatal ontogeny of the cochlea and flight ability in Jamaican fruit bats (Phyllostomidae) with implications for the evolution of echolocation. J Anat 226:301-308. Cheney JA, Ton D, Konow N, Riskin DK, Breuer KS, Swartz SM. 2014. Hindlimb motion during steady flight of the lesser dog-faced fruit bat, Cynopterus brachyotis. PLoS ONE 9:e98093. Clavel J, Escarguel G, Merceron G. 2015. mvmorph: an r package for fitting multivariate evolutionary models to morphometric data. Methods in Ecology and Evolution 6:1311- 1319. Cobb SN, O’Higgins P. 2004. Hominins Do Not Share a Common Postnatal Facial Ontogenetic Shape Trajectory. J Exp Zool B Mol Dev Evol 302:302-321. Collyer ML, Sekora DJ, Adams DC. 2015. A method for analysis of phenotypic change for phenotypes described by high-dimensional data. Heredity (Edinb) 115:357-365. Cooper KL, Tabin CJ. 2008. Understanding of bat wing evolution takes flight. Genes Dev 22:121-124. Cooper L, Cretekos CJ, Sears KE. 2012. The evolution and development of mammalian flight. Wiley Interdisciplinary Reviews: Developmental Biology 1:773-779. Cooper LN, Sears KE, Armfield BA, Kala B, Hubler M, Thewissen JGM. 2017. Review and experimental evaluation of the embryonic development and evolutionary history of development and hyperphalangy in dolphins (Cetacea: Mammalia). Genesis 56: e23076. Cretekos CJ, Wang Y, Green ED, Martin JF, Rasweiler JJt, Behringer RR. 2008. Regulatory divergence modifies limb length between mammals. Genes Dev 22:141- 151. Cretekos CJ, Weatherbee SD, Chen CH, Badwaik NK, Niswander L, Behringer RR, Rasweiler JJt. 2005. Embryonic staging system for the short-tailed fruit bat, Carollia perspicillata, a model organism for the mammalian order Chiroptera, based upon timed pregnancies in captive-bred animals. Dev Dyn 233:721-738. Cubo J, Casinos A. 1998. The variation of the cross-sectional shape in the long bones of birds and mammals. Annales des Sciences Naturelles - Zoologie et Biologie Animale 19:51-62. Currey JD, Alexander MR. 1985. The thickness of the walls of tubular bones. Journal of Zoology 206:453-468. Datzmann T, von Helversen O, Mayer F. 2010. Evolution of nectarivory in phyllostomid bats (Phyllostomidae Gray, 1825, Chiroptera: Mammalia). BMC Evol Biol 10: 165. Davalos LM, Velazco PM, Warsi OM, Smits PD, Simmons NB. 2014. Integrating incomplete fossils by isolating conflicting signal in saturated and non-independent morphological characters. Syst Biol 63:582-600. 118 de Camargo NF, de Oliveira HFM. 2012. Sexual Dimorphism in Sturnira lilium (Chiroptera, Phyllostomidae): Can Pregnancy and Pup Carrying Be Responsible for Differences in Wing Shape? PLoS ONE 7. Dececchi TA, Larsson HC, Habib MB. 2016. The wings before the bird: an evaluation of flapping-based locomotory hypotheses in bird antecedents. PeerJ 4:e2159. Denzinger A, Schnitzler HU. 2013. Bat guilds, a concept to classify the highly diverse foraging and echolocation behaviors of microchiropteran bats. Front Physiol 4:164. Dickinson M. 2008. Animal locomotion: a new spin on bat flight. Curr Biol 18:R468-470. Dietz CL. 1973. Bat walking behavior. Journal of mammalogy 54: 790-792. Dobson GE. 1875. On the genus Scotophilus, with description of a new genus and species allied thereto. Proceedings of the Zoological Society of London 24:368-373. Dumont ER. 2010. Bone density and the lightweight skeletons of birds. Proceedings of the Royal Society B: Biological Sciences 277:2193-2198. Eckalbar WL, Schlebusch SA, Mason MK, Gill Z, Parker AV, Booker BM, Nishizaki S, Muswamba-Nday C, Terhune E, Nevonen KA, et al. 2016. Transcriptomic and epigenomic characterization of the developing bat wing. Nature genetics 48:528-536. Eghbali H, Shahabi S, Najafi N, Mehdizadeh R, Yousefi S, Sharifi M. 2017. Postnatal growth, wing development and age estimations in the Mediterranean horseshoe bat Rhinolophus euryale (Chiroptera: Rhinolophidae) in Kerend cave, western Iran. Mammalia 82: 276-287. Eiting TP, Gunnell GF. 2009. Global Completeness of the Bat Fossil Record. Journal of Mammalian Evolution 16:151-173. Ellington CP, vandenBerg C, Willmott AP, Thomas ALR. 1996. Leading-edge vortices in insect flight. Nature 384:626-630. Esquerré D, Sherratt E, Keogh JS. 2017. Evolution of extreme ontogenetic allometric diversity and heterochrony in pythons, a clade of giant and dwarf snakes. Evolution 71:2829-2844. Falkowski PG, Katz ME, Milligan AJ, Fennel K, Cramer BS, Aubry MP, Berner RA, Novacek MJ, Zapol WM. 2005. The rise of oxygen over the past 205 million years and the evolution of large placental mammals. Science 309:2202-2204. Faure PA, Barclay RMR. 1994. Substrate-Gleaning Versus Aerial-Hawking - Plasticity in the Foraging and Echolocation Behavior of the Long-Eared Bat, Myotis-Evotis. Journal of Comparative Physiology a-Sensory Neural and Behavioral Physiology 174:651-660. Felsenstein J. 1973. Maximum-likelihood estimation of evolutionary trees from continuous characters. American Journal of Human Genetics 25:471-492. Fenton MB. 2010. Convergences in the diversification of bats. Current Zoology (formerly Acta Zoologica Sinica) 56: 454-468. Figueirido B, Janis CM, Perez-Claros JA, De Renzi M, Palmqvist P. 2012. Cenozoic climate change influences mammalian evolutionary dynamics. Proc Natl Acad Sci U S A 109:722-727. Foley NM, Thong VD, Soisook P, Goodman SM, Armstrong KN, Jacobs DS, Puechmaille SJ, Teeling EC. 2015. How and why overcome the impediments to resolution: lessons from rhinolophid and hipposiderid bats. Mol Biol Evol 32:313-333. Fong JJ, Brown JM, Fujita MK, Boussau B. 2012. A Phylogenomic Approach to Vertebrate Phylogeny Supports a Turtle-Archosaur Affinity and a Possible Paraphyletic Lissamphibia. PLoS ONE 7.

119

Frederich B, Vandewalle P. 2011. Bipartite life cycle of coral reef fishes promotes increasing shape disparity of the head skeleton during ontogeny: an example from damselfishes (Pomacentridae). BMC Evolutionary Biology 11: 82. Frelat MA, Katina S, Weber GW, Bookstein FL. 2012. Technical note: A novel geometric morphometric approach to the study of long bone shape variation. American Journal of Physical Anthropology 149:628-638. Friedman ST, Martinez CM, Price SA, Wainwright PC. 2019. The influence of size on body shape diversification across Indo‐Pacific shore fishes. Evolution 73: 1873-1884. Gardiner JD, Codd JR, Nudds RL. 2011. An association between ear and tail morphologies of bats and their foraging style. Canadian Journal of Zoology 89:90-99. Gaudioso PJ, Diaz MM, Barquez RM. 2017. Morphology of the axial skeleton of seven bat genera (Chiroptera: Phyllostomidae). An Acad Bras Cienc 89:2341-2358. Geiger M, Forasiepi AM, Koyabu D, Sánchez-Villagra MR. 2014. Heterochrony and post- natal growth in mammals--an examination of growth plates in limbs. J Evol Biol 27:98- 115. Gerber S. 2013. On the Relationship between the Macroevolutionary Trajectories of Morphological Integration and Morphological Disparity. PLoS ONE 8. Gerber S, Eble GJ, Neige P. 2008. Allometric space and allometric disparity: a developmental perspective in the macroevolutionary analysis of morphological disparity. Evolution 62:1450-1457. Giannini N, Goswami A, Sánchez-Villagra MR. 2006. Development of integumentary structures in Rousettus amplexicaudatus (Mammalia: Chiroptera: Pteropodidae) during late-embryonic and fetal stages. Journal of mammalogy 95:993-1001. Giannini NP, Abdala F, Flores DA. 2004. Comparative postnatal ontogeny of the skull in Dromiciops gliroides (Marsupialia : Microbiotheriidae). American Museum Novitates:1- 17. Goswami A. 2007. Cranial modularity and sequence heterochrony in mammals. Evol Dev 9:290-298. Goswami A, Binder WJ, Meachen J, O'Keefe FR. 2015. The fossil record of phenotypic integration and modularity: A deep-time perspective on developmental and evolutionary dynamics. Proc Natl Acad Sci U S A 112:4891-4896. Goswami A, Smaers JB, Soligo C, Polly PD. 2014. The macroevolutionary consequences of phenotypic integration: from development to deep time. Philos Trans R Soc Lond B Biol Sci 369:20130254. Goswami A, Weisbecker V, Sánchez-Villagra MR. 2009. Developmental modularity and the marsupial-placental dichotomy. J Exp Zool B Mol Dev Evol 312B:186-195. Gould SJ. 1998. On embryos and ancestors. Nat Hist 107:20-22, 58-64. Gunnell GF, Simmons NB, Seiffert ER. 2014. New Myzopodidae (Chiroptera) from the late Paleogene of Egypt: emended family diagnosis and biogeographic origins of Noctilionoidea. PLoS ONE 9:e86712. Gunnell GF, Worsham SR, Seiffert ER, Simons EL. 2009. Vampyravus orientalis Schlosser (Chiroptera) from the Early Oligocene (Rupelian), Fayum, Egypt - body mass, humeral morphology and affinitiesd. Acta Chiropterologica 11:271-278. Habib MB, Ruff CB. 2008. The effects of locomotion on the structural characteristics of avian limb bones. Zoological Journal of the Linnean Society 153:601-624. Hafner M, Remsen Jr J, Lanyon S. 1984. Bird Versus Mammal Morphological Diversity. Evolution 38: 1154-1156. 120

Hahsler M, Chelluboina S, Hornik K, Buchta C. 2011. The arules R-Package Ecosystem: Analyzing Interesting Patterns from Large Transaction Data Sets. J. Mach. Learn. Res. 12:2021-2025. Hakansson J, Hedenstrom A, Winter Y, Johansson LC. 2015. The wake of hovering flight in bats. J R Soc Interface 12:20150357. Hall BK. 2000. Evo-devo or devo-evo--does it matter. Evol Dev 2:177-178. Halley AC. 2016. Prenatal brain-body allometry in mammals. Brain Behavior Evolution 88:14-24. Hallgrímsson B, Jamniczky HA, Young NM, Rolian C, Schmidt-Ott U, Marcucio RS. 2012. The generation of variation and the developmental basis for evolutionary novelty. J Exp Zool B Mol Dev Evol 318:501-517. Hammer O, Harper DAT, Ryan PD. 2001. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica 4:9pp. Hand SJ, Beck RMD, Archer M, Simmons NB, Gunnell GF, Scofield RP, Tennyson AJD, De Pietri VL, Salisbury SW, Worthy TH. 2018. A new, large-bodied omnivorous bat (Noctilionoidea: Mystacinidae) reveals lost morphological and ecological diversity since the Miocene in New Zealand. Sci Rep 8:235. Hand SJ, Lee DE, Worthy TH, Archer M, Worthy JP, Tennyson AJ, Salisbury SW, Scofield RP, Mildenhall DC, Kennedy EM, et al. 2015. Miocene Fossils Reveal Ancient Roots for New Zealand's Endemic Mystacina (Chiroptera) and Its Rainforest Habitat. PLoS ONE 10:e0128871. Hand SJ, Sigé B. 2018. A new archaic bat (Chiroptera: Archaeonycteridae) from an Early Eocene forest in the Paris Basin. Historical Biology 30:227-236. Hand SJ, Sigé B, Archer M, Gunnell GF, Simmons NB. 2015. A New Early Eocene (Ypresian) Bat from Pourcy, Paris Basin, France, with Comments on Patterns of Diversity in the Earliest Chiropterans. Journal of Mammalian Evolution 22:343-354. Hand SJ, Weisbecker V, Beck RM, Archer M, Godthelp H, Tennyson AJ, Worthy TH. 2009. Bats that walk: a new evolutionary hypothesis for the terrestrial behaviour of New Zealand's endemic mystacinids. BMC Evol Biol 9:169. Hand SJ, Worthy TH, Archer M, Worthy JP, Tennyson AJ, Scofield RP. 2013. Miocene mystacinids (Chiroptera, Noctilionoidea) indicate a long history for endemic bats in New Zealand. Journal of Vertebrate Paleontology 33:1442-1448. Hanot P, Herrel A, Guintard C, Cornette R. 2018. The impact of artificial selection on morphological integration in the appendicular skeleton of domestic horses. J Anat. Hanot P, Herrel A, Guintard C, Cornette R. 2017. Morphological integration in the appendicular skeleton of two domestic taxa: the horse and donkey. Proc Biol Sci 284. Hansen TF. 1997. STABILIZING SELECTION AND THE COMPARATIVE ANALYSIS OF ADAPTATION. Evolution 51:1341-1351. Harmon LJ, Losos JB, Jonathan Davies T, Gillespie RG, Gittleman JL, Bryan Jennings W, Kozak KH, McPeek MA, Moreno-Roark F, Near TJ, et al. 2010. Early bursts of body size and shape evolution are rare in comparative data. Evolution 64:2385-2396. Hautier L, Bennett NC, Viljoen H, Howard L, Milinkovitch MC, Tzika AC, Goswami A, Asher RJ. 2013. Patterns of ossification in southern versus northern placental mammals. Evolution 67:1994-2010. Hautier L, Weisbecker V, Sánchez-Villagra MR, Goswami A, Asher RJ. 2010. Skeletal development in sloths and the evolution of mammalian vertebral patterning.

121

Proceedings of the National Academy of Sciences of the United States of America 107:18903-18908. Hedenstrom A, Johansson LC. 2015. Bat flight: aerodynamics, kinematics and flight morphology. Journal of Experimental Biology 218:653-663. Hedenstrom A, Johansson LC, Spedding GR. 2009. Bird or bat: comparing airframe design and flight performance. Bioinspir Biomim 4:015001. Hedrick BP, Mutumi GL, Munteanu VD, Sadier A, Davies KTJ, Rossiter SJ, Sears KE, Dávalos LM, Dumont E. 2019. Morphological Diversification under High Integration in a Hyper Diverse Mammal Clade. Journal of Mammalian Evolution 27, 563–575. Henderson DM. 2010. Pterosaur body mass estimates from three-dimensional mathematical slicing. Journal of Vertebrate Paleontology 30:768-785. Hendrikse JL, Parsons TE, Hallgrimsson B. 2007. Evolvability as the proper focus of evolutionary developmental biology. Evol Dev 9:393-401. Hockman D, Mason MK, Jacobs DS, Illing N. 2009. The role of early development in mammalian limb diversification: a descriptive comparison of early limb development between the Natal long-fingered bat (Miniopterus natalensis) and the mouse (Mus musculus). Dev Dyn 238:965-979. Hornung RW, Reed LD. 1990. Estimation of Average Concentration in the Presence of Nondetectable Values. Applied Occupational and Environmental Hygiene 5:46-51. Houssaye A, Botton-Divet L. 2018. From land to water: evolutionary changes in long bone microanatomy of otters (Mammalia: Mustelidae). Biological Journal of the Linnean Society 125:240-249. Houssaye A, Waskow K, Hayashi S, Cornette R, Lee AH, Hutchinson JR. 2016. Biomechanical evolution of solid bones in large animals: a microanatomical investigation. Biological Journal of the Linnean Society 117:350-371. Iriarte-Diaz J, Riskin DK, Breuer KS, Swartz SM. 2012. Kinematic plasticity during flight in fruit bats: individual variability in response to loading. PLoS ONE 7:e36665. Iriarte-Diaz J, Riskin DK, Willis DJ, Breuer KS, Swartz SM. 2011. Whole-body kinematics of a fruit bat reveal the influence of wing inertia on body accelerations. J Exp Biol 214:1546-1553. Jarrin VP, Flores C, Salcedo J. 2010. Morphological variation in the short-tailed fruit bat (Carollia) in Ecuador, with comments on the practical and philosophical aspects of boundaries among species. Integr Zool 5:226-240. Johansson LC, Engel S, Kelber A, Heerenbrink MK, Hedenstrom A. 2013. Multiple leading edge vortices of unexpected strength in freely flying hawkmoth. Sci Rep 3:3264. Johansson LC, Hakansson J, Jakobsen L, Hedenstrom A. 2016. Ear-body lift and a novel thrust generating mechanism revealed by the complex wake of brown long-eared bats (Plecotus auritus). Sci Rep 6:24886. Johansson LC, Wolf M, Hedenstrom A. 2010. A quantitative comparison of bird and bat wakes. J R Soc Interface 7:61-66. Jones G. 1999. Scaling of echolocation call parameters in bats. Journal of Experimental Biology 202:3359-3367. Jones G, Teeling EC. 2006. The evolution of echolocation in bats. Trends Ecol Evol 21:149-156. Jones KE, Safi K. 2011. Ecology and evolution of mammalian biodiversity Introduction. Philosophical Transactions of the Royal Society B-Biological Sciences 366:2451-2461. 122

Jones PL, Page RA, Ratcliffe JM. 2016. To Scream or to Listen? Prey Detection and Discrimination in Animal-Eating Bats. In: Springer, editor. Bat Bioacoustics. p. 93-116. Kapusta A, Suh A, Freschotte C. 2017. Dynamics of genome size evolution in birds and mammals. Proceedings of the National Academy of Science 114:E1460–E1469. Kilbourne BM, Hutchinson JR. 2019. Morphological diversification of biomechanical traits: mustelid locomotor specializations and the of long bone cross- sectional morphology. BMC Evolutionary Biology 19:37. Kirkpatrick S. 1994. Scale Effects on the Stresses and Safety Factors in the Wing Bones of Birds and Bats. Journal of Experimental Biology 190:195-215. Klingenberg CP. 2009. Morphometric integration and modularity in configurations of landmarks: tools for evaluating a priori hypotheses. Evol Dev 11:405-421. Klingenberg CP. 2014. Studying morphological integration and modularity at multiple levels: concepts and analysis. Philos Trans R Soc Lond B Biol Sci 369:20130249. Klingenberg CP. 2010. There's something afoot in the evolution of ontogenies. BMC Evol Biol 10:221. Konow N, Cheney Jorn A, Roberts Thomas J, Waldman JRS, Swartz Sharon M. 2015. Spring or string: does tendon elastic action influence wing muscle mechanics in bat flight? Proceedings of the Royal Society B: Biological Sciences 282:20151832. Koyabu D. 2017. Prenatal postcranial development in two species of sympatric Japanese wood mice (Apodemus argenteus and A-speciosus): a comparison of arboreal versus terrestrial congeners. Journal of Veterinary Medical Science 79:952-956. Koyabu D, Endo H, Mitgutsch C, Suwa G, Catania KC. 2011. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. BMC EvoDevo: 2: 21. Koyabu D, Son N. 2014. Patterns of postcranial ossification and sequence heterochrony in bats: Life histories and developmental trade‐offs. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 322:607-618. Koyabu D, Werneburg I, Morimoto N, Zollikofer CPE, Forasiepi A, Endo H, Kimura J, Ohdachi SD, Son NT, Sánchez-Villagra MR. 2014. Mammalian skull heterochrony reveals modular evolution and a link between cranial development and brain size. Nat Commun 5:3625. Kunz TH, Adams RA, Wood WR. 2009. Methods for assessing size at birth and postnatal growth and development in bats. In: Kunz TH, Parsons S, editors. Ecological and behavioral methods for the study of bats. Baltimore: The Johns Hopkins University Press. p. 273–314. Law CJ. 2019. Evolutionary shifts in extant mustelid (Mustelidae: Carnivora) cranial shape, body size and body shape coincide with the Mid-Miocene Climate Transition. Biology Letters 15:20190155. Lee AH, Simons E. 2015. Wing bone laminarity is not an adaptation for torsional resistance in bats. PeerJ 3: e823. Lei M, Dong D. 2016. Phylogenomic analyses of bat subordinal relationships based on transcriptome data. Sci Rep 6:27726. Lentink D, Dickson WB, van Leeuwen JL, Dickinson MH. 2009. Leading-Edge Vortices Elevate Lift of Autorotating Plant Seeds. Science 324:1438-1440.

123

Lindhe Norberg UM, Winter Y. 2006. Wing beat kinematics of a nectar-feeding bat, Glossophaga soricina, flying at different flight speeds and Strouhal numbers. J Exp Biol 209:3887-3897. López-Aguirre C, Hand SJ, Koyabu D, Son NT, Wilson LAB. 2019a. Postcranial heterochrony, modularity, integration and disparity in the prenatal ossification in bats (Chiroptera). BMC Evolutionary Biology 19:75. López-Aguirre C, Hand SJ, Koyabu D, Son NT, Wilson LAB. 2019b. Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera). Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 332:36-49. López-Aguirre C, Hand SJ, Laffan SW, Archer M. 2018. Phylogenetic diversity, types of endemism and the evolutionary history of New World bats. Ecography 41:1955-1966. Louzada NSV, Nogueira MR, Pessôa LM. 2019. Comparative morphology and scaling of the femur in yangochiropteran bats. Journal of Anatomy 235:124-150. Love AC, Raff RA. 2003. Knowing your ancestors: themes in the history of evo-devo. Evol Dev 5:327-330. Luo ZX, Yuan CX, Meng QJ, Ji Q. 2011. A eutherian mammal and divergence of marsupials and placentals. Nature 476:442-445. MacAyeal LC, Riskin DK, Swartz SM, Breuer KS. 2011. Climbing flight performance and load carrying in lesser dog-faced fruit bats (Cynopterus brachyotis). J Exp Biol 214:786- 793. Machado FA, Zahn TMG, Marroig G. 2018. Evolution of morphological integration in the skull of Carnivora (Mammalia): Changes in Canidae lead to increased evolutionary potential of facial traits. Evolution. Machado JP, Johnson WE, Gilbert MT, Zhang G, Jarvis ED, O'Brien SJ, Antunes A. 2016. Bone-associated gene evolution and the origin of flight in birds. BMC Genomics 17:371. Macintosh AA, Pinhasi R, Stock JT. 2014. Lower limb skeletal biomechanics track long- term decline in mobility across similar to 6150 years of agriculture in Central Europe. Journal of Archaeological Science 52:376-390. Macintosh AA, Pinhasi R, Stock JT. 2017. Prehistoric women's manual labor exceeded that of athletes through the first 5500 years of farming in Central Europe. Science Advances 3: eaao3893. Maestri R, Monteiro LR, Fornel R, Upham NS, Patterson BD, de Freitas TRO. 2017. The ecology of a continental evolutionary radiation: Is the radiation of sigmodontine rodents adaptive? Evolution 71:610-632. Marcé-Nogué J, Püschel TA, Kaiser TM. 2017. A biomechanical approach to understand the ecomorphological relationship between primate mandibles and diet. Scientific Reports 7:8364. Marchetti K, Price T, Richman A. 1995. Correlates of Wing Morphology with Foraging Behaviour and Migration Distance in the Genus Phylloscopus. Journal of Avian Biology 26:177-181. Marelli CA, Simons EL. 2014. Microstructure and cross-sectional shape of limb bones in Great Horned Owls and Red-tailed Hawks: how do these features relate to differences in flight and hunting behavior? PLoS ONE 9:e106094.

124

Marinello MM, Bernard E. 2014. Wing morphology of Neotropical bats: a quantitative and qualitative analysis with implications for habitat use. Canadian Journal of Zoology 92:141-147. Marroig G, Cheverud JM. 2005. Size as a line of least evolutionary resistance: Diet and adaptive morphological radiation in new world monkeys. Evolution 59:1128-1142. Marroig G, Shirai LT, Porto A, de Oliveira FB, De Conto V. 2009. The Evolution of Modularity in the Mammalian Skull II: Evolutionary Consequences. Evolutionary Biology 36:136-148. Marroig G, Vivo M, Cheverud JM. 2003. Cranial evolution in sakis (Pithecia, Platyrrhini) II: evolutionary processes and morphological integration. Journal of Evolutionary Biology 17:144-155. Maxwell EE, Harrison LB. 2009. Methods for the analysis of developmental sequence data. Evol Dev 11:109-119. Maxwell EE, Harrison LB, Larsson HCE. 2010. Assessing the phylogenetic utility of sequence heterochrony: evolution of avian ossification sequences as a case study. Zoology 113:57-66. McElroy EJ, Hickey KL, Reilly SM. 2008. The correlated evolution of biomechanics, gait and foraging mode in lizards. Journal of Experimental Biology 211:1029-1040. McGowan AJ, Dyke GJ. 2007. A morphospace-based test for competitive exclusion among flying vertebrates: did birds, bats and pterosaurs get in each other's space? J Evol Biol 20:1230-1236. McNamara KJ. 2012. Heterochrony: the Evolution of Development. Evolution: Education and Outreach 5:203-218. Milinkovitch MC, Tzika A. 2007. Escaping the mouse trap: The selection of new evo- devo model species. Journal of Experimental Zoology Part B-Molecular and Developmental Evolution 308b:337-346. Morales AE, Ruedi M, Field K, Carstens BC. 2019. Diversification rates have no effect on the convergent evolution of foraging strategies in the most speciose genus of bats, Myotis. Evolution 73:2263-2280. Mosimann JE, James FC. 1979. New Statistical Methods for Allometry with Application to Florida Red-Winged Blackbirds. Evolution 33:444-459. Moyers Arévalo RL, Amador LI, Almeida FC, Giannini NP. 2018. Evolution of Body Mass in Bats: Insights from a Large Supermatrix Phylogeny. Journal of Mammalian Evolution 27: 123-138. Muijres FT, Christoffer Johansson L, Winter Y, Hedenstrom A. 2014. Leading edge vortices in lesser long-nosed bats occurring at slow but not fast flight speeds. Bioinspir Biomim 9:025006. Muijres FT, Johansson LC, Barfield R, Wolf M, Spedding GR, Hedenstrom A. 2008. Leading-Edge Vortex Improves Lift in Slow-Flying Bats. Science 319: 1250-1253. Muijres FT, Johansson LC, Bowlin MS, Winter Y, Hedenstrom A. 2012. Comparing aerodynamic efficiency in birds and bats suggests better flight performance in birds. PLOS ONE 7:e37335. Muijres FT, Johansson LC, Hedenstrom A. 2012. Leading edge vortex in a slow-flying passerine. Biol Lett 8:554-557. Muijres FT, Johansson LC, Winter Y, Hedenstrom A. 2011. Comparative aerodynamic performance of flapping flight in two bat species using time-resolved wake visualization. J R Soc Interface 8:1418-1428. 125

Muñoz MM, Hu Y, Anderson PSL, Patek SN. 2018. Strong biomechanical relationships bias the tempo and mode of morphological evolution. eLife 7:e37621. Nojiri T, Werneburg I, Son NT, Tu VT, Sasaki T, Maekawa Y, Koyabu D. 2018. Prenatal cranial bone development of Thomas's horseshoe bat (Rhinolophus thomasi): with special reference to petrosal morphology. J Morphol 279: 809-827. Norberg RA, Rayner JMV. 1987. Ecological Morphology and Flight in Bats (Mammalia; Chiroptera): Wing Adaptations, Flight Performance, Foraging Strategy and Echolocation. Philos Trans R Soc Lond B Biol Sci 316:335-427. Norberg UM. 1990. Vertebrate Flight, Mechanics, Physiology, Morphology, Ecology and Evolution: Springer-Verlag. Norberg UM, Norberg RA. 2012. Scaling of wingbeat frequency with body mass in bats and limits to maximum bat size. J Exp Biol 215:711-722. Organ CL, Cooper LN, Hieronymus TL. 2015. Macroevolutionary developmental biology: Embryos, fossils, and phylogenies. Dev Dyn 244:1184-1192. Organ CL, Shedlock AM. 2009. Palaeogenomics of pterosaurs and the evolution of small genome size in flying vertebrates. Biol Lett 5:47-50. Oyston JW, Hughes M, Wagner PJ. 2015. What limits the morphological disparity of clades? Interface Focus 5: 20150042. Panyutina A, Korzun LP, Kuznetsov AN. 2015. Flight of Mammals: From Terrestrial Limbs to Wings: Springer, New York. Papadimitriou HM, Swartz SM, Kunz TH. 1996. Ontogenetic and anatomic variation in mineralization of the wing skeleton of the Mexican free-tailed bat, brasiliensis. Journal of Zoology 240:411-426. Patel BA, Ruff CB, Simons EL, Organ JM. 2013. Humeral cross-sectional shape in suspensory primates and sloths. Anat Rec (Hoboken) 296:545-556. Pavlicev M, Cheverud JM, Wagner GP. 2009. Measuring Morphological Integration Using Eigenvalue Variance. Evolutionary Biology 36:157-170. Pennell MW, Eastman JM, Slater GJ, Brown JW, Uyeda JC, FitzJohn RG, Alfaro ME, Harmon LJ. 2014. geiger v2.0: an expanded suite of methods for fitting macroevolutionary models to phylogenetic trees. Bioinformatics 30:2216-2218. Petit F, Sears KE, Ahituv N. 2017. Limb development: a paradigm of gene regulation. Nature Reviews Genetics 18: 245-258. Pieretti J, Gehrke AR, Schneider I, Adachi N, Nakamura T, Shubin NH. 2015. Organogenesis in deep time: A problem in genomics, development, and paleontology. Proc Natl Acad Sci U S A 112:4871-4876. Pimiento C, Cantalapiedra JL, Shimada K, Field DJ, Smaers JB. 2019. Evolutionary pathways toward gigantism in sharks and rays. Evolution 73:588-599. Piras P, Salvi D, Ferrara G, Maiorino L, Delfino M, Pedde L, Kotsakis T. 2011. The role of post-natal ontogeny in the evolution of phenotypic diversity in Podarcis lizards. Journal of Evolutionary Biology 24:2705-2720. Ponssa ML, Candioti MFV. 2012. Patterns of skull development in anurans: size and shape relationship during postmetamorphic cranial ontogeny in five species of the Leptodactylus fuscus Group (Anura: Leptodactylidae). Zoomorphology 131:349-362. Porto A, de Oliveira FB, Shirai LT, De Conto V, Marroig G. 2008. The Evolution of Modularity in the Mammalian Skull I: Morphological Integration Patterns and Magnitudes. Evolutionary Biology 36:118-135.

126

Porto A, Shirai LT, de Oliveira FB, Marroig G. 2013. Size Variation, Growth Strategies, and the Evolution of Modularity in the Mammalian Skull. Evolution 67:3305-3322. Pratt IV, Johnston JD, Walker E, Cooper DML. 2018. Interpreting the three-dimensional orientation of vascular canals and cross-sectional geometry of cortical bone in birds and bats. Journal of Anatomy 232: 931-942. Prentice KC, Ruta M, Benton MJ. 2011. Evolution of morphological disparity in pterosaurs. Journal of Systematic Palaeontology 9:337-353. Raff RA. 1996. The shape of life: genes, development, and the evolution of animal form: University of Chicago Press. Ramirez-Chaves HE, Wroe SW, Selwood L, Hinds LA, Leigh C, Koyabu D, Kardjilov N, Weisbecker V. 2016. Mammalian development does not recapitulate suspected key transformations in the evolutionary detachment of the mammalian middle ear. Proc Biol Sci 283. Randau M, Goswami A. 2018. Shape Covariation (or the Lack Thereof) Between Vertebrae and Other Skeletal Traits in Felids: The Whole is Not Always Greater than the Sum of Parts. Evolutionary Biology 45: 196-210. Randau M, Goswami A. 2017. Unravelling intravertebral integration, modularity and disparity in Felidae (Mammalia). Evol Dev 19:85-95. Rayner JM. 1988. The evolution of vertebrate flight. Biological Journal of the Linnean Society 34:269-287. Revell LJ. 2012. phytools: an R package for phylogenetic comparative biology (and other things). Methods in Ecology and Evolution 3:217-223. Reyes-Amaya N, Jerez A, Flores D. 2017. Morphology and Postnatal Development of Lower Hindlimbs in Desmodus rotundus (Chiroptera: Phyllostomidae): A Comparative Study. Anat Rec (Hoboken) 300:2150-2165. Riskin DK, Bergou A, Breuer KS, Swartz SM. 2012. Upstroke wing flexion and the inertial cost of bat flight. Proc Biol Sci 279:2945-2950. Riskin DK, Bertram JEA, Hermanson JW. 2005. Testing the hindlimb-strength hypothesis: non-aerial locomotion by Chiroptera is not constrained by the dimensions of the femur or tibia. Journal of Experimental Biology 208:1309-1319. Riskin DK, Fenton MB. 2001. Sticking ability in Spix's disk-winged bat, Thyroptera tricolor (Microchiroptera : Thyropteridae). Canadian Journal of Zoology 79:2261-2267. Riskin DK, Iriarte-Diaz J, Middleton KM, Breuer KS, Swartz SM. 2010. The effect of body size on the wing movements of pteropodid bats, with insights into thrust and lift production. J Exp Biol 213:4110-4122. Riskin DK, Parsons S, Schutt WA, Jr., Carter GG, Hermanson JW. 2006. Terrestrial locomotion of the New Zealand short-tailed bat Mystacina tuberculata and the common vampire bat Desmodus rotundus. J Exp Biol 209:1725-1736. Riskin DK, Racey PA. 2010. How do sucker-footed bats hold on, and why do they roost head-up? Biological Journal of the Linnean Society 99:233-240. Riskin DK, Willis DJ, Iriarte-Diaz J, Hedrick TL, Kostandov M, Chen J, Laidlaw DH, Breuer KS, Swartz SM. 2008. Quantifying the complexity of bat wing kinematics. J Theor Biol 254:604-615. Rojas D, Vale Á, Ferrero V, Navarro L. 2012. The role of frugivory in the diversification of bats in the Neotropics. Journal of Biogeography 39:1948-1960.

127

Ross D, Marcot JD, Betteridge KJ, Nascone-Yoder N, Bailey CS, Sears KE. 2013. Constraints on Mammalian forelimb development: insights from developmental disparity. Evolution 67:3645-3652. Rossoni DM, Assis APA, Giannini NP, Marroig G. 2017. Intense preceded the invasion of new adaptive zones during the radiation of New World leaf- nosed bats. Scientific Reports 7: 11076. Rossoni DM, Costa BMA, Giannini NP, Marroig G. 2019. A multiple peak adaptive landscape based on feeding strategies and roosting ecology shaped the evolution of cranial covariance structure and morphological differentiation in phyllostomid bats. Evolution 73:961-981. Ruff C. 1987. Sexual dimorphism in human lower limb bone structure: relationship to subsistence strategy and sexual division of labor. Journal of 16:391- 416. Ruth EB. 1932. A study of the development of the mammalian pelvis. Anat Rec (Hoboken) 53:207-225. Sánchez-Villagra MR. 2002. Comparative patterns of postcranial ontogeny in therian mammals: an analysis of relative timing of ossification events. J Exp Zool 294:264-273. Sánchez-Villagra MR, Goswami A, Weisbecker V, Mock O, Kuratani S. 2008. Conserved relative timing of cranial ossification patterns in early mammalian evolution. Evol Dev 10:519-530. Sánchez-Villagra MR, Goswami A, Weisbecker V, Mock O, Kuratani S. 2008. Conserved relative timing of cranial ossification patterns in early mammalian evolution. Evol Dev 10:519-530. Sánchez-Villagra MR, Segura V, Geiger M, Heck L, Veitschegger K, Flores D. 2017. On the lack of a universal pattern associated with mammalian domestication: differences in skull growth trajectories across phylogeny. R Soc Open Sci 4:170876. Santana SE, Cheung E. 2016. Go big or go fish: morphological specializations in carnivorous bats. Proceedings of the Royal Society B: Biological Sciences 283:20160615. Santana SE, Dumont ER, Davis JL. 2010. Mechanics of bite force production and its relationship to diet in bats. Functional Ecology 24:776-784. Santana SE, Grosse IR, Dumont ER. 2012b. Dietary hardness, loading behavior, and the evolution of skull form in bats. Evolution 66:2587-2598. Santana SE, Strait S, Dumont ER. 2011. The better to eat you with: functional correlates of tooth structure in bats. The better to eat you with: functional correlates of tooth structure in bats. Sato K, Sakamoto KQ, Watanuki Y, Takahashi A, Katsumata N, Bost CA, Weimerskirch H. 2009. Scaling of soaring seabirds and implications for flight abilities of giant pterosaurs. PLoS ONE 4:e5400. Schliemann H, Goodman SM. 2011. A new study on the structure and function of the adhesive organs of the Old World sucker-footed bat (Myzopoda: Myzopodidae) of Madagascar. Verh. Naturwiss. Ver. Hamburg. Schmieder DA, Benitez HA, Borissov IM, Fruciano C. 2015. Bat Species Comparisons Based on External Morphology: A Test of Traditional versus Geometric Morphometric Approaches. PLOS ONE 10:e0127043. Schulz M. 2000. Diet and foraging behavior of the golden-tipped bat, Kerivoula papuensis: A spider specialist? Journal of mammalogy 81:948-957. 128

Sears KE. 2014. Differences in Growth Generate the Diverse Palate Shapes of New World Leaf-Nosed Bats (Order Chiroptera, Family Phyllostomidae). Evolutionary Biology 41: 12-21. Sears KE, Behringer RR, Rasweiler JJ, Niswander LA. 2006. Development of bat flight: Morphologic and of bat wing digits. Proceedings of the National Academy of Sciences of the United States of America 103:6581-6586. Sears KE, Behringer RR, Rasweiler JJt, Niswander L. 2007. The Evolutionary and Developmental Basis of Parallel Reduction in Mammalian Zeugopod Elements. Am Nat 169:105-117. Shi JJ, Rabosky DL. 2015. dynamics during the global radiation of extant bats. Evolution 69:1528-1545. Shubin NH, Tabin CJ, Carroll S. 1997. Fossils, genes and the evolution of animal limbs. Nature 388:639–648. Simmons NB, Seymour KL, Habersetzer J, Gunnell GF. 2008. Primitive Early Eocene bat from Wyoming and the evolution of flight and echolocation. Nature 451:818-821. Simons EL. 2010. Forelimb skeletal morphology and flight mode evolution in pelecaniform birds. Zoology (Jena) 113:39-46. Simons EL, Hieronymus TL, O'Connor PM. 2011. Cross sectional geometry of the forelimb skeleton and flight mode in pelecaniform birds. J Morphol 272:958-971. Simons EL, O'Connor P M. 2012. Bone laminarity in the avian forelimb skeleton and its relationship to flight mode: testing functional interpretations. Anat Rec (Hoboken) 295:386-396. Slater GJ, Friscia AR. 2019. Hierarchy in adaptive radiation: A case study using the Carnivora (Mammalia). Evolution 73:524-539. Slater GJ, Price SA, Santini F, Alfaro ME. 2010. Diversity versus disparity and the radiation of modern cetaceans. Proceedings of the Royal Society B: Biological Sciences 277:3097-3104. Smith KK. 2001. Heterochrony revisited: the evolution of developmental sequences. Biological Journal of the Linnean Society 73:169-186. Smith KK. 2002. Sequence heterochrony and the evolution of development. J Morphol 252:82-97. Smith KK. 2003. Time's arrow: heterochrony and the evolution of development. International Journal of Developmental Biology 47:613-621. Spielmann M, Brancati F, Krawitz PM, Robinson PN, Ibrahim DM, Franke M, Hecht J, Lohan S, Dathe K, Nardone AM, et al. 2012. Homeotic Arm-to-Leg Transformation Associated with Genomic Rearrangements at the PITX1 Locus. American Journal of Human Genetics 91:629-635. Spoor CF, Zonneveld FW, Macho GA. 1993. Linear Measurements of Cortical Bone and Dental Enamel by Computed-Tomography - Applications and Problems. American Journal of Physical Anthropology 91:469-484. Springer MS. 2013. : bats united, microbats divided. Curr Biol 23:R999- 1001. Stanchak KE, Arbour JH, Santana SE. 2019. Anatomical diversification of a skeletal novelty in bat feet. Evolution 73: 1591-1603. Stanchak KE, Santana SE. 2018. Assessment of the Hindlimb Membrane Musculature of Bats: Implications for Active Control of the Calcar. The Anatomical Record 301:441- 448. 129

Stange M, Aguirre-Fernandez G, Salzburger W, Sánchez-Villagra MR. 2018. Study of morphological variation of northern Neotropical Ariidae reveals conservatism despite macrohabitat transitions. BMC Evol Biol 18:38. Stroud JT, Losos JB. 2016. Ecological Opportunity and Adaptive Radiation. Annual Review of Ecology, Evolution, and Systematics 47:507-532. Swartz SM, Bennett MB, Carrier DR. 1992. Wing bone stresses in free flying bats and the evolution of skeletal design for flight. Nature 359: 726-729. Swartz SM, Konow N. 2015. Advances in the study of bat flight: the wing and the wind. Canadian Journal of Zoology 93: 977-990. Swartz SM, Middleton KM. 2008. Biomechanics of the bat limb skeleton: Scaling, material properties and mechanics. Cells Tissues Organs 187:59-84. Teeling EC, Madsen O, Van Den Bussche RA, de Jong WW, Stanhope M. 2002. paraphyly and the convergent evolution of a key innovation in Old World rhinolophoid microbats. Nature 99:1431-1436. Teeling EC, Scally M, Kao D, Romagnoli M, Springer MS, Stanhope M. 2000. Molecular evidence regarding the origin of echolocation and flight in bats. Nature 403:188–192. Teeling EC, Springer MS, Madsen O, Bates P, O’Brien J, Murphy WJ. 2005. A Molecular Phylogeny for Bats Illuminates Biogeography and the Fossil Record. Science 307:581- 584. Thean T, Kardjilov N, Asher RJ. 2017. Inner ear development in cetaceans. J Anat 230:249-261. Thiagavel J, Cechetto C, Santana SE, Jakobsen L, Warrant EJ, Ratcliffe JM. 2018. Auditory opportunity and visual constraint enabled the evolution of echolocation in bats. Nat Commun 9:98. Tokita M. 2015. How the pterosaur got its wings. Biol Rev Camb Philos Soc 90:1163- 1178. Tokita M, Abe T, Suzuki K. 2012. The developmental basis of bat wing muscle. Nature Communications 3: 1302. Tsagkogeorga G, Parker J, Stupka E, Cotton JA, Rossiter SJ. 2013. Phylogenomic analyses elucidate the evolutionary relationships of bats. Curr Biol 23:2262-2267. Uyeda JC, Caetano DS, Pennell MW. 2015. Comparative Analysis of Principal Components Can be Misleading. Systematic Biology 64:677-689. Veselka N, McErlain DD, Holdsworth DW, Eger JL, Chhem RK, Mason MJ, Brain KL, Faure PA, Fenton MB. 2010. A bony connection signals laryngeal echolocation in bats. Nature 463:939-942. Voeten D, Cubo J, de Margerie E, Roper M, Beyrand V, Bures S, Tafforeau P, Sánchez S. 2018. Wing bone geometry reveals active flight in Archaeopteryx. Nat Commun 9:923. Von Busse R, Hedenstrom A, Winter Y, Johansson LC. 2012. Kinematics and wing shape across flight speed in the bat, Leptonycteris yerbabuenae. Biol Open 1:1226-1238. Wagner GP, Altenberg L. 1996. Perspective: Complex adaptations and the evolution of evolvability. Evolution 50:967-976. Wagner GP, Pavlicev M, Cheverud JM. 2007. The road to modularity. Nature Reviews Genetics 8:921-931. Wallin J, Wilting J, Koseki H, Fritsch R, Christ B, Balling R. 1994. The role of Pax-1 in axial skeleton development. Development 120:1109-1121. Wang X, McGowan AJ, Dyke GJ. 2011. Avian wing proportions and flight styles: first step towards predicting the flight modes of mesozoic birds. PLOS ONE 6:e28672. 130

Wang XL, Kellner AWA, Jiang SX, Cheng X, Wang Q, Ma YX, Paidoula Y, Rodrigues T, Chen H, Sayao JM, et al. 2017. Egg accumulation with 3D embryos provides insight into the life history of a pterosaur. Science 358:1197-1201. Wang Z, Han N, Racey PA, Ru BH, He GM. 2010. A comparative study of prenatal development in Miniopterus schreibersii fuliginosus, Hipposideros armiger H. pratti. BMC Developmental Biology 10: 10. Wang Z, Zhu T, Xue H, Fang N, Zhang J, Zhang L, Pang J, Teeling EC, Zhang S. 2017. Prenatal development supports a single origin of laryngeal echolocation in bats. Nature Ecology & Evolution 1:0021. Warrick DR, Tobalske BW, Powers DR. 2009. Lift production in the hovering hummingbird. Proc Biol Sci 276:3747-3752. Watts P, Mitchell E, Swartz S. 2001. A computational model for estimating the mechanics of horizontal flapping flight in bats: Model description and validation. Journal of Experimental Biology 204:2873-2898. Weisbecker V, Goswami A, Wroe S, Sánchez-Villagra MR. 2008. Ossification heterochrony in the therian postcranial skeleton and the marsupial-placental dichotomy. Evolution 62:2027-2041. Werneburg I, Geiger M. 2017. Ontogeny of domestic dogs and the developmental foundations of carnivoran domestication. Journal of Mammalian Evolution 24:323-343. Werneburg I, Laurin M, Koyabu D, Sánchez-Villagra MR. 2016. Evolution of organogenesis and the origin of altriciality in mammals. Evolution & Development 18:229-244. Werneburg I, Tzika AC, Hautier L, Asher RJ, Milinkovitch MC, Sánchez‐Villagra MR. 2013. Development and embryonic staging in non‐model organisms: the case of an afrotherian mammal. Journal of Anatomy 222:2-18. Wilson L, Humphrey LT. 2015. A Virtual geometric morphometric approach to the quantification of long bone bilateral asymmetry and cross‐sectional shape. American Journal of Physical Anthropology 4: 541-556. Wilson LA. 2013a. Allometric disparity in rodent evolution. Ecol Evol 3:971-984. Wilson LA. 2011a. Comparison of prenatal and postnatal ontogeny: cranial allometry in the African striped mouse (Rhabdomys pumilio). Journal of mammalogy 92:407-420. Wilson LA, Ives R, Cardoso HF, Humphrey LT. 2015. Shape, size, and maturity trajectories of the human ilium. Am J Phys Anthropol 156:19-34. Wilson LA, Sánchez-Villagra MR. 2010a. Diversity trends and their ontogenetic basis: an exploration of allometric disparity in rodents. Proc Biol Sci 277:1227-1234. Wilson LAB. 2011c. The contribution of developmental palaeontology to extensions of evolutionary theory. Acta Zoologica 94:254-260. Wilson LAB. 2013b. Cranial Suture Closure Patterns in Sciuridae: Heterochrony and Modularity. Journal of Mammalian Evolution 21:257-268. Wilson LAB. 2018. The evolution of ontogenetic allometric trajectories in mammalian domestication. Evolution 72: 867-877. Wilson LAB, Sánchez-Villagra MR. 2010b. Diversity trends and their ontogenetic basis: an exploration of allometric disparity in rodents. Proceedings of the Royal Society B- Biological Sciences 277:1227-1234. Wilson LAB, Schradin C, Mitgutsch C, Galliari FC, Mess A, Sánchez-Villagra MR. 2010. Skeletogenesis and sequence heterochrony in rodent evolution, with particular

131 emphasis on the African striped mouse, Rhabdomys pumilio (Mammalia). Organisms Diversity & Evolution 10:243-258. Witton MP. 2008. A new approach to determining pterosaur body mass and its implications for pterosaur flight. Zitteliana 28:143-158. Witton MP, Habib MB. 2010. On the size and flight diversity of giant pterosaurs, the use of birds as pterosaur analogues and comments on pterosaur flightlessness. PLoS ONE 5:e13982. Wolf M, Johansson LC, von Busse R, Winter Y, Hedenstrom A. 2010. Kinematics of flight and the relationship to the vortex wake of a Pallas' long tongued bat (Glossophaga soricina). J Exp Biol 213:2142-2153. Wyant KA, Adams RA. 2007. Prenatal growth and development in the Angolan free- tailed bat, Mops condylurus (Chiroptera : Molossidae). Journal of mammalogy 88:1248-1251. Young NM, HallgrÍmsson B. 2005. Serial Homology and the Evolution of Mammalian Limb Covariation Structure. Evolution 59: 2691-2704. Zeileis A, Grothendieck G, Ryan JA, Ulrich JM, Andrews F. 2018. zoo: S3 Infrastructure for Regular and Irregular Time Series (Z's Ordered Observations). Version 1.8-1: R Development Core Team. Zelditch ML, Calamari ZT, Swiderski DL. 2016. Disparate Postnatal Ontogenies Do Not Add to the Shape Disparity of Infants. Evolutionary Biology 43:188-207. Zelditch ML, Li J, Tran LA, Swiderski DL. 2015. Relationships of diversity, disparity, and their evolutionary rates in squirrels (Sciuridae). Evolution 69:1284-1300. Zelditch ML, Lundrigan BL, Garland T, Jr. 2004. Developmental regulation of skull morphology. I. Ontogenetic dynamics of variance. Evol Dev 6:194-206. Zelditch ML, Lundrigan BL, Sheets HD, Garland T. 2003. Do precocial mammals develop at a faster rate? A comparison of rates of skull development in Sigmodon fulviventer and Mus musculus domesticus. Journal of Evolutionary Biology 16:708-720. Zelditch ML, Mezey J, Sheets HD, Lundrigan BL, Garland T, Jr. 2006. Developmental regulation of skull morphology II: ontogenetic dynamics of covariance. Evol Dev 8:46- 60.

132

Chapter 4 Prenatal developmental trajectories of fluctuating asymmetry in bat humeri3 Abstract

Fluctuating asymmetry (random fluctuations between right and left sides of the body) has been interpreted as an index to quantify both the developmental instabilities and homeostatic capabilities of organisms, linking the phenotypic and genotypic aspects of morphogenesis. However, most studies have been methodologically limited to study model organisms in juvenile and adult stages, missing prenatal trajectories of asymmetry that could better elucidate the developmental pathways controlling symmetric morphogenesis. In this study we quantified the presence and magnitude of asymmetry during the prenatal development of bats, focusing on the humerus. We deconstructed levels of asymmetry by measuring longitudinal and cross-sectional asymmetry of the humerus, using a combination of linear measurements and geometric morphometrics. We statistically tested the presence of different types of asymmetry and calculated the magnitude of size-controlled fluctuating asymmetry to assess developmental instability. Statistical support for the presence of fluctuating asymmetry was found for both longitudinal and cross-sectional asymmetry, explaining on average 24.61% of asymmetric variation. Significant directional asymmetry accounted for less than 1% of asymmetric variation. Both measures of fluctuating

3 A version of this manuscript has been submitted and it is currently under consideration for publication by Evolution & Development. * I certify that this manuscript was a direct result of my research towards this PhD, including the design, analysis and interpretation of results, and that reproduction in this thesis does not breach copyright regulations. I acknowledge my co-authors for their supervisory guidance and their contributions to the publication. 133 asymmetry decreased throughout ontogeny, but only cross-sectional asymmetry was significantly different across developmental stages. Finally, we did not find a correspondence between developmental patterns of longitudinal and cross-sectional asymmetry, indicating decoupled processes. We hypothesise fluctuating asymmetry is canalised to decrease across development, buffering developmental instability. We suggest functional differences between bone elongation and cross-sectional bone deposition linked to the newborn bat’s ecology (clinging to mother and the later acquisition of flight) may explain differences between longitudinal and cross-sectional asymmetry.

Introduction

Body plan symmetry was an important evolutionary driver of animal diversification

(Genikhovich and Technau 2017); the vast majority (more than 99%) of living animals show bilateral symmetry (i.e. body plans that have two orthogonal body axes and a left and right side that are mirror images) (Holló 2015). Bilateral symmetry evolved early in animal evolution, during the Ediacaran period, around 560 Mya (Pecoits, et al. 2012). It has been argued that bilaterality represented two major adaptive advantages during animal evolution that favoured the diversification of bilaterians by: 1) facilitating the functional specialisation of regions of the body (e.g. anterior end specialised for sensorial capabilities; Genikhovich and Technau (2017), and 2) increasing manoeuvrability capabilities during locomotion in three-dimensional space (i.e. manoeuvrability hypothesis; Holló and Novák (2012). Still, the debate around the evolution of bilaterality remains open and several novel hypotheses have been proposed (Balavoine and Adoutte 2003; Holló 2015; Genikhovich and Technau 2017).

134

Parallel to the study of body plan evolution, the study of how body symmetry is maintained has become an area of increasing interest (Kellner and Alford 2003; Leamy and Klingenberg 2005; Dongen 2006). For body plans that are naturally symmetrical, deviations from an “ideal” state have been interpreted as a signal of reduced fitness

(Dongen 2006). Accordingly, quantifying the deviation from perfect symmetry can indicate the amount of stress an organism undergoes and its homeostatic capacity (i.e. buffering of instabilities to maintain fitness) (Gummer and Brigham 1995; Albert and

Greene 1999; Aparicio and Bonal 2002; Coda, et al. 2017). The main regulatory mechanisms that influence phenotypic symmetry occur ontogenically, as it is during development that the genotypic and phenotypic mechanisms involved in morphogenesis can be destabilised by genetic or environmental stressors (i.e. developmental noise) (Hallgrímsson 1999; Hallgrímsson, et al. 2003; Kellner and Alford

2003; Carter and Houle 2011). Evolutionary studies have provided evidence for the heritability of an organism’s capacity to buffer developmental noise (developmental stability/instability, DI), suggesting that natural selection can act as a regulator of phenotypic asymmetry (Carter and Houle 2011; Tocts, et al. 2016).

Phenotypic asymmetry in animals with bilateral asymmetry can be easily quantified by computing morphometric differences between the right and left side of the body (R –

L) (Palmer 1994). However, advances in the theoretical framework associating body asymmetry with DI has enabled recognition of three different types of asymmetry, each with a different biological interpretation (Klingenberg 2015): fluctuating asymmetry (FA) is characterised as random deviations from “ideal” perfectly symmetrical phenotypes (Klingenberg 2015), directional asymmetry (DA) is described

135 as a natural tendency to have consistently asymmetrical phenotypes (i.e. one side always large than the other; (Klingenberg 2015), and antisymmetry (AS) represents a pattern where symmetrical phenotypes are least-favoured and asymmetry is equally distributed across both sides (Klingenberg 2015). All three types of asymmetry have also been described in mathematical terms: FA is characterised by a normal distribution of asymmetry values (R – L) along a value mean of zero, DA is described by a normal distribution of asymmetry values along a mean different to zero, and AS is identified where asymmetry values have a bimodal distribution and most values are different from zero (Palmer 1994; Klingenberg and McIntyre 1998; Klingenberg 2015).

Combining the biological and mathematical interpretations of FA, DA and AS, FA has been regularly used as a possible indicator of DI, although some studies argue that DA and AS can also be indicators of DI (Palmer 1994; Leamy and Klingenberg 2005). FA and

DI became of great interest to evolutionary biologists between the 1990s and early

2000s, followed by a steep decline in the number of studies (Dongen 2006). This decrease was possibly due to concerns suggesting the theoretical framework was insufficient to detect a real FA-DI link, which was accompanied by increasingly complex mathematical models postulated to test it (Palmer 1994; Dongen 2006; Klingenberg

2015).

Despite the fact that the conceptual framework behind the FA-DI link relies heavily on developmental mechanisms to explain its evolutionary basis, most studies are focused on adult specimens, possibly due to methodological limitations (Kellner and Alford

2003). Studies on the ontogeny of FA have postulated different hypotheses describing developmental patterns of asymmetry and possible evolutionary mechanisms shaping

136 them (Kellner and Alford 2003); small fluctuations during early growth are magnified during later morphogenesis (i.e. magnification of asymmetry hypothesis) (Kellner and

Alford 2003), side-biased environmental influences can skew growth towards asymmetrical phenotypes (i.e. directional external cues hypothesis) (Kellner and Alford

2003), accumulative growth of independent subunits will homogenise morphogenesis, reducing asymmetry throughout development (i.e. coin-toss hypothesis) (Kellner and

Alford 2003), or developmental feedback mechanisms will stabilise asymmetric growth between structures by either promoting or constraining growth (i.e. compensatory growth hypothesis) (Kellner and Alford 2003). Studies on the developmental basis of

FA have also been restricted in scope (most focused on postnatal development) and study groups (invertebrates and captive populations) (Hallgrímsson 1999; Klingenberg and Nijhout 1999; Hallgrímsson, et al. 2003; Leamy and Klingenberg 2005; Breuker, et al. 2006; Blackburn 2011; Carter and Houle 2011; Perchalski, et al. 2018), limiting our understanding of variation in wild non-model species and potential insight into the mechanisms controlling FA in early development. Based on the limited studies available, decreasing magnitudes of FA across prenatal development have been reported in support of morphogenetic mechanisms of compensatory growth

(Hallgrímsson 1999; Hallgrímsson, et al. 2003; Kellner and Alford 2003).

Bats (Chiroptera) are the second most speciose mammal group in the world, showing a remarkable ecological diversification (e.g. widest dietary range of all mammals), partly due to evolutionary novelties that are exceptional within Mammalia (i.e. echolocation, flight and upside-down roosting) (Adams 2008; Simmons, et al. 2008; Shi and Rabosky

2015). Uncovering the historical trajectories that led to the ecological diversification

137 and specialisation in bats has been greatly limited by a markedly incomplete fossil record (Eiting and Gunnell 2009; Brown, et al. 2019). Evolutionary developmental biology (EvoDevo) has emerged as a promising approach to study the evolution of flight and echolocation in bats, while circumventing limitations in the fossil record

(Sears, et al. 2006; Adams 2008; Cooper, et al. 2012; Camacho, et al. 2019; López-

Aguirre, Hand, et al. 2019a, b). Studies have provided evidence for the ontogenetic mechanisms behind forelimb specialisation in bats and the evolution of vertebrate flight (Adams 2008; Cooper and Tabin 2008; Cooper, et al. 2012; Tokita, et al. 2012;

Adams and Shaw 2013), the developmental basis of echolocation (Wang, Zhu, et al.

2017; Nojiri, et al. 2018), ecology-driven deviations in chiropteran development from general mammalian patterns and the phylogenetic signal in postcranial development

(Adams 1992; Koyabu and Son 2014; López-Aguirre, Hand, et al. 2019a, b). Moreover,

EvoDevo has provided knowledge to inform different evolutionary hypotheses that discuss the ontogenetic basis of phenotypic divergence during animal evolution (e.g.

“funnel model” and the “hourglass model”) (Irie and Kuratani 2014; Cordero, et al.

2020). Funnel model of development suggests that phenotypic disparity is constrained during early development, followed by an increase of disparity across development

(Irie and Kuratani 2014). On the other hand, the hourglass model argues that phenotypic disparity is higher during early and late development, with an intermediate period (i.e. phylotypic period) where disparity is constrained (Irie and Kuratani 2014;

Cordero, et al. 2020).

Asymmetry in bats has been studied at the species and population level (Juste, et al.

2001a, b; Lüpold, et al. 2004; Voigt, et al. 2005; López-Aguirre and Pérez-Torres 2015),

138 all studies being based on the analysis of adult specimens only. Patterns and magnitude of FA in bats have been associated with differential reproductive success

(Lüpold, et al. 2004; Voigt, et al. 2005), suggesting favours symmetric individuals in Saccopteryx bilineata (Voigt, et al. 2005). FA has also been implemented as an effective measure of fitness in Carollia perspicillata, revealing a significant link between asymmetry and reproductive potential (Monteiro, et al. 2019). Resistance to environmental stress and resilience to anthropogenic habitat change has been assessed on the basis of the presence and magnitude of FA, suggesting high resilience in Neotropical bat species (de Figueiredo, et al. 2015; Castillo-Figueroa 2018). Research also indicates that variation in levels of FA across morphological traits could depend on functional importance, with FA decreasing in traits under higher functional demands

(Gummer and Brigham 1995; López-Aguirre and Pérez-Torres 2015; Robaina, et al.

2017). Compensatory growth has been reported in the wing of the vampire bat

Desmodus rotundus as a way to maintain wingspan symmetry (Ueti, et al. 2015), while sex-based differences in the magnitudes of FA have been reported in the wing of D. rotundus and the cranium of Artibeus lituratus (López-Aguirre and Pérez-Torres 2015;

Ueti, et al. 2015).

The objective of this study was to assess the ontogenetic trajectory of phenotypic asymmetry in bats. We tested the compensatory growth hypothesis of asymmetry by analysing the presence and magnitude of FA in the prenatal morphogenesis of the humerus in bats, representing the first developmental study of FA in Chiroptera. We focused on the humerus because it represents a clear example of multiple functional demands acting on a single bone (i.e. withstanding torsional and bending stress,

139 increasing muscle insertion area and controlling manoeuvrability of the wing) (Swartz, et al. 1992; Panyutina, et al. 2015). Further, prenatal limb FA has been described as an accurate indicator of DI in human foetuses (Broek, et al. 2017). We quantify asymmetry based on bone elongation and cross-sectional cortical bone deposition

(using geometric morphometrics; GMM) as a way to exemplify the complexity of bone development. Humeral cross-sectional shape has been found to reflect ecological differences across bat taxa (López-Aguirre, Wilson, et al. 2019). Humeral length asymmetry is commonly used in bat studies (Gummer and Brigham 1995; Voigt, et al.

2005; de Figueiredo, et al. 2015; Ueti, et al. 2015; Robaina, et al. 2017; Castillo-

Figueroa 2018), whereas cross-sectional asymmetry is commonly measured in other mammals (Macintosh, et al. 2013; Wilson and Humphrey 2015; Kubicka, et al. 2018;

Perchalski, et al. 2018). Based on previous findings of compensatory growth in the wing of bats and previous evidence of decreasing FA across development in terrestrial mammals (Hallgrímsson 1999; Hallgrímsson, et al. 2003; Ueti, et al. 2015), we hypothesise that asymmetry of the humerus will decrease throughout ontogeny.

Moreover, given the different functional pressures on the bat humerus, we predict that cross-sectional and longitudinal asymmetry will be correlated, optimising the performance of the humerus after birth.

Methods

Sampling

66 prenatal specimens from 11 bat species (Aselliscus dongbacana, A. stoliczkanus,

Cynopterus sphinx, Hesperoptenus blanfordi, Hipposideros larvatus, Kerivoula hardwickii, Miniopterus schreibersii, Myotis sp., Rhinolophus pearsoni, R. pusillus and

140

R. thomasi) were collected through taxonomic fieldwork in Vietnam by V.T.T. and D.K. under the collection permit No. 972/UBND-TH issued by Tuyen Quang Provincial

People's Committee and research and ethics permit No. 322/STTNSV of the Institute of

Ecology and Biological Resources, Vietnam Academy of Sciences. Specimens were fixed in Serra's fixative (ethanol, formalin, and glacial acetic acid mixed 6:3:1 by volume) for

48 hours, then transferred and preserved in 70% ethanol. 3D scanning of the embryos and foetuses was performed using a microfocal X‐ray CT system at the University

Museum, University of Tokyo (TXS225‐ACTIS; TESCO; Tokyo, Japan) with 70 kV source voltage and 114 μA source currents at a resolution of 36 μm. Osseous skeletal elements were segmented using the thresholding and manual segmenting tools in

MIMICS v. 20 software (Materialise NV, Leuven, Belgium). To achieve finer segmentation in early embryos, thresholding Ct values of osseous and cartilaginous tissues was performed using the Half Maximum Height method (i.e., gradual change in computerised tomography values at the boundary of a structure) (Spoor, et al. 1993).

After selecting specimens with at least partially ossified and unbroken humeri, three individuals were excluded from the final sample. Left and right humeri were segmented and exported as STL files for further processing and analysis.

Data collection

Humeral development was described based on bone elongation (i.e. humeral length;

HL) and cortical bone deposition (i.e. cross-sectional shape and size). To estimate cross-sectional and humeral length measurements, 3D humeri models were imported into Rhinoceros 5.0 (Robert McNeel & Associates, Seattle, WA). To remove the non- shape effects of translation, rotation and scale, all humeri models were aligned to a

141 standard position in 3D space, following a protocol for long-bone cross-sectional geometry (see Wilson and Humphrey 2015). HL was automatically measured as the length of the long axis of a rectilinear box (i.e. bounding box) enclosing the model created using the BoundingBox command in Rhinoceros 5.0, preventing measurement error (ME). Cross-sections of the left and right humeri were extracted at the midshaft

(i.e. 50% of HL), for a total of 126 cross-sections (63 x 2 [left and right side]). Humeral cross-sectional shape and size were quantified using geometric morphometrics.

Following the method described in Wilson and Humphrey (2015), we used a set of 16 equiangular landmarks semi-automatically placed along the periosteal surface of each cross-section. Cross-sectional GMM enables quantification of bone shape while circumventing the lack of identifiable homologous landmarks early in prenatal development. To control for the effect of matching bilateral symmetry (i.e. left and right sides of the body are mirror), landmark coordinates of cross-sections of the right side were reflected along the y axis by multiplying the coordinates of that axis by -1.

Given the lack of Carnegie staging systems for many non-model taxa (nine of the 11 species in our sample), staging of developmental series was based on crown to rump length and bone ossification sequence, as described in (López-Aguirre, Hand, et al.

2019a) and following general patterns described in bat development (Cretekos, et al.

2005). All developmental stages were represented by at least three individuals

(Appendix Table 6).

Estimation of asymmetry

Total asymmetry (TA) in bilateral organisms can be described as the difference between both sides of the body (e.g. right – left) (Palmer 1994), which we estimated

142 based on HL (longitudinal asymmetry) and cross-sectional shape (cross-sectional asymmetry). Individual longitudinal TA was quantified as the signed difference between right and left humeri, negative and positive values indicating directionality of asymmetry (Palmer 1994). Individual longitudinal FA was described as the size- corrected signed difference between sides (i.e. FA6 index), so as to truly represent DI and not developmental bone growth (Palmer 1994).

In order to estimate individual cross-sectional asymmetry, left and right (reflected) cross-sections were replicated and reflected (Wilson and Humphrey 2015), for a combined dataset of 252 cross-sections. A Generalised Procrustes Analysis (GPA) of the combined dataset was performed to decompose cross-sectional morphology into its shape and size (centroid size; CS) components (Woodard and Neustupa 2016;

Neubauer, et al. 2020). Next we performed a multivariate Procrustes ANOVA that accounted for body plan symmetry (bilateral) and subdivided shape and CS variation into symmetrical and asymmetrical variation (Klingenberg and McIntyre 1998;

Klingenberg 2015), while also providing a statistical test for DA, FA and ME (see section below). GPA and Procrustes ANOVA were performed using the gpagen and bilat.symmetry functions implemented in R package Geomorph 3.2 (Adams, et al.

2013; Adams, et al. 2017). Individual cross-sectional TA was estimated as the

Procrustes distance between right and left cross-sections (Sherratt, et al. 2017).

Individual cross-sectional FA (CFA) was computed as the sum of principal component

(PC) scores of a Principal Component Analysis (PCA) of the FA shape coordinates (left plus right sections) retrieved from the Procrustes ANOVA (Woodard and Neustupa

2016).

143

Probability density functions were used to test the normal distribution and zero mean assumptions in raw asymmetry values, as expected for the presence of FA.

Data analysis

Presence of FA and DA in longitudinal asymmetry was statistically tested with an

ANOVA, using side (left or right), individual and replicate as factors

(FA6~Side+Individual+Side:Individual) (Monteiro, et al. 2019; Rivera and Neely 2020).

The side factor provides a statistical test for DA, whereas the side-individual interaction provides statistical tests for FA. ME was not computed for longitudinal asymmetry because the automated protocol implemented to obtain HL ensures no ME could affect our data. In addition to decomposing shape and CS variation in its symmetrical and asymmetrical components (see section above), Procrustes ANOVA can also provide a statistical test of FA, DA and ME (Woodard and Neustupa 2016;

Neubauer, et al. 2020). Hence, Procrustes ANOVA was used to estimate the statistical significance of cross-sectional FA, DA and ME for the combined dataset, following the same protocol as the ANOVA described above. Presence of FA, DA and AS was also tested by visually examining the distribution of signed values of FA6 (longitudinal asymmetry) and CFA (PC scores of the PCA of FA shape coordinates) (Neubauer, et al.

2020).

Developmental trajectories of FA6 and CFA were explored using boxplots, while statistical differences across developmental stages were tested using two ANOVAs, using the stage of each individual as a factor and unsigned FA6 and CFA (FA6~Stage and CFA~Stage).

144

Lastly, we tested the association of FA6 and CFA across development using a linear regression model based on ordinary least squares.

Results

Symmetric and asymmetric shape variation

PCAs (morphospaces) of symmetric and asymmetric components of cross-sectional shape summarised variation in 15 and 18 PCs, respectively. The first two PCs of the symmetric component of cross-sectional shape explained 65.46% of the variation, whereas the first two components of the asymmetric component explained 52.06% of the variation. Both symmetric and asymmetric components of shape variation revealed great overlap between developmental stages in morphospace (Fig. 4.1).

Symmetric cross-sectional shape disparity showed greater variability in later stages of development (stages 7-10), whereas asymmetric shape disparity was higher in intermediate stages (stages 4-5), with a tendency to decrease across development.

Figure 4.1. PCAs of symmetric (A) and asymmetric (B) components of humeral cross-sectional shape variation across prenatal development. Colours represent developmental stages. Stages 1 to 10 represent early to late prenatal development.

145

Longitudinal and cross-sectional asymmetry

Procrustes ANOVA of cross-sectional morphology found statistical support for the presence of FA both in shape and CS, and support for the presence of DA in shape only

(Table 4.1). DA accounted for only 0.3% of CS variation and FA for 43.77% and 30.04% of shape and CS variation, respectively. Individual variation explained the highest proportion of variation, accounting for 55.43% of shape variation and 69.66% of CS variation. ANOVA of asymmetry on HL supported the presence of both DA and FA in longitudinal asymmetry (Table 4.1). Nevertheless, FA accounted only for 0.03% of variation, DA for less than 0.0001% of the variation and individual variation for 99.97% of variability.

Table 4.1. Procrustes ANOVA (Shape, CS) and ANOVA (HL) statistical tests of significance of fluctuating asymmetry (FA), directional asymmetry (DA) and measurement error (ME) in cross-sectional shape, centroid size (CS) and humeral length (HL). Side factor tests for DA, Individual:Side interaction tests for FA and replicate tests for ME.

Df SS MS Rsq F Pr(>F) Shape Individual 62 0.16377 0.0026415 0.55433 1.2663 1 Side 1 0.002339 0.0023395 0.00792 1.1216 0.3394 Individual:Side 62 0.129326 0.0020859 0.43775 1.6685 <0.0001 Replicate 126 0 0 0 Total 251 0.295435 CS Individual 62 2.86E-05 4.61E-07 0.69665 1021240520 <0.0001 Side 1 1.22E-07 1.22E-07 0.00297 270355113 <0.0001 Individual:Side 62 1.23E-05 1.99E-07 0.30037 440325704 <0.0001 Replicate 126 0 0 0 Total 251 4.10E-05 HL Individual 62 1.07E+03 1.73E+01 0.99971 4.23E+29 <0.0001 Side 1 7.13E-03 7.13E-03 6.66E-06 1.74E+26 <0.0001 Side:Individual 62 3.08E-01 4.97E-03 0.00029 1.22E+26 <0.0001 Residuals 125 5.11E-27 4.09E-29 4.8E-30 Distribution of values of signed FA6 and PC1-2 scores of asymmetric cross-sectional variation demonstrate a normal distribution with a mean near zero (Fig. 4.2), 146 supporting the presence of FA in longitudinal and cross-sectional humeral asymmetry throughout development. 42.86% of individuals had negative values of FA6

(longitudinal asymmetry; Fig. 4.2A), indicating that a narrow majority of individuals had larger right humeri (57.14%). On average, 46.03% of individuals showed negative values of cross-sectional asymmetry (42.86% for PC1 and 49.21% for PC2 scores; Fig.

4.2B-C). Across datasets (cross-sectional and HL), only three individuals were found to have perfectly symmetrical humeri (i.e. R-L= 0), for HL in developmental stages 5, 7 and 8.

Figure 4.2. Probability density functions of distribution of values of longitudinal (signed FA6, A) and cross-sectional FA (PC1-2 scores, B-C). Barplots show individual asymmetry values from which density functions were estimated.

Boxplots of FA6 and CFA revealed independent trajectories of cross-sectional and longitudinal humeral FA across development (Fig. 4.3). Unsigned FA6 indicated higher values of longitudinal FA early in prenatal development, followed by a sharp decrease in FA6 values after stage 1. FA6 values remained relatively stable from stage 2 onwards with a noticeable increase in variability in stages 8 and 9. Developmental patterns of

147

CFA suggested intermediate magnitudes of FA in early stages of prenatal development, followed by an increase between stages 3 and 6. CFA consistently decreased from stage 7 onwards, reaching its lowest values in stages 9 and 10. Despite the differences in the trajectories of FA6 and CFA throughout development, our results indicate a trend of decreasing FA throughout development. ANOVAs for differences in FA6 and

CFA across developmental stages found statistically significant differences in CFA but not FA6 (Table 4.2).

Figure 4.3. Boxplots of longitudinal (unsigned FA6, left) and cross-sectional (CFA, right) FA values across development. Stages 1 to 10 represent early to late prenatal development.

Table 4.2. ANOVA test results for statistically significant differences in longitudinal (FA6) and cross-sectional (CFA) humeral FA across developmental stages.

Df SS MS Rsq F Pr(>F) FA6 1 0 0 -0.0164 0 0.993 CFA 1 0.00871 0.008706 0.1058 8.332 0.0054 Scatterplots of longitudinal and cross-sectional asymmetry did not indicate a clear association between CFA and FA6 throughout development, with a slightly negative tendency (Fig. 4.4). A linear regression model of CFA and FA6 confirmed a non- significant negative correlation between both dimensions of humeral FA (R2= -0.191,

P= 0.849). 148

Figure 4.4. Scatterplots of association between longitudinal (unsigned FA6) and cross-sectional (CFA) fluctuating asymmetry across prenatal development. Dot colours represent developmental stages, stages 1 to 10 representing early to late prenatal development. Discussion

This study is the first to demonstrate the presence of FA and DA during bat prenatal development. Our results showed statistically significant differences in FA in humeral morphology (both cross-sectional and longitudinal) across prenatal development, supporting our hypothesis of decreasing asymmetry throughout ontogeny. However, our results also show that both measures of FA (FA6 and CFA) did not correlate across development, showing decoupled ontogenetic trajectories, contrary to our prediction.

Previous studies have found decreasing trajectories of asymmetry across prenatal development in the postcranium of humans and mice, suggesting the presence of homeostatic mechanisms canalising FA ontogenetically (Hallgrímsson, et al. 2003;

Dongen, et al. 2017). Compensatory growth between left and right sides of the body in response to increased asymmetry has been discussed as a mechanism to reduce DI

(Emlen, et al. 1993; Aparicio 1998; Ueti, et al. 2015). 149

FA in bats has been an area of increasing interest in recent decades (Gummer and

Brigham 1995; Juste, et al. 2001a, b; Lüpold, et al. 2004; Voigt, et al. 2005; de

Figueiredo, et al. 2015; López-Aguirre and Pérez-Torres 2015; Ueti, et al. 2015;

Robaina, et al. 2017; Castillo-Figueroa 2018; Monteiro, et al. 2019). A set of two studies explored patterns of FA in insular populations of fruit bat species Eidolon helvum and Rousettus egyptiacus of the gulf of Guinea (Juste, et al. 2001a, b). Juste, et al. (2001a) found similar patterns of FA across populations of both species, discussing the interpretation of Population Asymmetry Parameters (i.e. consistent patterns of FA for a set of characters across populations of the same species; PAP) and suggested the scalability of PAP above the species level. Juste, et al. (2001b) found consistent magnitudes of multivariate FA and significant integration of asymmetry across traits and species, hypothesising high canalisation in the developmental pathways controlling phenotypic asymmetry that are shared between the two species.

The presence and magnitude of FA has also been associated with reproductive success and sexual selection in bats (Voigt, et al. 2005; Monteiro, et al. 2019). Analysing forearm length asymmetry, it has been suggested that sexual selection favours more symmetrical males in the polygynous greater sac-winged bat Saccopteryx bilineata, canalising FA (Voigt, et al. 2005). The number of offspring produced by males of S. bilineata was found to significantly decrease with increasing forearm asymmetry

(Voigt, et al. 2005). Increases in forearm asymmetry in the Neotropical frugivorous bat

Carollia perspicillata have also been linked with a significant decrease in survival probability and the probability of more than one pregnancy per reproductive season

(Monteiro, et al. 2019). However, Lüpold, et al. (2004) did not find a significant

150 association between FA and other measures of individual fitness and allometry in bat genitalia.

Studies have also linked presence and magnitude of FA to anthropogenic perturbations and habitat degradation, analysing four Neotropical phyllostomid species (i.e. Artibeus lituratus, A. planirostris, C. perspicillata and Sturnira lilium) (de Figueiredo, et al. 2015;

Castillo-Figueroa 2018). However, based on the statistical analyses in both studies, we have detected potential limitations in their results concerning the relative contributions of each kind of asymmetry AA, DA and FA to overall asymmetry, an issue previously raised in multiple studies (Dongen 2006). Neither study explicitly tested the presence of the three kinds of asymmetry, casting doubt on their interpretation that

FA is an accurate index of resilience and adaptation of species’ to perturbations.

Robaina, et al. (2017) and López-Aguirre and Pérez-Torres (2015) evaluated whether differences in asymmetry across traits reflect functional adaptations and importance.

Robaina, et al. (2017) may have insufficiently assessed the presence and magnitude of all types of asymmetry before drawing conclusions on the validity of FA to reflect functional importance. Our results suggest that special attention to the statistical framework used to describe the biological and theoretical interpretation of asymmetry is warranted.

Asymmetry across multiple traits has shown decoupled patterns, reflecting differences in functional importance and indicating independent developmental mechanisms controlling phenotypic symmetry across different structures of the body of bats

(Gummer and Brigham 1995; Robaina, et al. 2017), turtles (Rivera and Neely 2020) and birds (Aparicio and Bonal 2002). Our results of decoupled patterns of longitudinal

(FA6) and cross-sectional (CFA) humeral asymmetry evidence the functional 151 differentiation of patterns of asymmetry, while also suggesting it could be applicable within single structures. Longitudinal and cross-sectional growth are hypothesised to respond to different selective pressures, with cross-sectional bone deposition potentially associated with biomechanical resistance against torsional and bending stresses (Steele and Mays 1995; Blackburn 2011; Perchalski, et al. 2018), and bone elongation correlating with maintenance of body proportions within a functional unit

(Ueti, et al. 2015).

Despite both FA6 and CFA showing decreasing magnitudes across development, only

CFA showed statistically significant differences across developmental stages.

Increasing postcranial morphological disparity and integration across prenatal development in bats has been reported in ossification sequences and metric growth

(López-Aguirre, Hand, et al. 2019a). Our results of decreased FA across prenatal development suggest ontogenetic mechanisms canalising symmetry and disparity could interact to promote variability while also buffering DI by constraining asymmetry. In order for novel morphotypes to be positively selected as phenotypic disparity increases across ontogeny, developmental mechanisms may be in place to buffer stress and reduce FA. Our results can also inform the discussion around developmental models of phenotypic divergence (e.g. funnel and hourglass models), suggesting that ontogenetic mechanisms that control morphogenetic divergence could also depend on mechanisms controlling DI (Irie and Kuratani 2014; Cordero, et al.

2020). Our study indicate that developmental trajectories of FA could also provide novel information to discuss the hourglass and funnel models of morphogenetic divergence.

152

We hypothesise that our results indicate a greater selective pressure to canalise cross- sectional asymmetry as a response to functional demands associated with a newborn bat’s ecology. In particular, there may be a greater requirement for humeral biomechanical robusticity to facilitate early roosting behaviours (in many bat species, newborns attach to the mothers using their feet and thumb (Koyabu and Son 2014) than to immediately fly, with bone elongation continuing during this period. Multiple studies have documented postnatal development of flight in bats, describing altricial development of the forelimb in bats followed by accelerated bone elongation (Hughes and Rayner 1993; Kunz and Robson 1995; Baptista, et al. 2000; Kunz, et al. 2009; Lin, et al. 2010; Jin, et al. 2011; Carter and Adams 2015; Eghbali, et al. 2017; Eghbali and

Sharifi 2018). First in most species that have been studied occur consistently in synchrony with weaning, usually a couple weeks after birth once adult body dimensions are reached (Hughes and Rayner 1993; Kunz and Robson 1995; Kunz, et al.

2009; Lin, et al. 2010; Eghbali, et al. 2017; Eghbali and Sharifi 2018). We hypothesise that because self-powered flight is not achieved immediately after birth, bone elongation asymmetry and compensatory growth to optimise wing proportions would be less constrained prenatally (Ueti, et al. 2015). Further studies quantifying developmental trajectories of asymmetry should focus on describing the complementary developmental process (pre- and postnatal).

Conclusions

We found significant support for the presence of FA and DA during the prenatal development of the humerus in bats. We also show that magnitudes of FA decrease across prenatal development, in line with previous studies in humans and model species, and we hypothesise this to be evidence of developmental canalisation of FA. 153

Moreover, we find evidence for decoupled patterns of longitudinal and cross-sectional asymmetry throughout prenatal humeral development. We suggest functional differences between bone elongation and cross-sectional bone deposition may be associated with the newborn’s ecology (i.e., pup roosting behaviour and the later acquisition of flight). To our knowledge, this study is the first to analyse asymmetry patterns in the development of bats, providing new information about developmental pathways controlling phenotypic asymmetry and DI in non-model taxa. We highlight the importance of assessing the correlation between FA and DI beyond patterns of total asymmetry FA and DI.

References Adams, D. C., M. L. Collyer, A. Kaliontzopoulou, and E. Sherratt. 2017. Geomorph: software for geometric morphometric analyses. R package. https://cran.r- project.org/package=geomorph. Adams, D. C., E. Otárola-Castillo, and E. Paradis. 2013. Geomorph: an R package for the collection and analysis of geometric morphometric shape data. Methods in Ecology and Evolution 4:393-399. Adams, R. A. 1992. Comparative Skeletogenesis of the Forearm of the Little Brown Bat (Myotis-Lucifugus) and the Norway Rat (Rattus-Norvegicus). Journal of Morphology 214:251-260. Adams, R. A. 2008. Morphogenesis in Bat Wings: Linking Development, Evolution and Ecology. Cells Tissues Organs 187:13-23. Adams, R. A. and J. B. Shaw. 2013. Time’s Arrow in the Evolutionary Development of Bat Flight. Bat evolution, ecology and conservation: 21-46. Albert, A. M. and D. L. Greene. 1999. Bilateral asymmetry in skeletal growth and maturation as an indicator of environmental stress. American Journal of Physical Anthropology 110:341-349. Aparicio, J. 1998. Patterns of fluctuating asymmetry in developing primary feathers: a test of the compensational growth hypothesis. Proceedings of the Royal Society of London. Series B: Biological Sciences 265:2353-2357. Aparicio, J. M. and R. Bonal. 2002. Why do some traits show higher fluctuating asymmetry than others? A test of hypotheses with tail feathers of birds. Heredity 89:139-144. Balavoine, G. and A. Adoutte. 2003. The Segmented Urbilateria: A Testable Scenario1. Integrative and Comparative Biology 43:137-147. Baptista, T. L., C. S. Richardson, and T. H. Kunz. 2000. Postnatal Growth and Age Estimation in Free-Ranging Bats: A Comparison of Longitudinal and Cross-Sectional Sampling Methods. Journal of Mammalogy 81:709-718.

154

Blackburn, A. 2011. Bilateral asymmetry of the humerus during growth and development. American Journal of Physical Anthropology 145:639-646. Breuker, C. J., J. S. Patterson, and C. P. Klingenberg. 2006. A single basis for developmental buffering of Drosophila wing shape. PLoS One 1:e7. Broek, C. M. A. t., J. Bots, M. Bugiani, F. Galis, and S. v. Dongen. 2017. Developmental Origins of Limb Developmental Instability in Human Fetuses: Many Abnormalities Make the Difference. Symmetry 9: 51. Brown, E. E., D. D. Cashmore, N. B. Simmons, and R. J. Butler. 2019. Quantifying the completeness of the bat fossil record. Palaeontology 62: 757-776. Camacho, J., A. Heyde, B.-A. S. Bhullar, D. Haelewaters, N. B. Simmons, and A. Abzhanov. 2019. Peramorphosis, an evolutionary developmental mechanism in neotropical bat skull diversity. Developmental Dynamics 248:1129-1143. Carter, A. J. R. and D. Houle. 2011. Artificial selection reveals heritable variation for developmental instability. Evolution 65:3558-3564. Carter, R. T. and R. A. Adams. 2015. Postnatal ontogeny of the cochlea and flight ability in Jamaican fruit bats (Phyllostomidae) with implications for the evolution of echolocation. J Anat 226:301-308. Castillo-Figueroa, D. 2018. Fluctuating asymmetry of three bat species in extensive livestock systems from Córdoba department, Colombia. Revista colombiana de ciencia animal recia 10:143-153. Coda, J. A., J. J. Martiínez, A. R. Steinmann, J. Priotto, and M. D. Gomez. 2017. Fluctuating asymmetry as an indicator of environmental stress in small mammals Mastozoología Neotropical 24:313-321. Cooper, K. L. and C. J. Tabin. 2008. Understanding of bat wing evolution takes flight. Genes Dev 22:121-124. Cooper, L., C. J. Cretekos, and K. E. Sears. 2012. The evolution and development of mammalian flight. Wiley Interdisciplinary Reviews: Developmental Biology 1:773-779. Cordero, G. A., M. R. Sánchez-Villagra, and I. Werneburg. 2020. An irregular hourglass pattern describes the tempo of phenotypic development in placental mammal evolution. Biol Letters 16:20200087. Cretekos, C. J., S. D. Weatherbee, C. H. Chen, N. K. Badwaik, L. Niswander, R. R. Behringer, and J. J. t. Rasweiler. 2005. Embryonic staging system for the short-tailed fruit bat, Carollia perspicillata, a model organism for the mammalian order Chiroptera, based upon timed pregnancies in captive-bred animals. Dev Dyn 233:721-738. de Figueiredo, D., B. Gonçalves Miller, E. S. Barbosa Leal, and M. A. Montes. 2015. Fluctuating asymmetry in populations of bats: species adapted to urban environments are not hampered by habitat degradation. Chiroptera Neotropical 21:1305-1311. Dongen, S. V. 2006. Fluctuating asymmetry and developmental instability in evolutionary biology: past, present and future. Journal of Evolutionary Biology 19:1727-1743. Dongen, S. V., C. T. Broek, J. Bots, and F. Galis. 2017. Changes of Fluctuating Asymmetry with Age in Human Fetuses and Young Infants. Symmetry 9: 44. Eghbali, H., S. Shahabi, N. Najafi, R. Mehdizadeh, S. Yousefi, and M. Sharifi. 2017. Postnatal growth, wing development and age estimations in the Mediterranean horseshoe bat Rhinolophus euryale (Chiroptera: Rhinolophidae) in Kerend cave, western Iran. Mammalia 82: 276-287.

155

Eghbali, H. and M. Sharifi. 2018. Postnatal Growth, Age Estimation, and Wing Development in Geoffroy's Bat Myotis emarginatus (Chiroptera: Vespertilionidae). Mammal Study 43:153 - 165. Eiting, T. P. and G. F. Gunnell. 2009. Global Completeness of the Bat Fossil Record. Journal of Mammalian Evolution 16:151-173. Emlen, J. M., D. C. Freeman, and J. H. Graham. 1993. Nonlinear growth dynamics and the origin of fluctuating asymmetry. Genetica 89:77. Genikhovich, G. and U. Technau. 2017. On the evolution of bilaterality. Development 144:3392. Gummer, D. L. and R. M. Brigham. 1995. Does fluctuating asymmetry reflect the importance of traits in little brown bats (Myotis lucifugus)? Canadian Journal of Zoology 73:990-992. Hallgrímsson, B. 1999. Ontogenetic Patterning of Skeletal Fluctuating Asymmetry in Rhesus Macaques and Humans: Evolutionary and Developmental Implications. International Journal of Primatology 20:121-151. Hallgrímsson, B., T. Miyake, K. Wilmore, and B. K. Hall. 2003. Embryological origins of developmental stability: size, shape and fluctuating asymmetry in prenatal random bred mice. Journal of experimental zoology. Part B, Molecular and developmental evolution 296:40-57. Holló, G. 2015. A new paradigm for animal symmetry. Interface Focus 5:20150032. Holló, G. and M. Novák. 2012. The manoeuvrability hypothesis to explain the maintenance of bilateral symmetry in animal evolution. Biology Direct 7:22. Hughes, P. and J. M. V. Rayner. 1993. The flight of pipistrelle bats Pipistrellus pipistrellus during pregnancy and lactation. Journal of Zoology 230:541-555. Irie, N. and S. Kuratani. 2014. The developmental hourglass model: a predictor of the basic body plan? Development 141:4649. Jin, L.-r., A.-q. Lin, K.-p. Sun, Y. Liu, and J. Feng. 2011. Postnatal development of morphological features and vocalization in the pomona leaf-nosed bat Hipposideros pomona. Acta Theriologica 56:13-22. Juste, J., C. López-González, and R. E. Strauss. 2001a. Analysis of asymmetries in the African fruit bats Eidolon helvum and Rousettus egyptiacus (Mammalia: Megachiroptera) from the islands of the Gulf of Guinea. I. Variance and size components of bilateral variation. Journal of Evolutionary Biology 14:663-671. Juste, J., C. López-González, and R. E. Strauss. 2001b. Analysis of asymmetries in the African fruit bats Eidolon helvum and Rousettus egyptiacus (Mammalia: Megachiroptera) from the islands of the Gulf of Guinea. II. Integration and levels of multivariate fluctuating asymmetry across a geographical range. Journal of Evolutionary Biology 14:672-680. Kellner, James R. and Ross A. Alford. 2003. The Ontogeny of Fluctuating Asymmetry. The American Naturalist 161:931-947. Klingenberg, C. P. 2015. Analyzing Fluctuating Asymmetry with Geometric Morphometrics: Concepts, Methods, and Applications. Symmetry 7. Klingenberg, C. P. and G. S. McIntyre. 1998. Geometric Morphometrics of Developmental Instability: Analyzing Patterns of Fluctuating Asymmetry with Procrustes Methods. Evolution 52:1363-1375. Klingenberg, C. P. and H. F. Nijhout. 1999. Genetics of fluctuating asymmetry: A developmental model of developmental instability. Evolution 53:358-375. 156

Koyabu, D. and N. Son. 2014. Patterns of postcranial ossification and sequence heterochrony in bats: Life histories and developmental trade‐offs. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 322:607-618. Kubicka, A. M., W. Nowaczewska, A. Balzeau, and J. Piontek. 2018. Bilateral asymmetry of the humerus in Neandertals, Australian aborigines and medieval humans. American Journal of Physical Anthropology 167:46-60. Kunz, T. H., R. A. Adams, and W. R. Wood. 2009. Methods for assessing size at birth and postnatal growth and development in bats. Pp. 273–314 in T. H. Kunz, and S. Parsons, eds. Ecological and behavioral methods for the study of bats. The Johns Hopkins University Press, Baltimore. Kunz, T. H. and S. K. Robson. 1995. Postnatal Growth and Development in the Mexican Free-Tailed Bat (Tadarida brasiliensis mexicana): Birth Size, Growth Rates, and Age Estimation. Journal of Mammalogy 76:769-783. Leamy, L. J. and C. P. Klingenberg. 2005. The Genetics and Evolution of Fluctuating Asymmetry. Annual Review of Ecology, Evolution, and Systematics 36:1-21. Lin, A., L. Jin, Y. Liu, K. Sun, and J. X. Feng. 2010. Postnatal Growth and Age Estimation in Horsfield's Leaf-Nosed Bat Hipposideros larvatus. Zoological studies 49: 789-796. López-Aguirre, C., S. J. Hand, D. Koyabu, N. T. Son, and L. A. B. Wilson. 2019a. Postcranial heterochrony, modularity, integration and disparity in the prenatal ossification in bats (Chiroptera). BMC Evolutionary Biology 19:75. López-Aguirre, C., S. J. Hand, D. Koyabu, N. T. Son, and L. A. B. Wilson. 2019b. Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera). Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 332:36-49. López-Aguirre, C. and J. Pérez-Torres. 2015. Asimetría cráneo-mandibular de Artibeus lituratus (Chiroptera, Phyllostomidae) en Colombia. Universitas Scientiarum 20:141- 152. López-Aguirre, C., L. A. Wilson, D. Koyabu, V. Tan Tu, and S. J. Hand. 2019c. Decoupled morphological and biomechanical evolution and diversification of the wing in bats. EcoEvoRxiv:31. Lüpold, S., A. G. McElligott, and D. J. Hosken. 2004. Bat genitalia: allometry, variation and good genes. Biological Journal of the Linnean Society 83:497-507. Macintosh, A. A., T. G. Davies, T. M. Ryan, C. N. Shaw, and J. T. Stock. 2013. Periosteal Versus True Cross-Sectional Geometry: A Comparison Along Humeral, Femoral, and Tibial Diaphyses. American Journal of Physical Anthropology 150:442-452. Monteiro, L. R., B. Mellado, M. R. Nogueira, and M. M. de Morais-Jr. 2019. Individual asymmetry as a predictor of fitness in the bat Carollia perspicillata. Journal of Evolutionary Biology 32:1207-1229. Neubauer, S., P. Gunz, N. A. Scott, J.-J. Hublin, and P. Mitteroecker. 2020. Evolution of brain lateralization: A shared hominid pattern of endocranial asymmetry is much more variable in humans than in great apes. Sci Adv 6:eaax9935. Nojiri, T., I. Werneburg, N. T. Son, V. T. Tu, T. Sasaki, Y. Maekawa, and D. Koyabu. 2018. Prenatal cranial bone development of Thomas's horseshoe bat (Rhinolophus thomasi): with special reference to petrosal morphology. J Morphol 279: 809-827. Palmer, A. R. 1994. Fluctuating asymmetry analyses: a primer. Pp. 335-364 in T. A. Markow, ed. Developmental Instability: Its Origins and Evolutionary Implications: Proceedings of the International Conference on Developmental Instability: Its Origins 157 and Evolutionary Implications, Tempe, Arizona, 14–15 June 1993. Springer Netherlands, Dordrecht. Panyutina, A., L. P. Korzun, and A. N. Kuznetsov. 2015. Flight of Mammals: From Terrestrial Limbs to Wings. Springer, New York. Pecoits, E., K. O. Konhauser, N. R. Aubet, L. M. Heaman, G. Veroslavsky, R. A. Stern, and M. K. Gingras. 2012. Bilaterian Burrows and Grazing Behavior at &gt;585 Million Years Ago. Science 336:1693. Perchalski, B., A. Placke, S. M. Sukhdeo, C. N. Shaw, J. H. Gosman, D. A. Raichlen, and T. M. Ryan. 2018. Asymmetry in the Cortical and Trabecular Bone of the Human Humerus During Development. Anat Rec (Hoboken) 301:1012-1025. Rivera, G. and C. M. D. Neely. 2020. Patterns of fluctuating asymmetry in the limbs of freshwater turtles: Are more functionally important limbs more symmetrical? Evolution 74:660-670. Robaina, Z., D. Denis, and M. Sánchez-Lozada. 2017. ¿Son regulados los niveles de asimetría fluctuante según la función de las estructuras morfológicas? Estudio de caso en Phyllonycteris poeyi (chiroptera: phyllostomidae)/A Are levels of fluctuating asymmetry regulated according to the function of. Poeyana; Núm. 505. Sears, K. E., R. R. Behringer, J. J. Rasweiler, and L. A. Niswander. 2006. Development of bat flight: Morphologic and molecular evolution of bat wing digits. P Natl Acad Sci USA 103:6581-6586. Sherratt, E., J. M. Serb, and D. C. Adams. 2017. Rates of morphological evolution, asymmetry and morphological integration of shell shape in scallops. BMC Evolutionary Biology 17:248. Shi, J. J. and D. L. Rabosky. 2015. Speciation dynamics during the global radiation of extant bats. Evolution 69:1528-1545. Simmons, N. B., K. L. Seymour, J. Habersetzer, and G. F. Gunnell. 2008. Primitive Early Eocene bat from Wyoming and the evolution of flight and echolocation. Nature 451:818-821. Spoor, C. F., F. W. Zonneveld, and G. A. Macho. 1993. Linear Measurements of Cortical Bone and Dental Enamel by Computed-Tomography - Applications and Problems. American Journal of Physical Anthropology 91:469-484. Steele, J. and S. Mays. 1995. Handedness and directional asymmetry in the long bones of the human upper limb. International Journal of Osteoarchaeology 5:39-49. Swartz, S. M., M. B. Bennett, and D. R. Carrier. 1992. Wing bone stresses in free flying bats and the evolution of skeletal design for flight. Nature 359: 726-729. Tocts, A. M. S., D. W. Johnson, and A. J. R. Carter. 2016. Strong nonlinear selection against fluctuating asymmetry in wild populations of a marine fish. Evolution 70:2899- 2908. Tokita, M., T. Abe, and K. Suzuki. 2012. The developmental basis of bat wing muscle. Nature Communications 3: 1302. Ueti, A., P. S. Pompeu, and R. L. Ferreira. 2015. Asymmetry compensation in a small vampire bat population in a cave: a case study in Brazil. Subterranean Biology 15. Voigt, C. C., G. Heckel, and F. Mayer. 2005. Sexual selection favours small and symmetric males in the polygynous greater sac-winged bat Saccopteryx bilineata (, Chiroptera). Behavioral Ecology and Sociobiology 57:457-464.

158

Wang, Z., T. Zhu, H. Xue, N. Fang, J. Zhang, L. Zhang, J. Pang, E. C. Teeling, and S. Zhang. 2017. Prenatal development supports a single origin of laryngeal echolocation in bats. Nature Ecology & Evolution 1: 0021. Wilson, L. and L. T. Humphrey. 2015. A Virtual geometric morphometric approach to the quantification of long bone bilateral asymmetry and cross‐sectional shape. American Journal of Physical Anthropology 4: 541-556. Woodard, K. and J. Neustupa. 2016. Morphometric Asymmetry of Frustule Outlines in the Pennate Diatom Luticola poulickovae (Bacillariophyceae). Symmetry 8.

159

Chapter 5 Variation in cross-sectional shape and biomechanical properties of the bat humerus under Wolff’s law

Abstract

Bats use their forelimbs in different ways, but flight is the most notable example of morphological adaptation. Foraging and roosting specialisations beyond flight have also been described in several bat lineages. Understanding postcranial evolution during the locomotory and foraging diversification of bats is fundamental to understanding bat evolution. We investigated whether different foraging and roosting behaviours influenced humeral cross-sectional shape and biomechanical variation, following Wolff’s law of bone remodelling. The effect of body size and phylogenetic relatedness was also tested, in order to evaluate multiple sources of variation. Our results suggest strong ecological signal and no phylogenetic structuring in shape and biomechanical variation in humeral phenotypes. Decoupled modes of scaling of shape and biomechanical variation were consistently indicated across foraging and roosting behaviours, suggesting divergent allometric trajectories. Terrestrial locomoting and upstand roosting species showed unique patterns of shape and biomechanical variation across all our analyses, suggesting that these rare behaviours among bats place unique functional demands on the humerus, canalising phenotypes. Our results suggest that complex and multiple adaptive pathways interplay in the postcranium, leading to the decoupling of different features and regions of skeletal elements optimised for different functional demands. Moreover, our results shed further light on the phenotypical diversification of the wing in bats and how adaptations besides flight could have shaped the evolution of the bat postcranium. 160

Introduction

Bats evolved different flight strategies varying mainly in manoeuvrability (i.e. understory to open-air foragers) and speed (i.e. hovering to fast-flying) (Canals, et al.

2011; Marinello and Bernard 2014). Hovering and slow, highly-manoeuvrable flight correlate with the evolution of nectarivory and frugivory in New World bats (Norberg and Rayner 1987; Amador, Almeida, et al. 2019), whereas fast flight is associated with the evolution of hawking and trawling (Canals, et al. 2011). Broadly speaking, six foraging guilds have been described in bats based on behavioural differences: trawling, gleaning, hawking, frugivory, carnivory and terrestrial locomotion (Norberg and Rayner

1987). Classifying foraging guilds (in terms of their flight patterns) for bats with frugivorous and carnivorous diets has proven difficult, given the high degree of behavioural plasticity they exhibit (Norberg and Rayner 1987). For example, frugivores perch to feed on foodstuff from the understory, canopy or ground, whereas carnivores shift from ground- and foliage-gleaning, to hawking and perch-hunting (Norberg and

Rayner 1987). Most studies classifying foraging guilds in bats based on flight patterns have been limited in the variables used to discriminate wing phenotypes (i.e. aspect ratio and wing loading) (Riskin, et al. 2010; Iriarte-Diaz, et al. 2012; Marinello and

Bernard 2014; Hedenstrom and Johansson 2015). Aspect ratio (AR) is a ratio of wingspan to wing area that is interpreted as a measure of the energetic cost and speed of flight (Rayner 1988). Wing loading (WL) is a ratio of body weight to wing area interpreted as a measure of manoeuvrability and loading capacity (Norberg and

Rayner 1987). These variables integrate the length, width and weight of the wing to characterise the overall structure of the wing and describe ecomorphological aerial

161 guilds (Norberg and Rayner 1987; Norberg 1990; Marinello and Bernard 2014;

Hedenstrom and Johansson 2015). However, information about the ecological and foraging plasticity of many taxa is still lacking, limiting the applicability of a non- ambiguous classification to an ordinal level (Norberg and Rayner 1987; Arita and

Fenton 1997; Bullen and McKenzie 2001, 2004; Denzinger and Schnitzler 2013).

Although the forelimb skeleton of bats is adapted primarily to facilitate flight, bats display a wide range of locomotor and ecological behaviours aside from flying that represent novel forelimb motion strategies (Dickinson 2008). Terrestrial locomotion and upstand roosting (i.e. head-up roosting inside furled leaves) are rare behaviours that represent a different use of the bat forelimb (Riskin, et al. 2005; Riskin, et al.

2006; Hand, et al. 2009; Fenton 2010; Schliemann and Goodman 2011). Terrestrial locomotion evolved independently in vampire bats (Desmodontinae), some molossid species (e.g. Cheiromeles) and species of the family Mystacinidae (Riskin, et al. 2005).

Terrestrial locomotion in vampire bats has been suggested to provide an evolutionary advantage, enabling them to chase prey that flees during a feeding event at the lowest energetic cost possible (Riskin, et al. 2005; Riskin, et al. 2006; Hand, et al. 2009).

Terrestrial locomotion in Mystacinidae has been considered a result of insular flightlessness (Riskin, et al. 2005; Riskin, et al. 2006). However, analysis of fossil mystacinid species from Australia revealed that mystacinid terrestrial locomotion evolved before geographic isolation in New Zealand, positing a new hypothesis driven by selective advantage or energetic benefit (Hand, et al. 2009; Hand, et al. 2013; Hand, et al. 2015; Hand, et al. 2018).

162

Upstand roosting also evolved convergently in the bat families Thyropteridae, from

South and Central America, and Myzopodidae, from Africa and Madagascar (Fenton

2010). Each family independently developed a unique set of adhesive structures on their wrists that enable upstand roosting (Riskin, et al. 2005; Schliemann and Goodman

2011; Davalos, et al. 2014). This novelty allowed these species to exploit new roosts, such as the inside of furled leaves (Riskin and Fenton 2001; Riskin and Racey 2010;

Schliemann and Goodman 2011; Davalos, et al. 2014; Boerma, et al. 2019).

Despite study of the evolutionary drivers of modern ecomorphological diversity in bats

(Rossoni, et al. 2017; Thiagavel, et al. 2018; Arbour, et al. 2019; Hedrick, et al. 2019;

Rossoni, et al. 2019), the relationship between ecological traits (e.g. diet and echolocation) and locomotory-related morphological adaptations remains unclear. The advent of three-dimensional imaging techniques has allowed for the development of new methods to study the morpho-biomechanical properties of bones, providing new insights into the evolution and ecology of functional performance (Simons, et al. 2011;

Patel, et al. 2013; Pratt, et al. 2018; Voeten, et al. 2018). Here we use 3D morphometric analysis to investigate the connection between postcranial morpho- biomechanical variability and ecological and biological diversity in modern bats. We focus on the cross-sectional geometry and biomechanical properties of the humerus

(widely used for locomotory inferences in flying vertebrates), a bone specialised in bats to withstand high mechanical loading and the attachment site for muscles crucial in flapping flight (Swartz, et al. 1992; Watts, et al. 2001). Cross-sectional morphology of long bones has been shown to differ markedly within functionally diverse and speciose lineages (e.g. Carnivora – (Kilbourne and Hutchinson 2019). Similarly, biomechanical

163 properties of long bones have been suggested to correlate with differences in locomotion and foraging (Wolff’s law) (Wolff 1986). Beam theory has been successfully applied to study biomechanical properties of long bones and their association with ecology and behaviour (Meers 2002). Beams are defined as structural elements with fairly straight long axes whose length is greater than width and depth. Long bones are typically modelled as a solid or hollow cylindrical beam subjected to bending by forces applied to them (Salathe, et al. 1989). Under such conditions, bending stress/strain at any location is a function of the magnitude and orientation of the forces and the cross- sectional and material properties of the bone. Bone geometry has been shown to grossly reflect loading patterns in the upper and lower limbs, particularly in the context of habitual locomotion patterns (e.g. rowing activity in humans; Macintosh, et al. 2014, 2017). Differences in long bone cross-sectional traits have been reported in birds (Habib and Ruff 2008), carnivorans and primates (Houssaye, et al. 2016;

Houssaye and Botton-Divet 2018) with differing locomotor modes.

With the objective of characterising differences in morpho-biomechanical properties of the bat humerus that could be attributed to behavioural differences (foraging strategies and upstand roosting), we aimed at addressing two questions: 1) Do cross- sectional shape and biomechanical properties scale with size similarly across foraging and roosting guilds? and 2) Do behavioural differences in foraging and roosting and phylogenetic relatedness shape patterns of morpho-biomechanical variation of the humerus? For the former, we tested differences in morpho-biomechanical humeral scaling across foraging and roosting guilds, using cross-sections extracted from the midshaft of the diaphysis. For the latter, we assessed whether behavioural differences

164 in foraging and roosting or phylogenetic relatedness could explain differences in humeral cross-sectional shape and biomechanical disparity. Given that bat flight and diet evolved associated with body size limitations (Moyers Arévalo, et al. 2018;

Amador, Almeida, et al. 2019), we predict a significant scaling of humeral shape and biomechanics with body size that will vary across foraging and roosting guilds. Also, based on the repeated and independent evolution of feeding and roosting strategies in bats, we hypothesise that foraging behaviour and upstand roosting, and not phylogeny, represent an adaptive opportunity that influenced phenotypic diversification during the evolution of the humerus in bats.

Materials and methods

Sample description and digitisation

55 adult specimens were analysed in this study, sampling a phylogenetic (37 species,

20 of 21 modern families and both bat suborders), foraging behaviour (all broad foraging categories described for bats and 85% of dietary habits) and body size range

(10-fold range in body mass) (Fig. 5.1). Both families in which upstand roosting (UR) is present (i.e. Myzopodidae and Thyropteridae) were also included in the sample.

Alcohol-preserved specimens were sourced from the Western Australian Museum and research collections at the University of New South Wales and City University of Hong

Kong (Appendix Table 7). The specimens were scanned at Musashino Art University using a microCT system (inspeXio SMX-90CT Plus, Shimadzu) with 90kv source voltage and 100mA source current at a resolution of 15μm, and a U-CT (Milabs, Utrecht) at

UNSW Sydney with 55kV and 0.17 mA, ultrafocused setting at a resolution of 30-50

μm. Additional species were sampled from whole-body scans sourced from Digimorph

165

(Appendix Table 7). Segmentation of humeri was performed using the thresholding tool in MIMICS v. 20 software (Materialise NV, Leuven, Belgium). To control the unwanted effect of bilateral asymmetry, only left humeri were sampled.

Figure 5.1. Phylogenetic reconstruction of the evolutionary relationships between the sampled species (top), based on Shi and Rabosky (2015). Colours of branches represent different foraging and roosting categories. Species classified based on body size, represented as discretised centroid size (CS) categories: small-sized (triangle), medium-sized (square) and large-sized species (diamond). Subordinal clades are marked in the phylogeny. Lineage through time plot (bottom) showing temporal accumulation of lineages in our sample.

166

Extraction of landmark and biomechanical data

3D humeri models were imported into Rhinoceros 5.0 (Robert McNeel & Associates,

Seattle, WA). To remove non-shape effects of translation, rotation and size, all bone models were aligned to a standard position in 3D space, following a protocol for long- bone cross-sectional geometry, equivalent to a Procrustes superimposition (Wilson and Humphrey 2015). To quantify differences in morpho-biomechanical properties of the bone, cross-sections of the humeri were extracted at 50% (midshaft) of the length of the bone in Rhinoceros 5.0. This location is widely used as the most informative when analysing biomechanical properties in long bones, supported on applications of beam theory to biological studies (Cubo and Casinos 1998; Simons, et al. 2011; Patel, et al. 2013). A total of 55 cross-sections were then aligned based on the position of the centroid of the cross-sections in the world coordinate system. A set of 16, equiangular landmarks were semi-automatically placed along the periosteal surface of each cross- section to describe the cross-sectional shape of the humerus (Fig. 5.2), following the method described in Wilson and Humphrey (2015).

For our biomechanical dataset we extracted six properties of each cross-section, usually used to describe aerial guilds in flying vertebrates (Wolf, et al. 2010; Simons, et al. 2011; Voeten, et al. 2018): maximum second moment of area (Imax), minimum second moment of area (Imin), second moment of area about the x axis (Ix), second moment of area about the y axis (Iy), polar moment of inertia of an area (J) and circularity (CircMaxR). In beam theory, the second moment of area is important to estimate the deflection and stress a beam can experience. Moments of area can be defined relative to: (1) anatomical orientation (e.g. Iy and Ix) and (2) principal axes of distribution of mass (e.g. Imax and Imin) (Lovejoy, et al. 1976; Ruff and Hayes 1983). 167

Accordingly, Ix measures bending strength in the anterior-posterior plane and Iy measures bending strength in the medial-lateral plane, whereas Imax and Imin measure maximum and minimum bending strength of the bone at that cross-section, respectively (Ruff 1987). J is a measure of resistance against torsional forces (Voeten, et al. 2018). CircMaxR is a ratio between the maximum to actual circumference of a cross section, measuring the overall robusticity of a long bone, based on how circular the cross section is (Wilson and Humphrey 2015). All biomechanical properties were quantified from individual cross-sections using outputs from automated calculations implemented in the AreaMoments command in Rhinoceros 5.0 (Wilson and Humphrey

2015).

Figure 5.2. Three-dimensional virtual reconstruction of the humerus of Austronomus australis, showing a schematic representation of the cross-sectioning and landmarking protocol used in this study. Landmarks were collected at the intersection of equiangular radii and the periosteal contour of the cross section.

168

Ecological and phylogenetic characterisation of species

We compiled taxonomic arrangement, body size and foraging behaviour data for all species. These parameters were assessed in relation to morpho-biomechanical properties of the humerus. Taxonomic arrangement was registered at subordinal level

(i.e. Yangochiroptera or Yinpterochiroptera) (see Appendix Table 7). Humeral length

(HL) was used as proxy for body size, based on the strong correlation (R2= 0.91, P <

0.001; see Appendix Figure 8) between humeral cross-sectional diameter (commonly used to reconstruct body size in fossil bats; (Gunnell, et al. 2009) and HL. Foraging strategies (FG) were classified in five categories (trawling, hawking, gleaning, carnivory, frugivory, terrestrial locomotion), following traditional classifications established for bats (Norberg and Rayner 1987; Denzinger and Schnitzler 2013). Species that exhibit

UR were also recorded to test the effect of this rare behaviour in humeral shape and biomechanical properties.

Statistical analyses

In order to assess whether patterns of morpho-biomechanical variation were divergent, each dataset (shape and biomechanical data) was analysed independently.

Differences in biomechanical traits across foraging and roosting guilds were visualised using boxplots. Kmult and K statistics of the “physignal” and “phylosig” functions in

Geomorph (Adams 2014) and phytools (Revell 2012) were used to test whether shape and biomechanical differences had a phylogenetic structure reflecting evolutionary relatedness. Shi and Rabosky (2015)’s species-level bat super-tree was used as the phylogenetic hypothesis of evolutionary kinship in our sample. Then, the scaling of cross-sectional shape and biomechanical properties with size was tested based on log-

169 transformed HL, using a Procrustes regression with the “procD.lm” function in

Geomorph (Adams, et al. 2013; Adams, et al. 2017). Biomechanical properties were log-transformed for all statistical analyses. Differences in the scaling of shape and biomechanical data between foraging and roosting guilds were tested using pairwise comparisons as implemented in the “pairwise” and “TukeyHSD” functions in R.

Allometric trajectories of biomechanical data were visualised with biplots using log- transformed HL as size. Given the strong statistical significance of scaling in shape and biomechanical data (with the exception of CircMaxR), the residuals of each regression were used as allometry-corrected shape and biomechanical data in subsequent analyses. Species with multiple individuals were analysed based on the mean shape of the specimens using the “mshape” function in Geomorph for R.

Using a Principal Component Analysis (PCA) to visualise and summarise the dimensionality in our data, we reconstructed the spaces of shape variation (hereafter, morphospace), using the “gm.prcomp” function in Geomorph. This was performed for allometry-corrected and phylogeny-allometry-corrected shape data. We estimated differences in the amount of allometry-corrected shape disparity between guilds, using the “morphol.disparity” function in Geomorph for R (Adams, et al. 2017). Regular disparity was used to account for the unequal number of individuals per foraging guild group. Pairwise comparisons in the allometry-corrected shape disparity between guilds were performed using the “pairwise” function in RRPP. All tests of statistical significance were based on the distribution of 10,000 iterations.

170

Results

Scaling of biomechanical and shape data

Procrustes regression analyses showed a significant effect of allometry on shape and all biomechanical variables, excepting CircMaxR (Table 5.1). Allometry explained

79.26% of shape variation and an average of 77.95% across biomechanical traits.

Biplots of the scaling of biomechanical data with log-transformed HL revealed decoupled trajectories across foraging and roosting guilds (Fig. 3). Ix, Iy, Imax, Imin and J showed similar positive allometric trajectories with different slopes across guilds (Fig.

5.3B-F). CircMaxR was the only trait that did not display an allometric relationship (Fig.

5.3A). Pairwise comparisons of scaling in shape and biomechanics revealed significant differences in Imax, Imin and shape between foraging guilds but not when including

UR (Appendix Table 8).

Table 5.1. Individual Procrustes linear regressions for the test of scaling in shape and biomechanical data with humeral length (HL), based on an isometric null hypothesis. Biomechanical properties and HL were log-transformed.

2 Df SS MS R F P CircMaxR 1 0.0000346 0.00003463 0.00147 0.0516 0.8268 J 1 85.503 85.503 0.9397 545.39 <0.001 Iy 1 83.212 83.212 0.92871 455.94 <0.001 Ix 1 85.959 85.959 0.94362 585.81 <0.001 Imax 1 84.897 84.897 0.9292 459.37 <0.001 Imin 1 83.367 83.367 0.93435 498.13 <0.001 Shape 1 52.189 52.189 0.79262 133.77 <0.001

171

Figure 5.3. Allometric trajectories of scaling of biomechanical traits with size (humeral length, HL) in the bat humeri, depending on foraging guild category (FG): carnivore (C), frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown colour represents species with upstand roosting (UR).

Ecological differences in morpho- and mechanospaces

Boxplots of biomechanical data showed differences across foraging and roosting guilds

(Fig. 5.4). Terrestrial locomoting species showed higher values of CircMaxR than any other guild, whereas upstand roosting species showed the lowest values (Fig. 5.4A).

CircMaxR showed low dispersion in our sample, with all taxa ranging between 0.89 and

0.97. Ix, Iy, Imax, Imin and J shared similar patterns with carnivore species having the highest values and with gleaning, trawling and upstand roosting species sharing the lowest values (Fig. 5.4B-F). Procrustes ANOVAs did not reject the null hypothesis of equal patterns of variation between foraging and roosting guilds for CircMaxR and indicated significant differences across guilds for Iy, Imax and shape (Table 5.2). Marginal statistical significance (P<0.01) was found for differences in J across guilds. Pairwise comparisons of differences in shape and biomechanical data across foraging and roosting guilds revealed significant differences between pairs of guilds in all traits

(Appendix Table 9). Procrustes ANOVAs evidenced that foraging and roosting 172 behaviour accounted for more allometry-corrected variation in shape (35.43%) than biomechanics (on average 28.80%).

Figure 5.4. Boxplots of values of biomechanical traits of the bat humeri, depending on foraging guild category (FG): carnivore (C), frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown data points represent species with upstand roosting (UR).

Table 5.2. Individual Procrustes ANOVA for the test of differences in shape and biomechanical data across foraging and roosting guilds.

Df SS MS R2 F Z P CircMaxR 6 0.0044719 0.00074532 0.2177 1.3914 0.71103 0.2463 J 6 1.5998 0.26664 0.29156 2.0578 1.2745 0.0891 Iy 6 2.1011 0.35018 0.32893 2.4508 1.5428 0.0511 Ix 6 1.4333 0.23888 0.27908 1.9356 1.1891 0.106 Imax 6 2.178 0.36299 0.3367 2.5381 1.5913 0.0429 Imin 6 1.6059 0.26765 0.27416 1.8885 1.1512 0.1152 Shape 6 4.8383 0.80638 0.35433 2.7439 1.8057 0.0422 Based on the PCAs performed for shape data (allometry-corrected and phylogeny- allometry-corrected), the first two principal components of the morpho- and mechanospaces explained an average of 97.42% of the variance in shape (97.43% and

97.4% for allometry-corrected and phylogeny-allometry-corrected shape, respectively). For both datasets, PC1 primarily divides gleaning and hawking species

173 from carnivore and frugivore species (each pair of foraging guilds in opposite ends of

PC1 and PC2; Fig. 5.5). Terrestrial locomoting, gleaning and upstand roosting species showed the lowest inter- and intraguild dispersion in morphospace, each guild clustered around the origin of PC1 and PC2. Austronomus australis, Phyllostomus hastatus, Pteronotus parnellii and Rousettus bidens occupied non-overlapping subspaces on opposite ends of PC1 and PC2, distant from any other taxa. A. australis and P. hastatus shared a slight mediolateral constriction (both occupying the negative- most end of PC2), R. bidens showed a more elliptical shape (positive end of PC1), whereas P. parnellii showed a more circular shape (negative end of PC1). Trawling and frugivore taxa showed the highest dispersion of any guild, the former overlapping with most guilds, and the latter occupying a relatively exclusive subspace. Variation in allometry-corrected shape (Fig. 5.5A) and phylogeny-allometry-corrected shape (Fig.

5.5B) showed similar dispersion across morphospace, evidencing the lack of phylogenetic signal in humeral cross-sectional shape. This pattern was preserved even after removing outlying species (i.e. A. australis, P. hastatus, P. parnellii and R. bidens;

Appendix Figure 9). UR taxa were consistently clustered together, irrespective of phylogenetic relatedness and allometry.

174

Figure 5.5. Morphospace of humerus allometry-corrected cross-sectional shape (A) and allometry-phylogeny- corrected (B) shape data. Data points are colour-coded by foraging guild category: carnivore (C), frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown data points represent species with upstand roosting (UR). Deformation grids show cross-sectional shape of species at opposite extremes of variation along PC1 and PC2.

Shape disparity was consistently higher in trawling and frugivore species, being the only guilds significantly different from the rest (Appendix Table 10), whereas terrestrial locomoting and gleaning species were the foraging guilds with the lowest disparity

(Fig. 5.6). Shape disparity was lower in upstand roosting taxa than in any foraging guild, although not statistically significant (Appendix Table 10).

175

Figure 5.6. Phenotypic disparity based on shape variance for humeral cross-sectional shape. Shape disparity was deconstructed based on foraging guild categories: carnivore (C), frugivore (F), gleaning (G), hawking (H), trawling (T) and terrestrial locomotor (TL). Brown data points represent species with upstand roosting (UR).

Discussion

Combining three-dimensional comparative morphology and functional biomechanics, this study represents a novel approach to investigate the correspondence between the ecological and biological drivers shaping humeral variation, by deconstructing its biomechanical and morphological components. Our results showed a strong link between phenotypic variation and ecological differences between taxa, corresponding with previous studies both in bats (Arbour, et al. 2019) and a wide variety of other vertebrates (Esquerré, et al. 2017; Maestri, et al. 2017; Stange, et al. 2018; Arbour, et al. 2019; Pimiento, et al. 2019; Stanchak, et al. 2019). Our findings of strong differences in the scaling of shape and biomechanics of the humerus supports the general notion of Wolff’s law of bone functional adaptation (Wolff 1986; Ruff, et al.

2006), contrary to common allometric trajectories found in crocodylians (Meers

2002). Moreover, we also found a marked decoupling between the scaling of humeral

176 shape and biomechanics across foraging and roosting guilds, where ecology and size were associated differently with shape and biomechanical properties. A strong correlation between biomechanical properties, size and ecological traits has been suggested to have influenced the morphological evolution of many vertebrate taxa

(McElroy, et al. 2008; Houssaye, et al. 2016; Muñoz, et al. 2018). Nevertheless, the scale of the biomechanics-form-ecology interaction on macroevolutionary trajectories requires further study.

Scaling of biomechanical and shape data

Our results reflect the role of size on vertebrate morphological diversification as a most-parsimonious evolutionary pathway (Marroig and Cheverud 2005; Friedman, et al. 2019). Foraging behaviour also explained a significant portion of trait variability, following similar trends found between the morphological evolution of the chiropteran skull and diet (Arbour, et al. 2019; Rossoni, et al. 2019). Upstand roosting was also shown to explain differences in the scaling and variation of shape and biomechanical properties, indicating that non-locomotory behaviours can also impact postcranial morphology. Similar patterns of allometry-corrected and size-dependent differences between shape and biomechanical data suggest foraging and roosting behaviour repatterned humeral morpho-biomechanical variation and its scaling. Femoral morphological diversity in bats has been described as size-dependent, linking biomechanical aspects of long bones (i.e. increase in robusticity) with increases in body mass (Louzada, et al. 2019). Our results strongly support this finding, as size was highly associated with almost all biomechanical properties, CircMaxR being the only exception, possibly due to all taxa having highly circular midshafts, resulting in little

177 variation. Body size evolution in bats is thought to have been constrained mainly by echolocation and flight (Jones 1999; Giannini 2012; Norberg and Norberg 2012;

Moyers Arévalo, et al. 2018). Echolocation appears to have imposed the strongest constraint, greatly limiting size increases in insectivorous bats, whereas non- echolocating bats (e.g. frugivorous pteropodids) evolved larger body sizes (Moyers

Arévalo, et al. 2018). Flight acted as a posteriori secondary constraint, limiting maximum body size in larger-than-average non-echolocating bats (e.g. pteropodids)

(Giannini 2012; Moyers Arévalo, et al. 2018; Amador, Almeida, et al. 2019). This pattern can be found in our sample, where larger species were either frugivorous or omnivorous, and animalivorous species were smaller. Nevertheless, ecological opportunity fostered adaptive radiation and morphological diversification in bats, leading to the colonisation of highly specialised niches, including the convergent evolution of larger size in animalivorous bats (Santana and Cheung 2016). Future studies should focus on increasing sampling to better capture intraguild morphological variability.

Ecological differences in shape and biomechanical properties

Diet-related morphological differences have been found both in closely-related mammal taxa as well as in sympatric taxa, indicating that these differences can have ecological and evolutionary drivers (Adams and Rohlf 2000; Marcé-Nogué, et al. 2017).

Dietary diversification is thought to have been a major driver of and morphological specialisation in bats, resulting in multiple cases of convergent evolution of cranial phenotypes (Datzmann, et al. 2010; Santana, et al. 2010; Santana, et al. 2012). Correspondence between previous results of cranial morphological

178 disparity and ours of humeral shape disparity suggests that diet and foraging behaviour may have shaped evolution of the cranium and postcranium along similar adaptive pathways.

Morphospace across allometry- and phylogeny-allometry-corrected shape data showed common general patterns of species dispersion based on foraging and roosting behaviour, with frugivore, carnivore and upstand roosting species occupying non-overlapping subspaces distant from a cluster of mostly gleaning and hawking species. Frugivore taxa tended to have more elliptical cross-sections (R. bidens), whereas carnivore taxa showed a mediolateral constriction in cross-sectional shape

(e.g. P. hastatus). Similar patterns of variation in both types of shape data indicate the overarching importance of foraging and roosting behaviour to explain differences in shape data, and that humeral cross-sectional shape variance is not explained by phylogeny. New approaches to comparative analyses of multidimensional phenotypic data have found support for more complex models of morphological disparity in mammals than traditional models based on a priori assumptions (Maestri, et al. 2017;

Thiagavel, et al. 2018; Arbour, et al. 2019; Rossoni, et al. 2019). Moreover, in our study shape showed a decrease in dispersion of species with rare behavioural ecology, such as terrestrial locomotion in vampire bats and mystacinids and upstand roosting in thyropterids and myzopodids, providing evidence for functionally canalised humeral phenotypes not associated with flight. In contrast, trawling species showed the highest morphospace dispersion and shape disparity, indicating diverging phenotypes successfully adapted to this behaviour. Similar to our morphospace results, shape disparity showed dissimilarity between foraging and roosting behaviours. Differences

179 in shape disparity based on behaviour underscore the importance of dietary and roosting strategies in similar patterns of cranial and postcranial morphological variation in bats. However, sample sizes in the foraging groups were highly unequal, limiting the generalisation of our disparity results. Changes in cranial morphology have been linked to the convergent evolution of foraging strategies in Myotis, highlighting the interplay between diet, foraging, and cranial and postcranial morphology (Morales, et al. 2019). Nevertheless, differences in disparity require further investigation to assess how these may related to intraguild variation in foraging patterns and habitat location. Further species-specific ecological studies could also help inform a refined foraging category classification, as some species commonly fluctuate between foraging strategies (e.g. Mystacina is a terrestrial locomotor as well as an aerial insectivore, frugivore, and carnivore; Lloyd 2001).

Mineralisation of the wing bones in bats shows a decreasing trend along the proximo- distal axis, with the humerus showing the highest levels of mineralisation

(Papadimitriou, et al. 1996; Watts, et al. 2001; Swartz and Middleton 2008). That pattern corresponds to the characterization of the armwing as the place of insertion of primary flight muscles of the wing (i.e. production of most of the force that powers up- and downstroke), whereas the handwing conferred higher manoeuvrability during bat flight evolution (Swartz, et al. 1992; Amador, Almeida, et al. 2019). Considering that flight can potentially limit body size evolvability in bats (Giannini 2012; Moyers

Arévalo, et al. 2018; Amador, Almeida, et al. 2019), variation in biomechanical allometry in the humerus suggests that higher kinematic demands in some guilds could have played a role in biomechanical diversification, optimising functional performance

180

(Wolff 1986; Meers 2002; Ruff, et al. 2006; Barak, et al. 2011). Similar patterns of foraging- and roosting-dependent allometry across different biomechanical traits suggests that biomechanical variation is homogeneous at the sampled midshaft, indicating highly similar patterns of cross-sectional variation across species (Cubo and

Casinos 1998). Our results seem to support Wolff’s law of bone functional adaptation, with shape and biomechanical variation and scaling both correlated with behavioural differences across taxa (Wolff 1986; Ruff, et al. 2006; Barak, et al. 2011; Kivell 2016).

In contrast, behaviour-based differences in shape disparity suggest that trawling and a frugivorous diet seem to be associated with more variability, whereas terrestrial locomotion, gleaning and upstand roosting with less variability. Less manoeuvrable flight (long, thin wings and restricted movement around elbow and shoulder) is generally correlated with fast flying aerial insectivory (Norberg and Rayner 1987), while more manoeuvrable flight (short, wide wings, greater movement in elbow and shoulder joints) is associated with gleaning insectivory and non-insectivory (e.g. frugivory, carnivory, sanguivory) (Norberg and Rayner 1987). As such, we suggest that aerial insectivory may have constrained shape disparity.

Considering the importance that manoeuvrability has had for the dietary diversification (i.e. foraging guilds) of bats and how it is highly dependent on shoulder and wrist joint mobility (Panyutina, et al. 2015; Amador, Almeida, et al. 2019), our results suggest that humeral shape is more functionally related to foraging and dietary strategies than phylogenetically constrained. We suggest that biomechanical scaling of the humerus is foraging-dependent, corresponding with the evolutionary role that the armwing had for the early acquisition of flight and how flight could have constrained

181 body size evolution in bats (Moyers Arévalo, et al. 2018; Amador, Simmons, et al.

2019). Moreover, decreased shape disparity and dispersion in morphospace in terrestrial locomotors indicates that this adaptation could have canalised humeral phenotypes despite the ecological differences between these taxa (Desmodus is a highly specialised blood feeder, whereas Mystacina is a dietary generalist). A similar pattern was found for upstand roosting species, where despite their phylogenetic distance and ecological differences (Thyroptera is a Neotropical gleaner and Myzopoda an African-insular hawker), they showed low shape disparity and close clustering in morphospace. Our results provide evidence for the decoupled variation and scaling of humeral shape and biomechanics across foraging and roosting guilds, emphasising the need to test not only multivariate and ecologically-informed hypotheses, but also to test them for different attributes (e.g. similar pattern of decoupled mopho- biomechanical variation found in carnivorans’ limb; Kilbourne and Hutchinson 2019).

The results presented in this study, although based on an ecologically comprehensive sample, are limited in their taxonomic sampling. Future studies could focus on exploring intra- and inter-guild variability by expanding the number of sample genera and species. More detailed species-specific ecological and behavioural descriptions would also inform more complex mechanistic hypotheses that can shed light on the phenotype-diet-locomotion interaction. Moreover, further studies should include information on other wing bone elements, as well as expanding the taxonomic sampling, to explore more detailed evolutionary hypotheses of locomotory evolution in bats. The analyses presented in this study could provide palaeobiological insights into fossil taxa, as they could be included within a broader study of shape space in modern taxa. 182

Conclusions

This study analyses the morpho-biomechanical disparity and scaling in the humerus of modern bats. We quantified differences in cross-sectional geometry of the humerus across bats with different foraging behaviours, providing novel information about the ecomorphology of locomotion in Chiroptera. We found support for differences in humeral morpho-biomechanical properties based on foraging and roosting behaviour in bats, in support of Wolff’s law of bone functional remodelling. This pattern does not appear to reflect phylogenetic structuring, but rather an ecological signature associated mainly with the evolution of divergent foraging and roosting behaviours and larger body size. We provide evidence for high correlation between humeral length and cross-sectional shape and biomechanics, with variations in allometric trajectories found across guilds. Our results suggest a correspondence between the evolutionary trajectories in cranial phenotypic diversification shown in previous studies

(Arbour, et al. 2019; Rossoni, et al. 2019), and the evolutionary trajectories found in humeral phenotypic variability in our study. We found canalisation in shape disparity associated with functional demands that novel locomotory (e.g. terrestrial locomotion) and roosting (e.g. upstand roosting) behaviours may have imposed on the wing of bats, shaping the ecomorphology of bat humeri beyond the evolution of flight.

References

Adams DC. 2014. A generalized K statistic for estimating phylogenetic signal from shape and other high-dimensional multivariate data. Syst Biol 63:685-697. Adams DC, Collyer ML, Kaliontzopoulou A, Sherratt E. 2017. Geomorph: software for geometric morphometric analyses. R package. Version 3.0.5: https://cran.r- project.org/package=geomorph. Adams DC, Otárola-Castillo E, Paradis E. 2013. Geomorph: an R package for the collection and analysis of geometric morphometric shape data. Methods in Ecology and Evolution 4:393-399. 183

Adams DC, Rohlf FJ. 2000. Ecological character displacement in Plethodon: Biomechanical differences found from a geometric morphometric study. Proceedings of the National Academy of Sciences 97:4106-4111. Amador LI, Almeida FC, Giannini NP. 2019. Evolution of Traditional Aerodynamic Variables in Bats (Mammalia: Chiroptera) within a Comprehensive Phylogenetic Framework. Journal of Mammalian Evolution 27: 549-561. Amador LI, Simmons NB, Giannini NP. 2019. Aerodynamic reconstruction of the primitive fossil bat Onychonycteris finneyi (Mammalia: Chiroptera). Biology Letters 15:20180857. Arbour JH, Curtis AA, Santana SE. 2019. Signatures of echolocation and dietary ecology in the adaptive evolution of skull shape in bats. Nat Commun 10:2036. Barak MM, Lieberman DE, Hublin J-J. 2011. A Wolff in sheep's clothing: Trabecular bone adaptation in response to changes in joint loading orientation. Bone 49:1141- 1151. Boerma DB, Barrantes JP, Chung C, Chaverri G, Swartz SM. 2019. Specialized landing maneuvers in Spix's disk-winged bats (Thyroptera tricolor) reveal linkage between roosting ecology and landing biomechanics. The Journal of Experimental Biology:jeb.204024. Canals M, Iriarte-Diaz J, Grossi B. 2011. Biomechanical, Respiratory and Cardiovascular Adaptations of Bats and the Case of the Small Community of Bats in Chile. In: Klika V, editor. Biomechanics in Applications: InTech. p. 299-322. Cubo J, Casinos A. 1998. The variation of the cross-sectional shape in the long bones of birds and mammals. Annales des Sciences Naturelles - Zoologie et Biologie Animale 19:51-62. Datzmann T, von Helversen O, Mayer F. 2010. Evolution of nectarivory in phyllostomid bats (Phyllostomidae Gray, 1825, Chiroptera: Mammalia). BMC Evol Biol 10: 165. Davalos LM, Velazco PM, Warsi OM, Smits PD, Simmons NB. 2014. Integrating incomplete fossils by isolating conflicting signal in saturated and non-independent morphological characters. Syst Biol 63:582-600. Denzinger A, Schnitzler HU. 2013. Bat guilds, a concept to classify the highly diverse foraging and echolocation behaviors of microchiropteran bats. Front Physiol 4:164. Dickinson M. 2008. Animal locomotion: a new spin on bat flight. Curr Biol 18:R468-470. Esquerré D, Sherratt E, Keogh JS. 2017. Evolution of extreme ontogenetic allometric diversity and heterochrony in pythons, a clade of giant and dwarf snakes. Evolution 71:2829-2844. Fenton MB. 2010. Convergences in the diversification of bats. Current Zoology (formerly Acta Zoologica Sinica) 56: 454-468. Friedman ST, Martinez CM, Price SA, Wainwright PC. 2019. The influence of size on body shape diversification across Indo‐Pacific shore fishes. Evolution 73: 1873-1884. Giannini NP. 2012. Toward an integrative theory on the origin of bat flight. In: Gunnell GF, Simmons NB, editors. Evolutionary History of Bats: Fossils, Molecules and Morphology. Cambridge: Cambridge University Press. p. 353-384. Gunnell GF, Worsham SR, Seiffert ER, Simons EL. 2009. Vampyravus orientalis Schlosser (Chiroptera) from the Early Oligocene (Rupelian), Fayum, Egypt - body mass, humeral morphology and affinitiesd. Acta Chiropterologica 11:271-278. Habib MB, Ruff CB. 2008. The effects of locomotion on the structural characteristics of avian limb bones. Zoological Journal of the Linnean Society 153:601-624. 184

Hand SJ, Beck RMD, Archer M, Simmons NB, Gunnell GF, Scofield RP, Tennyson AJD, De Pietri VL, Salisbury SW, Worthy TH. 2018. A new, large-bodied omnivorous bat (Noctilionoidea: Mystacinidae) reveals lost morphological and ecological diversity since the Miocene in New Zealand. Sci Rep 8:235. Hand SJ, Lee DE, Worthy TH, Archer M, Worthy JP, Tennyson AJ, Salisbury SW, Scofield RP, Mildenhall DC, Kennedy EM, et al. 2015. Miocene Fossils Reveal Ancient Roots for New Zealand's Endemic Mystacina (Chiroptera) and Its Rainforest Habitat. PLoS ONE 10:e0128871. Hand SJ, Weisbecker V, Beck RM, Archer M, Godthelp H, Tennyson AJ, Worthy TH. 2009. Bats that walk: a new evolutionary hypothesis for the terrestrial behaviour of New Zealand's endemic mystacinids. BMC Evol Biol 9:169. Hand SJ, Worthy TH, Archer M, Worthy JP, Tennyson AJ, Scofield RP. 2013. Miocene mystacinids (Chiroptera, Noctilionoidea) indicate a long history for endemic bats in New Zealand. Journal of Vertebrate Paleontology 33:1442-1448. Hedenstrom A, Johansson LC. 2015. Bat flight: aerodynamics, kinematics and flight morphology. Journal of Experimental Biology 218:653-663. Hedrick BP, Mutumi GL, Munteanu VD, Sadier A, Davies KTJ, Rossiter SJ, Sears KE, Dávalos LM, Dumont E. 2019. Morphological Diversification under High Integration in a Hyper Diverse Mammal Clade. Journal of Mammalian Evolution 27, 563–575. Houssaye A, Botton-Divet L. 2018. From land to water: evolutionary changes in long bone microanatomy of otters (Mammalia: Mustelidae). Biological Journal of the Linnean Society 125:240-249. Houssaye A, Waskow K, Hayashi S, Cornette R, Lee AH, Hutchinson JR. 2016. Biomechanical evolution of solid bones in large animals: a microanatomical investigation. Biological Journal of the Linnean Society 117:350-371. Iriarte-Diaz J, Riskin DK, Breuer KS, Swartz SM. 2012. Kinematic plasticity during flight in fruit bats: individual variability in response to loading. PLoS ONE 7:e36665. Jones G. 1999. Scaling of echolocation call parameters in bats. Journal of Experimental Biology 202:3359-3367. Kilbourne BM, Hutchinson JR. 2019. Morphological diversification of biomechanical traits: mustelid locomotor specializations and the macroevolution of long bone cross- sectional morphology. BMC Evolutionary Biology 19:37. Kivell TL. 2016. A review of trabecular bone functional adaptation: what have we learned from trabecular analyses in extant hominoids and what can we apply to fossils? Journal of Anatomy 228:569-594. Louzada NSV, Nogueira MR, Pessôa LM. 2019. Comparative morphology and scaling of the femur in yangochiropteran bats. Journal of Anatomy 235:124-150. Lovejoy CO, Burstein AH, Heiple KG. 1976. The biomechanical analysis of bone strength: A method and its application to platycnemia. American Journal of Physical Anthropology 44:489-505. Lloyd BD. 2001. Advances in New Zealand mammalogy 1990–2000: short-tailed bats. Journal of the Royal Society of New Zealand 31: 59–81. Macintosh AA, Pinhasi R, Stock JT. 2014. Lower limb skeletal biomechanics track long- term decline in mobility across similar to 6150 years of agriculture in Central Europe. Journal of Archaeological Science 52:376-390.

185

Macintosh AA, Pinhasi R, Stock JT. 2017. Prehistoric women's manual labor exceeded that of athletes through the first 5500 years of farming in Central Europe. Science Advances 3: eaao3893. Maestri R, Monteiro LR, Fornel R, Upham NS, Patterson BD, de Freitas TRO. 2017. The ecology of a continental evolutionary radiation: Is the radiation of sigmodontine rodents adaptive? Evolution 71:610-632. Marcé-Nogué J, Püschel TA, Kaiser TM. 2017. A biomechanical approach to understand the ecomorphological relationship between primate mandibles and diet. Scientific Reports 7:8364. Marinello MM, Bernard E. 2014. Wing morphology of Neotropical bats: a quantitative and qualitative analysis with implications for habitat use. Canadian Journal of Zoology 92:141-147. Marroig G, Cheverud JM. 2005. Size as a line of least evolutionary resistance: Diet and adaptive morphological radiation in new world monkeys. Evolution 59:1128-1142. McElroy EJ, Hickey KL, Reilly SM. 2008. The correlated evolution of biomechanics, gait and foraging mode in lizards. Journal of Experimental Biology 211:1029-1040. Meers MB. 2002. Cross-sectional geometric properties of the crocodylian humerus: an exception to Wolff's Law? Journal of Zoology 258:405-418. Morales AE, Ruedi M, Field K, Carstens BC. 2019. Diversification rates have no effect on the convergent evolution of foraging strategies in the most speciose genus of bats, Myotis. Evolution 73:2263-2280. Moyers Arévalo RL, Amador LI, Almeida FC, Giannini NP. 2018. Evolution of Body Mass in Bats: Insights from a Large Supermatrix Phylogeny. Journal of Mammalian Evolution 27: 123-138. Muñoz MM, Hu Y, Anderson PSL, Patek SN. 2018. Strong biomechanical relationships bias the tempo and mode of morphological evolution. eLife 7:e37621. Norberg RA, Rayner JMV. 1987. Ecological Morphology and Flight in Bats (Mammalia; Chiroptera): Wing Adaptations, Flight Performance, Foraging Strategy and Echolocation. Philos Trans R Soc Lond B Biol Sci 316:335-427. Norberg UM. 1990. Vertebrate Flight, Mechanics, Physiology, Morphology, Ecology and Evolution: Springer-Verlag. Norberg UM, Norberg RA. 2012. Scaling of wingbeat frequency with body mass in bats and limits to maximum bat size. J Exp Biol 215:711-722. Panyutina A, Korzun LP, Kuznetsov AN. 2015. Flight of Mammals: From Terrestrial Limbs to Wings: Springer, New York. Papadimitriou HM, Swartz SM, Kunz TH. 1996. Ontogenetic and anatomic variation in mineralization of the wing skeleton of the Mexican free-tailed bat, Tadarida brasiliensis. Journal of Zoology 240:411-426. Patel BA, Ruff CB, Simons EL, Organ JM. 2013. Humeral cross-sectional shape in suspensory primates and sloths. Anat Rec (Hoboken) 296:545-556. Pimiento C, Cantalapiedra JL, Shimada K, Field DJ, Smaers JB. 2019. Evolutionary pathways toward gigantism in sharks and rays. Evolution 73:588-599. Pratt IV, Johnston JD, Walker E, Cooper DML. 2018. Interpreting the three-dimensional orientation of vascular canals and cross-sectional geometry of cortical bone in birds and bats. Journal of Anatomy 232: 931-942. Rayner JM. 1988. The evolution of vertebrate flight. Biological Journal of the Linnean Society 34:269-287. 186

Revell LJ. 2012. phytools: an R package for phylogenetic comparative biology (and other things). Methods in Ecology and Evolution 3:217-223. Riskin DK, Bertram JEA, Hermanson JW. 2005. Testing the hindlimb-strength hypothesis: non-aerial locomotion by Chiroptera is not constrained by the dimensions of the femur or tibia. Journal of Experimental Biology 208:1309-1319. Riskin DK, Fenton MB. 2001. Sticking ability in Spix's disk-winged bat, Thyroptera tricolor (Microchiroptera : Thyropteridae). Canadian Journal of Zoology 79:2261-2267. Riskin DK, Iriarte-Diaz J, Middleton KM, Breuer KS, Swartz SM. 2010. The effect of body size on the wing movements of pteropodid bats, with insights into thrust and lift production. J Exp Biol 213:4110-4122. Riskin DK, Parsons S, Schutt WA, Jr., Carter GG, Hermanson JW. 2006. Terrestrial locomotion of the New Zealand short-tailed bat Mystacina tuberculata and the common vampire bat Desmodus rotundus. J Exp Biol 209:1725-1736. Riskin DK, Racey PA. 2010. How do sucker-footed bats hold on, and why do they roost head-up? Biological Journal of the Linnean Society 99:233-240. Rossoni DM, Assis APA, Giannini NP, Marroig G. 2017. Intense natural selection preceded the invasion of new adaptive zones during the radiation of New World leaf- nosed bats. Scientific Reports 7: 11076. Rossoni DM, Costa BMA, Giannini NP, Marroig G. 2019. A multiple peak adaptive landscape based on feeding strategies and roosting ecology shaped the evolution of cranial covariance structure and morphological differentiation in phyllostomid bats. Evolution 73:961-981. Ruff C. 1987. Sexual dimorphism in human lower limb bone structure: relationship to subsistence strategy and sexual division of labor. Journal of Human Evolution 16:391- 416. Ruff C, Holt B, Trinkaus E. 2006. Who's afraid of the big bad Wolff?: "Wolff's law" and bone functional adaptation. Am J Phys Anthropol 129:484-498. Ruff CB, Hayes WC. 1983. Cross-sectional geometry of Pecos Pueblo femora and tibiae—A biomechanical investigation: I. Method and general patterns of variation. American Journal of Physical Anthropology 60:359-381. Salathe EP, Arangio GA, Salathe EP. 1989. An application of beam theory to determine the stress and deformation of long bones. Journal of Biomechanics 22:189-199. Santana SE, Cheung E. 2016. Go big or go fish: morphological specializations in carnivorous bats. Proceedings of the Royal Society B: Biological Sciences 283:20160615. Santana SE, Dumont ER, Davis JL. 2010. Mechanics of bite force production and its relationship to diet in bats. Functional Ecology 24:776-784. Santana SE, Grosse IR, Dumont ER. 2012. Dietary hardness, loading behavior, and the evolution of skull form in bats. Evolution 66:2587-2598. Schliemann H, Goodman SM. 2011. A new study on the structure and function of the adhesive organs of the Old World sucker-footed bat (Myzopoda: Myzopodidae) of Madagascar. Verh. Naturwiss. Ver. Hamburg. Shi JJ, Rabosky DL. 2015. Speciation dynamics during the global radiation of extant bats. Evolution 69:1528-1545. Simons EL, Hieronymus TL, O'Connor PM. 2011. Cross sectional geometry of the forelimb skeleton and flight mode in pelecaniform birds. J Morphol 272:958-971.

187

Stanchak KE, Arbour JH, Santana SE. 2019. Anatomical diversification of a skeletal novelty in bat feet. Evolution 73: 1591-1603. Stange M, Aguirre-Fernandez G, Salzburger W, Sánchez-Villagra MR. 2018. Study of morphological variation of northern Neotropical Ariidae reveals conservatism despite macrohabitat transitions. BMC Evol Biol 18:38. Swartz SM, Bennett MB, Carrier DR. 1992. Wing bone stresses in free flying bats and the evolution of skeletal design for flight. Nature 359: 726-729. Swartz SM, Middleton KM. 2008. Biomechanics of the bat limb skeleton: Scaling, material properties and mechanics. Cells Tissues Organs 187:59-84. Thiagavel J, Cechetto C, Santana SE, Jakobsen L, Warrant EJ, Ratcliffe JM. 2018. Auditory opportunity and visual constraint enabled the evolution of echolocation in bats. Nat Commun 9:98. Voeten D, Cubo J, de Margerie E, Roper M, Beyrand V, Bures S, Tafforeau P, Sánchez S. 2018. Wing bone geometry reveals active flight in Archaeopteryx. Nat Commun 9:923. Watts P, Mitchell E, Swartz S. 2001. A computational model for estimating the mechanics of horizontal flapping flight in bats: Model description and validation. Journal of Experimental Biology 204:2873-2898. Wilson L, Humphrey LT. 2015. A Virtual geometric morphometric approach to the quantification of long bone bilateral asymmetry and cross‐sectional shape. American Journal of Physical Anthropology 4: 541-556. Wolf M, Johansson LC, von Busse R, Winter Y, Hedenstrom A. 2010. Kinematics of flight and the relationship to the vortex wake of a Pallas' long tongued bat (Glossophaga soricina). J Exp Biol 213:2142-2153. Wolff J. 1986. Concept of the Law of Bone Remodelling. In: Wolff J, editor. The Law of Bone Remodelling. Berlin, Heidelberg: Springer Berlin Heidelberg. p. 1-1.

188

Chapter 6 Phylogeny and foraging behaviour shape the modular morphological variation in bat humeri4 Abstract

Bats show a remarkable ecological diversity that is reflected both in dietary and foraging guilds. Cranial ecomorphological adaptations linked to diet have been widely studied in bats, using a variety of anatomical, computational and mathematical approaches. However, foraging-related ecomorphological adaptations and the concordance between cranial and postcranial morphological adaptations remains unexamined in bats and limited to the interpretation of traditional aerodynamic properties of the wing (e.g. wing loading and aspect ratio). For this reason, the postcranial ecomorphological diversity in bats and its drivers remain understudied.

Using 3D virtual modelling and geometric morphometrics, we explored the phylogenetic, ecological and biological drivers of humeral morphology in bats, evaluating the presence and magnitude of modularity and integration. To explore decoupled patterns of variation across the bone, we analysed whole-bone shape, diaphyseal and epiphyseal shape. We also tested whether traditional aerodynamic wing traits correlate with humeral shape. By studying 37 species from 20 families

(covering all FG and 85% of dietary guilds), we found similar patterns of variation in whole-bone and diaphyseal shape and unique variation patterns in epiphyseal shape.

Phylogeny, diet and foraging guild significantly correlated with shape variation at all

4 A version of this manuscript has been submitted and it is currently under consideration with major revisions for publication by Journal of Anatomy. * I certify that this manuscript was a direct result of my research towards this PhD, including the design, analysis and interpretation of results, and that reproduction in this thesis does not breach copyright regulations. I acknowledge my co-authors for their supervisory guidance and their contributions to the publication. 189 levels, whereas size only had a significant effect on epiphyseal morphology. We found a significant phylogenetic signal in all levels of humeral shape. Epiphyseal shape significantly correlated with wing aspect ratio. Statistical support for a diaphyseal- epiphyseal modular partition of the humerus suggests a functional partition of shape variability. Our study is the first to show within-structure modular morphological variation in the appendicular skeleton of any living tetrapod. Our results suggest that diaphyseal shape correlates more with phylogeny, whereas epiphyseal shape correlates with diet and foraging guild.

Introduction

Animal locomotion is a key component of the ecological interactions that shape ecosystem functioning (Denzinger and Schnitzler 2013; da Silva, et al. 2014).

Locomotory strategy has an important role in the evolvability and adaptability of taxa

(Witton 2015; Dececchi, et al. 2016) by shaping biological traits both at a micro- and macroscale (Patel, et al. 2013; Martin-Serra, et al. 2014; Verde Arregoitia, et al. 2017;

Medina, et al. 2018; Luo, et al. 2019). Locomotion has also been a major evolutionary driver in animals, enabling occupation of novel ecological niches in some cases

(Simmons, et al. 2008; Sallan and Friedman 2012) and an evolutionary constraint limiting adaptability in other cases (McInroe, et al. 2016; Gutarra, et al. 2019).

Phenotypic specialisations have evolved (e.g. increased bone density and higher metabolic rates), fulfilling the functional demands associated with locomotory strategies (Voigt, et al. 2012; Carter and Adams 2016; Dececchi, et al. 2016). Moreover, phenotypic adaptations for locomotion also have a phylogenetic component, reflecting evolutionary relationships among taxa (Hand, et al. 2009; Fabre, et al. 2015). The link

190 between phylogeny, ecology and morphology in studies of bat locomotion has proven to be variable and sometimes inconsistent with predictions based on various ecomorphological hypotheses (Diogo 2017).

Bats are remarkably diverse from both a taxonomic (second most speciose mammal group after rodents) and ecological standpoint (widest dietary range among mammals)

(Hedrick, et al. 2019), as well as being the only mammals capable of self-powered flight

(Rayner 1988). Despite possessing a hyper-specialised postcranium adapted for flight

(e.g. elongated metacarpals, increased bone density and keeled sternum) (Panyutina, et al. 2015), bats show a remarkable repertoire of locomotory and foraging behaviours, including hovering flight, terrestrial locomotion, water-surface trawling and long-distance migration (Norberg and Rayner 1987; Hand, et al. 2009; Weller, et al. 2016). A key aspect for understanding how animals evolve and diversify is understanding form-to-function inter-dependence (Ferry-Graham, et al. 2002).

Locomotory and foraging strategies have allowed for the incursion and colonisation of novel niches that are unique to bats among mammals, for example, terrestrial locomotion coevolved with sanguivory in vampires and hovering flight with nectar feeding (Amador, Almeida, et al. 2019). However, most studies on functional morphology in bats have focused on cranial morphology and adaptations that can be linked directly to diet and echolocation (Monteiro and Nogueira 2011; Rossoni, et al.

2017; Arbour, et al. 2019; Rossoni, et al. 2019). Both diet and echolocation have linked cranial phenotypic diversification and evolutionary adaptive radiations in bats

(Santana, et al. 2012a; Santana and Cheung 2016; Rossoni, et al. 2017; Arbour, et al.

2019; Hedrick, et al. 2019; Rossoni, et al. 2019), shedding light on the

191 macroevolutionary trajectories that shaped modern bat diversity (Dumont, et al. 2012;

Dumont, et al. 2014). Tooth row complexity, cranial shape and size, nose-leaf morphology and biomechanical performance have all been linked to the colonisation of dietary and echolocating niches during major diversification events (Monteiro and

Nogueira 2011; Santana, et al. 2011; Arbour, et al. 2019; Rossoni, et al. 2019; Brokaw and Smotherman 2020).

Studies on postcranial functional morphology in bats, on the other hand, have mostly focused on the evolution of flight (Norberg and Rayner 1987; Simmons, et al. 2008;

Amador, Giannini, et al. 2018; Stanchak, et al. 2019). Using traditional metrics of aerodynamic properties and descriptive anatomy of the wing of living and a few complete fossils, evolutionary trajectories have been reconstructed (Norberg and

Rayner 1987; Amador, Almeida, et al. 2019; Amador, Simmons, et al. 2019), providing support for an arboreal ancestor for bats (Smith 1977; Rayner 1988; Bishop 2008;

Simmons, et al. 2008). Studies reconstructing the aerofoil of Onychonycteris finneyi

(one of the most complete bat fossils ever found) using traditional aerodynamic modelling concluded that it had a primitive locomotor apparatus and was capable of self-powered flight (Amador, Simmons, et al. 2019). That study described O. finneyi’s armwing (i.e. stylopod and zeugopod) as biomechanically adapted for unsophisticated self-powered flight, suggesting the handwing’s development was a key innovation during early bat flight evolution (Amador, Simmons, et al. 2019). Enhanced flight manoeuvrability, and the origin of many modern families, has been linked to the subsequent evolution of morphological adaptations in the handwing (Amador,

Simmons, et al. 2019). Another study gathered data on the aerodynamic properties of

192 the wing of a wide variety of modern bats to reconstruct the evolutionary trajectory of bat wing aerodynamics (Amador, Almeida, et al. 2019). It hypothesised an Oligo-

Miocene aerial diversification in bats that was associated with dietary specialisations, loss of echolocation in one lineage and optimal adaptation to environmental change

(Amador, Almeida, et al. 2019). Body size evolution has also been linked to the evolution of flight in bats, flight acting as a selective constraint in limiting maximum body size in lineages of larger modern bats (e.g. pteropodids) (Moyers Arévalo, et al.

2018).

Foraging guilds based on morphological traits have also being described for bats by interpreting differences in the aerodynamic properties of the wing (Norberg and

Rayner 1987; Bullen and McKenzie 2001; Denzinger and Schnitzler 2013; Amador,

Almeida, et al. 2019), depicting foraging differences across a bivariate gradient of aspect ratio (AR) and wing loading (WL). AR is a wingspan to wing area ratio that is correlated with the energetic cost and speed of flight, higher AR values being interpreted as decreased energetic costs (Rayner 1988). WL is a body weight to wing area ratio that can be used to assess manoeuvrability and mass-carrying ability, higher

WL values corresponding to increased vulnerability (Norberg and Rayner 1987). As a result, wing morphology has been associated with different foraging strategies ranging in flight speed and manoeuvrability (Norberg and Rayner 1987). Frugivores show below-average AR and average WL, reflecting that frugivory does not require fast and agile flight (Norberg and Rayner 1987). Fast hawkers (aerial pursuit of prey at high speeds) usually show high values of WL, whereas slow hawkers (understory and slow aerial pursuit of prey) show low WL (Norberg and Rayner 1987). Carnivorous bats

193 usually experience higher demands for take-off and considerable prey load-carrying within clutter, and they tend to show low WL and AR (Norberg and Rayner 1987). AR and WL have also been traditionally used to study wing shape and flight evolution in birds and pterosaurs (McGowan and Dyke 2007; Habib and Ruff 2008; Bell, et al. 2011).

Recent studies have challenged a of the postcranium, where postcranial and cranial evolution are partitioned, each driven by a different adaptive regime, postcranial evolution associated only with locomotion and cranial evolution with diet, respectively (Morales, et al. 2019; Hedrick, et al. 2020). Studies in flying vertebrates have also suggested that adaptations for flight are more multi-faceted than previously assumed, indicating that traditional metrics used to describe wing morphological specialisations (e.g. AR and WL) can limit our understanding of vertebrate flight ecology (Chin, et al. 2017; Baliga, et al. 2019). Moreover, the need to test traditional interpretations of morphological adaptations to flight at a single structure or trait level has also become evident (Amador, Giannini, et al. 2018; Baliga, et al. 2019; Stanchak, et al. 2019). Computational and mathematical approaches have allowed for the application of biomechanical and kinematic principles to the study of flight ecology and evolution (Dececchi, et al. 2016; Chin, et al. 2017; Baliga, et al.

2019). The advent of Geometric Morphometrics (GMM) and phylogenetic comparative methods have also enabled the development of theoretical frameworks from which to interpret patterns of phenotypic diversification (Klingenberg 2014; Adams and Collyer

2018a). The interaction between morphological disparity, integration (i.e. high covariation between traits) and modularity (i.e. modules of highly correlated trait within a structure) is a major recent development in evolutionary theory (Gerber 2013;

194

Klingenberg 2013; Felice, et al. 2018). Integration and modularity have been shown to either increase or constrain phenotypic variation, shaping evolutionary patterns and ecological adaptations (Zelditch, et al. 2016; Felice, et al. 2018; López-Aguirre, Hand, et al. 2019b).

GMM and phylogenetic comparative methods have also been applied to study the bat postcranium, revising our understanding of bat postcranial morphology (Louzada, et al.

2019), development (López-Aguirre, Hand, et al. 2019b, a) and evolution (López-

Aguirre, Wilson, et al. 2019; Stanchak, et al. 2019). Prenatal development of the postcranium indicates a positive interaction between integration and disparity across development, while also revealing differences in allometric trajectories between bat suborders (López-Aguirre, Hand, et al. 2019a, b). Calcar development and histology in bats has also indicated a kinematic role in flight performance, highlighting the importance of previously underestimated morphological traits during the evolution of mammalian flight (Stanchak, et al. 2019). Bat hindlimb morphology has also been found to reflect taxonomic and locomotory-related differences, shedding new light on form-to-function interplay in bats (Louzada, et al. 2019).

Here, we investigate patterns of phenotypic disparity in the bat wing by studying humeral shape and the possible concordance between cranial and postcranial morphological adaptations. The humerus is a bone uniquely specialised in bats to perform under multiple functional demands: withstand high mechanical loading

(Swartz, et al. 1992; Watts, et al. 2001), increase muscle insertion area for muscles associated with wingbeat (Tokita, et al. 2012), and control manoeuvrability of the wing by rotation of the shoulder and elbow joints (Boerma, Breuer, et al. 2019). Diaphyseal

195 and epiphyseal morphology responds to different functional needs (i.e. resistance to torsion and bending in the diaphysis and joint range of motion in the epiphyses)

(Cooper and Tabin 2008; Cooper, et al. 2012). We aim to test whether humeral epiphyseal and diaphyseal morphological disparities vary independently, responding to different functional constraints. Using GMM to study 3D virtual models of the humerus we tested the relationship between phenotypic disparity and phylogeny (evolutionary relatedness), ecology (diet and foraging strategy) and biology (body size). In order to test decoupled patterns of morphological variation across functionally dissimilar sections of the humerus, we decomposed analyses of humeral morphology into whole- bone, diaphyseal and epiphyseal morphology. We tested for the presence of functional modules (epiphyses and diaphysis) in the humerus and the level of association between those. Furthermore, we assessed whether traditional metrics of wing aerodynamics are related to patterns and magnitudes of humeral morphological disparity. We hypothesise that the humerus is composed of two functional modules

(diaphysis and epiphyses), one relating to ecology (epiphyses) and the other relating to body size (diaphysis). We also hypothesise that, given the strong association between differences in traditional aerodynamics and foraging strategies, AR and WL will also have a strong effect on epiphyseal morphology.

Methods

Sample specimens

Our sample for this study comprised 55 adult specimens, selected to optimise coverage of phylogenetic (37 species, 20 families and both bat suborders), foraging behaviour (all broad foraging categories described for bats) and body size (10-fold

196 range in body mass) diversity (Fig. 6.1). This sample represents 95% of all modern families, 85% of dietary habits and all biomes where the order occurs. Alcohol- preserved specimens were sourced from the Western Australian Museum and research collections at the University of New South Wales (UNSW) and Institute of

Ecology and Biological Resources of Vietnamese Academy of Science and Technology

(IEBR; Appendix Table 7). Specimens sourced from IEBR were scanned at Musashino

Art University using a microCT system (inspeXio SMX-90CT Plus, Shimadzu) with 90kv source voltage and 100mA source current at a resolution of 15μm. Specimens sourced from UNSW were scanned at the same institution using a U-CT (Milabs, Utrecht) with

55kV and 0.17 mA, ultrafocused setting at a resolution of 30-50 μm. Additional species were sampled from whole-body scans sourced from Digimorph (Appendix Table 7). 3D virtual models of the humeri were created by segmenting the raw DICOM data using the thresholding tool in MIMICS v. 20 software (Materialise NV, Leuven, Belgium). In order to control for unwanted bilateral asymmetry affecting our analyses, only left humeri were digitised.

197

Figure 6.1. Phylogenetic relationships between sampled taxa based on Shi and Rabosky (2015)’s phylogeny. Branch colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= terrestrial locomotion). 3D models of humeri illustrate humeral diversity in sampled taxa. Represented taxa clockwise from bottem left to bottom right: Desmodus rotundus, Furipterus horrens, Nycteris grandis, Macroglossus minimus, Myotis daubentoni, Molossus molossus.

Morphometric data

To quantify whole bone humeral morphology, a landmarking protocol was developed using IDAV Landmark Editor (UC Davis, USA). The protocol comprised 221 landmarks and was implemented by the lead author to keep landmark placement consistent and avoid user bias error (Appendix Table 11). Epiphyseal morphology was described with

31 homologous landmarks in traits found in all specimens (Fig. 6.2). Proximal epiphysis was defined as the area containing the caput humeri, taberculum majus and minus, cristae pectoralis and taberculi minoris. Distal epiphysis was defined as the area containing the fossae olecrani and radialis, spina entepicondyli, epicondylus lateralis, 198 and the condylus lateralis and medialis. Given the smooth curving surface of the diaphysis, diaphyseal morphology was described with 160 equidistant sliding semi- landmarks across four curves (40 landmarks per curve) placed between homologous landmarks from the distal and proximal epiphyses (Fig. 6.2). 30 sliding semi-landmarks were placed along three curves (10 landmarks per curve) to describe the crista pectoralis, crista tuberculi minoris and crista tuberculi majoris. With this arrangement, we ensured that despite semi-landmarks not being homologous, the placement of the curves defining their positions along the humerus was homologous for all specimens.

Figure 6.2. Landmarking protocol used to quantify humeral morphology. From left to right humeri are presented in anterior (far left), medial (centre left), posterior (centre right) and lateral (far right) views. Proximal (top right) and distal (bottom right) epiphyses are also presented. Homologous landmarks are represented by numbers 0-30 and curves used to place semi-landmarks are represented by C0-C5.

Phylogenetic, ecological and biological traits

To test different sources of variation in our morphometric data, we compiled information on three types of traits: phylogeny (i.e. subordinal arrangement),

199 biological (i.e. body size), and ecological (i.e. foraging behaviour and diet) for all species. Phylogeny was codified at subordinal level (i.e. Yangochiroptera or

Yinpterochiroptera) (see Appendix Table 12). Centroid Size (CS), retrieved from the

Generalised Procrustes Analysis of landmark coordinates (PGA; see statistical analysis section), was used as proxy for body size, implemented in “gpagen” in the Geomorph R package (Adams, et al. 2013; Adams, et al. 2017). Foraging strategies (FG) were classified in five categories (trawling, hawking, gleaning, carnivory, frugivory, terrestrial locomotion), following traditional classifications established for bats (Norberg and

Rayner 1987; Denzinger and Schnitzler 2013). Frugivory and carnivory were included as foraging strategies given that species with these diets show a plasticity in their foraging behaviours that is unique across bats: hovering flight to feeding perches in frugivores, and ground- and foliage-gleaning to perch-hunting in carnivores (Norberg and Rayner 1987; Santana and Cheung 2016). Dietary categories comprised frugivory, insectivory, carnivory, piscivory, sanguivory and omnivory, and followed broad dietary classifications used in multiple studies (Nogueira, et al. 2009; Monteiro and Nogueira

2011; Santana, et al. 2012a; Arbour, et al. 2019). Considering the existing interpretations of morphological adaptations for different foraging guilds in bats based on aerodynamic properties of wing shape, we tested whether humeral morphology relates to patterns of WL and AR in our sample. WL and AR values were taken from the literature for 31 species in our sample.

Statistical analysis

For all statistical analyses, CS was log-transformed and species with multiple individuals were analysed based on the mean shape of the specimens using the

200

“mshape” function in Geomorph 3.2.1 for R 3.6.0 (Adams, et al. 2013; Adams, et al.

2017). For all phylogenetic analysis, Shi and Rabosky (2015)’s species-level bat super- tree was used as the phylogenetic hypothesis of evolutionary kinship in our sample. To estimate how accurately our landmarking protocol captures shape variation, we used the Landmark Sampling Evaluation Curve (LaSEC) approach, developed in the R package LaMBDA 0.1 (Watanabe 2018). This approach estimates the fit of a “parent” landmarking protocol from the original dataset by comparing it with the fit of subsampled landmarking protocols using Procrustes Sum of Squares. The original dataset was subsampled by sequentially adding one landmark at a time starting from 0

(each addition representing a new subsample), comparing the fit of each subsample to the to the parent protocol (Watanabe 2018). An optimal landmarking protocol is expected to reach stationarity in its fit before the parental level of complexity is reached.

We tested patterns of humeral shape variation based on different subordinal, ecological and biological hypotheses using Procrustes ANOVAs (PLM) with the

“procD.lm” function in Geomorph 3.2.1 (Adams, et al. 2013; Adams, et al. 2017). We compared the fit of four different statistical hypotheses against a null hypothesis in which shape is not correlated with any independent variable (shape ~ 1), using the

“anova” function in the RRPP 0.5.2 R package: differences based on body size (shape ~

CS), phylogeny (shape ~ suborder), foraging behaviour (shape~ FG), and diet (shape ~ dietary category). Next, we tested the effect of evolutionary history on our four hypotheses of shape variation (Adams and Collyer 2018b). For this, phylogenetic

Procrustes ANOVAs (PGLS) were used to test how much of morphological variation of

201 the humerus can be explained by diet, FG and body size (CS), after controlling for evolutionary kinship, using the “procD.pgls” function in Geomorph 3.2.1 (Adams, et al.

2013; Adams, et al. 2017). Model fit comparison followed the one used for Procrustes

ANOVA described above, using the “anova” function in RRPP 0.5.2. PLM and PGLS were used in order to test correspondence between humeral morphology and traditional wing shape aerodynamic properties (AR and WL) with the “procD.lm” and

“procD.pgls” functions in Geomorph 3.2.1.

The multivariate Kmult (K−) statistic was used to test whether morphological variation had a phylogenetic structure reflecting evolutionary relatedness, using the “physignal” function in Geomorph 3.2.1 (Adams 2014). K- evaluates the degree of phylogenetic signal in our dataset compared to what would be expected under a Brownian motion model of evolution based on 1000 iterations. Phylogenetic signal was tested for the whole bone and each module separately.

We performed a Principal Component Analysis (PCA) to reduce the dimensionality of our data and visualise patters of variation across taxa, reconstructing the morphospaces of humeral, diaphyseal and epiphyseal morphological variation, using the “plotTangentSpace” function in Geomorph 3.2.1. Given the strong and significant phylogenetic signal in our dataset, we also reconstructed morphospaces of humeral, diaphyseal and epiphyseal morphological variation that accounted for phylogenetic relationships (Uyeda, et al. 2015), using a phylogenetic PCA (pPCA) as implemented in the “phyl.pca” function in the phytools 0.6-99 R package. To visualise the magnitude of shape variation across principal components (PCs) explained by individual landmarks,

202 we estimated heatmaps of landmark shape variation by comparing the minimum and maximum of each PC, using landvR 0.3 (Guillerme and Weisbecker 2019).

Morphological disparity, integration and modularity

Humeral shape disparity was quantified at whole-bone, diaphyseal and epiphyseal levels among foraging guilds using Procrustes variance (i.e. sum of diagonal elements of covariance matrix divided by number of specimens by group) using the function

“morphol.disparity” from the R package Geomorph 3.2.1 (Adams, et al. 2013; Adams, et al. 2017). We used modularity and integration approaches to test the a priori hypothesis that the humerus represents two functional units (modules); one module for the diaphysis and the other for the epiphyses. Integration refers to the level of association in morphological variation within a structure, whereas modularity refers to how that association is distributed within the structure. modularity reflects covariation being unevenly distributed within a structure, forming modules of highly correlated traits that show lower magnitudes of between-module covariation than within-module covariation. To assess whether the diaphysis and epiphyses of the humerus represent two independent modules, we implemented the “modularity.test” function in

Geomorph 3.2.1 (Adams, et al. 2013; Adams, et al. 2017), which quantifies the degree of modularity using the Covariation Ratio (CR) coefficient (Adams and Peres-Neto

2016). Values <1 indicate greater within module covariance relative to between module (i.e. lower values provide greater support for modularity hypothesis).

Statistical significance was assessed by comparing the observed CR value against a distribution of simulated CR values obtained by randomly assigning landmarks to either module for 1000 iterations. Lastly, to estimate the amount of covariation

203 between the epiphyses and diaphysis modules, a two-block Partial Least Squares (PLS) analysis was used, implemented in the “integration.test” function in Geomorph 3.2.1, and its statistical significance was based on 999 iterations (Adams, et al. 2013; Adams, et al. 2017). Modularity and integration were also tested after accounting for phylogenetic relatedness using the “phylo.modularity” and “phylo.integration” functions in Geomorph.

Results

Landmarking accuracy

LaSEC analyses suggest that our landmarking protocol was effective in capturing humeral morphology at all levels (whole bone, diaphyseal and epiphyseal), sampling curves showing asymptotic trajectories reaching fit values of 1 before reaching parental level of landmarking complexity (Appendix Figure 11).

Humeral morphological variation

Model comparison of shape variation based on PLM revealed that all four models tested (CS, suborders, diet and FG) performed better than the null model of shape variation not explained by any independent variable, with the exception of CS which was not significant for whole-bone and diaphyseal shape variation (Table 6.1). On average, diet explained the highest proportion of shape variation (27.49%), followed by FG (25.88%) and suborders (10.91%). CS only accounted for 4.09% of shape variation, on average.

204

Table 6.1. Procrustes ANOVA (PLM) for different hypotheses of shape variation. Significance test was based on 1000 iterations.

WHOLE-BONE ResDf Df RSS SS MS Rsq F Z P log(CS) 32 1 0.338 0.009 0.009 0.026 0.841 -0.026 0.492 Phylogeny 32 1 0.305 0.042 0.042 0.120 4.354 3.271 0.001 Diet 28 5 0.260 0.087 0.017 0.250 1.866 1.808 0.027 FG 28 5 0.267 0.080 0.016 0.230 1.675 1.898 0.034 DIAPHYSIS ResDf Df RSS SS MS Rsq F Z P log(CS) 32 1 0.453 0.011 0.011 0.024 0.777 -0.135 0.529 Phylogeny 32 1 0.407 0.057 0.057 0.122 4.465 3.187 0.002 Diet 28 5 0.354 0.110 0.022 0.237 1.742 1.574 0.062 FG 28 5 0.364 0.100 0.020 0.217 1.548 1.560 0.063 EPIPHYSIS ResDf Df RSS SS MS Rsq F Z P log(CS) 32 1 0.027 0.002 0.002 0.074 2.546 2.480 0.019 Phylogeny 32 1 0.026 0.002 0.002 0.085 2.978 3.015 0.005 Diet 28 5 0.019 0.010 0.002 0.338 2.855 4.704 0.001 FG 28 5 0.019 0.009 0.002 0.330 2.753 4.795 0.001 Tests of phylogenetic signal revealed a significant effect of evolutionary kinship on patterns of shape variation in our sample, suggesting that closely-related taxa are morphologically more similar than expected under a Brownian motion model of evolution (Table 6.2). Both K- and ZCR values were higher for epiphysis shape, revealing a stronger phylogenetic constraint, whereas whole-bone and diaphysis shape had similar values of K- and ZCR.

Table 6.2. Kmult- statistic test of phylogenetic signal on shape data. Significance test was based on 1000 iterations.

K- P ZCR Whole-bone 0.9425 0.005 2.4979 Diaphysis 0.9341 0.009 2.1878 Epiphyses 1.0479 0.001 6.9649 PGLS regressions were performed to test the fit of our models while accounting for the strong phylogenetic signal in our data, revealing that some of our models still performed better than the null model. Contrary to our PLM results, PGLS revealed

205 divergent model fit in epiphyseal shape and similar model fit for diaphyseal and whole- bone shape (Table 6.3). For whole-bone and diaphyseal shape, only ecological differences were marginally significant, accounting for 20.5% and 19.15% of shape variation, respectively. Contrastingly, CS was a significant model only for diaphyseal shape variation. Diet explained the highest proportion of diaphyseal shape variation

(31.9%), followed by FG (29.2%) and CS (6%). Given the close interaction between diet and FG (frugivory and carnivory were also classified as foraging categories), and the functional association between foraging behaviours and wing morphology, all further analyses focused on FG. WL was not correlated with humeral shape at any level of variation, whereas AR correlated significantly with epiphyseal shape (Table 6.4).

Table 6.3. Phylogenetic procrustes ANOVA for different hypotheses of shape variation. Significance test was based on 1000 iterations.

WHOLE-BONE ResDf Df RSS SS MS Rsq F Z P log(CS) 32 1 0.007 0.000 0.000 0.024 0.800 -0.107 0.518 Diet 28 5 0.005 0.001 0.000 0.218 1.560 1.297 0.108 FG 28 5 0.005 0.001 0.000 0.192 1.330 1.040 0.142 DIAPHYSIS ResDf Df RSS SS MS Rsq F Z P log(CS) 32 1 0.009 0.000 0.000 0.023 0.742 -0.207 0.553 Diet 28 5 0.007 0.002 0.000 0.209 1.477 1.123 0.144 FG 28 5 0.007 0.002 0.000 0.182 1.243 0.786 0.210 EPIPHYSIS ResDf Df RSS SS MS Rsq F Z P log(CS) 32 1 0.000 0.000 0.000 0.060 2.042 2.045 0.034 Diet 28 5 0.000 0.000 0.000 0.319 2.621 4.386 0.001 FG 28 5 0.000 0.000 0.000 0.292 2.314 4.348 0.001

206

Table 6.4. Procrustes ANOVA for hypotheses of covariation between humeral shape and wing aerodynamic properties (WL and AR). Significance test was based on 1000 iterations.

WHOLE BONE Df SS MS Rsq F Z P WL 1 0.007 0.007 0.023 0.676 -0.518 0.698 AR 1 0.010 0.010 0.030 0.909 0.012 0.513 DIAPHYSIS Df SS MS Rsq F Z P WL 1 0.009 0.009 0.022 0.649 -0.543 0.704 AR 1 0.012 0.012 0.029 0.859 -0.071 0.544 EPIPHYSES Df SS MS Rsq F Z P WL 1 0.001 0.001 0.027 0.817 -0.297 0.597 AR 1 0.003 0.003 0.106 3.455 3.244 0.002 Levels of morphological disparity across different FG showed similarities between whole-bone and diaphyseal shape, and unique patterns of variation in epiphyseal shape disparity (Fig. 6.3). For whole-bone and diaphyseal shape, frugivory showed the highest values of disparity (0.014 and 0.02 respectively), followed by gleaning (0.009 and 0.012 respectively) and hawking (0.002 and 0.01 respectively), carnivory (0.004 and 0.005 respectively) and terrestrial locomotion (0.007 and 0.01 respectively).

Trawling showed the lowest values of whole-bone and diaphyseal shape disparity

(0.002 for both; Table 6.2). However, no statistical significance was found among pairwise comparisons (Appendix Table 13). Hawking had the highest epiphyseal shape disparity (6.69E-04), followed by gleaning (5.89E-04) and terrestrial locomotion (4.56E-

04), carnivory (4.08E-04), frugivory (3.61E-04) and Trawling with the lowest epiphyseal shape disparity (3.50E-04; Fig. 6.3). Pairwise comparisons revealed that hawking was significantly different from carnivory, frugivory and trawling (Appendix Table 13).

207

Figure 6.3. Humeral shape disparity of whole-bone (left), diaphyseal (centre) and epiphyseal (right) morphology. Shape disparity was decomposed based on foraging guild categories: C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= terrestrial locomotion.

Morphospace reconstruction

PCA of whole-bone humeral morphology resulted in the first four principal components of morphospace accounting for 69.34% of shape variance (Fig. 6.4). PC1 primarily divides species between suborders, with a shape variation trend of increasing sigmoidal curvature of the diaphysis, as well as a reduction of the proximal projection of the tuberculum majus. Frugivory and hawking showed the highest dispersion across

PC2, whereas carnivory and trawling showed the lowest dispersion, driven by differences in the shape of the crista pectoralis. Frugivory showed lower dispersion across PC3 (characterised by greater humeral torsion) and PC4 (changes in shape of taberculum minus and spina entepicondyli). The first two components of pPCA revealed a greater overlap in morphospace between suborders, and lower dispersion within each FG. Austronomus australis expanded the distribution of Yangochiroptera both in PCA and pPCA.

208

Figure 6.4. Morphospace (PCA, A and C) and phylogenetically-corrected morphospace (pPCA, B and D) based on whole-bone shape data. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= terrestrial locomotion), and polygon colours suborder (purple= Yangochiroptera, yellow= Yinpterochiroptera). Landmark heatmaps of shape change represent magnitude of shape variation across each PC by the comparing the minimum and maximum of each component. Humeri 3D models represent position of landmark heatmaps; red colours representing greater variation and yellow colours lower variation.

The first four principal components of diaphyseal and epiphyseal morphospaces explained on average 64.03% of shape variation (71.39% and 56.68%, respectively).

Diaphyseal and whole-bone morphospaces showed analogous patterns of variation across PC1 and PC2, with FGs distributed similarly across morphospace (Figs. 6.4A, C and 6.5A, C). Pteropodid and phyllostomid frugivores clustered on opposite sides of

PC1 and PC2, whereas carnivores and trawlers clustered forming small morphospaces towards the centre. Higher dispersion of carnivores and gleaners was evident across

PC3 and PC4. Suborders showed higher dispersion and overlap across the first four

209 components of diaphyseal morphospace, compared to diaphyseal and whole-bone morphospaces. Moreover, FGs showed less overlap in epiphyseal morphospace, with carnivores and frugivores occupying non-overlapping sections of morphospace.

Figure 6.5. Diaphyseal (left) and epiphyseal (right) morphospaces of humeral morphology based on PCAs of shape data. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= terrestrial locomotion), and polygon colours suborder (purple= Yangochiroptera, yellow= Yinpterochiroptera).

Phylogenetically-corrected morphospaces of diaphyseal and epiphyseal shape show greater dispersion of suborders across morphospace, particularly in Yangochiroptera

(Fig 6.6). In diaphyseal morphospace, Yinpterochiroptera and Yangochiroptera dispersed across opposite principal components (pPC1 and pPC2, respectively).

Pteropodid and phyllostomid frugivores occupied opposite ends of diaphyseal

210 morphospace across pPC1 and pPC2, probably reflecting the sigmoidal shape diaphysis in pteropodids versus the straight shaft in phyllostomids. Gleaners and hawkers showed the highest overlap across pPC1 to pPC4, reflecting foraging plasticity and similarities in hunting styles between both groups. Carnivores and frugivores showed the highest discrimination across diaphyseal morphospace. FGs were relatively separated across pPCA diaphyseal morphospace relative to PCA morphospace. TL species clustered closer together in pPCA epiphyseal morphospace, compared to whole-bone and diaphyseal morphospaces. Gleaners and hawkers shared similar patterns of variation once the effect of phylogeny was removed, overlapping more across pPC1-pPC4 than any other FG. Carnivores, frugivores, trawlers and terrestrial locomotors showed clear differences in diaphyseal morphospace once phylogenetic signal was removed.

211

Figure 6.6. Diaphyseal (A and C) and epiphyseal (B and D) phylogenetically-corrected morphospaces of humeral morphology based on pPCAs of shape data. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= terrestrial locomotion), and polygon colours suborder (purple= Yangochiroptera, yellow= Yinpterochiroptera).

Morphological modularity and integration

Tests of modularity and phylogenetic modularity both rejected the null hypothesis of no modularity in humeral shape, supporting our hypothesis of a functional diaphysis and epiphyses modular partition. Modularity tests that accounted for phylogenetic relationships found greater support for our modularity hypothesis (CR= 0.67, P= 0.001, z= -13.66) than those that did not (CR= 0.79, P= 0.001, z= -8.68). Integration was also statistically significant for both tests, with phylogenetic integration showing a slightly lower magnitude of integration (r= 0.83, P= 0.001, z= 4.53) than non-phylogenetic integration (r= 0.87, P= 0.001, z= 4.89). PLS results of integration placed TL and frugivore taxa at opposite ends of the axes, TL species occupying the positive extreme

212 and frugivores the opposite extreme (Fig. 6.7). The first axis of PLS explained 83% of shape covariation between modules.

Figure 6.7. PLS biplot of first two axes of diaphyseal and epiphyseal shape covariation. Dot colours represent foraging guild categories (C= carnivores, F= frugivores, G= gleaners, H=hawkers, T=trawlers and TL= terrestrial locomotion). Discussion

By combining GMM and phylogenetic comparative methods to study humeral morphology, we found a strong association between humeral shape and ecology (i.e. diet and FG) and phylogeny (subordinal arrangement) of bats, indicating that humeral morphological disparity has both an ecological and evolutionary signal. The magnitude of the effect of diet and FG in humeral shape mirrors patterns of cranial morphological disparity in bats, providing evidence for a correspondence between cranial and postcranial morphological disparity (Monteiro and Nogueira 2011; Arbour, et al. 2019;

Hedrick, et al. 2019; Brokaw and Smotherman 2020; Leiser-Miller and Santana 2020).

Our results reveal differences between epiphyseal and diaphyseal shape in their relationship to ecology and phylogeny. We found that ecology had a greater effect on epiphyseal morphology, whereas phylogeny and size (i.e. CS) had a greater effect on

213 diaphyseal shape, also reflected in statistical support for our modularity hypothesis dividing the humerus into two functional modules (diaphysis and epiphyses).

Integration tests found significant interaction between diaphyseal and epiphyseal shape variation with differing magnitude across FG. We found only epiphyseal shape correlated with wing AR, traditionally used to study bat wing/flight capabilities

(Norberg and Rayner 1987; Rayner 1988; Amador, Almeida, et al. 2019; Amador,

Simmons, et al. 2019). Our results indicate multi-faceted patterns of shape variation in bats where ecology, size and phylogeny interplay to shape the modular morphology of the humerus (Kilbourne and Hutchinson 2019; López-Aguirre, Wilson, et al. 2019).

Moreover, we found a significant association between ecomorphological adaptations of the humeral epiphyses and wing shape, revealing the interplay between epiphyseal shape, wing shape and control of mobility during flight (Swartz, et al. 2007; Bergou, et al. 2015; Boerma, Breuer, et al. 2019).

Drivers of humeral morphological variation

Forelimb ecomorphological diversity has been associated with the ecological and taxonomic diversification of mammals, showing an evolutionary trajectory of increasing forelimb disparity (Lungmus and Angielczyk 2019). Our PLM results revealed a significant effect of evolutionary history and ecology on humeral shape at the three levels studied (i.e. whole-bone, diaphyseal and epiphyseal), a result expected following studies of morphological variation in bats (Monteiro and Nogueira 2011; Rossoni, et al.

2017; Arbour, et al. 2019; Brokaw and Smotherman 2020), other mammals (Law 2019) and other vertebrates (Wilson 2013a; Gill, et al. 2014; Vidal-García and Scott Keogh

2017; Hedrick, et al. 2020). Morphological adaptations of the humerus and shoulder

214 joint in bats have been studied to inform both functional performance capabilities and systematics in the order (Schlosser-Sturm and Schliemann 1995; Hand, et al. 2009).

Previous analyses of femoral morphological disparity in Yangochiropteran bats reveal a similar pattern to ours, detecting differences between major taxonomic groups, as well as between species with different ecologies (Louzada, et al. 2019). Foraging strategies have been strongly associated with differences in bat femoral morphology (Louzada, et al. 2019), revealing parallel morpho-biomechanical traits between species with different foraging strategies (e.g. trawlers and walking bats). Convergent functional demands in trawling and terrestrial locomotion could have led to shared morpho- biomechanical adaptations (e.g. robust diaphyses) to develop in these two foraging strategies (Louzada, et al. 2019). Bat humeri exhibit a range of morphological, biomechanical and histological adaptations responding to demands for muscle insertion (enlarged crista pectoralis and crista tuberculi; Panyutina, et al. (2015)), shoulder and elbow joint mobility (taberculum majus of proximal epiphysis and spinous process of distal epiphysis; Panyutina, et al. (2015)), and resistance to stress and strain (higher mineralisation compared to other wing bones and more circular diaphyseal cross-sectional geometry (Swartz and Middleton 2008)). Broadly speaking, functional demands driven by loading regimes and manoeuvrability exert multiple selective pressures across the humerus (López-Aguirre, Wilson, et al. 2019). The humeral diaphysis is functionally adapted to withstand torsion and bending stresses during flight (Swartz, et al. 1992), whereas the humeral epiphyses show adaptations associated with control of wingbeat and manoeuvrability (Cubo and Casinos 1998;

Simons, et al. 2011; Patel, et al. 2013).

215

Support for different models of shape variation for diaphyseal and epiphyseal morphology (based on PGLS) may reflect multiple functional pressures acting on different parts of the humerus in bats. Over 30% of epiphyseal shape variation explained by diet and FG indicate that this region of the humerus could experience stronger selective pressures associated with adaptability to different foraging strategies. Descriptive and studies of the shoulder joint have identified three broad categories of shoulder joint specialisations linked to specific flight and locomotory capabilities (Schlosser-Sturm and Schliemann 1995): 1)

Generalist shoulder joint with a single articular surface, commonly found in pteropodid frugivores, 2) specialised shoulder joint with a single articular face described in mormoopid, noctilionid and emballonurid animalivores, and 3) specialised shoulder joint with a secondary articulation reported in some noctilionoid families (e.g.

Phyllostomidae, Thyropteridae and Furipteridae), and all vespertilionoid families studied (i.e. Natalidae, Molossidae and Vespertilionidae). Shoulder joint specialisations have a functional role in locomotory performance in limiting humeral rotation due to pronation during downstroke (Schlosser-Sturm and Schliemann 1995). Specialisations of the shoulder joint imply adaptations of the proximal epiphysis of the humerus, indicating that humeral morphology can work as a proxy to study such adaptations

(Schlosser-Sturm and Schliemann 1995; Hand, et al. 2009).

Statistically significant phylogenetic signal in our datasets follow similar patterns previously reported for cranial morphology in bats (Arbour, et al. 2019; Hedrick, et al.

2019). The magnitude of phylogenetic signal in postcranial morphological diversity in mammals remains unclear, with studies reporting a significant effect in carnivorans

216 and marsupials (Martín-Serra, et al. 2014; Martin-Serra, et al. 2017; Janis, et al. 2020), but non-significant at macroevolutionary scales (Lungmus and Angielczyk 2019).

Further studies are needed to understand the macroevolutionary patterns of morphological disparity in the mammalian postcranium and the role of ecology and phylogeny influencing those patterns. We found epiphyseal shape to be highly correlated with AR (wing area) but found no significant correlation of humeral morphology with WL wing loading (weight divided by wing area). Considering the many studies using aerodynamic metrics of the wing to describe functional morphology in bats (Norberg and Rayner 1987; Rayner 1988), it is reasonable to expect that epiphyseal ecomorphology would be correlated with overall wing shape, reflecting daily functional demands on the shoulder and elbow joints that would act similarly at both scales. Epiphyseal morphology is directly involved in wingbeat as it controls movement, which in turn has a direct impact of the energetic cost (Riskin, et al. 2012), flight speed and manoeuvrability of flight (Iriarte-Diaz, et al. 2011; Bergou, et al. 2015; Boerma, Breuer, et al. 2019). Recent studies in birds have posited that traditional metrics of wing morphology fall short in explaining the complexity of phenotypic adaptations for flight (Baliga, et al. 2019), highlighting the need to apply novel and multi-disciplinary approaches to this question. Our results indicate that phenotypic adaptations for flight in bats are multi-faceted, acting at different scales

(e.g. from single structure to functional unit), and that the implementation of different approaches can inform different aspects of those adaptations.

Strong similarities in whole-bone and diaphyseal morphospaces and levels of disparity reveal the influence of diaphyseal morphology on overall disparity in humeral shape.

217

Across whole-bone and diaphyseal morphospaces, Austronomus australis separated from the rest of yangochiropterans, which tended to cluster together, revealing the humeral morphology of this species adapted for higher speeds and minimising drag

(Bullen and McKenzie 2007). Gracile humerus and radius in A. australis have been associated with its interceptor foraging behaviour, which depends on optimal generation of a leading-edge vortex (Bullen and McKenzie 2007). Higher diaphyseal disparity in frugivores parallels the role that frugivory had in the adaptive radiation of cranial phenotypes in bats (Rossoni, et al. 2017; Arbour, et al. 2019; Rossoni, et al.

2019). Phyllostomid and pteropodid frugivores occupied opposite regions of morphospace, suggesting divergent patterns of morphological specialisation for frugivory that parallel the convergent evolution of frugivory and nectarivory in the two groups of bats (Datzmann, et al. 2010; Monteiro and Nogueira 2011; Rojas, et al.

2012). Compared to other bats, pteropodid humeri show markedly simple diaphyses, with low cristae and an S shaped shaft (Panyutina, et al. 2015). Greater dispersion of gleaners and hawkers across diaphyseal morphospace (overlapping with most foraging strategies) could suggest lower evolutionary diversification, resulting in more generalist morphotypes. Ancestral reconstructions have theorised that insectivory and aerial hawking were the ancestral states in Chiroptera (Amador, Simmons, et al. 2019), which could indicate that modern gleaning and hawking morphotypes retain an ancestral, less divergent morphotype. Higher epiphyseal disparity in gleaners and hawkers could reflect a secondary aerodynamic diversification in bats associated with adaptations to manoeuvrability (Amador, Almeida, et al. 2019; Amador, Simmons, et al. 2019). Oligo-Miocene environmental changes coincide with the diversification of most modern bat families, linking the expansion of open-mosaic ecosystems with 218 increased functional demands for manoeuvrability, along with sustained and fast flight

(Amador, Almeida, et al. 2019; Amador, Simmons, et al. 2019). Carnivore and trawling species consistently clustered together with species of the same guild across morphospace, reflecting cranial morphological specialisations correlated with carnivory and piscivory in bats (Santana and Cheung 2016).

Morphological modularity and integration

Contrary to our study, most modularity hypotheses have focused on sets of bones that are tightly associated to form a single structure (e.g. cranium or complete limbs)

(Hallgrimsson, et al. 2004; Porto, et al. 2008; Marroig, et al. 2009; Goswami and Polly

2010; Santana and Lofgren 2013; Martín-Serra, et al. 2014; Martin-Serra, et al. 2017), with notable exceptions being studies on mandibular modularity (Jojic, et al. 2012;

Garcia, et al. 2014; López-Aguirre, et al. 2015). Within-bone modularity has been widely studied and demonstrated for mandibular morphology in mammals (Atchley and Hall 1991; Hall 2003; Zelditch, et al. 2008; Polanski 2011; Jojic, et al. 2012), although the prevalence of mandibular modularity in other remains to be tested (Parsons, et al. 2012). Within-bone modularity has been reported in felid vertebrae, suggesting a developmental signal reflecting ontogenetic similarities

(Randau and Goswami 2017). Postcranial modularity in mammals has been studied between groups of bones representing functional units, such as the vertebral column in felids (Randau and Goswami 2018), and the appendicular skeleton (Goswami, et al.

2014a; Martín-Serra, et al. 2014; Conaway, et al. 2018; Diogo, et al. 2019). Special interest has been taken in studying developmental modularity and how it reflects the

219 evolutionary history of mammals (Goswami, et al. 2009; Young, et al. 2010; López-

Aguirre, Hand, et al. 2019a).

Our study provides evidence for significant within-bone modularity in the appendicular skeleton, the first reported for any vertebrate. A significant diaphyseal-epiphyseal partition of the humerus has been suggested to indicate a functional signal in the modular morphological variation of fossil primates (Marchi, et al. 2016). Epiphyseal and diaphyseal morphological specialisations have been associated with functional modularity in the appendicular skeleton of mammals (Goswami, et al. 2014a; Marchi, et al. 2016). Epiphyseal adaptations in the humerus of bats have been linked to shoulder and elbow joint mobility, shaping the performance and manoeuvrability of the wing during flight and landing (Vaughan 1959; Altenbach 1979; Schlosser-Sturm and Schliemann 1995; Boerma, Barrantes, et al. 2019; Boerma, Breuer, et al. 2019). On the other hand, specialisations of diaphyseal morphology have been useful to describe functional adaptations to biomechanical loading stresses engendered during flight

(Swartz, et al. 1992; Swartz and Middleton 2008; Krause, et al. 2014; López-Aguirre,

Wilson, et al. 2019). Novel locomotory and foraging strategies (e.g. terrestrial locomotion) in bats could have also canalised morphological adaptations in the epiphysis of the humerus (Norberg and Rayner 1987; Riskin, et al. 2006; Hand, et al.

2009). At least eight ossification centres have been identified during humeral development in mammals (Kwong, et al. 2014; Wisniewski, et al. 2017). The humeral shaft is ossified prenatally in mammals (Wisniewski, et al. 2017), whereas the epiphyseal plate remains cartilaginous at birth to allow longitudinal growth of the bone (Kwong, et al. 2014). Multiple secondary ossification centres have been identified during the postnatal ossification of the epiphyses (Kwong, et al. 2014), indicating that a 220 developmental modularity hypothesis would be more complex than our diaphyseal- epiphyseal partition. Evidence of developmental modularity in the appendicular skeleton of mammals has been reported in primates and carnivorans (Young and

HallgrÍmsson 2005; Lawler 2008; Martín-Serra, et al. 2014; Martin-Serra, et al. 2017;

Conaway, et al. 2018).

Integration between the diaphysis and epiphyses varied between foraging strategies.

Our PLS plot showed that frugivore, carnivore, walking and trawling species occupied non-overlapping morphospaces. Ecological differences between mammal taxa have been suggested to structure the magnitude of morphological integration between functionally correlated traits (Makedonska, et al. 2012). However, a similar test for postcranial integration in mammals is still missing at a broader scale. Dispersion of species with similar foraging strategies in our PLS plot does not reflect either size similarities or phylogenetic relationships (e.g. clustering of pteropodid and phyllostomid frugivores and walking bats). Significant integration between diaphysis and the epiphyses of the humerus show that ecomorphological adaptations shape morphology at different scales (within and between modules) within a single structure

(Hallgrimsson, et al. 2002; Young and HallgrÍmsson 2005; Young, et al. 2010).

Differences in patterns of integration between foraging strategies seem to reflect functional differences in manoeuvrability and loading.

Conclusions

Our study explores the drivers (phylogenetic, ecological and biological) of within structure morphological modularity and integration in bat humeri, elucidating the interaction between different types of traits exerting multiple selective pressures in a single bone. To our knowledge, this study is the first to find significant patterns of 221 within structure modularity in the appendicular skeleton of any tetrapod, highlighting the need to further explore decoupled patterns of phenotypic variation within a single structure. Humeral morphological disparity was found to reflect the foraging strategies and diets of species, following previous findings on cranial morphology, and suggesting an interplay between cranial and postcranial morphological variation. Diaphyseal and epiphyseal morphology varied independently, the former better reflecting the effect of evolutionary kinship, and the latter the effect of diet, FG and size. Statistical support for a diaphysis-epiphyses modular partition of humeral shape variation also reinforces our hypothesis of decoupled patterns of morphological variation in bat humeri.

Frugivores evidenced greater diaphyseal disparity, whereas animalivores had greater epiphyseal disparity, suggesting a correlation between epiphyseal shape and control of manoeuvrability during flight. Also, we found a significant association between shape and AR only for the epiphyses, revealing an association between wing shape and epiphyseal morphology which governs range of movement at shoulder and elbow joint range of motion during flight.

References

Adams, D. C. 2014. A generalized K statistic for estimating phylogenetic signal from shape and other high-dimensional multivariate data. Syst Biol 63:685-697. Adams, D. C. and M. L. Collyer. 2018a. Multivariate Phylogenetic Comparative Methods: Evaluations, Comparisons, and Recommendations. Syst Biol 67:14-31. Adams, D. C. and M. L. Collyer. 2018b. Phylogenetic ANOVA: Group-clade aggregation, biological challenges, and a refined permutation procedure. Evolution 72:1204-1215. Adams, D. C., M. L. Collyer, A. Kaliontzopoulou, and E. Sherratt. 2017. Geomorph: software for geometric morphometric analyses. R package. https://cran.r- project.org/package=geomorph. Adams, D. C., E. Otárola-Castillo, and E. Paradis. 2013. Geomorph: an R package for the collection and analysis of geometric morphometric shape data. Methods in Ecology and Evolution 4:393-399. Adams, D. C. and P. Peres-Neto. 2016. Evaluating modularity in morphometric data: challenges with the RV coefficient and a new test measure. Methods in Ecology and Evolution 7:565-572. 222

Altenbach, J. S. 1979. Locomotor morphology of the vampire bat, Desmodus rotundus. Special Publication No. 6, American Society of Mammalogists. Amador, L. I., F. C. Almeida, and N. P. Giannini. 2019a. Evolution of Traditional Aerodynamic Variables in Bats (Mammalia: Chiroptera) within a Comprehensive Phylogenetic Framework. Journal of Mammalian Evolution 27: 549-561. Amador, L. I., N. P. Giannini, N. B. Simmons, and V. Abdala. 2018. Morphology and Evolution of Sesamoid Elements in Bats (Mammalia: Chiroptera). American Museum Novitates 2018:1-40, 40. Amador, L. I., N. B. Simmons, and N. P. Giannini. 2019b. Aerodynamic reconstruction of the primitive fossil bat Onychonycteris finneyi (Mammalia: Chiroptera). Biol Letters 15:20180857. Arbour, J. H., A. A. Curtis, and S. E. Santana. 2019. Signatures of echolocation and dietary ecology in the adaptive evolution of skull shape in bats. Nat Commun 10:2036. Atchley, W. R. and B. K. Hall. 1991. A MODEL FOR DEVELOPMENT AND EVOLUTION OF COMPLEX MORPHOLOGICAL STRUCTURES. Biological Reviews 66:101-157. Baliga, V. B., I. Szabo, and D. L. Altshuler. 2019. Range of motion in the avian wing is strongly associated with flight behavior and body mass. Sci Adv 5:eaaw6670. Bell, E., B. Andres, and A. Goswami. 2011. Integration and dissociation of limb elements in flying vertebrates: a comparison of pterosaurs, birds and bats. J Evol Biol 24:2586-2599. Bergou, A. J., S. M. Swartz, H. Vejdani, D. K. Riskin, L. Reimnitz, G. Taubin, and K. S. Breuer. 2015. Falling with Style: Bats Perform Complex Aerial Rotations by Adjusting Wing Inertia. PLoS Biol 13:e1002297. Bishop, K. L. 2008. The evolution of flight in bats: narrowing the field of plausible hypotheses. Q Rev Biol 83:153-169. Boerma, D. B., J. P. Barrantes, C. Chung, G. Chaverri, and S. M. Swartz. 2019a. Specialized landing maneuvers in Spix's disk-winged bats (Thyroptera tricolor) reveal linkage between roosting ecology and landing biomechanics. The Journal of Experimental Biology:jeb.204024. Boerma, D. B., K. S. Breuer, T. L. Treskatis, and S. M. Swartz. 2019b. Wings as inertial appendages: how bats recover from aerial stumbles. The Journal of Experimental Biology:jeb.204255. Brokaw, A. F. and M. Smotherman. 2020. Role of ecology in shaping external nasal morphology in bats and implications for olfactory tracking. PLoS One 15:e0226689. Bullen, R. and N. L. McKenzie. 2001. Bat airframe design: flight performance, stability and control in relation to foraging ecology. Australian Journal of Zoology 49:235-261. Bullen, R. D. and N. L. McKenzie. 2007. Bat wing airfoil and planform structures relating to aerodynamic cleanliness. Australian Journal of Zoology 55:237-247. Carter, R. T. and R. A. Adams. 2016. Integrating Ontogeny of Echolocation and Locomotion Gives Unique Insights into the Origin of Bats. Journal of Mammalian Evolution 23:413–421. Chin, D. D., L. Y. Matloff, A. K. Stowers, E. R. Tucci, and D. Lentink. 2017. Inspiration for wing design: how forelimb specialization enables active flight in modern vertebrates. J R Soc Interface 14. Conaway, M. A., L. Schroeder, and N. von Cramon-Taubadel. 2018. Morphological integration of anatomical, developmental, and functional postcranial modules in the crab-eating macaque (Macaca fascicularis). Am J Phys Anthropol 166:661-670. 223

Cooper, K. L. and C. J. Tabin. 2008. Understanding of bat wing evolution takes flight. Genes Dev 22:121-124. Cooper, L., C. J. Cretekos, and K. E. Sears. 2012. The evolution and development of mammalian flight. Wiley Interdisciplinary Reviews: Developmental Biology 1:773-779. Cubo, J. and A. Casinos. 1998. The variation of the cross-sectional shape in the long bones of birds and mammals. Annales des Sciences Naturelles - Zoologie et Biologie Animale 19:51-62. da Silva, J. M., A. Herrel, G. J. Measey, B. Vanhooydonck, and K. A. Tolley. 2014. Linking microhabitat structure, morphology and locomotor performance traits in a recent radiation of dwarf chameleons. Functional Ecology 28:702-713. Datzmann, T., O. von Helversen, and F. Mayer. 2010. Evolution of nectarivory in phyllostomid bats (Phyllostomidae Gray, 1825, Chiroptera: Mammalia). BMC Evol Biol 10: 165. Dececchi, T. A., H. C. Larsson, and M. B. Habib. 2016. The wings before the bird: an evaluation of flapping-based locomotory hypotheses in bird antecedents. PeerJ 4:e2159. Denzinger, A. and H. U. Schnitzler. 2013. Bat guilds, a concept to classify the highly diverse foraging and echolocation behaviors of microchiropteran bats. Front Physiol 4:164. Diogo, R. 2017. Etho-Eco-Morphological Mismatches, an Overlooked Phenomenon in Ecology, Evolution and Evo-Devo That Supports ONCE (Organic Nonoptimal Constrained Evolution) and the Key Evolutionary Role of Organismal Behavior. Frontiers in Ecology and Evolution 5. Diogo, R., J. M. Ziermann, C. Smith, M. Alghamdi, J. S. M. Fuentes, and A. Duerinckx. 2019. First use of anatomical networks to study modularity and integration of heads, forelimbs and hindlimbs in abnormal anencephalic and cyclopic vs normal human development. Sci Rep 9:7821. Dumont, E. R., L. M. Davalos, A. Goldberg, S. E. Santana, K. Rex, and C. C. Voigt. 2012. Morphological innovation, diversification and invasion of a new adaptive zone. Proc Biol Sci 279:1797-1805. Dumont, E. R., K. Samadevam, I. Grosse, O. M. Warsi, B. Baird, and L. M. Davalos. 2014. Selection for mechanical advantage underlies multiple cranial optima in new world leaf-nosed bats. Evolution 68:1436-1449. Fabre, A.-C., R. Cornette, A. Goswami, and S. Peigné. 2015. Do constraints associated with the locomotor habitat drive the evolution of forelimb shape? A case study in musteloid carnivorans. Journal of Anatomy 226:596-610. Felice, R. N., M. Randau, and A. Goswami. 2018. A fly in a tube: Macroevolutionary expectations for integrated phenotypes. Evolution 72:2580-2594. Ferry-Graham, L., D. Bolnick, and P. Wainwright. 2002. Using Functional Morphology to Examine the Ecology and Evolution of Specialization. Integrative and Comparative Biology 42:265–277. Garcia, G., E. Hingst-Zaher, R. Cerqueira, and G. Marroig. 2014. Quantitative Genetics and Modularity in Cranial and Mandibular Morphology of Calomys expulsus. Evolutionary Biology 41:619-636. Gerber, S. 2013. On the Relationship between the Macroevolutionary Trajectories of Morphological Integration and Morphological Disparity. Plos One 8.

224

Gill, P. G., M. A. Purnell, N. Crumpton, K. R. Brown, N. J. Gostling, M. Stampanoni, and E. J. Rayfield. 2014. Dietary specializations and diversity in feeding ecology of the earliest stem mammals. Nature 512:303-305. Goswami, A. and P. D. Polly. 2010. The Influence of Modularity on Cranial Morphological Disparity in Carnivora and Primates (Mammalia). Plos One 5:e9517. Goswami, A., J. B. Smaers, C. Soligo, and P. D. Polly. 2014. The macroevolutionary consequences of phenotypic integration: from development to deep time. Philos Trans R Soc Lond B Biol Sci 369:20130254. Goswami, A., V. Weisbecker, and M. R. Sánchez-Villagra. 2009. Developmental modularity and the marsupial-placental dichotomy. J Exp Zool B Mol Dev Evol 312B:186-195. Guillerme, T. and V. Weisbecker. 2019. landvR: Tools for measuring landmark position variation. Available from:https://www.github.com/TGuillerme/landvR/tree/v0.2.Zenodo. https://doi.org/10.5281/zenodo.2620785. Gutarra, S., B. C. Moon, I. A. Rahman, C. Palmer, S. Lautenschlager, A. J. Brimacombe, and M. J. Benton. 2019. Effects of body plan evolution on the hydrodynamic drag and energy requirements of swimming in ichthyosaurs. Proceedings of the Royal Society B: Biological Sciences 286:20182786. Habib, M. B. and C. B. Ruff. 2008. The effects of locomotion on the structural characteristics of avian limb bones. Zoological Journal of the Linnean Society 153:601- 624. Hall, B. K. 2003. Unlocking the Black Box between Genotype and Phenotype: Cell Condensations as Morphogenetic (modular) Units. Biology and Philosophy 18:219-247. Hallgrimsson, B., K. Willmore, C. Dorval, and D. M. Cooper. 2004. Craniofacial variability and modularity in macaques and mice. J Exp Zool B Mol Dev Evol 302:207- 225. Hallgrimsson, B., K. Willmore, and B. K. Hall. 2002. Canalization, developmental stability, and morphological integration in primate limbs. Am J Phys Anthropol Suppl 35:131-158. Hand, S. J., V. Weisbecker, R. M. Beck, M. Archer, H. Godthelp, A. J. Tennyson, and T. H. Worthy. 2009. Bats that walk: a new evolutionary hypothesis for the terrestrial behaviour of New Zealand's endemic mystacinids. BMC Evol Biol 9:169. Hedrick, B. P., B. V. Dickson, E. R. Dumont, and S. E. Pierce. 2020. The evolutionary diversity of locomotor innovation in rodents is not linked to proximal limb morphology. Sci Rep 10:717. Hedrick, B. P., G. L. Mutumi, V. D. Munteanu, A. Sadier, K. T. J. Davies, S. J. Rossiter, K. E. Sears, Dávalos LM, Dumont E. 2019. Morphological Diversification under High Integration in a Hyper Diverse Mammal Clade. Journal of Mammalian Evolution 27, 563–575. Iriarte-Diaz, J., D. K. Riskin, D. J. Willis, K. S. Breuer, and S. M. Swartz. 2011. Whole- body kinematics of a fruit bat reveal the influence of wing inertia on body accelerations. J Exp Biol 214:1546-1553. Janis, C. M., J. G. Napoli, C. Billingham, and A. Martín-Serra. 2020. Proximal Humerus Morphology Indicates Divergent Patterns of Locomotion in Extinct Giant Kangaroos. Journal of Mammalian Evolution.

225

Jojic, V., J. Blagojevic, and M. Vujosevic. 2012. Two-module organization of the mandible in the yellow-necked mouse: a comparison between two different morphometric approaches. J Evol Biol 25:2489-2500. Kilbourne, B. M. and J. R. Hutchinson. 2019. Morphological diversification of biomechanical traits: mustelid locomotor specializations and the macroevolution of long bone cross-sectional morphology. BMC Evolutionary Biology 19:37. Klingenberg, C. P. 2013. Cranial integration and modularity: insights into evolution and development from morphometric data. Hystrix, the Italian Journal of Mammalogy 24:43-58. Klingenberg, C. P. 2014. Studying morphological integration and modularity at multiple levels: concepts and analysis. Philos Trans R Soc Lond B Biol Sci 369:20130249. Krause, D. W., S. Hoffmann, J. R. Wible, E. C. Kirk, J. A. Schultz, W. von Koenigswald, J. R. Groenke, J. B. Rossie, P. M. O'Connor, E. R. Seiffert, E. R. Dumont, W. L. Holloway, R. R. Rogers, L. J. Rahantarisoa, A. D. Kemp, and H. Andriamialison. 2014. First cranial remains of a gondwanatherian mammal reveal remarkable mosaicism. Nature 515:512-517. Kwong, S., S. Kothary, and L. L. Poncinelli. 2014. Skeletal development of the proximal humerus in the pediatric population: MRI features. AJR Am J Roentgenol 202:418-425. Law, C. J. 2019. Evolutionary shifts in extant mustelid (Mustelidae: Carnivora) cranial shape, body size and body shape coincide with the Mid-Miocene Climate Transition. Biol Letters 15:20190155. Lawler, R. R. 2008. Morphological integration and natural selection in the postcranium of wild verreaux's sifaka (Propithecus verreauxi verreauxi). Am J Phys Anthropol 136:204-213. Leiser-Miller, L. B. and S. E. Santana. 2020. Morphological Diversity in the Sensory System of Phyllostomid Bats: Implications for Acoustic and Dietary Ecology. Funct Ecol. 2020; 34: 1416– 1427. López-Aguirre, C., S. J. Hand, D. Koyabu, N. T. Son, and L. A. B. Wilson. 2019a. Postcranial heterochrony, modularity, integration and disparity in the prenatal ossification in bats (Chiroptera). BMC Evolutionary Biology 19:75. López-Aguirre, C., S. J. Hand, D. Koyabu, N. T. Son, and L. A. B. Wilson. 2019b. Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera). Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 332:36-49. López-Aguirre, C., J. Perez-Torres, and L. A. Wilson. 2015. Cranial and mandibular shape variation in the genus Carollia (Mammalia: Chiroptera) from Colombia: biogeographic patterns and morphological modularity. PeerJ 3:e1197. López-Aguirre, C., L. A. Wilson, D. Koyabu, V. Tan Tu, and S. J. Hand. 2019c. Decoupled morphological and biomechanical evolution and diversification of the wing in bats. EcoEvoRxiv:31. Louzada, N. S. V., M. R. Nogueira, and L. M. Pessôa. 2019. Comparative morphology and scaling of the femur in yangochiropteran bats. Journal of Anatomy 235:124-150. Lungmus, J. K. and K. D. Angielczyk. 2019. Antiquity of forelimb ecomorphological diversity in the mammalian stem lineage (Synapsida). Proc Natl Acad Sci U S A 116:6903-6907. Luo, B., S. E. Santana, Y. Pang, M. Wang, Y. Xiao, and J. Feng. 2019. Wing morphology predicts geographic range size in vespertilionid bats. Sci Rep 9:4526. 226

Makedonska, J., B. W. Wright, and D. S. Strait. 2012. The effect of dietary adaption on cranial morphological integration in capuchins (order Primates, genus Cebus). PLoS One 7:e40398. Marchi, D., C. B. Ruff, A. Capobianco, K. L. Rafferty, M. B. Habib, and B. A. Patel. 2016. The locomotion of Babakotia radofilai inferred from epiphyseal and diaphyseal morphology of the humerus and femur. J Morphol 277:1199-1218. Marroig, G., L. T. Shirai, A. Porto, F. B. de Oliveira, and V. De Conto. 2009. The Evolution of Modularity in the Mammalian Skull II: Evolutionary Consequences. Evolutionary Biology 36:136-148. Martin-Serra, A., B. Figueirido, and P. Palmqvist. 2014. A three-dimensional analysis of morphological evolution and locomotor performance of the carnivoran forelimb. PLoS One 9:e85574. Martin-Serra, A., B. Figueirido, and P. Palmqvist. 2017. Non-decoupled morphological evolution of the fore- and hindlimb of sabretooth predators. J Anat. Martín-Serra, A., B. Figueirido, and P. Palmqvist. 2014. A Three-Dimensional Analysis of Morphological Evolution and Locomotor Performance of the Carnivoran Forelimb. PLOS ONE 9:e85574. McGowan, A. J. and G. J. Dyke. 2007. A morphospace-based test for competitive exclusion among flying vertebrates: did birds, bats and pterosaurs get in each other's space? J Evol Biol 20:1230-1236. McInroe, B., H. C. Astley, C. Gong, S. M. Kawano, P. E. Schiebel, J. M. Rieser, H. Choset, R. W. Blob, and D. I. Goldman. 2016. Tail use improves performance on soft substrates in models of early vertebrate land locomotors. Science 353:154-158. Medina, I., G. M. Cooke, and T. J. Ord. 2018. Walk, swim or fly? Locomotor mode predicts genetic differentiation in vertebrates. Ecology Letters 21:638-645. Monteiro, L. R. and M. R. Nogueira. 2011. Evolutionary patterns and processes in the radiation of phyllostomid bats. BMC Evol Biol 11:137. Morales, A. E., M. Ruedi, K. Field, and B. C. Carstens. 2019. Diversification rates have no effect on the convergent evolution of foraging strategies in the most speciose genus of bats, Myotis. Evolution 73:2263-2280. Moyers Arévalo, R. L., L. I. Amador, F. C. Almeida, and N. P. Giannini. 2018. Evolution of Body Mass in Bats: Insights from a Large Supermatrix Phylogeny. Journal of Mammalian Evolution 27: 123-138. Nogueira, M. R., A. L. Peracchi, and L. R. Monteiro. 2009. Morphological correlates of bite force and diet in the skull and mandible of phyllostomid bats. Functional Ecology 23:715-723. Norberg, R. A. and J. M. V. Rayner. 1987. Ecological Morphology and Flight in Bats (Mammalia; Chiroptera): Wing Adaptations, Flight Performance, Foraging Strategy and Echolocation. Philos Trans R Soc Lond B Biol Sci 316:335-427. Panyutina, A., L. P. Korzun, and A. N. Kuznetsov. 2015. Flight of Mammals: From Terrestrial Limbs to Wings. Springer, New York. Parsons, K. J., E. Marquez, and R. C. Albertson. 2012. Constraint and opportunity: the genetic basis and evolution of modularity in the cichlid mandible. Am Nat 179:64-78. Patel, B. A., C. B. Ruff, E. L. Simons, and J. M. Organ. 2013. Humeral cross-sectional shape in suspensory primates and sloths. Anat Rec (Hoboken) 296:545-556. Polanski, J. M. 2011. Morphological Integration of the Modern Human Mandible during Ontogeny. Int J Evol Biol 2011:545879. 227

Porto, A., F. B. de Oliveira, L. T. Shirai, V. De Conto, and G. Marroig. 2008. The Evolution of Modularity in the Mammalian Skull I: Morphological Integration Patterns and Magnitudes. Evolutionary Biology 36:118-135. Randau, M. and A. Goswami. 2017. Unravelling intravertebral integration, modularity and disparity in Felidae (Mammalia). Evol Dev 19:85-95. Randau, M. and A. Goswami. 2018. Shape Covariation (or the Lack Thereof) Between Vertebrae and Other Skeletal Traits in Felids: The Whole is Not Always Greater than the Sum of Parts. Evolutionary Biology 45: 196-210. Rayner, J. M. 1988. The evolution of vertebrate flight. Biological Journal of the Linnean Society 34:269-287. Riskin, D. K., A. Bergou, K. S. Breuer, and S. M. Swartz. 2012. Upstroke wing flexion and the inertial cost of bat flight. Proc Biol Sci 279:2945-2950. Riskin, D. K., S. Parsons, W. A. Schutt, Jr., G. G. Carter, and J. W. Hermanson. 2006. Terrestrial locomotion of the New Zealand short-tailed bat Mystacina tuberculata and the common vampire bat Desmodus rotundus. J Exp Biol 209:1725-1736. Rojas, D., Á. Vale, V. Ferrero, and L. Navarro. 2012. The role of frugivory in the diversification of bats in the Neotropics. Journal of Biogeography 39:1948-1960. Rossoni, D. M., A. P. A. Assis, N. P. Giannini, and G. Marroig. 2017. Intense natural selection preceded the invasion of new adaptive zones during the radiation of New World leaf-nosed bats. Scientific Reports 7: 11076. Rossoni, D. M., B. M. A. Costa, N. P. Giannini, and G. Marroig. 2019. A multiple peak adaptive landscape based on feeding strategies and roosting ecology shaped the evolution of cranial covariance structure and morphological differentiation in phyllostomid bats. Evolution 73:961-981. Sallan, L. C. and M. Friedman. 2012. Heads or tails: staged diversification in vertebrate evolutionary radiations. Proceedings of the Royal Society B: Biological Sciences 279:2025-2032. Santana, S. E. and E. Cheung. 2016. Go big or go fish: morphological specializations in carnivorous bats. Proceedings of the Royal Society B: Biological Sciences 283:20160615. Santana, S. E., I. R. Grosse, and E. R. Dumont. 2012. Dietary hardness, loading behavior, and the evolution of skull form in bats. Evolution 66:2587-2598. Santana, S. E. and S. E. Lofgren. 2013. Does nasal echolocation influence the modularity of the mammal skull? J Evol Biol 26:2520-2526. Santana, S. E., S. Strait, and E. R. Dumont. 2011. The better to eat you with: functional correlates of tooth structure in bats. The better to eat you with: functional correlates of tooth structure in bats. Schlosser-Sturm, E. and H. Schliemann. 1995. Morphology and function of the shoulder joint of bats (Mammalia: Chiroptera)1. Journal of Zoological Systematics and Evolutionary Research 33:88-98. Shi, J. J. and D. L. Rabosky. 2015. Speciation dynamics during the global radiation of extant bats. Evolution 69:1528-1545. Simmons, N. B., K. L. Seymour, J. Habersetzer, and G. F. Gunnell. 2008. Primitive Early Eocene bat from Wyoming and the evolution of flight and echolocation. Nature 451:818-821. Simons, E. L., T. L. Hieronymus, and P. M. O'Connor. 2011. Cross sectional geometry of the forelimb skeleton and flight mode in pelecaniform birds. J Morphol 272:958-971. 228

Smith, J. D. 1977. Comments on Flight and the Evolution of Bats. Pp. 427-437 in M. K. Hecht, P. C. Goody, and B. M. Hecht, eds. Major Patterns in Vertebrate Evolution. Springer US, Boston, MA. Stanchak, K. E., J. H. Arbour, and S. E. Santana. 2019. Anatomical diversification of a skeletal novelty in bat feet. Evolution 73: 1591-1603. Swartz, S., J. Iriarte-Diaz, D. Riskin, X. Tian, A. Song, and K. Breuer. 2007. Wing Structure and the Aerodynamic Basis of Flight in Bats. American institute of Aeronautics and Austronautics. Swartz, S. M., M. B. Bennett, and D. R. Carrier. 1992. Wing bone stresses in free flying bats and the evolution of skeletal design for flight. Nature 359: 726-729. Swartz, S. M. and K. M. Middleton. 2008. Biomechanics of the bat limb skeleton: Scaling, material properties and mechanics. Cells Tissues Organs 187:59-84. Tokita, M., T. Abe, and K. Suzuki. 2012. The developmental basis of bat wing muscle. Nature Communications 3: 1302. Uyeda, J. C., D. S. Caetano, and M. W. Pennell. 2015. Comparative Analysis of Principal Components Can be Misleading. Systematic Biology 64:677-689. Vaughan, T. A. 1959. Functional morphology of three bats : Eumops, Myotis, Macrotus. University of Kansas, Lawrence. Verde Arregoitia, L. D., D. O. Fisher, and M. Schweizer. 2017. Morphology captures diet and locomotor types in rodents. R Soc Open Sci 4:160957. Vidal-García, M. and J. Scott Keogh. 2017. Phylogenetic conservatism in skulls and evolutionary lability in limbs – morphological evolution across an ancient frog radiation is shaped by diet, locomotion and burrowing. BMC Evolutionary Biology 17:165. Voigt, C. C., I. M. Borrisov, and S. L. Voigt-Heucke. 2012. Terrestrial locomotion imposes high metabolic requirements on bats. J Exp Biol 215:4340-4344. Watanabe, A. 2018. How many landmarks are enough to characterize shape and size variation? PLOS ONE 13:e0198341. Watts, P., E. Mitchell, and S. Swartz. 2001. A computational model for estimating the mechanics of horizontal flapping flight in bats: Model description and validation. Journal of Experimental Biology 204:2873-2898. Weller, T. J., K. T. Castle, F. Liechti, C. D. Hein, M. R. Schirmacher, and P. M. Cryan. 2016. First Direct Evidence of Long-distance Seasonal Movements and Hibernation in a Migratory Bat. Sci Rep 6:34585. Wilson, L. A. 2013. Allometric disparity in rodent evolution. Ecol Evol 3:971-984. Wisniewski, M., M. Baumgart, M. Grzonkowska, B. Malkowski, A. Wilinska-Jankowska, Z. Siedlecki, and M. Szpinda. 2017. Ossification center of the humeral shaft in the human fetus: a CT, digital, and statistical study. Surg Radiol Anat 39:1107-1116. Witton, M. P. 2015. Were early pterosaurs inept terrestrial locomotors? PeerJ 3:e1018. Young, N. M. and B. HallgrÍmsson. 2005. Serial Homology and the Evolution of Mammalian Limb Covariation Structure. Evolution 59: 2691-2704. Young, N. M., G. P. Wagner, and B. Hallgrimsson. 2010. Development and the evolvability of human limbs. Proc Natl Acad Sci U S A 107:3400-3405. Zelditch, M. L., Z. T. Calamari, and D. L. Swiderski. 2016. Disparate Postnatal Ontogenies Do Not Add to the Shape Disparity of Infants. Evolutionary Biology 43: 188- 207. Zelditch, M. L., A. R. Wood, R. M. Bonett, and D. L. Swiderski. 2008. Modularity of the rodent mandible: integrating bones, muscles, and teeth. Evol Dev 10:756-768. 229

230

Chapter 7 General conclusions Main findings

This doctoral research aimed to study the phenotypic diversity of the bat wing by assessing the potential developmental, ecological and evolutionary drivers shaping it.

With this body of research, I explored different dimensions of postcranial diversity in bats, implementing novel methodological and statistical tools. The research underlined the value of applying geometric morphometrics and cross-sectional geometry to the study of bat wing morphology and biomechanics beyond traditional aerodynamic and morphological metrics. Overall, I found significant evidence for a multivariate macroevolutionary scenario where developmental, phylogenetic and ecological traits interact to shape the ecomorphological diversity of the wing in bats. The results found in this study share some similarities with previous studies on cranial macroevolution in bats (Monteiro and Nogueira 2011; Santana, et al. 2012a; Arbour, et al. 2019; Rossoni, et al. 2019), highlighting an understudied correspondence between cranial and postcranial morphological variation. Specifically, I focused on two objectives: 1) assessment of the developmental processes and mechanics controlling postcranial morphological diversity and specialisation, and 2) quantification of forelimb morphological and biomechanical diversity in modern bats by reconstructing ecology- based patterns of variation of the humerus. In chapter 2, I first combined data of sequence heterochrony and metric growth of 24 postcranial bones from 14 bat species to quantitatively compare the prenatal development of bats with that of birds and non-volant mammals, in the most comprehensive study of chiropteran prenatal ossification to date. I found sequence heterochrony across vertebrate groups, showing 231 that bat postcranial development shares similarities with a general mammalian ossification pattern, except for bones functionally associated with flight which showed similarities with other flying vertebrate groups (i.e. precocial pedal phalanges, altricial sternum and heterochrony between the fore- and hindlimb). I also tested nine different modularity hypotheses of postcranial development (both in metric growth and ossification sequence datasets), of which only an axial-appendicular partition of the postcranium was found to be significant both for bats and non-volant mammals.

An axial-appendicular modular partition of postcranial development, which was common for all mammals studied, suggests a common mammalian developmental pattern shaped by the developmental origin of the different components of the postcranium (all appendicular bones derive from the lateral plate mesoderm;

Goswami, et al. (2009). These results also indicate that flight did not over-write developmental integration patterns in bats, which instead retain the general mammalian condition.

Using metric growth data, I also reconstructed developmental trajectories of morphological disparity and integration, testing whether morphological integration constrains morphological variability. My results indicate that both integration and disparity increase across development, rejecting the hypothesis that integration constrains disparity and suggesting a positive correlation prenatally (Goswami, et al.

2014a; Felice, et al. 2018). By testing for differences in the disparity-integration interaction across functionally distinctive units (e.g. axial and appendicular skeleton), I found that a positive correlation is only evident from a temporal perspective rather than a morphofunctional one. Overall, these results indicate that bats have retained a

232 mammalian developmental pattern (both in modular partition and ossification patterns), with the exception of bones functionally associated with flight (i.e. handwing and sternum), where bats were more similar to birds in their development.

In chapter 3, I used metric growth data to reconstruct the allometric trajectories of nine bat species in order to assess the developmental basis of the postcranial phenotypic diversity in bats. I found statistically significant allometric growth in most of the species, while also finding greater variability within Yinpterochiroptera than within Yangochiroptera and between suborders. Using a phylogenetic hypothesis to test the presence of phylogenetic signal in prenatal allometry, I found that evolutionary kinship shapes differences in allometry, with species of the same suborder showing greater similarity than between suborders. I also used the allometric trajectories to reconstruct a developmental morphospace, where each suborder occupied non-overlapping subspaces that evidenced deviations in the development of specific bones for each group (i.e. slower than average growth of the cervical, thoracic and lumbar vertebrae in Yinpterochiroptera and faster than average growth of manual phalanges in Yangochiroptera).

I also used a phylogenetic hypothesis to perform an ancestral state reconstruction of shape change with size, using log-shape ratios to deconstruct linear measurements into shape and size. Greater changes in shape with size were found in larger species, indicating that size could act as an evolutionary buffer promoting phenotypic diversification in bats. This finding was supported by the higher magnitudes of allometry-corrected phenotypic disparity I found in larger-bodied bat species.

Combining all findings in this chapter, I provided evidence for an ontogenetic basis to 233 the postcranial morphological diversity in modern bats that reflects evolutionary relationships among major lineages. Heterochronies in ossification sequences have been reported between major mammalian lineages (Hautier, et al. 2013). Moreover, greater shape changes with size and greater allometry-corrected disparity in larger species indicate that differences found among bat species might reflect size- dependent evolutionary constraints, where variability in body size and allometric patterns are associated such that greater body size may facilitate higher magnitudes of disparity (Marroig and Cheverud 2005b).

In chapter 4, I explored the ontogenetic mechanisms controlling morphogenesis of the wing in bats by assessing the presence and magnitude of phenotypic asymmetry in the humerus. I focused on the humerus because it represents a structure recognised to be under multiple functional adaptive pressures (i.e. increasing insertion area for key muscles during wingbeat, greater resistance to torsional and bending stress and control of wing manoeuvrability). Specifically, I tested the hypothesis that asymmetry is developmentally constrained to decrease across development, as a result of compensatory growth (Ueti, et al. 2015). By decomposing humeral asymmetry into three types of previously described asymmetries (i.e. fluctuating asymmetry, directional asymmetry and antisymmetry), I was able to find a clear link between fluctuating asymmetry and developmental instability. I measured two different dimensions of asymmetry (longitudinal and cross-sectional) in order to assess whether they show similar temporal trajectories.

Fluctuating asymmetry was statistically significant for both longitudinal and cross- sectional asymmetry, explaining on average for 25% of asymmetric variation during

234 humeral development. Longitudinal and cross-sectional asymmetry decreased throughout ontogeny, supporting the hypothesis of mechanisms canalising asymmetry to buffer developmental instability (Kellner and Alford 2003). Moreover, I found decoupled developmental trajectories between longitudinal and cross-sectional asymmetry, suggesting that multiple patterns of asymmetry can occur within a single structure. I hypothesise that statistically significant levels of decreasing cross-sectional asymmetry across developmental stages indicate greater canalisation of biomechanical asymmetry, possibly as a result of selective pressures associated with bat newborn ecology (i.e. greater bone robusticity to ensure flight development is completed postnatally and the attachment of newborns to the mothers; Adams (2008); Koyabu and Son (2014).

Next, I reconstructed the variation patterns and ecological drivers of the morphological and biomechanical diversification of the humerus in modern bats, studying cross- sectional geometry and whole-bone external morphology of the bone (chapters 5 and

6, respectively).

In chapter 5, I used phylogenetic comparative methods (Adams and Collyer 2019), geometric morphometrics (Wilson and Humphrey 2015) and cross-sectional geometry

(Simons, et al. 2011; Voeten, et al. 2018) to quantify the biomechanical properties and periosteal morphology of the humeral diaphysis and investigate whether different functional demands influenced cross-sectional shape and biomechanical disparity and scaling. Significant ecological signal and no phylogenetic structuring in humeral phenotypic variability were evident in the cross-sectional dataset. A decoupling in the modes of shape and biomechanical scaling and variation was found to respond to

235 ecological specialisations, in support of Wolff’s law of functional bone remodelling

(Wolff 1986; Meers 2002; Ruff, et al. 2006). By reconstructing allometric trajectories and magnitudes of disparity of morpho-biomechanical properties of the humerus, it was evident that foraging strategies could have shaped the phenotypic evolution of the postcranium (Amador, Almeida, et al. 2019). Moreover, significant shifts in the magnitudes of morphological disparity and how it scales with size were found in taxa with rare ecologies within Chiroptera (e.g. terrestrial locomotion and upstand roosting).

In chapter 6, I quantified the morphological disparity of the humerus in modern bats, by dissecting patterns of whole-bone, diaphyseal and epiphyseal morphological variation. Traditional aerodynamic metrics used to analyse wing morphology were also tested for an association with humeral shape (Bullen and McKenzie 2001; Bullen and

McKenzie 2007). The role of phylogeny, diet, foraging strategy and size were tested as explainers of humeral shape variation at all levels of morphological variation analysed.

Overall, whole-bone and diaphyseal shape showed similar patterns of variation, suggesting that whole-bone shape variation is mostly driven by its diaphyseal component, possibly obscuring the pattern and magnitude of epiphyseal shape variation. A significant modular partition of humeral shape variation into diaphyseal and epiphyseal morphology also highlights the potential of analysing these components separately to better describe humeral ecomorphology. All three levels of humeral shape variation responded to phylogeny, diet and foraging guild, whereas only epiphyseal morphology correlated with size. Only epiphyseal shape correlated with any wing aerodynamic metric (i.e. aspect ratio). With this study I provided the

236 first evidence of within-structure modular morphological variation in the appendicular skeleton of a living tetrapod.

Future studies

The study of ecomorphological evolution has experienced a resurgence of interest in recent decades with the advent of novel analytical and computational capabilities that have revolutionised the scope and depth of studies. In mammals, greater effort has been focused on studying cranial morphological evolution in order to better understand the evolution of attributes such as diet (e.g. dental complexity) and cognitive ability (e.g. brain mass) (Monteiro and Nogueira 2011; Cardini and Polly

2013; Prufrock, et al. 2016; Rossoni, et al. 2017; Arbour, et al. 2019; Hedrick, et al.

2019; Law 2019; Rossoni, et al. 2019; Neubauer, et al. 2020). Postcranial ecomorphological evolution has been less studied. Some studies have analysed how locomotion shaped diversification in some clades (Riskin, et al. 2005; Young 2006;

Lawler 2008; Hand, et al. 2009; Martin-Serra, et al. 2014; Jones, et al. 2015), but other major locomotory transitions during mammalian evolution remain comparatively less understood (e.g. self-powered flight in bats). Due to the relatively poor bat fossil record (~20% has been described) (Eiting and Gunnell 2009; Brown, et al. 2019), furthering our understanding of the evolutionary novelty of mammalian self-powered flight relies heavily on the application of Evolutionary Developmental Biology

(EvoDevo), geometric morphometrics and phylogenetic comparative methods to reconstruct macroevolutionary trajectories.

The evolutionary role that integration and modularity play during phenotypic diversification is a promising area of research that will help elucidate historical

237 ecomorphological changes (Wagner, et al. 2007; Porto, et al. 2008; Marroig, et al.

2009; Goswami, et al. 2014a; Felice, et al. 2018). Much of what we currently know of the integration-disparity evolutionary dynamic is based on cranial morphology, providing evidence for both a positive (i.e. integration promotes disparity) and a negative (i.e. integration constrains disparity) association (Goswami, et al. 2014a;

Felice, et al. 2018). Phenotypic integration in the postcranium of mammals has been relatively less studied, although it has gained increasing interest recently (Hanot, et al.

2017, 2018; Mallet, et al. 2020), applying methodologies that have been mostly applied to cranial forms. Exploring the role that different locomotory modes have on shaping magnitudes of postcranial integration in mammals is an area with particular potential, as most such studies are limited to non-volant vertebrates. Moreover, future studies should further explore results presented in this thesis of decoupled patterns of within-structure phenotypic variation, as evidenced by patterns of fluctuating asymmetry, morpho-biomechanical disparity and epiphyseal-diaphyseal modular covariation. For example, testing decoupled patterns of morphological variation in the femur could reflect similar evolutionary pathways of autopodial ecomorphological specialisation.

Comparing macroevolutionary trends and the possible ecological and biological drivers of postcranial phenotypic variation across flying vertebrate groups could elucidate common processes and mechanisms that facilitated the evolution of vertebrate flight

(Rayner 1988). Early studies have shown the potential of similar methods to the ones applied in this thesis to test evolutionary hypothesis of niche partitioning and limb covariation in bats, birds and pterosaurs (McGowan and Dyke 2007; Bell, et al. 2011).

Understanding the magnitude of association between wing bones in flying vertebrates, 238 in comparison to terrestrial , could shed light on whether postcranial integration promoted ecological diversity and speciation. This could potentially inform how morphological variability impacts ecological resilience and adaptability in the

Anthropocene. Evaluating differences in fluctuating asymmetry between flying vertebrates with different foraging and locomotory strategies would reveal multipatterned evolutionary patterns within vertebrate aerial niches. Many of the results and conclusions presented in this thesis could be used to inform studies necessary to enhance our knowledge of mammalian flight and vertebrate flight in general. Moreover, improving our understanding of the ecomorphology of bats would inform the diversity of functional roles bats have in modern ecosystems.

Summary

This project helped elucidate previously unexplored levels of phenotypic diversity in the wing of bats. Developmental mechanisms and processes were linked to postcranial phenotypic morphological disparity that correlate with a variety of ecological and biological traits of bats beyond flight. Deviations from general mammalian developmental patterns were detected in bones associated with flight, suggesting common morphogenetic processes in the evolution of flight across different vertebrate lineages. Allometric trajectories of postcranial development reconstructed evolutionary relationships between major lineages. The humerus was used as a study case for detailed exploration of variation patterns and homeostatic developmental mechanisms of phenotypic diversification. Decoupled modes of variation and scaling were evident across different foraging guilds, as well as across different phenotypic dimensions of the bone (i.e. morphological and biomechanical). Previously unreported

239 patterns of morphological modularity were found in the bat humerus, recording within-single-structure modularity for the first time in the appendicular skeleton of any tetrapod. This thesis opens new scope for research to better understand the evolution of mammalian flight by combining developmental, ecological and phylogenetic data.

References

Adams DC. 2014. A generalized K statistic for estimating phylogenetic signal from shape and other high-dimensional multivariate data. Syst Biol 63:685-697. Adams DC, Collyer ML. 2018a. Multivariate Phylogenetic Comparative Methods: Evaluations, Comparisons, and Recommendations. Syst Biol 67:14-31. Adams DC, Collyer ML. 2018b. Phylogenetic ANOVA: Group-clade aggregation, biological challenges, and a refined permutation procedure. Evolution 72:1204-1215. Adams DC, Collyer ML. 2019. Phylogenetic Comparative Methods and the Evolution of Multivariate Phenotypes. Annual Review of Ecology, Evolution, and Systematics 50:405-425. Adams DC, Collyer ML, Kaliontzopoulou A, Sherratt E. 2017. Geomorph: software for geometric morphometric analyses. R package. Version 3.0.5: https://cran.r- project.org/package=geomorph. Adams DC, Nistri A. 2010. Ontogenetic convergence and evolution of foot morphology in European cave salamanders (Family: Plethodontidae). BMC Evolutionary Biology 10. Adams DC, Otárola-Castillo E, Paradis E. 2013a. Geomorph: an R package for the collection and analysis of geometric morphometric shape data. Methods in Ecology and Evolution 4:393-399. Adams DC, Otárola-Castillo E, Paradis E. 2013b. geomorph: anrpackage for the collection and analysis of geometric morphometric shape data. Methods in Ecology and Evolution 4:393-399. Adams DC, Peres-Neto P. 2016. Evaluating modularity in morphometric data: challenges with the RV coefficient and a new test measure. Methods in Ecology and Evolution 7:565-572. Adams DC, Rohlf FJ. 2000. Ecological character displacement in Plethodon: Biomechanical differences found from a geometric morphometric study. Proceedings of the National Academy of Sciences 97:4106-4111. Adams RA. 1992. Comparative Skeletogenesis of the Forearm of the Little Brown Bat (Myotis-Lucifugus) and the Norway Rat (Rattus-Norvegicus). Journal of Morphology 214:251-260. Adams RA. 2008. Morphogenesis in Bat Wings: Linking Development, Evolution and Ecology. Cells Tissues Organs 187:13-23. Adams RA. 2000. Wing ontogeny, shifting niche dimensions, and adaptive landscapes. In: Adams RA, Pedersen SC, editors. Ontogeny, Functional Ecology and Evolution of Bats. Cambridge: Cambridge University Press. p. 275-316. Adams RA, Shaw JB. 2013. Time’s Arrow in the Evolutionary Development of Bat Flight. Bat evolution, ecology and conservation: 21-46.

240

Albert AM, Greene DL. 1999. Bilateral asymmetry in skeletal growth and maturation as an indicator of environmental stress. American Journal of Physical Anthropology 110:341-349. Aldridge HDJN. 1987. Turning Flight of Bats. Journal of Experimental Biology 128:419- 425. Altenbach, J. S. 1979. Locomotor morphology of the vampire bat, Desmodus rotundus. Special Publication No. 6, American Society of Mammalogists. Amador LI, Almeida FC, Giannini NP. 2019. Evolution of Traditional Aerodynamic Variables in Bats (Mammalia: Chiroptera) within a Comprehensive Phylogenetic Framework. Journal of Mammalian Evolution 27: 549-561.. Amador LI, Arevalo RLM, Almeida FC, Catalano SA, Giannini NP. 2018. Bat Systematics in the Light of Unconstrained Analyses of a Comprehensive Molecular Supermatrix. Journal of Mammalian Evolution 25:37-70. Amador LI, Giannini NP, Simmons NB, Abdala V. 2018. Morphology and Evolution of Sesamoid Elements in Bats (Mammalia: Chiroptera). American Museum Novitates 2018:1-40, 40. Amador LI, Simmons NB, Giannini NP. 2019. Aerodynamic reconstruction of the primitive fossil bat Onychonycteris finneyi (Mammalia: Chiroptera). Biology Letters 15:20180857. Amson E, Arnold P, van Heteren AH, Canoville A, Nyakatura JA. 2017. Trabecular architecture in the forelimb epiphyses of extant xenarthrans (Mammalia). Frontiers in Zoology 14:52. Aparicio J. 1998. Patterns of fluctuating asymmetry in developing primary feathers: a test of the compensational growth hypothesis. Proceedings of the Royal Society of London. Series B: Biological Sciences 265:2353-2357. Aparicio JM, Bonal R. 2002. Why do some traits show higher fluctuating asymmetry than others? A test of hypotheses with tail feathers of birds. Heredity 89:139-144. Arbour JH, Curtis AA, Santana SE. 2019. Signatures of echolocation and dietary ecology in the adaptive evolution of skull shape in bats. Nat Commun 10:2036. Arita HT, Fenton MB. 1997. Flight and echlocation in the ecology and evolution of bats. Trends in Ecology & Evolution 12:53-58. Arita HT, Vargas-Barón J, Villalobos F. 2014. Latitudinal gradients of genus richness and endemism and the diversification of New World bats. Ecography 37: 1024-1033. Armfield BA, Zheng Z, Bajpai S, Vinyard CJ, Thewissen J. 2013. Development and evolution of the unique cetacean dentition. PeerJ 1:e24. Atchley WR, Hall BK. 1991. A MODEL FOR DEVELOPMENT AND EVOLUTION OF COMPLEX MORPHOLOGICAL STRUCTURES. Biological Reviews 66:101-157. Bahlman JW, Price‐Waldman RM, Lippe HW, Breuer KS, Swartz SM. 2016. Simplifying a wing: diversity and functional consequences of digital joint reduction in bat wings. Journal of Anatomy 229:114-127. Balavoine G, Adoutte A. 2003. The Segmented Urbilateria: A Testable Scenario1. Integrative and Comparative Biology 43:137-147. Baliga VB, Szabo I, Altshuler DL. 2019. Range of motion in the avian wing is strongly associated with flight behavior and body mass. Science Advances 5:eaaw6670. Baptista TL, Richardson CS, Kunz TH. 2000. Postnatal Growth and Age Estimation in Free-Ranging Bats: A Comparison of Longitudinal and Cross-Sectional Sampling Methods. Journal of mammalogy 81:709-718. 241

Bastide P, Ane C, Robin S, Mariadassou M. 2018. Inference of Adaptive Shifts for Multivariate Correlated Traits. Syst Biol 67:662-680. Bell E, Andres B, Goswami A. 2011. Integration and dissociation of limb elements in flying vertebrates: a comparison of pterosaurs, birds and bats. J Evol Biol 24:2586- 2599. Bergou AJ, Swartz SM, Vejdani H, Riskin DK, Reimnitz L, Taubin G, Breuer KS. 2015. Falling with Style: Bats Perform Complex Aerial Rotations by Adjusting Wing Inertia. PLoS Biol 13:e1002297. Bininda-Emonds OR, Cardillo M, Jones KE, MacPhee RD, Beck RM, Grenyer R, Price SA, Vos RA, Gittleman JL, Purvis A. 2007. The delayed rise of present-day mammals. Nature 446:507-512. Bininda-Emonds OR, Jeffery JE, Sánchez-Villagra MR, Hanken J, Colbert M, Pieau C, Selwood L, Ten Cate C, Raynaud A, Osabutey CK, et al. 2007. Forelimb-hindlimb developmental timing changes across tetrapod phylogeny. BMC Evol Biol 7:182. Bishop KL. 2008. The evolution of flight in bats: narrowing the field of plausible hypotheses. Q Rev Biol 83:153-169. Blackburn A. 2011. Bilateral asymmetry of the humerus during growth and development. American Journal of Physical Anthropology 145:639-646. Boerma DB, Barrantes JP, Chung C, Chaverri G, Swartz SM. 2019. Specialized landing maneuvers in Spix's disk-winged bats (Thyroptera tricolor) reveal linkage between roosting ecology and landing biomechanics. The Journal of Experimental Biology:jeb.204024. Boerma DB, Breuer KS, Treskatis TL, Swartz SM. 2019. Wings as inertial appendages: how bats recover from aerial stumbles. The Journal of Experimental Biology:jeb.204255. Boessenecker RW, Fordyce RE. 2015. Anatomy, feeding ecology, and ontogeny of a transitional baleen whale: a new genus and species of Eomysticetidae (Mammalia: Cetacea) from the Oligocene of New Zealand. PeerJ 3:e1129. Bohmer C, Amson E, Arnold P, van Heteren AH, Nyakatura JA. 2018. Homeotic transformations reflect departure from the mammalian 'rule of seven' cervical vertebrae in sloths: inferences on the Hox code and morphological modularity of the mammalian neck. BMC Evol Biol 18:84. Boonman A, Bumrungsri S, Yovel Y. 2014. Nonecholocating fruit bats produce biosonar clicks with their wings. Current biology : CB 24:2962-2967. Breuker CJ, Patterson JS, Klingenberg CP. 2006. A single basis for developmental buffering of Drosophila wing shape. PLoS ONE 1:e7. Broek CMAt, Bots J, Bugiani M, Galis F, Dongen Sv. 2017. Developmental Origins of Limb Developmental Instability in Human Fetuses: Many Abnormalities Make the Difference. Symmetry 9: 51. Brokaw AF, Smotherman M. 2020. Role of ecology in shaping external nasal morphology in bats and implications for olfactory tracking. PLoS ONE 15:e0226689. Brown EE, Cashmore DD, Simmons NB, Butler RJ. 2019. Quantifying the completeness of the bat fossil record. Palaeontology 62: 757-776. Buchholtz EA, Stepien CC. 2009. Anatomical transformation in mammals: developmental origin of aberrant cervical anatomy in tree sloths. Evolution & Development 11:69-79.

242

Bullen R, McKenzie NL. 2001. Bat airframe design: flight performance, stability and control in relation to foraging ecology. Australian Journal of Zoology 49:235-261. Bullen RD, McKenzie NL. 2007. Bat wing airfoil and planform structures relating to aerodynamic cleanliness. Australian Journal of Zoology 55:237-247. Burke AC, Nowicki JL. 2003. A new view of patterning domains in the vertebrate mesoderm. Developmental Cell 4:159-165. Butler RJ, Benson RBJ, Barrett PM. 2013. Pterosaur diversity: Untangling the influence of sampling biases, Lagerstätten, and genuine biodiversity signals. Palaeogeography, Palaeoclimatology, Palaeoecology 372:78-87. Camacho J, Heyde A, Bhullar B-AS, Haelewaters D, Simmons NB, Abzhanov A. 2019. Peramorphosis, an evolutionary developmental mechanism in neotropical bat skull diversity. Developmental Dynamics 248:1129-1143. Canals M, Iriarte-Diaz J, Grossi B. 2011. Biomechanical, Respiratory and Cardiovascular Adaptations of Bats and the Case of the Small Community of Bats in Chile. In: Klika V, editor. Biomechanics in Applications: InTech. p. 299-322. Canty A, Ripley BS. 2017. boot: Bootstrap R (S-Plus) Functions. Version 1.3-20. Cardini A, O’Higgins P. 2005. Post-natal ontogeny of the mandible and ventral cranium in Marmota species (Rodentia, Sciuridae): allometry and phylogeny. Zoomorphology 124:189-203. Cardini A, Polly PD. 2013. Larger mammals have longer faces because of size-related constraints on skull form. Nature Communications 4: 2458. Carter AJR, Houle D. 2011. Artificial selection reveals heritable variation for developmental instability. Evolution 65:3558-3564. Carter RT, Adams RA. 2016. Integrating Ontogeny of Echolocation and Locomotion Gives Unique Insights into the Origin of Bats. Journal of Mammalian Evolution 23:413– 421. Carter RT, Adams RA. 2015. Postnatal ontogeny of the cochlea and flight ability in Jamaican fruit bats (Phyllostomidae) with implications for the evolution of echolocation. J Anat 226:301-308. Castillo-Figueroa D. 2018. Fluctuating asymmetry of three bat species in extensive livestock systems from Córdoba department, Colombia. Revista colombiana de ciencia animal recia 10:143-153. Cheney JA, Ton D, Konow N, Riskin DK, Breuer KS, Swartz SM. 2014. Hindlimb motion during steady flight of the lesser dog-faced fruit bat, Cynopterus brachyotis. PLoS ONE 9:e98093. Chin DD, Matloff LY, Stowers AK, Tucci ER, Lentink D. 2017. Inspiration for wing design: how forelimb specialization enables active flight in modern vertebrates. J R Soc Interface 14. Clavel J, Escarguel G, Merceron G. 2015. mvmorph: an r package for fitting multivariate evolutionary models to morphometric data. Methods in Ecology and Evolution 6:1311- 1319. Cobb SN, O’Higgins P. 2004. Hominins Do Not Share a Common Postnatal Facial Ontogenetic Shape Trajectory. J Exp Zool B Mol Dev Evol 302:302-321. Coda JA, Martiínez JJ, Steinmann AR, Priotto J, Gomez MD. 2017. Fluctuating asymmetry as an indicator of environmental stress in small mammals Mastozoología Neotropical 24:313-321.

243

Collyer ML, Sekora DJ, Adams DC. 2015. A method for analysis of phenotypic change for phenotypes described by high-dimensional data. Heredity (Edinb) 115:357-365. Conaway MA, Schroeder L, von Cramon-Taubadel N. 2018. Morphological integration of anatomical, developmental, and functional postcranial modules in the crab-eating macaque (Macaca fascicularis). Am J Phys Anthropol 166:661-670. Cooper KL, Tabin CJ. 2008. Understanding of bat wing evolution takes flight. Genes Dev 22:121-124. Cooper L, Cretekos CJ, Sears KE. 2012. The evolution and development of mammalian flight. Wiley Interdisciplinary Reviews: Developmental Biology 1:773-779. Cooper LN, Sears KE, Armfield BA, Kala B, Hubler M, Thewissen JGM. 2017. Review and experimental evaluation of the embryonic development and evolutionary history of flipper development and hyperphalangy in dolphins (Cetacea: Mammalia). Genesis 56: e23076. Cordero GA, Sánchez-Villagra MR, Werneburg I. 2020. An irregular hourglass pattern describes the tempo of phenotypic development in placental mammal evolution. Biology Letters 16:20200087. Cretekos CJ, Wang Y, Green ED, Martin JF, Rasweiler JJt, Behringer RR. 2008. Regulatory divergence modifies limb length between mammals. Genes Dev 22:141- 151. Cretekos CJ, Weatherbee SD, Chen CH, Badwaik NK, Niswander L, Behringer RR, Rasweiler JJt. 2005. Embryonic staging system for the short-tailed fruit bat, Carollia perspicillata, a model organism for the mammalian order Chiroptera, based upon timed pregnancies in captive-bred animals. Dev Dyn 233:721-738. Cubo J, Casinos A. 1998. The variation of the cross-sectional shape in the long bones of birds and mammals. Annales des Sciences Naturelles - Zoologie et Biologie Animale 19:51-62. Currey JD, Alexander MR. 1985. The thickness of the walls of tubular bones. Journal of Zoology 206:453-468. da Silva JM, Herrel A, Measey GJ, Vanhooydonck B, Tolley KA. 2014. Linking microhabitat structure, morphology and locomotor performance traits in a recent radiation of dwarf chameleons. Functional Ecology 28:702-713. Datzmann T, von Helversen O, Mayer F. 2010. Evolution of nectarivory in phyllostomid bats (Phyllostomidae Gray, 1825, Chiroptera: Mammalia). BMC Evol Biol 10: 165. Davalos LM, Velazco PM, Warsi OM, Smits PD, Simmons NB. 2014. Integrating incomplete fossils by isolating conflicting signal in saturated and non-independent morphological characters. Syst Biol 63:582-600. de Camargo NF, de Oliveira HFM. 2012. Sexual Dimorphism in Sturnira lilium (Chiroptera, Phyllostomidae): Can Pregnancy and Pup Carrying Be Responsible for Differences in Wing Shape? PLoS ONE 7. de Figueiredo D, Gonçalves Miller B, Barbosa Leal ES, Montes MA. 2015. Fluctuating asymmetry in populations of bats: species adapted to urban environments are not hampered by habitat degradation. Chiroptera Neotropical 21:1305-1311. Dececchi TA, Larsson HC, Habib MB. 2016. The wings before the bird: an evaluation of flapping-based locomotory hypotheses in bird antecedents. PeerJ 4:e2159. Denzinger A, Schnitzler HU. 2013. Bat guilds, a concept to classify the highly diverse foraging and echolocation behaviors of microchiropteran bats. Front Physiol 4:164. Dickinson M. 2008. Animal locomotion: a new spin on bat flight. Curr Biol 18:R468-470. 244

Dietz CL. 1973. Bat walking behavior. Journal of mammalogy 54: 790-792. Diogo R. 2017. Etho-Eco-Morphological Mismatches, an Overlooked Phenomenon in Ecology, Evolution and Evo-Devo That Supports ONCE (Organic Nonoptimal Constrained Evolution) and the Key Evolutionary Role of Organismal Behavior. Frontiers in Ecology and Evolution 5. Diogo R, Ziermann JM, Smith C, Alghamdi M, Fuentes JSM, Duerinckx A. 2019. First use of anatomical networks to study modularity and integration of heads, forelimbs and hindlimbs in abnormal anencephalic and cyclopic vs normal human development. Sci Rep 9:7821. Dobson GE. 1875. On the genus Scotophilus, with description of a new genus and species allied thereto. Proceedings of the Zoological Society of London 24:368-373. Dongen SV. 2006. Fluctuating asymmetry and developmental instability in evolutionary biology: past, present and future. Journal of Evolutionary Biology 19:1727-1743. Dongen SV, Broek CT, Bots J, Galis F. 2017. Changes of Fluctuating Asymmetry with Age in Human Fetuses and Young Infants. Symmetry 9: 44. Dumont ER. 2010. Bone density and the lightweight skeletons of birds. Proceedings of the Royal Society B: Biological Sciences 277:2193-2198. Dumont ER, Davalos LM, Goldberg A, Santana SE, Rex K, Voigt CC. 2012. Morphological innovation, diversification and invasion of a new adaptive zone. Proc Biol Sci 279:1797- 1805. Dumont ER, Samadevam K, Grosse I, Warsi OM, Baird B, Davalos LM. 2014. Selection for mechanical advantage underlies multiple cranial optima in new world leaf-nosed bats. Evolution 68:1436-1449. Eckalbar WL, Schlebusch SA, Mason MK, Gill Z, Parker AV, Booker BM, Nishizaki S, Muswamba-Nday C, Terhune E, Nevonen KA, et al. 2016. Transcriptomic and epigenomic characterization of the developing bat wing. Nature genetics 48:528-536. Eghbali H, Shahabi S, Najafi N, Mehdizadeh R, Yousefi S, Sharifi M. 2017. Postnatal growth, wing development and age estimations in the Mediterranean horseshoe bat Rhinolophus euryale (Chiroptera: Rhinolophidae) in Kerend cave, western Iran. Mammalia 82: 276-287. Eghbali H, Sharifi M. 2018. Postnatal Growth, Age Estimation, and Wing Development in Geoffroy's Bat Myotis emarginatus (Chiroptera: Vespertilionidae). Mammal Study 43:153 - 165. Eiting TP, Gunnell GF. 2009. Global Completeness of the Bat Fossil Record. Journal of Mammalian Evolution 16:151-173. Ellington CP, vandenBerg C, Willmott AP, Thomas ALR. 1996. Leading-edge vortices in insect flight. Nature 384:626-630. Emlen JM, Freeman DC, Graham JH. 1993. Nonlinear growth dynamics and the origin of fluctuating asymmetry. Genetica 89:77. Esquerré D, Sherratt E, Keogh JS. 2017. Evolution of extreme ontogenetic allometric diversity and heterochrony in pythons, a clade of giant and dwarf snakes. Evolution 71:2829-2844. Fabre A-C, Cornette R, Goswami A, Peigné S. 2015. Do constraints associated with the locomotor habitat drive the evolution of forelimb shape? A case study in musteloid carnivorans. Journal of Anatomy 226:596-610.

245

Falkowski PG, Katz ME, Milligan AJ, Fennel K, Cramer BS, Aubry MP, Berner RA, Novacek MJ, Zapol WM. 2005. The rise of oxygen over the past 205 million years and the evolution of large placental mammals. Science 309:2202-2204. Faure PA, Barclay RMR. 1994. Substrate-Gleaning Versus Aerial-Hawking - Plasticity in the Foraging and Echolocation Behavior of the Long-Eared Bat, Myotis-Evotis. Journal of Comparative Physiology a-Sensory Neural and Behavioral Physiology 174:651-660. Felice RN, Randau M, Goswami A. 2018. A fly in a tube: Macroevolutionary expectations for integrated phenotypes. Evolution 72:2580-2594. Felsenstein J. 1973. Maximum-likelihood estimation of evolutionary trees from continuous characters. American Journal of Human Genetics 25:471-492. Fenton MB. 2010. Convergences in the diversification of bats. Current Zoology (formerly Acta Zoologica Sinica) 56:454-468. Ferry-Graham L, Bolnick D, Wainwright P. 2002. Using Functional Morphology to Examine the Ecology and Evolution of Specialization. Integrative and Comparative Biology 42:265–277. Figueirido B, Janis CM, Perez-Claros JA, De Renzi M, Palmqvist P. 2012. Cenozoic climate change influences mammalian evolutionary dynamics. Proc Natl Acad Sci U S A 109:722-727. Foley NM, Thong VD, Soisook P, Goodman SM, Armstrong KN, Jacobs DS, Puechmaille SJ, Teeling EC. 2015. How and why overcome the impediments to resolution: lessons from rhinolophid and hipposiderid bats. Mol Biol Evol 32:313-333. Fong JJ, Brown JM, Fujita MK, Boussau B. 2012. A Phylogenomic Approach to Vertebrate Phylogeny Supports a Turtle-Archosaur Affinity and a Possible Paraphyletic Lissamphibia. PLoS ONE 7. Frederich B, Vandewalle P. 2011. Bipartite life cycle of coral reef fishes promotes increasing shape disparity of the head skeleton during ontogeny: an example from damselfishes (Pomacentridae). BMC Evolutionary Biology 11. Frelat MA, Katina S, Weber GW, Bookstein FL. 2012. Technical note: A novel geometric morphometric approach to the study of long bone shape variation. American Journal of Physical Anthropology 149:628-638. Friedman ST, Martinez CM, Price SA, Wainwright PC. 2019. The influence of size on body shape diversification across Indo‐Pacific shore fishes. Evolution 73: 1873-1884. Garcia G, Hingst-Zaher E, Cerqueira R, Marroig G. 2014. Quantitative Genetics and Modularity in Cranial and Mandibular Morphology of Calomys expulsus. Evolutionary Biology 41:619-636. Gardiner JD, Codd JR, Nudds RL. 2011. An association between ear and tail morphologies of bats and their foraging style. Canadian Journal of Zoology 89:90-99. Gaudioso PJ, Diaz MM, Barquez RM. 2017. Morphology of the axial skeleton of seven bat genera (Chiroptera: Phyllostomidae). An Acad Bras Cienc 89:2341-2358. Geiger M, Forasiepi AM, Koyabu D, Sánchez-Villagra MR. 2014. Heterochrony and post- natal growth in mammals--an examination of growth plates in limbs. J Evol Biol 27:98- 115. Genikhovich G, Technau U. 2017. On the evolution of bilaterality. Development 144:3392. Gerber S. 2013. On the Relationship between the Macroevolutionary Trajectories of Morphological Integration and Morphological Disparity. PLoS ONE 8.

246

Gerber S, Eble GJ, Neige P. 2008. Allometric space and allometric disparity: a developmental perspective in the macroevolutionary analysis of morphological disparity. Evolution 62:1450-1457. Giannini N, Goswami A, Sánchez-Villagra MR. 2006. Development of integumentary structures in Rousettus amplexicaudatus (Mammalia: Chiroptera: Pteropodidae) during late-embryonic and fetal stages. Journal of mammalogy 95:993-1001. Giannini NP. 2012. Toward an integrative theory on the origin of bat flight. In: Gunnell GF, Simmons NB, editors. Evolutionary History of Bats: Fossils, Molecules and Morphology. Cambridge: Cambridge University Press. p. 353-384. Giannini NP, Abdala F, Flores DA. 2004. Comparative postnatal ontogeny of the skull in Dromiciops gliroides (Marsupialia : Microbiotheriidae). American Museum Novitates:1- 17. Gill PG, Purnell MA, Crumpton N, Brown KR, Gostling NJ, Stampanoni M, Rayfield EJ. 2014. Dietary specializations and diversity in feeding ecology of the earliest stem mammals. Nature 512:303-305. Goswami A. 2007. Cranial modularity and sequence heterochrony in mammals. Evol Dev 9:290-298. Goswami A, Binder WJ, Meachen J, O'Keefe FR. 2015. The fossil record of phenotypic integration and modularity: A deep-time perspective on developmental and evolutionary dynamics. Proc Natl Acad Sci U S A 112:4891-4896. Goswami A, Polly PD. 2010. The Influence of Modularity on Cranial Morphological Disparity in Carnivora and Primates (Mammalia). PloS one 5:e9517. Goswami A, Smaers JB, Soligo C, Polly PD. 2014. The macroevolutionary consequences of phenotypic integration: from development to deep time. Philos Trans R Soc Lond B Biol Sci 369:20130254. Goswami A, Weisbecker V, Sánchez-Villagra MR. 2009. Developmental modularity and the marsupial-placental dichotomy. J Exp Zool B Mol Dev Evol 312B:186-195. Gould SJ. 1998. On embryos and ancestors. Nat Hist 107:20-22, 58-64. Guillerme T, Weisbecker V. 2019. landvR: Tools for measuring landmark position variation. Available from:https://www.github.com/TGuillerme/landvR/tree/v0.2.Zenodo. https://doi.org/10.5281/zenodo.2620785. Gummer DL, Brigham RM. 1995. Does fluctuating asymmetry reflect the importance of traits in little brown bats (Myotis lucifugus)? Canadian Journal of Zoology 73:990-992. Gunnell GF, Simmons NB, Seiffert ER. 2014. New Myzopodidae (Chiroptera) from the late Paleogene of Egypt: emended family diagnosis and biogeographic origins of Noctilionoidea. PLoS ONE 9:e86712. Gunnell GF, Worsham SR, Seiffert ER, Simons EL. 2009. Vampyravus orientalis Schlosser (Chiroptera) from the Early Oligocene (Rupelian), Fayum, Egypt - body mass, humeral morphology and affinitiesd. Acta Chiropterologica 11:271-278. Gutarra S, Moon BC, Rahman IA, Palmer C, Lautenschlager S, Brimacombe AJ, Benton MJ. 2019. Effects of body plan evolution on the hydrodynamic drag and energy requirements of swimming in ichthyosaurs. Proceedings of the Royal Society B: Biological Sciences 286:20182786. Habib MB, Ruff CB. 2008. The effects of locomotion on the structural characteristics of avian limb bones. Zoological Journal of the Linnean Society 153:601-624.

247

Hafner M, Remsen Jr J, Lanyon S. 1984. Bird Versus Mammal Morphological Diversity. Evolution 38: 1154-1156. Hahsler M, Chelluboina S, Hornik K, Buchta C. 2011. The arules R-Package Ecosystem: Analyzing Interesting Patterns from Large Transaction Data Sets. J. Mach. Learn. Res. 12:2021-2025. Hakansson J, Hedenstrom A, Winter Y, Johansson LC. 2015. The wake of hovering flight in bats. J R Soc Interface 12:20150357. Hall BK. 2000. Evo-devo or devo-evo--does it matter. Evol Dev 2:177-178. Hall BK. 2003. Unlocking the Black Box between Genotype and Phenotype: Cell Condensations as Morphogenetic (modular) Units. Biology and Philosophy 18:219-247. Halley AC. 2016. Prenatal brain-body allometry in mammals. Brain Behavior Evolution 88:14-24. Hallgrímsson B. 1999. Ontogenetic Patterning of Skeletal Fluctuating Asymmetry in Rhesus Macaques and Humans: Evolutionary and Developmental Implications. International Journal of Primatology 20:121-151. Hallgrímsson B, Jamniczky HA, Young NM, Rolian C, Schmidt-Ott U, Marcucio RS. 2012. The generation of variation and the developmental basis for evolutionary novelty. J Exp Zool B Mol Dev Evol 318:501-517. Hallgrímsson B, Miyake T, Wilmore K, Hall BK. 2003. Embryological origins of developmental stability: size, shape and fluctuating asymmetry in prenatal random bred mice. Journal of experimental zoology. Part B, Molecular and developmental evolution 296:40-57. Hallgrimsson B, Willmore K, Dorval C, Cooper DM. 2004. Craniofacial variability and modularity in macaques and mice. J Exp Zool B Mol Dev Evol 302:207-225. Hallgrimsson B, Willmore K, Hall BK. 2002. Canalization, developmental stability, and morphological integration in primate limbs. Am J Phys Anthropol Suppl 35:131-158. Hammer O, Harper DAT, Ryan PD. 2001. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica 4:9pp. Hand SJ, Beck RMD, Archer M, Simmons NB, Gunnell GF, Scofield RP, Tennyson AJD, De Pietri VL, Salisbury SW, Worthy TH. 2018. A new, large-bodied omnivorous bat (Noctilionoidea: Mystacinidae) reveals lost morphological and ecological diversity since the Miocene in New Zealand. Sci Rep 8:235. Hand SJ, Lee DE, Worthy TH, Archer M, Worthy JP, Tennyson AJ, Salisbury SW, Scofield RP, Mildenhall DC, Kennedy EM, et al. 2015. Miocene Fossils Reveal Ancient Roots for New Zealand's Endemic Mystacina (Chiroptera) and Its Rainforest Habitat. PLoS ONE 10:e0128871. Hand SJ, Sigé B. 2018. A new archaic bat (Chiroptera: Archaeonycteridae) from an Early Eocene forest in the Paris Basin. Historical Biology 30:227-236. Hand SJ, Sigé B, Archer M, Gunnell GF, Simmons NB. 2015. A New Early Eocene (Ypresian) Bat from Pourcy, Paris Basin, France, with Comments on Patterns of Diversity in the Earliest Chiropterans. Journal of Mammalian Evolution 22:343-354. Hand SJ, Weisbecker V, Beck RM, Archer M, Godthelp H, Tennyson AJ, Worthy TH. 2009. Bats that walk: a new evolutionary hypothesis for the terrestrial behaviour of New Zealand's endemic mystacinids. BMC Evol Biol 9:169. Hand SJ, Worthy TH, Archer M, Worthy JP, Tennyson AJ, Scofield RP. 2013. Miocene mystacinids (Chiroptera, Noctilionoidea) indicate a long history for endemic bats in New Zealand. Journal of Vertebrate Paleontology 33:1442-1448. 248

Hanot P, Herrel A, Guintard C, Cornette R. 2018. The impact of artificial selection on morphological integration in the appendicular skeleton of domestic horses. J Anat. Hanot P, Herrel A, Guintard C, Cornette R. 2017. Morphological integration in the appendicular skeleton of two domestic taxa: the horse and donkey. Proc Biol Sci 284. Hansen TF. 1997. STABILIZING SELECTION AND THE COMPARATIVE ANALYSIS OF ADAPTATION. Evolution 51:1341-1351. Harmon LJ, Losos JB, Jonathan Davies T, Gillespie RG, Gittleman JL, Bryan Jennings W, Kozak KH, McPeek MA, Moreno-Roark F, Near TJ, et al. 2010. EARLY BURSTS OF BODY SIZE AND SHAPE EVOLUTION ARE RARE IN COMPARATIVE DATA. Evolution 64:2385- 2396. Hautier L, Bennett NC, Viljoen H, Howard L, Milinkovitch MC, Tzika AC, Goswami A, Asher RJ. 2013. Patterns of ossification in southern versus northern placental mammals. Evolution 67:1994-2010. Hautier L, Weisbecker V, Sánchez-Villagra MR, Goswami A, Asher RJ. 2010. Skeletal development in sloths and the evolution of mammalian vertebral patterning. Proceedings of the National Academy of Sciences of the United States of America 107:18903-18908. Hedenstrom A, Johansson LC. 2015. Bat flight: aerodynamics, kinematics and flight morphology. Journal of Experimental Biology 218:653-663. Hedenstrom A, Johansson LC, Spedding GR. 2009. Bird or bat: comparing airframe design and flight performance. Bioinspir Biomim 4:015001. Hedrick BP, Dickson BV, Dumont ER, Pierce SE. 2020. The evolutionary diversity of locomotor innovation in rodents is not linked to proximal limb morphology. Sci Rep 10:717. Hedrick BP, Mutumi GL, Munteanu VD, Sadier A, Davies KTJ, Rossiter SJ, Sears KE, Dávalos LM, Dumont E. 2019. Morphological Diversification under High Integration in a Hyper Diverse Mammal Clade. Journal of Mammalian Evolution 27, 563–575. Henderson DM. 2010. Pterosaur body mass estimates from three-dimensional mathematical slicing. Journal of Vertebrate Paleontology 30:768-785. Hendrikse JL, Parsons TE, Hallgrimsson B. 2007. Evolvability as the proper focus of evolutionary developmental biology. Evol Dev 9:393-401. Hockman D, Mason MK, Jacobs DS, Illing N. 2009. The role of early development in mammalian limb diversification: a descriptive comparison of early limb development between the Natal long-fingered bat (Miniopterus natalensis) and the mouse (Mus musculus). Dev Dyn 238:965-979. Holló G. 2015. A new paradigm for animal symmetry. Interface Focus 5:20150032. Holló G, Novák M. 2012. The manoeuvrability hypothesis to explain the maintenance of bilateral symmetry in animal evolution. Biology Direct 7:22. Hornung RW, Reed LD. 1990. Estimation of Average Concentration in the Presence of Nondetectable Values. Applied Occupational and Environmental Hygiene 5:46-51. Houssaye A, Botton-Divet L. 2018. From land to water: evolutionary changes in long bone microanatomy of otters (Mammalia: Mustelidae). Biological Journal of the Linnean Society 125:240-249. Houssaye A, Waskow K, Hayashi S, Cornette R, Lee AH, Hutchinson JR. 2016. Biomechanical evolution of solid bones in large animals: a microanatomical investigation. Biological Journal of the Linnean Society 117:350-371.

249

Hughes P, Rayner JMV. 1993. The flight of pipistrelle bats Pipistrellus pipistrellus during pregnancy and lactation. Journal of Zoology 230:541-555. Iriarte-Diaz J, Riskin DK, Breuer KS, Swartz SM. 2012. Kinematic plasticity during flight in fruit bats: individual variability in response to loading. PLoS ONE 7:e36665. Iriarte-Diaz J, Riskin DK, Willis DJ, Breuer KS, Swartz SM. 2011. Whole-body kinematics of a fruit bat reveal the influence of wing inertia on body accelerations. J Exp Biol 214:1546-1553. Irie N, Kuratani S. 2014. The developmental hourglass model: a predictor of the basic body plan? Development 141:4649. Janis CM, Napoli JG, Billingham C, Martín-Serra A. 2020. Proximal Humerus Morphology Indicates Divergent Patterns of Locomotion in Extinct Giant Kangaroos. Journal of Mammalian Evolution. Jarrin VP, Flores C, Salcedo J. 2010. Morphological variation in the short-tailed fruit bat (Carollia) in Ecuador, with comments on the practical and philosophical aspects of boundaries among species. Integr Zool 5:226-240. Jin L-r, Lin A-q, Sun K-p, Liu Y, Feng J. 2011. Postnatal development of morphological features and vocalization in the pomona leaf-nosed bat Hipposideros pomona. Acta Theriologica 56:13-22. Johansson LC, Engel S, Kelber A, Heerenbrink MK, Hedenstrom A. 2013. Multiple leading edge vortices of unexpected strength in freely flying hawkmoth. Sci Rep 3:3264. Johansson LC, Hakansson J, Jakobsen L, Hedenstrom A. 2016. Ear-body lift and a novel thrust generating mechanism revealed by the complex wake of brown long-eared bats (Plecotus auritus). Sci Rep 6:24886. Johansson LC, Wolf M, Hedenstrom A. 2010. A quantitative comparison of bird and bat wakes. J R Soc Interface 7:61-66. Jojic V, Blagojevic J, Vujosevic M. 2012. Two-module organization of the mandible in the yellow-necked mouse: a comparison between two different morphometric approaches. J Evol Biol 25:2489-2500. Jones G. 1999. Scaling of echolocation call parameters in bats. Journal of Experimental Biology 202:3359-3367. Jones G, Teeling EC. 2006. The evolution of echolocation in bats. Trends Ecol Evol 21:149-156. Jones KE, Safi K. 2011. Ecology and evolution of mammalian biodiversity Introduction. Philosophical Transactions of the Royal Society B-Biological Sciences 366:2451-2461. Jones KE, Smaers JB, Goswami A. 2015. Impact of the terrestrial-aquatic transition on disparity and rates of evolution in the carnivoran skull. BMC Evolutionary Biology 15:8. Jones PL, Page RA, Ratcliffe JM. 2016. To Scream or to Listen? Prey Detection and Discrimination in Animal-Eating Bats. In: Springer, editor. Bat Bioacoustics. p. 93-116. Juste J, López-González C, Strauss RE. 2001a. Analysis of asymmetries in the African fruit bats Eidolon helvum and Rousettus egyptiacus (Mammalia: Megachiroptera) from the islands of the Gulf of Guinea. I. Variance and size components of bilateral variation. Journal of Evolutionary Biology 14:663-671. Juste J, López-González C, Strauss RE. 2001b. Analysis of asymmetries in the African fruit bats Eidolon helvum and Rousettus egyptiacus (Mammalia: Megachiroptera) from the islands of the Gulf of Guinea. II. Integration and levels of multivariate fluctuating asymmetry across a geographical range. Journal of Evolutionary Biology 14:672-680. 250

Kapusta A, Suh A, Freschotte C. 2017. Dynamics of genome size evolution in birds and mammals. Proceedings of the National Academy of Science 114:E1460–E1469. Kellner James R, Alford Ross A. 2003. The Ontogeny of Fluctuating Asymmetry. The American Naturalist 161:931-947. Kilbourne BM, Hutchinson JR. 2019. Morphological diversification of biomechanical traits: mustelid locomotor specializations and the macroevolution of long bone cross- sectional morphology. BMC Evolutionary Biology 19:37. Kirkpatrick S. 1994. Scale Effects on the Stresses and Safety Factors in the Wing Bones of Birds and Bats. Journal of Experimental Biology 190:195-215. Klingenberg CP. 2015. Analyzing Fluctuating Asymmetry with Geometric Morphometrics: Concepts, Methods, and Applications. Symmetry 7. Klingenberg CP. 2013. Cranial integration and modularity: insights into evolution and development from morphometric data. Hystrix, the Italian Journal of Mammalogy 24:43-58. Klingenberg CP. 2009. Morphometric integration and modularity in configurations of landmarks: tools for evaluating a priori hypotheses. Evol Dev 11:405-421. Klingenberg CP. 2014. Studying morphological integration and modularity at multiple levels: concepts and analysis. Philos Trans R Soc Lond B Biol Sci 369:20130249. Klingenberg CP. 2010. There's something afoot in the evolution of ontogenies. BMC Evol Biol 10:221. Klingenberg CP, McIntyre GS. 1998. Geometric Morphometrics of Developmental Instability: Analyzing Patterns of Fluctuating Asymmetry with Procrustes Methods. Evolution 52:1363-1375. Klingenberg CP, Nijhout HF. 1999. Genetics of fluctuating asymmetry: A developmental model of developmental instability. Evolution 53:358-375. Konow N, Cheney Jorn A, Roberts Thomas J, Waldman JRS, Swartz Sharon M. 2015. Spring or string: does tendon elastic action influence wing muscle mechanics in bat flight? Proceedings of the Royal Society B: Biological Sciences 282:20151832. Koyabu D. 2017. Prenatal postcranial development in two species of sympatric Japanese wood mice (Apodemus argenteus and A-speciosus): a comparison of arboreal versus terrestrial congeners. Journal of Veterinary Medical Science 79:952-956. Koyabu D, Endo H, Mitgutsch C, Suwa G, Catania KC. 2011. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. Heterochrony and developmental modularity of cranial osteogenesis in lipotyphlan mammals. BMC EvoDevo 2: 21. Koyabu D, Son N. 2014. Patterns of postcranial ossification and sequence heterochrony in bats: Life histories and developmental trade‐offs. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 322:607-618. Koyabu D, Werneburg I, Morimoto N, Zollikofer CPE, Forasiepi A, Endo H, Kimura J, Ohdachi SD, Son NT, Sánchez-Villagra MR. 2014. Mammalian skull heterochrony reveals modular evolution and a link between cranial development and brain size. Nat Commun 5:3625. Krause DW, Hoffmann S, Wible JR, Kirk EC, Schultz JA, von Koenigswald W, Groenke JR, Rossie JB, O'Connor PM, Seiffert ER, et al. 2014. First cranial remains of a gondwanatherian mammal reveal remarkable mosaicism. Nature 515:512-517.

251

Kubicka AM, Nowaczewska W, Balzeau A, Piontek J. 2018. Bilateral asymmetry of the humerus in Neandertals, Australian aborigines and medieval humans. American Journal of Physical Anthropology 167:46-60. Kunz TH, Adams RA, Wood WR. 2009. Methods for assessing size at birth and postnatal growth and development in bats. In: Kunz TH, Parsons S, editors. Ecological and behavioral methods for the study of bats. Baltimore: The Johns Hopkins University Press. p. 273–314. Kunz TH, Robson SK. 1995. Postnatal Growth and Development in the Mexican Free- Tailed Bat (Tadarida brasiliensis mexicana): Birth Size, Growth Rates, and Age Estimation. Journal of mammalogy 76:769-783. Kwong S, Kothary S, Poncinelli LL. 2014. Skeletal development of the proximal humerus in the pediatric population: MRI features. AJR Am J Roentgenol 202:418-425. Law CJ. 2019. Evolutionary shifts in extant mustelid (Mustelidae: Carnivora) cranial shape, body size and body shape coincide with the Mid-Miocene Climate Transition. Biology Letters 15:20190155. Lawler RR. 2008. Morphological integration and natural selection in the postcranium of wild verreaux's sifaka (Propithecus verreauxi verreauxi). Am J Phys Anthropol 136:204- 213. Leamy LJ, Klingenberg CP. 2005. The Genetics and Evolution of Fluctuating Asymmetry. Annual Review of Ecology, Evolution, and Systematics 36:1-21. Lee AH, Simons E. 2015. Wing bone laminarity is not an adaptation for torsional resistance in bats. PeerJ 3: e823. Lei M, Dong D. 2016. Phylogenomic analyses of bat subordinal relationships based on transcriptome data. Sci Rep 6:27726. Leiser-Miller LB, Santana SE. 2020. Morphological Diversity in the Sensory System of Phyllostomid Bats: Implications for Acoustic and Dietary Ecology. Funct Ecol. 2020; 34: 1416– 1427 Lentink D, Dickson WB, van Leeuwen JL, Dickinson MH. 2009. Leading-Edge Vortices Elevate Lift of Autorotating Plant Seeds. Science 324:1438-1440. Lin A, Jin L, Liu Y, Sun K, Feng JX editors.; 2010. Lindhe Norberg UM, Winter Y. 2006. Wing beat kinematics of a nectar-feeding bat, Glossophaga soricina, flying at different flight speeds and Strouhal numbers. J Exp Biol 209:3887-3897. López-Aguirre C, Hand SJ, Koyabu D, Son NT, Wilson LAB. 2019a. Postcranial heterochrony, modularity, integration and disparity in the prenatal ossification in bats (Chiroptera). BMC Evolutionary Biology 19:75. López-Aguirre C, Hand SJ, Koyabu D, Son NT, Wilson LAB. 2019b. Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera). Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 332:36-49. López-Aguirre C, Hand SJ, Laffan SW, Archer M. 2018. Phylogenetic diversity, types of endemism and the evolutionary history of New World bats. Ecography 41:1955-1966. López-Aguirre C, Pérez-Torres J. 2015. Asimetría cráneo-mandibular de Artibeus lituratus (Chiroptera, Phyllostomidae) en Colombia. Universitas Scientiarum 20:141- 152.

252

López-Aguirre C, Perez-Torres J, Wilson LA. 2015. Cranial and mandibular shape variation in the genus Carollia (Mammalia: Chiroptera) from Colombia: biogeographic patterns and morphological modularity. PeerJ 3:e1197. López-Aguirre C, Wilson LA, Koyabu D, Tan Tu V, Hand SJ. 2019. Decoupled morphological and biomechanical evolution and diversification of the wing in bats. EcoEvoRxiv:31. Louzada NSV, Nogueira MR, Pessôa LM. 2019. Comparative morphology and scaling of the femur in yangochiropteran bats. Journal of Anatomy 235:124-150. Love AC, Raff RA. 2003. Knowing your ancestors: themes in the history of evo-devo. Evol Dev 5:327-330. Lungmus JK, Angielczyk KD. 2019. Antiquity of forelimb ecomorphological diversity in the mammalian stem lineage (Synapsida). Proc Natl Acad Sci U S A 116:6903-6907. Luo B, Santana SE, Pang Y, Wang M, Xiao Y, Feng J. 2019. Wing morphology predicts geographic range size in vespertilionid bats. Sci Rep 9:4526. Luo ZX, Yuan CX, Meng QJ, Ji Q. 2011. A Jurassic eutherian mammal and divergence of marsupials and placentals. Nature 476:442-445. Lüpold S, McElligott AG, Hosken DJ. 2004. Bat genitalia: allometry, variation and good genes. Biological Journal of the Linnean Society 83:497-507. MacAyeal LC, Riskin DK, Swartz SM, Breuer KS. 2011. Climbing flight performance and load carrying in lesser dog-faced fruit bats (Cynopterus brachyotis). J Exp Biol 214:786- 793. Machado FA, Zahn TMG, Marroig G. 2018. Evolution of morphological integration in the skull of Carnivora (Mammalia): Changes in Canidae lead to increased evolutionary potential of facial traits. Evolution. Machado JP, Johnson WE, Gilbert MT, Zhang G, Jarvis ED, O'Brien SJ, Antunes A. 2016. Bone-associated gene evolution and the origin of flight in birds. BMC Genomics 17:371. Macintosh AA, Davies TG, Ryan TM, Shaw CN, Stock JT. 2013. Periosteal Versus True Cross-Sectional Geometry: A Comparison Along Humeral, Femoral, and Tibial Diaphyses. American Journal of Physical Anthropology 150:442-452. Macintosh AA, Pinhasi R, Stock JT. 2014. Lower limb skeletal biomechanics track long- term decline in mobility across similar to 6150 years of agriculture in Central Europe. Journal of Archaeological Science 52:376-390. Macintosh AA, Pinhasi R, Stock JT. 2017. Prehistoric women's manual labor exceeded that of athletes through the first 5500 years of farming in Central Europe. Science Advances 3: eaao3893. Maestri R, Monteiro LR, Fornel R, Upham NS, Patterson BD, de Freitas TRO. 2017. The ecology of a continental evolutionary radiation: Is the radiation of sigmodontine rodents adaptive? Evolution 71:610-632. Makedonska J, Wright BW, Strait DS. 2012. The effect of dietary adaption on cranial morphological integration in capuchins (order Primates, genus Cebus). PLoS ONE 7:e40398. Mallet C, Billet G, Houssaye A, Cornette R. 2020. A first glimpse at the influence of body mass in the morphological integration of the limb long bones: an investigation in modern rhinoceroses. Journal of Anatomy n/a.

253

Marcé-Nogué J, Püschel TA, Kaiser TM. 2017. A biomechanical approach to understand the ecomorphological relationship between primate mandibles and diet. Scientific Reports 7:8364. Marchetti K, Price T, Richman A. 1995. Correlates of Wing Morphology with Foraging Behaviour and Migration Distance in the Genus Phylloscopus. Journal of Avian Biology 26:177-181. Marchi D, Ruff CB, Capobianco A, Rafferty KL, Habib MB, Patel BA. 2016. The locomotion of Babakotia radofilai inferred from epiphyseal and diaphyseal morphology of the humerus and femur. J Morphol 277:1199-1218. Marelli CA, Simons EL. 2014. Microstructure and cross-sectional shape of limb bones in Great Horned Owls and Red-tailed Hawks: how do these features relate to differences in flight and hunting behavior? PLoS ONE 9:e106094. Marinello MM, Bernard E. 2014. Wing morphology of Neotropical bats: a quantitative and qualitative analysis with implications for habitat use. Canadian Journal of Zoology 92:141-147. Marroig G, Cheverud JM. 2005b. Size as a line of least evolutionary resistance: Diet and adaptive morphological radiation in new world monkeys. Evolution 59:1128-1142. Marroig G, Shirai LT, Porto A, de Oliveira FB, De Conto V. 2009. The Evolution of Modularity in the Mammalian Skull II: Evolutionary Consequences. Evolutionary Biology 36:136-148. Marroig G, Vivo M, Cheverud JM. 2003. Cranial evolution in sakis (Pithecia, Platyrrhini) II: evolutionary processes and morphological integration. Journal of Evolutionary Biology 17:144-155. Martin-Serra A, Figueirido B, Palmqvist P. 2017. Non-decoupled morphological evolution of the fore- and hindlimb of sabretooth predators. J Anat. Martin-Serra A, Figueirido B, Palmqvist P. 2014. A three-dimensional analysis of morphological evolution and locomotor performance of the carnivoran forelimb. PLoS ONE 9:e85574. Martín-Serra A, Figueirido B, Palmqvist P. 2014. A Three-Dimensional Analysis of Morphological Evolution and Locomotor Performance of the Carnivoran Forelimb. PLoS ONE 9:e85574. Maxwell EE, Harrison LB. 2009. Methods for the analysis of developmental sequence data. Evol Dev 11:109-119. Maxwell EE, Harrison LB, Larsson HCE. 2010. Assessing the phylogenetic utility of sequence heterochrony: evolution of avian ossification sequences as a case study. Zoology 113:57-66. McElroy EJ, Hickey KL, Reilly SM. 2008. The correlated evolution of biomechanics, gait and foraging mode in lizards. Journal of Experimental Biology 211:1029-1040. McGowan AJ, Dyke GJ. 2007. A morphospace-based test for competitive exclusion among flying vertebrates: did birds, bats and pterosaurs get in each other's space? J Evol Biol 20:1230-1236. McInroe B, Astley HC, Gong C, Kawano SM, Schiebel PE, Rieser JM, Choset H, Blob RW, Goldman DI. 2016. Tail use improves performance on soft substrates in models of early vertebrate land locomotors. Science 353:154-158. McNamara KJ. 2012. Heterochrony: the Evolution of Development. Evolution: Education and Outreach 5:203-218.

254

Medina I, Cooke GM, Ord TJ. 2018. Walk, swim or fly? Locomotor mode predicts genetic differentiation in vertebrates. Ecology Letters 21:638-645. Meers MB. 2002. Cross-sectional geometric properties of the crocodylian humerus: an exception to Wolff's Law? Journal of Zoology 258:405-418. Milinkovitch MC, Tzika A. 2007. Escaping the mouse trap: The selection of new evo- devo model species. Journal of Experimental Zoology Part B-Molecular and Developmental Evolution 308b:337-346. Monteiro LR, Mellado B, Nogueira MR, de Morais-Jr MM. 2019. Individual asymmetry as a predictor of fitness in the bat Carollia perspicillata. Journal of Evolutionary Biology 32:1207-1229. Monteiro LR, Nogueira MR. 2011. Evolutionary patterns and processes in the radiation of phyllostomid bats. BMC Evol Biol 11:137. Morales AE, Ruedi M, Field K, Carstens BC. 2019. Diversification rates have no effect on the convergent evolution of foraging strategies in the most speciose genus of bats, Myotis. Evolution 73:2263-2280. Mosimann JE, James FC. 1979. New Statistical Methods for Allometry with Application to Florida Red-Winged Blackbirds. Evolution 33:444-459. Moyers Arévalo RL, Amador LI, Almeida FC, Giannini NP. 2018. Evolution of Body Mass in Bats: Insights from a Large Supermatrix Phylogeny. Journal of Mammalian Evolution 27: 123-138. Muijres FT, Christoffer Johansson L, Winter Y, Hedenstrom A. 2014. Leading edge vortices in lesser long-nosed bats occurring at slow but not fast flight speeds. Bioinspir Biomim 9:025006. Muijres FT, Johansson LC, Barfield R, Wolf M, Spedding GR, Hedenstrom A. 2008. Leading-Edge Vortex Improves Lift in Slow-Flying Bats. Science 319: 1250-1253. Muijres FT, Johansson LC, Bowlin MS, Winter Y, Hedenstrom A. 2012. Comparing aerodynamic efficiency in birds and bats suggests better flight performance in birds. PLOS ONE 7:e37335. Muijres FT, Johansson LC, Hedenstrom A. 2012. Leading edge vortex in a slow-flying passerine. Biol Lett 8:554-557. Muijres FT, Johansson LC, Winter Y, Hedenstrom A. 2011. Comparative aerodynamic performance of flapping flight in two bat species using time-resolved wake visualization. J R Soc Interface 8:1418-1428. Muñoz MM, Hu Y, Anderson PSL, Patek SN. 2018. Strong biomechanical relationships bias the tempo and mode of morphological evolution. eLife 7:e37621. Neubauer S, Gunz P, Scott NA, Hublin J-J, Mitteroecker P. 2020. Evolution of brain lateralization: A shared hominid pattern of endocranial asymmetry is much more variable in humans than in great apes. Science Advances 6:eaax9935. Nogueira MR, Peracchi AL, Monteiro LR. 2009. Morphological correlates of bite force and diet in the skull and mandible of phyllostomid bats. Functional Ecology 23:715- 723. Nojiri T, Werneburg I, Son NT, Tu VT, Sasaki T, Maekawa Y, Koyabu D. 2018. Prenatal cranial bone development of Thomas's horseshoe bat (Rhinolophus thomasi): with special reference to petrosal morphology. J Morphol 279: 809-827. Norberg RA, Rayner JMV. 1987. Ecological Morphology and Flight in Bats (Mammalia; Chiroptera): Wing Adaptations, Flight Performance, Foraging Strategy and Echolocation. Philos Trans R Soc Lond B Biol Sci 316:335-427. 255

Norberg UM. 1990. Vertebrate Flight, Mechanics, Physiology, Morphology, Ecology and Evolution: Springer-Verlag. Norberg UM, Norberg RA. 2012. Scaling of wingbeat frequency with body mass in bats and limits to maximum bat size. J Exp Biol 215:711-722. Organ CL, Cooper LN, Hieronymus TL. 2015. Macroevolutionary developmental biology: Embryos, fossils, and phylogenies. Dev Dyn 244:1184-1192. Organ CL, Shedlock AM. 2009. Palaeogenomics of pterosaurs and the evolution of small genome size in flying vertebrates. Biol Lett 5:47-50. Oyston JW, Hughes M, Wagner PJ. 2015. What limits the morphological disparity of clades? Interface Focus 5: 20150042. Palmer AR. 1994. Fluctuating asymmetry analyses: a primer. In: Markow TA, editor. Developmental Instability: Its Origins and Evolutionary Implications: Proceedings of the International Conference on Developmental Instability: Its Origins and Evolutionary Implications, Tempe, Arizona, 14–15 June 1993. Dordrecht: Springer Netherlands. p. 335-364. Panyutina A, Korzun LP, Kuznetsov AN. 2015. Flight of Mammals: From Terrestrial Limbs to Wings: Springer, New York. Papadimitriou HM, Swartz SM, Kunz TH. 1996. Ontogenetic and anatomic variation in mineralization of the wing skeleton of the Mexican free-tailed bat, Tadarida brasiliensis. Journal of Zoology 240:411-426. Parsons KJ, Marquez E, Albertson RC. 2012. Constraint and opportunity: the genetic basis and evolution of modularity in the cichlid mandible. Am Nat 179:64-78. Patel BA, Ruff CB, Simons EL, Organ JM. 2013. Humeral cross-sectional shape in suspensory primates and sloths. Anat Rec (Hoboken) 296:545-556. Pavlicev M, Cheverud JM, Wagner GP. 2009. Measuring Morphological Integration Using Eigenvalue Variance. Evolutionary Biology 36:157-170. Pecoits E, Konhauser KO, Aubet NR, Heaman LM, Veroslavsky G, Stern RA, Gingras MK. 2012. Bilaterian Burrows and Grazing Behavior at &gt;585 Million Years Ago. Science 336:1693. Pennell MW, Eastman JM, Slater GJ, Brown JW, Uyeda JC, FitzJohn RG, Alfaro ME, Harmon LJ. 2014. geiger v2.0: an expanded suite of methods for fitting macroevolutionary models to phylogenetic trees. Bioinformatics 30:2216-2218. Perchalski B, Placke A, Sukhdeo SM, Shaw CN, Gosman JH, Raichlen DA, Ryan TM. 2018. Asymmetry in the Cortical and Trabecular Bone of the Human Humerus During Development. Anat Rec (Hoboken) 301:1012-1025. Petit F, Sears KE, Ahituv N. 2017. Limb development: a paradigm of gene regulation. Nature Reviews Genetics 18: 245-258. Pieretti J, Gehrke AR, Schneider I, Adachi N, Nakamura T, Shubin NH. 2015. Organogenesis in deep time: A problem in genomics, development, and paleontology. Proc Natl Acad Sci U S A 112:4871-4876. Pimiento C, Cantalapiedra JL, Shimada K, Field DJ, Smaers JB. 2019. Evolutionary pathways toward gigantism in sharks and rays. Evolution 73:588-599. Piras P, Salvi D, Ferrara G, Maiorino L, Delfino M, Pedde L, Kotsakis T. 2011. The role of post-natal ontogeny in the evolution of phenotypic diversity in Podarcis lizards. Journal of Evolutionary Biology 24:2705-2720. Polanski JM. 2011. Morphological Integration of the Modern Human Mandible during Ontogeny. Int J Evol Biol 2011:545879. 256

Ponssa ML, Candioti MFV. 2012. Patterns of skull development in anurans: size and shape relationship during postmetamorphic cranial ontogeny in five species of the Leptodactylus fuscus Group (Anura: Leptodactylidae). Zoomorphology 131:349-362. Porto A, de Oliveira FB, Shirai LT, De Conto V, Marroig G. 2008. The Evolution of Modularity in the Mammalian Skull I: Morphological Integration Patterns and Magnitudes. Evolutionary Biology 36:118-135. Porto A, Shirai LT, de Oliveira FB, Marroig G. 2013. Size Variation, Growth Strategies, and the Evolution of Modularity in the Mammalian Skull. Evolution 67:3305-3322. Pratt IV, Johnston JD, Walker E, Cooper DML. 2018. Interpreting the three-dimensional orientation of vascular canals and cross-sectional geometry of cortical bone in birds and bats. Journal of Anatomy 232: 931-942. Prentice KC, Ruta M, Benton MJ. 2011. Evolution of morphological disparity in pterosaurs. Journal of Systematic Palaeontology 9:337-353. Prufrock KA, Boyer DM, Silcox MT. 2016. The first major primate extinction: An evaluation of paleoecological dynamics of North American stem primates using a homology free measure of tooth shape. Am J Phys Anthropol 159:683-697. Raff RA. 1996. The shape of life: genes, development, and the evolution of animal form: University of Chicago Press. Ramirez-Chaves HE, Wroe SW, Selwood L, Hinds LA, Leigh C, Koyabu D, Kardjilov N, Weisbecker V. 2016. Mammalian development does not recapitulate suspected key transformations in the evolutionary detachment of the mammalian middle ear. Proc Biol Sci 283. Randau M, Goswami A. 2018. Shape Covariation (or the Lack Thereof) Between Vertebrae and Other Skeletal Traits in Felids: The Whole is Not Always Greater than the Sum of Parts. Evolutionary Biology 45: 196-210. Randau M, Goswami A. 2017. Unravelling intravertebral integration, modularity and disparity in Felidae (Mammalia). Evol Dev 19:85-95. Rayner JM. 1988. The evolution of vertebrate flight. Biological Journal of the Linnean Society 34:269-287. Revell LJ. 2012. phytools: an R package for phylogenetic comparative biology (and other things). Methods in Ecology and Evolution 3:217-223. Reyes-Amaya N, Jerez A, Flores D. 2017. Morphology and Postnatal Development of Lower Hindlimbs in Desmodus rotundus (Chiroptera: Phyllostomidae): A Comparative Study. Anat Rec (Hoboken) 300:2150-2165. Riskin DK, Bergou A, Breuer KS, Swartz SM. 2012. Upstroke wing flexion and the inertial cost of bat flight. Proc Biol Sci 279:2945-2950. Riskin DK, Bertram JEA, Hermanson JW. 2005. Testing the hindlimb-strength hypothesis: non-aerial locomotion by Chiroptera is not constrained by the dimensions of the femur or tibia. Journal of Experimental Biology 208:1309-1319. Riskin DK, Fenton MB. 2001. Sticking ability in Spix's disk-winged bat, Thyroptera tricolor (Microchiroptera : Thyropteridae). Canadian Journal of Zoology 79:2261-2267. Riskin DK, Iriarte-Diaz J, Middleton KM, Breuer KS, Swartz SM. 2010. The effect of body size on the wing movements of pteropodid bats, with insights into thrust and lift production. J Exp Biol 213:4110-4122. Riskin DK, Parsons S, Schutt WA, Jr., Carter GG, Hermanson JW. 2006. Terrestrial locomotion of the New Zealand short-tailed bat Mystacina tuberculata and the common vampire bat Desmodus rotundus. J Exp Biol 209:1725-1736. 257

Riskin DK, Racey PA. 2010. How do sucker-footed bats hold on, and why do they roost head-up? Biological Journal of the Linnean Society 99:233-240. Riskin DK, Willis DJ, Iriarte-Diaz J, Hedrick TL, Kostandov M, Chen J, Laidlaw DH, Breuer KS, Swartz SM. 2008. Quantifying the complexity of bat wing kinematics. J Theor Biol 254:604-615. Rivera G, Neely CMD. 2020. Patterns of fluctuating asymmetry in the limbs of freshwater turtles: Are more functionally important limbs more symmetrical? Evolution 74:660-670. Robaina Z, Denis D, Sánchez-Lozada M. 2017. ¿Son regulados los niveles de asimetría fluctuante según la función de las estructuras morfológicas? Estudio de caso en Phyllonycteris poeyi (chiroptera: phyllostomidae)/Are levels of fluctuating asymmetry regulated according to the function of. Poeyana; Núm. 505. Rojas D, Vale Á, Ferrero V, Navarro L. 2012. The role of frugivory in the diversification of bats in the Neotropics. Journal of Biogeography 39:1948-1960. Ross D, Marcot JD, Betteridge KJ, Nascone-Yoder N, Bailey CS, Sears KE. 2013. Constraints on Mammalian forelimb development: insights from developmental disparity. Evolution 67:3645-3652. Rossoni DM, Assis APA, Giannini NP, Marroig G. 2017. Intense natural selection preceded the invasion of new adaptive zones during the radiation of New World leaf- nosed bats. Scientific Reports 7: 11076. Rossoni DM, Costa BMA, Giannini NP, Marroig G. 2019. A multiple peak adaptive landscape based on feeding strategies and roosting ecology shaped the evolution of cranial covariance structure and morphological differentiation in phyllostomid bats. Evolution 73:961-981. Ruff C. 1987. Sexual dimorphism in human lower limb bone structure: relationship to subsistence strategy and sexual division of labor. Journal of Human Evolution 16:391- 416. Ruff C, Holt B, Trinkaus E. 2006. Who's afraid of the big bad Wolff?: "Wolff's law" and bone functional adaptation. Am J Phys Anthropol 129:484-498. Ruth EB. 1932. A study of the development of the mammalian pelvis. Anat Rec (Hoboken) 53:207-225. Sallan LC, Friedman M. 2012. Heads or tails: staged diversification in vertebrate evolutionary radiations. Proceedings of the Royal Society B: Biological Sciences 279:2025-2032. Sánchez-Villagra MR. 2002. Comparative patterns of postcranial ontogeny in therian mammals: an analysis of relative timing of ossification events. J Exp Zool 294:264-273. Sánchez-Villagra MR, Goswami A, Weisbecker V, Mock O, Kuratani S. 2008. Conserved relative timing of cranial ossification patterns in early mammalian evolution. Evol Dev 10:519-530. Sánchez-Villagra MR, Goswami A, Weisbecker V, Mock O, Kuratani S. 2008. Conserved relative timing of cranial ossification patterns in early mammalian evolution. Evol Dev 10:519-530. Sánchez-Villagra MR, Segura V, Geiger M, Heck L, Veitschegger K, Flores D. 2017. On the lack of a universal pattern associated with mammalian domestication: differences in skull growth trajectories across phylogeny. R Soc Open Sci 4:170876.

258

Santana SE, Cheung E. 2016. Go big or go fish: morphological specializations in carnivorous bats. Proceedings of the Royal Society B: Biological Sciences 283:20160615. Santana SE, Dumont ER, Davis JL. 2010. Mechanics of bite force production and its relationship to diet in bats. Functional Ecology 24:776-784. Santana SE, Grosse IR, Dumont ER. 2012a. Dietary hardness, loading behavior, and the evolution of skull form in bats. Evolution 66:2587-2598. Santana SE, Lofgren SE. 2013. Does nasal echolocation influence the modularity of the mammal skull? J Evol Biol 26:2520-2526. Santana SE, Strait S, Dumont ER. 2011. The better to eat you with: functional correlates of tooth structure in bats. The better to eat you with: functional correlates of tooth structure in bats. Sato K, Sakamoto KQ, Watanuki Y, Takahashi A, Katsumata N, Bost CA, Weimerskirch H. 2009. Scaling of soaring seabirds and implications for flight abilities of giant pterosaurs. PLOS ONE 4:e5400. Schliemann H, Goodman SM. 2011. A new study on the structure and function of the adhesive organs of the Old World sucker-footed bat (Myzopoda: Myzopodidae) of Madagascar. Verh. Naturwiss. Ver. Hamburg. Schlosser-Sturm E, Schliemann H. 1995. Morphology and function of the shoulder joint of bats (Mammalia: Chiroptera)1. Journal of Zoological Systematics and Evolutionary Research 33:88-98. Schmieder DA, Benitez HA, Borissov IM, Fruciano C. 2015. Bat Species Comparisons Based on External Morphology: A Test of Traditional versus Geometric Morphometric Approaches. PLOS ONE 10:e0127043. Schulz M. 2000. Diet and foraging behavior of the golden-tipped bat, Kerivoula papuensis: A spider specialist? Journal of mammalogy 81:948-957. Sears KE. 2014. Differences in Growth Generate the Diverse Palate Shapes of New World Leaf-Nosed Bats (Order Chiroptera, Family Phyllostomidae). Evolutionary Biology 41: 12-21. Sears KE, Behringer RR, Rasweiler JJ, Niswander LA. 2006. Development of bat flight: Morphologic and molecular evolution of bat wing digits. Proceedings of the National Academy of Sciences of the United States of America 103:6581-6586. Sears KE, Behringer RR, Rasweiler JJt, Niswander L. 2007. The Evolutionary and Developmental Basis of Parallel Reduction in Mammalian Zeugopod Elements. Am Nat 169:105-117. Sherratt E, Serb JM, Adams DC. 2017. Rates of morphological evolution, asymmetry and morphological integration of shell shape in scallops. BMC Evolutionary Biology 17:248. Shi JJ, Rabosky DL. 2015. Speciation dynamics during the global radiation of extant bats. Evolution 69:1528-1545. Shubin NH, Tabin CJ, Carroll S. 1997. Fossils, genes and the evolution of animal limbs. Nature 388:639–648. Simmons NB, Seymour KL, Habersetzer J, Gunnell GF. 2008. Primitive Early Eocene bat from Wyoming and the evolution of flight and echolocation. Nature 451:818-821. Simons EL. 2010. Forelimb skeletal morphology and flight mode evolution in pelecaniform birds. Zoology (Jena) 113:39-46.

259

Simons EL, Hieronymus TL, O'Connor PM. 2011. Cross sectional geometry of the forelimb skeleton and flight mode in pelecaniform birds. J Morphol 272:958-971. Simons EL, O'Connor P M. 2012. Bone laminarity in the avian forelimb skeleton and its relationship to flight mode: testing functional interpretations. Anat Rec (Hoboken) 295:386-396. Slater GJ, Friscia AR. 2019. Hierarchy in adaptive radiation: A case study using the Carnivora (Mammalia). Evolution 73:524-539. Slater GJ, Price SA, Santini F, Alfaro ME. 2010. Diversity versus disparity and the radiation of modern cetaceans. Proceedings of the Royal Society B: Biological Sciences 277:3097-3104. Smith JD. 1977. Comments on Flight and the Evolution of Bats. In: Hecht MK, Goody PC, Hecht BM, editors. Major Patterns in Vertebrate Evolution. Boston, MA: Springer US. p. 427-437. Smith KK. 2001. Heterochrony revisited: the evolution of developmental sequences. Biological Journal of the Linnean Society 73:169-186. Smith KK. 2002. Sequence heterochrony and the evolution of development. J Morphol 252:82-97. Smith KK. 2003. Time's arrow: heterochrony and the evolution of development. International Journal of Developmental Biology 47:613-621. Spielmann M, Brancati F, Krawitz PM, Robinson PN, Ibrahim DM, Franke M, Hecht J, Lohan S, Dathe K, Nardone AM, et al. 2012. Homeotic Arm-to-Leg Transformation Associated with Genomic Rearrangements at the PITX1 Locus. American Journal of Human Genetics 91:629-635. Spoor CF, Zonneveld FW, Macho GA. 1993. Linear Measurements of Cortical Bone and Dental Enamel by Computed-Tomography - Applications and Problems. American Journal of Physical Anthropology 91:469-484. Springer MS. 2013. Phylogenetics: bats united, microbats divided. Curr Biol 23:R999- 1001. Stanchak KE, Arbour JH, Santana SE. 2019. Anatomical diversification of a skeletal novelty in bat feet. Evolution 73: 1591-1603. Stanchak KE, Santana SE. 2018. Assessment of the Hindlimb Membrane Musculature of Bats: Implications for Active Control of the Calcar. The Anatomical Record 301:441- 448. Stange M, Aguirre-Fernandez G, Salzburger W, Sánchez-Villagra MR. 2018. Study of morphological variation of northern Neotropical Ariidae reveals conservatism despite macrohabitat transitions. BMC Evol Biol 18:38. Steele J, Mays S. 1995. Handedness and directional asymmetry in the long bones of the human upper limb. International Journal of Osteoarchaeology 5:39-49. Stroud JT, Losos JB. 2016. Ecological Opportunity and Adaptive Radiation. Annual Review of Ecology, Evolution, and Systematics 47:507-532. Swartz S, Iriarte-Diaz J, Riskin D, Tian X, Song A, Breuer K. 2007. Wing Structure and the Aerodynamic Basis of Flight in Bats. Swartz SM, Bennett MB, Carrier DR. 1992. Wing bone stresses in free flying bats and the evolution of skeletal design for flight. Nature 359: 726-729. Swartz SM, Konow N. 2015. Advances in the study of bat flight: the wing and the wind. Canadian Journal of Zoology 93: 977-990.

260

Swartz SM, Middleton KM. 2008. Biomechanics of the bat limb skeleton: Scaling, material properties and mechanics. Cells Tissues Organs 187:59-84. Teeling EC, Madsen O, Van Den Bussche RA, de Jong WW, Stanhope M. 2002. Microbat paraphyly and the convergent evolution of a key innovation in Old World rhinolophoid microbats. Nature 99:1431-1436. Teeling EC, Scally M, Kao D, Romagnoli M, Springer MS, Stanhope M. 2000. Molecular evidence regarding the origin of echolocation and flight in bats. Nature 403:188–192. Teeling EC, Springer MS, Madsen O, Bates P, O’Brien J, Murphy WJ. 2005. A Molecular Phylogeny for Bats Illuminates Biogeography and the Fossil Record. Science 307:581- 584. Thean T, Kardjilov N, Asher RJ. 2017. Inner ear development in cetaceans. J Anat 230:249-261. Thiagavel J, Cechetto C, Santana SE, Jakobsen L, Warrant EJ, Ratcliffe JM. 2018. Auditory opportunity and visual constraint enabled the evolution of echolocation in bats. Nat Commun 9:98. Tocts AMS, Johnson DW, Carter AJR. 2016. Strong nonlinear selection against fluctuating asymmetry in wild populations of a marine fish. Evolution 70:2899-2908. Tokita M. 2015. How the pterosaur got its wings. Biol Rev Camb Philos Soc 90:1163- 1178. Tokita M, Abe T, Suzuki K. 2012. The developmental basis of bat wing muscle. Nature Communications 3: 1302. Tsagkogeorga G, Parker J, Stupka E, Cotton JA, Rossiter SJ. 2013. Phylogenomic analyses elucidate the evolutionary relationships of bats. Curr Biol 23:2262-2267. Ueti A, Pompeu PS, Ferreira RL. 2015. Asymmetry compensation in a small vampire bat population in a cave: a case study in Brazil. Subterranean Biology 15. Uyeda JC, Caetano DS, Pennell MW. 2015. Comparative Analysis of Principal Components Can be Misleading. Systematic Biology 64:677-689. Vaughan TA. 1959. Functional morphology of three bats : Eumops, Myotis, Macrotus. Lawrence: University of Kansas. Verde Arregoitia LD, Fisher DO, Schweizer M. 2017. Morphology captures diet and locomotor types in rodents. R Soc Open Sci 4:160957. Veselka N, McErlain DD, Holdsworth DW, Eger JL, Chhem RK, Mason MJ, Brain KL, Faure PA, Fenton MB. 2010. A bony connection signals laryngeal echolocation in bats. Nature 463:939-942. Vidal-García M, Scott Keogh J. 2017. Phylogenetic conservatism in skulls and evolutionary lability in limbs – morphological evolution across an ancient frog radiation is shaped by diet, locomotion and burrowing. BMC Evolutionary Biology 17:165. Voeten D, Cubo J, de Margerie E, Roper M, Beyrand V, Bures S, Tafforeau P, Sánchez S. 2018. Wing bone geometry reveals active flight in Archaeopteryx. Nat Commun 9:923. Voigt CC, Borrisov IM, Voigt-Heucke SL. 2012. Terrestrial locomotion imposes high metabolic requirements on bats. J Exp Biol 215:4340-4344. Voigt CC, Heckel G, Mayer F. 2005. Sexual selection favours small and symmetric males in the polygynous greater sac-winged bat Saccopteryx bilineata (Emballonuridae, Chiroptera). Behavioral Ecology and Sociobiology 57:457-464. Von Busse R, Hedenstrom A, Winter Y, Johansson LC. 2012. Kinematics and wing shape across flight speed in the bat, Leptonycteris yerbabuenae. Biol Open 1:1226-1238.

261

Wagner GP, Altenberg L. 1996. Perspective: Complex adaptations and the evolution of evolvability. Evolution 50:967-976. Wagner GP, Pavlicev M, Cheverud JM. 2007. The road to modularity. Nature Reviews Genetics 8:921-931. Wallin J, Wilting J, Koseki H, Fritsch R, Christ B, Balling R. 1994. The role of Pax-1 in axial skeleton development. Development 120:1109-1121. Wang X, McGowan AJ, Dyke GJ. 2011. Avian wing proportions and flight styles: first step towards predicting the flight modes of mesozoic birds. PLOS ONE 6:e28672. Wang XL, Kellner AWA, Jiang SX, Cheng X, Wang Q, Ma YX, Paidoula Y, Rodrigues T, Chen H, Sayao JM, et al. 2017. Egg accumulation with 3D embryos provides insight into the life history of a pterosaur. Science 358:1197-1201. Wang Z, Han N, Racey PA, Ru BH, He GM. 2010. A comparative study of prenatal development in Miniopterus schreibersii fuliginosus, Hipposideros armiger and H. pratti. Bmc Developmental Biology 10: 10. Wang Z, Zhu T, Xue H, Fang N, Zhang J, Zhang L, Pang J, Teeling EC, Zhang S. 2017. Prenatal development supports a single origin of laryngeal echolocation in bats. Nature Ecology & Evolution 1: 0021. Warrick DR, Tobalske BW, Powers DR. 2009. Lift production in the hovering hummingbird. Proc Biol Sci 276:3747-3752. Watanabe A. 2018. How many landmarks are enough to characterize shape and size variation? PLoS ONE 13:e0198341. Watts P, Mitchell E, Swartz S. 2001. A computational model for estimating the mechanics of horizontal flapping flight in bats: Model description and validation. Journal of Experimental Biology 204:2873-2898. Weisbecker V, Goswami A, Wroe S, Sánchez-Villagra MR. 2008. Ossification heterochrony in the therian postcranial skeleton and the marsupial-placental dichotomy. Evolution 62:2027-2041. Weller TJ, Castle KT, Liechti F, Hein CD, Schirmacher MR, Cryan PM. 2016. First Direct Evidence of Long-distance Seasonal Movements and Hibernation in a Migratory Bat. Sci Rep 6:34585. Werneburg I, Geiger M. 2017. Ontogeny of domestic dogs and the developmental foundations of carnivoran domestication. Journal of Mammalian Evolution 24:323-343. Werneburg I, Laurin M, Koyabu D, Sánchez-Villagra MR. 2016. Evolution of organogenesis and the origin of altriciality in mammals. Evolution & Development 18:229-244. Werneburg I, Tzika AC, Hautier L, Asher RJ, Milinkovitch MC, Sánchez‐Villagra MR. 2013. Development and embryonic staging in non‐model organisms: the case of an afrotherian mammal. Journal of Anatomy 222:2-18. Wilson L, Humphrey LT. 2015. A Virtual geometric morphometric approach to the quantification of long bone bilateral asymmetry and cross‐sectional shape. American Journal of Physical Anthropology 4: 541-556. Wilson LA. 2013a. Allometric disparity in rodent evolution. Ecol Evol 3:971-984. Wilson LA. 2011a. Comparison of prenatal and postnatal ontogeny: cranial allometry in the African striped mouse (Rhabdomys pumilio). Journal of mammalogy 92:407-420. Wilson LA, Ives R, Cardoso HF, Humphrey LT. 2015. Shape, size, and maturity trajectories of the human ilium. Am J Phys Anthropol 156:19-34.

262

Wilson LA, Sánchez-Villagra MR. 2010a. Diversity trends and their ontogenetic basis: an exploration of allometric disparity in rodents. Proc Biol Sci 277:1227-1234. Wilson LAB. 2011b. Comparison of prenatal and postnatal ontogeny: cranial allometry in the African striped mouse (Rhabdomys pumilio). Journal of mammalogy 92:407-420. Wilson LAB. 2011c. The contribution of developmental palaeontology to extensions of evolutionary theory. Acta Zoologica 94:254-260. Wilson LAB. 2013b. Cranial Suture Closure Patterns in Sciuridae: Heterochrony and Modularity. Journal of Mammalian Evolution 21:257-268. Wilson LAB. 2018. The evolution of ontogenetic allometric trajectories in mammalian domestication. Evolution 72: 867-877. Wilson LAB, Sánchez-Villagra MR. 2010b. Diversity trends and their ontogenetic basis: an exploration of allometric disparity in rodents. Proceedings of the Royal Society B- Biological Sciences 277:1227-1234. Wilson LAB, Schradin C, Mitgutsch C, Galliari FC, Mess A, Sánchez-Villagra MR. 2010. Skeletogenesis and sequence heterochrony in rodent evolution, with particular emphasis on the African striped mouse, Rhabdomys pumilio (Mammalia). Organisms Diversity & Evolution 10:243-258. Wisniewski M, Baumgart M, Grzonkowska M, Malkowski B, Wilinska-Jankowska A, Siedlecki Z, Szpinda M. 2017. Ossification center of the humeral shaft in the human fetus: a CT, digital, and statistical study. Surg Radiol Anat 39:1107-1116. Witton MP. 2008. A new approach to determining pterosaur body mass and its implications for pterosaur flight. Zitteliana 28:143-158. Witton MP. 2015. Were early pterosaurs inept terrestrial locomotors? PeerJ 3:e1018. Witton MP, Habib MB. 2010. On the size and flight diversity of giant pterosaurs, the use of birds as pterosaur analogues and comments on pterosaur flightlessness. PLoS ONE 5:e13982. Wolf M, Johansson LC, von Busse R, Winter Y, Hedenstrom A. 2010. Kinematics of flight and the relationship to the vortex wake of a Pallas' long tongued bat (Glossophaga soricina). J Exp Biol 213:2142-2153. Wolff J. 1986. Concept of the Law of Bone Remodelling. In: Wolff J, editor. The Law of Bone Remodelling. Berlin, Heidelberg: Springer Berlin Heidelberg. p. 1-1. Woodard K, Neustupa J. 2016. Morphometric Asymmetry of Frustule Outlines in the Pennate Diatom Luticola poulickovae (Bacillariophyceae). Symmetry 8. Wyant KA, Adams RA. 2007. Prenatal growth and development in the Angolan free- tailed bat, Mops condylurus (Chiroptera : Molossidae). Journal of mammalogy 88:1248-1251. Young NM. 2006. Function, ontogeny and canalization of shape variance in the primate scapula. J Anat 209:623-636. Young NM, HallgrÍmsson B. 2005. Serial Homology and the Evolution of Mammalian Limb Covariation Structure. Evolution 59: 2691-2704. Young NM, Wagner GP, Hallgrimsson B. 2010. Development and the evolvability of human limbs. Proc Natl Acad Sci U S A 107:3400-3405. Zeileis A, Grothendieck G, Ryan JA, Ulrich JM, Andrews F. 2018. zoo: S3 Infrastructure for Regular and Irregular Time Series (Z's Ordered Observations). Version 1.8-1: R Development Core Team. Zelditch ML, Calamari ZT, Swiderski DL. 2016. Disparate Postnatal Ontogenies Do Not Add to the Shape Disparity of Infants. Evolutionary Biology 43:188-207. 263

Zelditch ML, Li J, Tran LA, Swiderski DL. 2015. Relationships of diversity, disparity, and their evolutionary rates in squirrels (Sciuridae). Evolution 69:1284-1300. Zelditch ML, Lundrigan BL, Garland T, Jr. 2004. Developmental regulation of skull morphology. I. Ontogenetic dynamics of variance. Evol Dev 6:194-206. Zelditch ML, Lundrigan BL, Sheets HD, Garland T. 2003. Do precocial mammals develop at a faster rate? A comparison of rates of skull development in Sigmodon fulviventer and Mus musculus domesticus. Journal of Evolutionary Biology 16:708-720. Zelditch ML, Mezey J, Sheets HD, Lundrigan BL, Garland T, Jr. 2006. Developmental regulation of skull morphology II: ontogenetic dynamics of covariance. Evol Dev 8:46- 60. Zelditch ML, Wood AR, Bonett RM, Swiderski DL. 2008. Modularity of the rodent mandible: integrating bones, muscles, and teeth. Evol Dev 10:756-768.

264

Appendices Appendix Figure 1

Flowchart summarising the methodology used in our study.

Appendix Figure 2

Linear discriminant analysis of staging system implemented for bats in this study.

265

Appendix Figure 3

PCA of linear measurements of 24 postcranial bones in bat foetuses. Bones are grouped reflecting the axial and appendicular modules found in our Kendall’s τ modularity analysis.

266

Appendix Figure 4

Generalized Linear Model (GLM) of integration (Eigenvalue dispersion) and disparity (bone size variance) values across all 24 bones measured and all bat species pooled. Each point represents a single skeletal element.

Appendix Figure 5

Scatterplot of integration (Eigenvalue dispersion) and disparity (bone size variance) values in bats against number of specimens in each stage.

267

Appendix Figure 6

GLM of integration (Eigenvalue dispersion) and disparity (bone size variance) values across developmental time in bats examined. Each point represents a developmental stage.

Appendix Figure 7

PCA of bat species based on metric growth data. Ellipses show the developmental space of both suborders.

268

Appendix Figure 8 Correlation between body size and humeral cross-sectional diameter

Appendix Figure 9 Morphospace of humeral cross-sectional shape based on PCA, after removing outlying taxa. Colour of dots represent different foraging guilds.

269

Appendix Figure 10 Sampling curves of LaSEC analyses evaluating the performance of the landmarking protocol to capture shape variation. 270

271

Appendix Table 1

Average values (mm) of 25 postcranial linear measurements of 66 specimens of 11 bat species and all species pooled. Variance in parenthesis below.

CRL HL CL SL FL RiL TL FiL RL UL StL MPL PPL 18.390 4.254 3.886 2.552 2.976 2.916 3.359 2.872 5.514 5.612 0.462 1.705 1.442 A. dongbacana (13.800) (2.120) (0.860) (0.776) (1.155) (0.619) (1.957) (2.015) (5.794) (9.438) (0.381) (0.953) (0.440) 13.740 2.369 2.828 1.508 1.665 1.983 1.844 1.385 2.884 2.661 0.093 0.333 0.458 A. stoliczkanus (28.100) (2.058) (0.915) (0.893) (1.143) (0.434) (1.723) (1.437) (3.959) (5.069) (0.065) (0.476) (0.512) 16.180 6.667 3.485 2.440 2.805 2.843 2.958 2.228 5.940 6.320 0.975 1.810 1.590 C. sphinx (118.000) (33.920) (5.786) (8.381) (7.732) (8.842) (11.708) (6.213) (50.878) (65.509) (1.808) (5.908) (4.347) 11.950 2.720 2.744 2.088 1.496 1.896 1.558 1.510 2.808 3.000 0.406 0.960 0.494 H. blanfordi (26.500) (3.131) (1.557) (2.323) (1.324) (0.996) (1.660) (1.641) (5.642) (6.523) (0.310) (1.258) (0.385) 17.030 4.160 3.618 2.133 2.285 2.638 2.258 2.055 4.318 4.900 0.000 0.900 0.633 H. larvatus (5.470) (1.005) (0.616) (0.540) (0.384) (0.376) (0.687) (0.697) (2.657) (4.316) (0.000) (1.029) (0.325) 8.780 1.141 1.594 0.671 0.711 1.009 0.516 0.471 0.904 1.131 0.000 0.093 0.077 K. hardwickii (5.230) (0.408) (0.214) (0.327) (0.216) (0.202) (0.343) (0.280) (0.699) (0.680) (0.000) (0.060) (0.042) 24.250 6.590 5.375 5.015 4.770 3.895 5.080 5.005 8.410 9.815 1.270 1.710 2.220 M. schreibersii (7.640) (0.192) (0.076) (0.036) (0.065) (0.048) (0.115) (0.151) (0.370) (0.510) (0.016) (0.005) (0.000) 10.510 2.005 2.110 1.470 1.120 1.595 1.085 0.980 1.850 1.815 0.000 0.000 0.000 Myotis sp (1.510) (0.490) (0.205) (0.088) (0.274) (0.068) (0.572) (0.480) (1.280) (0.832) (0.000) (0.000) (0.000) R. pearsoni 9.140 2.140 2.290 1.410 1.380 1.700 1.360 1.160 2.170 2.670 0.000 0.000 0.140 R. pusillus 8.000 1.380 2.190 0.630 0.780 1.620 0.540 0.340 1.050 1.310 0.000 0.000 0.000 14.350 3.068 2.875 1.920 1.900 2.218 2.106 1.880 3.170 3.713 0.000 0.376 0.671 R. thomasi (21.000) (1.539) (0.969) (0.797) (0.876) (0.773) (0.868) (0.742) (2.170) (3.150) (0.000) (0.239) (0.181) 14.570 3.140 3.050 1.930 2.020 2.230 2.180 1.850 3.640 3.880 0.250 0.760 0.760 Chiroptera (32.600) (4.560) (1.580) (1.710) (1.860) (1.270) (2.670) (2.190) (8.360) (10.990) (0.300) (1.160) (0.790)

272

Appendix Table 1 continued McL MtL TaL CaL CvL TvL LvL SvL CavL IlL IsL PuL 3.237 0.977 0.555 0.139 3.606 2.431 1.679 1.162 0.521 1.293 0.857 0.942 A. dongbacana (2.549) (0.178) (0.144) (0.052) (0.466) (0.207) (0.163) (0.284) (0.204) (0.228) (0.295) (0.434) 1.515 0.445 0.151 0.000 2.468 1.687 1.027 0.501 0.161 0.750 0.260 0.167 A. stoliczkanus (2.019) (0.221) (0.065) (0.000) (1.776) (0.796) (0.381) (0.349) (0.073) (0.280) (0.158) (0.138) 4.035 0.805 0.663 0.000 2.200 1.400 0.858 0.633 0.350 1.360 0.800 1.075 C. sphinx (25.613) (1.051) (0.711) (0.000) (6.970) (2.761) (1.074) (0.564) (0.168) (2.457) (1.007) (1.814) 2.212 0.312 0.335 0.096 2.426 1.592 0.808 0.660 0.268 0.854 0.380 0.600 H. blanfordi (5.570) (0.148) (0.154) (0.046) (2.510) (0.987) (0.545) (0.381) (0.080) (0.424) (0.273) (0.461) 2.595 0.378 0.360 0.000 3.240 2.283 1.688 0.745 0.225 0.928 0.305 0.385 H. larvatus (2.988) (0.069) (0.119) (0.000) (1.038) (0.209) (0.060) (0.563) (0.083) (0.292) (0.372) (0.385) 0.293 0.027 0.031 0.000 0.668 0.449 0.173 0.133 0.056 0.204 0.061 0.000 K. hardwickii (0.600) (0.005) (0.007) (0.000) (1.104) (0.615) (0.209) (0.124) (0.022) (0.108) (0.026) (0.000) 6.200 1.830 1.705 0.200 4.810 2.615 1.755 1.715 0.890 2.870 1.920 2.525 M. schreibersii (0.125) (0.088) (0.000) (0.080) (0.005) (0.001) (0.008) (0.000) (0.001) (0.029) (0.005) (0.008) 0.905 0.035 0.100 0.000 2.215 1.525 0.555 0.290 0.000 0.675 0.000 0.185 Myotis sp (1.638) (0.002) (0.020) (0.000) (0.084) (0.110) (0.616) (0.168) (0.000) (0.068) (0.000) (0.068) R. pearsoni 0.950 0.140 0.130 0.000 2.140 1.380 0.820 0.480 0.000 0.580 0.000 0.000 R. pusillus 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 2.008 0.377 0.148 0.037 2.823 1.898 1.392 0.843 0.145 0.931 0.140 0.194 R. thomasi (1.284) (0.051) (0.024) (0.012) (1.219) (0.497) (0.412) (0.253) (0.038) (0.195) (0.102) (0.100) 2.190 0.520 0.330 0.050 2.650 1.750 1.130 0.710 0.260 0.940 0.430 0.490 Chiroptera (4.500) (0.310) (0.210) (0.020) (2.190) (0.940) (0.540) (0.400) (0.120) (0.540) (0.360) (0.530)

273

Appendix Table 2

Description of developmental stages 1–10 based on CRL ranges and ossification events that characterise each stage. “X” represents the ossification onset of a given bone.

Stage CRL HL CL SL FL RiL TL FiL RL UL StL MPL PPL McL MtL TaL CaL CvL TvL LvL SvL CavL IlL IsL PuL 1 2.96-5.1 X X X X X X 2 6.04-6.58 X X X X X X X X 3 6.88-8.71 X X X X X X X X X X 4 9.09-10.75 X X X X X X X X X X X X X X X X X X 5 11.17-11.9 X X X X X X X X X X X X X X X X X X X X 6 13.3-13.5 X X X X X X X X X X X X X X X X X X 7 13.73-15.01 X X X X X X X X X X X X X X X X X X X X X 8 15.54-19.74 X X X X X X X X X X X X X X X X X X X X X X 9 19.98-21.08 X X X X X X X X X X X X X X X X X X X X X X X 10 21.2-30.52 X X X X X X X X X X X X X X X X X X X X X X X X

Stage CRL Specific features 1 2.96-5.1 Ossification onset of humerus, clavicle, scapula, femur, ribs and tibia 2 6.04-6.58 Ossification onset of radius and ulna 3 6.88-8.71 Ossification onset of illium and fibula 4 9.09-10.75 Ossification onset of pedal phalanges, metacarpals, metatarsals, tarsals, and non-caudal vertebrae 5 11.17-11.9 Ossification onset of ischium and pubis 6 13.3-13.5 7 13.73-15.01 Ossification onset of caudal vertebrae 8 15.54-19.74 Ossification onset of manual phalanges 9 19.98-21.08 Ossification onset of sternum 10 21.2-30.52 Ossification onset of carpals

274

Appendix Table 3

PCA loadings of relative ossification rank for 24 postcranial bones

PC 1 PC 2 PC 3 PC 4 PC 5 Humeri 0.37224 0.018485 -0.06207 -0.0122 -0.06593 Radius 0.3472 0.07574 0.02129 -0.00062 -0.00996 Ulna 0.32142 0.10613 0.07515 0.051041 0.059637 Carpals -0.03919 0.073991 -0.06951 0.19961 0.039924 Metacarpals 0.061604 0.089839 0.26136 0.19357 -0.31129 Manual phalanges -0.15987 0.094066 0.46062 0.15561 0.046823 Femur 0.33988 0.051027 0.090974 -0.05786 -0.03183 Tibia 0.32306 0.096476 0.088293 -0.02201 0.093496 Fibula 0.29492 0.11298 0.052054 0.000232 0.12979 Tarsals 0.04074 0.15126 -0.11469 0.61272 -0.31739 Metatarsals 0.10162 0.11436 0.26351 0.06382 -0.28971 Pedal phalanges -0.06262 0.063135 0.43929 0.098958 -0.26471 Clavicle 0.37094 -0.05783 -0.06154 -0.02526 -0.04942 Scapula 0.16646 0.075399 0.15225 -0.05673 0.23165 Cervical v. -0.02175 0.33231 -0.00711 -0.1716 0.13399 Thoracic v. -0.043 0.34855 -0.0231 -0.15125 0.006875 Lumbar v. -0.02597 0.45989 -0.08981 -0.11732 -0.11836 Sacral v. -0.07655 0.43768 -0.08585 -0.10509 -0.2124 Caudal v. -0.22191 0.39583 -0.18424 -0.13674 -0.04881 Ilium 0.01145 0.099892 0.24952 -0.014 0.22026 Ischium -0.10094 0.15848 0.31226 0.032832 0.48709 Pubis -0.13657 -0.02818 0.34447 0.003725 0.048189 Sternum 0.017253 0.21135 -0.23385 0.61233 0.43105 Ribs 0.17674 0.11999 0.018518 -0.1633 0.021456

275

Appendix Table 4

Bone size variance (disparity) and eigenvalue dispersion (integration) values of 24 postcranial bones of nine bat species included in this study. Left and right boxes include bones of the appendicular and axial module respectively. Int = Integration, Disp = disparity

Appendicular HL FL TL FiL RL UL MPL PPL McL MtL TaL A. dongbacana Int 0.316 0.04 0.107 0.052 0.991 1.215 0.045 0.081 0.113 0.185 0.287 Disp 2.12 1.155 1.957 2.015 5.794 9.438 0.953 0.44 2.549 0.178 0.144 A. stoliczkanus Int 0.119 0.02 0.045 0.015 0.285 0.274 0.068 0.053 0.046 0.052 0.09 Disp 2.058 1.143 1.723 1.437 3.959 5.069 0.476 0.512 2.019 0.221 0.065 C. sphinx Int 8.728 0.073 0.345 0.123 9.856 14.85 0.048 0.154 2.367 1.119 1.377 Disp 33.92 7.732 11.7 6.213 50.878 65.509 5.908 4.347 25.613 1.051 0.711 H. blanfordi Int 0.509 0.012 0.03 0.022 0.979 1.219 0.04 0.235 0.648 0.372 0.411 Disp 3.131 1.324 1.66 1.641 5.642 6.523 1.258 0.385 5.57 0.148 0.154 H. larvatus Int 1.667 0.057 0.076 0.032 2.407 3.967 0.507 0.458 0.625 0.766 0.675 Disp 1.005 0.384 0.687 0.697 2.657 4.316 1.029 0.325 2.988 0.069 0.119 K. hardwickii Int 0.095 0.014 0.01 0.005 0.079 0.121 0.025 0.029 0.033 0.046 0.044 Disp 0.408 0.216 0.343 0.28 0.699 0.68 0.06 0.042 0.6 0.005 0.007 M. schreibersii Int 8.686 1.167 2.067 1.969 23.076 38.67 3.866 2.32 6.318 2.838 4.115 Disp 0.192 0.065 0.115 0.151 0.37 0.51 0.005 0 0.125 0.088 0 Myotis sp Int 1.313 0.097 0.229 0.118 1.921 1.382 0.992 0.992 0.603 0.86 0.64 Disp 0.49 0.274 0.572 0.48 1.28 0.832 0 0 1.638 0.002 0.02 R. thomasi Int 0.312 0.04 0.056 0.022 0.401 0.711 0.15 0.079 0.063 0.158 0.188 Disp 1.539 0.876 0.868 0.742 2.17 3.15 0.239 0.181 1.284 0.051 0.024

276

Axial CaL StL CL SL RiL CvL TvL LvL SvL CavL IlL IsL PuL A. dongbacana Int 0.388 0.276 0.186 0.006 0.029 0.124 0.017 0.058 0.115 0.262 0.088 0.162 0.138 Disp 0.052 0.381 0.86 0.776 0.619 0.466 0.207 0.163 0.284 0.204 0.228 0.295 0.434 A. stoliczkanus Int 0.141 0.112 0.155 0.009 0.034 0.132 0.025 0.016 0.045 0.101 0.021 0.078 0.097 Disp 0 0.065 0.915 0.893 0.434 1.776 0.796 0.381 0.349 0.073 0.28 0.158 0.138 C. sphinx Int 2.926 0.727 0.449 0.053 0.374 0.087 0.532 1.148 1.578 2.159 0.428 1.156 0.74 Disp 0 1.808 5.786 8.381 8.842 6.97 2.761 1.074 0.564 0.168 2.457 1.007 1.814 H. blanfordi Int 0.566 0.304 0.297 0.173 0.06 0.376 0.054 0.13 0.162 0.417 0.097 0.319 0.172 Disp 0.046 0.31 1.557 2.323 0.996 2.51 0.987 0.545 0.381 0.08 0.424 0.273 0.461 H. larvatus Int 1.284 1.284 0.953 0.035 0.172 0.806 0.055 0.081 0.228 0.845 0.238 0.738 0.609 Disp 0 0 0.616 0.54 0.376 1.038 0.209 0.06 0.563 0.083 0.292 0.372 0.385 K. hardwickii Int 0.058 0.058 0.19 0.02 0.055 0.091 0.04 0.015 0.018 0.035 0.012 0.033 0.058 Disp 0 0 0.214 0.327 0.202 1.104 0.615 0.209 0.124 0.022 0.108 0.026 0 M. schreibersii Int 10.532 5.488 2.793 1.603 0.049 0.965 1.402 4.177 4.076 7.714 0.623 3.12 1.381 Disp 0.08 0.016 0.076 0.036 0.048 0.005 0.001 0.008 0 0.001 0.029 0.005 0.008 Myotis sp Int 0.992 0.992 1.042 0.155 0.213 0.933 0.217 0.008 0.183 0.992 0.064 0.992 0.401 Disp 0 0 0.205 0.088 0.068 0.084 0.11 0.616 0.168 0 0.068 0 0.068 R. thomasi Int 0.259 0.293 0.202 0.044 0.069 0.223 0.025 0.007 0.053 0.23 0.039 0.229 0.207 Disp 0.012 0 0.969 0.797 0.773 1.219 0.497 0.412 0.253 0.038 0.195 0.102 0.1

277

Appendix Table 5 PC loadings of linear measurements used in this study

PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 HL -0.816 0.056 -0.02005 -0.02827 0.016356 2.54E-02 -1.58E-02 -0.00507 1.49E-02 CL -0.79414 0.012 -0.01438 0.021812 -0.02424 0.026296 -0.02027 0.009809 -8.15E-03 SL -0.40256 -0.00783 0.067513 0.060084 0.037733 -0.00484 -0.01344 -0.00483 -7.57E-03 FL -0.39013 0.016139 -0.03995 0.006718 -0.01429 0.02319 0.032462 0.008644 1.23E-02 RiL -0.50911 -0.00901 -0.02132 0.012598 -0.00537 0.0154 -0.00317 -0.00253 5.03E-03 TL -0.40587 -0.00251 -0.0294 0.016464 -0.02998 -0.0179 0.043425 0.015094 -4.91E-03 FiL -0.24376 -0.00338 0.036386 0.003846 0.05692 -0.00247 0.023274 0.01016 -7.86E-03 RL -0.86368 0.069078 -0.02622 -0.00286 -0.03123 0.020344 0.018335 0.023837 -1.88E-03 UL -0.83703 0.118678 -0.00579 -0.03557 0.104283 -0.02435 0.000928 -0.03033 3.97E-03 StL 0.780322 0.106565 0.021106 -0.0264 0.000526 0.042728 -0.02524 0.007873 -1.71E-02 MPL 0.550108 0.164947 -0.05424 0.04439 0.015639 -0.0189 -0.03908 0.021444 2.05E-03 PPL 0.468021 0.084299 -0.11721 0.009847 -0.02912 -0.03696 -0.00829 -0.04019 1.25E-02 McL -0.3288854 0.0807845 0.0473787 -0.0749537 -0.0309049 -0.0575354 -0.0148645 0.0294416 -1.76E-02 MtL 0.598165 -0.00618 -0.0591 -0.02297 -0.03589 0.007216 0.015544 -0.03081 -3.71E-02 TaL 0.691697 0.010066 0.045054 -0.03463 0.035266 0.046254 0.038337 -0.00931 5.61E-03 CaL 0.951296 -0.15627 -0.01506 -0.02442 0.059988 0.004646 -0.0133 0.011634 -9.39E-04 CvL -0.66974 -0.15286 0.048882 0.041231 -0.04225 0.005514 -0.03253 -0.03055 -6.04E-03 TvL -0.36059 -0.16968 0.011978 0.022299 0.004633 0.017405 -0.01553 -0.00012 -4.38E-06 LvL 0.012087 -0.19889 -0.07539 -0.0972 -0.01673 -0.00316 -0.01096 0.012764 1.45E-02 SvL 0.379573 -0.0843 0.037065 0.032613 -0.00014 -0.091 0.011466 0.008465 9.09E-03 CavL 0.760001 -0.05169 -0.02235 0.042634 0.039204 0.007896 -0.00231 0.012845 -4.14E-03 IlL 0.146281 -0.00566 0.057195 0.005292 -0.02258 -0.01834 0.034576 -0.01848 -3.96E-03 IsL 0.678117 0.042429 -0.00362 0.073875 -0.03057 0.0235 0.010813 0.012526 1.48E-02 PuL 0.605901 0.086475 0.131526 -0.04643 -0.05724 0.009672 -0.01433 -0.01233 2.25E-02

278

Appendix Table 6 Species list with information on sex, familial and subordinal taxonomic arrangement and diet ID Species Individual Stage signedR.L unsignedR.L signedFA6 unsignedFA6 TA FA TA2 As_do_1_hum AD 1 6 0.19 0.19 -0.0031847 0.00318471 0.12519126 0.12519126 0.04818958 As_do_10_hum AD 2 7 -0.1 0.1 0.04590164 0.04590164 0.11862218 0.11862218 0.04035127 As_do_11_hum AD 3 10 0.04 0.04 0.02076125 0.02076125 0.04607092 0.04607092 0.01622656 As_do_12_hum AD 4 9 0.02 0.02 -0.0133333 0.01333333 0.09726894 0.09726894 0.03527519 As_do_13_hum AD 5 8 0.06 0.06 -0.009828 0.00982801 0.07695157 0.07695157 0.02962982 As_do_14_hum AD 6 8 0.03 0.03 -0.0040733 0.00407332 0.14328147 0.14328147 0.05105322 As_do_2_hum AD 7 9 -0.01 0.01 0.01392111 0.01392111 0.04394614 0.04394614 0.01310796 As_do_3_hum AD 8 4 -0.04 0.04 -0.0131579 0.0131579 0.08229525 0.08229525 0.03197219 As_do_4_hum AD 9 8 -0.05 0.05 -0.0145985 0.01459854 0.12630717 0.12630717 0.04192468 As_do_5_hum AD 10 10 -0.05 0.05 -0.0123077 0.01230769 0.09793844 0.09793844 0.03032002 As_do_6_hum AD 11 10 0 0 -0.0175219 0.0175219 0.10195867 0.10195867 0.04834986 As_do_7_hum AD 12 9 0.02 0.02 0.05633803 0.05633803 0.0644177 0.0644177 0.02309819 As_do_8_hum AD 13 8 0.01 0.01 -0.0223464 0.02234637 0.05700106 0.05700106 0.01801506 As_do_9_hum AD 14 8 0.07 0.07 -0.0100756 0.01007557 0.07576798 0.07576798 0.02815318 As_sto_1_hum AS 15 7 -0.09 0.09 -0.0202605 0.02026049 0.16473293 0.16473293 0.08154381 As_sto_10_hum AS 16 8 -0.03 0.03 -0.0029369 0.00293686 0.07066842 0.07066842 0.0219219 As_sto_11_hum AS 17 7 -0.08 0.08 -0.0106195 0.01061947 0.09253168 0.09253168 0.03686557 As_sto_12_hum AS 18 8 -0.09 0.09 0.00177778 0.00177778 0.13469058 0.13469058 0.05244264 As_sto_15_hum AS 19 4 -0.01 0.01 -0.008658 0.00865801 0.03378598 0.03378598 0.01394947 As_sto_2_hum AS 20 8 -0.01 0.01 -0.0884956 0.08849558 0.04621615 0.04621615 0.01962249 As_sto_3_hum AS 21 5 -0.03 0.03 0.01904762 0.01904762 0.12523467 0.12523467 0.04809194 As_sto_4_hum AS 22 8 -0.06 0.06 0.04444444 0.04444444 0.09475241 0.09475241 0.02894949 As_sto_5_hum AS 23 10 -0.03 0.03 0.0037244 0.0037244 0.09388336 0.09388336 0.034498 As_sto_6_hum AS 24 4 0.07 0.07 0.00858369 0.00858369 0.10070784 0.10070784 0.03136277 As_sto_7_hum AS 25 9 -0.08 0.08 -0.0307692 0.03076923 0.08241703 0.08241703 0.02535429

279

As_sto_8_hum AS 26 7 -0.14 0.14 -0.006192 0.00619195 0.09199225 0.09199225 0.03701794 As_sto_9_hum AS 27 7 0.03 0.03 0.02027788 0.02027788 0.05382441 0.05382441 0.01952226 Cy_sp_1_hum CS 28 8 0.03 0.03 0.02120141 0.02120141 0.09034398 0.09034398 0.03025819 Cy_sp_4_hum CS 29 10 -0.27 0.27 0.02006689 0.02006689 0.06229499 0.06229499 0.02269788 Cy_sp_5_hum CS 30 2 0.01 0.01 0.03968254 0.03968254 0.09525069 0.09525069 0.02943233 Cy_sp_6_hum CS 31 2 0 0 0.00891531 0.00891531 0.11613055 0.11613055 0.05031574 Hes_bla_1_hum HB 32 5 -0.01 0.01 0.00853971 0.00853971 0.13443716 0.13443716 0.08355797 Hes_bla_2_hum HB 33 4 -0.02 0.02 0.0020429 0.0020429 0.16727509 0.16727509 0.0737996 Hes_bla_3_hum HB 34 8 0.04 0.04 0.02531646 0.02531646 0.10437521 0.10437521 0.03789591 Hes_bla_4_hum HB 35 8 0.02 0.02 0 0 0.08083069 0.08083069 0.03580869 Hes_bla_5_hum HB 36 1 0.01 0.01 -0.0796646 0.07966457 0.09606402 0.09606402 0.03309422 Hip_lar_4_hum HL 37 8 0.08 0.08 0.07751938 0.07751938 0.0912262 0.0912262 0.04269354 Hip_lar_5_hum HL 38 8 0.04 0.04 -0.0578035 0.05780347 0.13537717 0.13537717 0.047777 Hip_lar_6_hum HL 39 7 0.01 0.01 -0.0051151 0.00511509 0.09799694 0.09799694 0.03369683 Hip_lar_7_hum HL 40 9 -0.01 0.01 0.06263048 0.06263048 0.07366237 0.07366237 0.02376659 Ker_hard_10_hum KH 41 2 0.05 0.05 -0.006993 0.00699301 0.10374086 0.10374086 0.03912258 Ker_hard_12_hum KH 42 3 -0.03 0.03 -0.0050125 0.00501253 0.15084568 0.15084568 0.06710866 Ker_hard_13_hum KH 43 4 -0.03 0.03 0.00278164 0.00278164 0.17313214 0.17313214 0.0827912 Ker_hard_3_hum KH 44 4 -0.02 0.02 0.02439024 0.02439024 0.10761088 0.10761088 0.04588921 Ker_hard_4_hum KH 45 5 0 0 0.0072904 0.0072904 0.07021098 0.07021098 0.0282595 Ker_hard_5_hum KH 46 3 -0.05 0.05 0.02898551 0.02898551 0.12363125 0.12363125 0.04633116 Ker_hard_6_hum KH 47 5 0.1 0.1 0 0 0.11251952 0.11251952 0.05595121 Min_schr_1_hum MS 48 10 -0.01 0.01 -0.0099305 0.00993049 0.06750469 0.06750469 0.02893657 Min_schr_2_hum MS 49 10 -0.16 0.16 -0.0318302 0.03183024 0.10040804 0.10040804 0.04215484 Myo_1_hum M 50 5 -0.03 0.03 -0.039886 0.03988604 0.17416543 0.17416543 0.10895725 Myo_2_hum M 51 4 0.02 0.02 0.01398601 0.01398601 0.15587009 0.15587009 0.07470592 Rhi_pea_1_hum Rpe 52 4 -0.03 0.03 0.01109878 0.01109878 0.10387342 0.10387342 0.04265414 Rhi_pus_1_hum Rpu 53 2 -0.08 0.08 0.0295203 0.0295203 0.12730678 0.12730678 0.04800463

280

Rhi_tho_1_hum RT 54 1 0.01 0.01 0.03522505 0.03522505 0.12103476 0.12103476 0.04724518 Rhi_tho_10_hum RT 55 6 -0.15 0.15 0.01724138 0.01724138 0.17786754 0.17786754 0.07102537 Rhi_tho_2_hum RT 56 8 0.02 0.02 0.00144823 0.00144823 0.11460549 0.11460549 0.04722346 Rhi_tho_3_hum RT 57 7 -0.06 0.06 0 0 0.0930922 0.0930922 0.03237144 Rhi_tho_4_hum RT 58 7 0.04 0.04 0.00407332 0.00407332 0.12309267 0.12309267 0.04141872 Rhi_tho_5_hum RT 59 8 0.07 0.07 0.02515723 0.02515723 0.13253399 0.13253399 0.04714821 Rhi_tho_6_hum RT 60 6 -0.01 0.01 -0.0024722 0.00247219 0.1346491 0.1346491 0.06067698 Rhi_tho_7_hum RT 61 7 -0.06 0.06 0.01369863 0.01369863 0.0737446 0.0737446 0.02593666 Rhi_tho_8_hum RT 62 9 0.05 0.05 -0.006006 0.00600601 0.1481859 0.1481859 0.05967132 Rhi_tho_9_hum RT 63 8 0.12 0.12 0.01192843 0.01192843 0.09117243 0.09117243 0.03196924

281

Appendix Table 7 Species list with information on sex, familial and subordinal taxonomic arrangement and diet Species ID Collection Sex Family Suborder Diet Aselliscus dongbacana NA CUHK Male Hipposideridae Yinpterochiroptera Insectivore Astronomus australis M18402 WAM Female Molossidae Yangochiroptera Insectivore Astronomus australis M10582 WAM Male Molossidae Yangochiroptera Insectivore Astronomus australis M14496 WAM Male Molossidae Yangochiroptera Insectivore Astronomus australis M10595 WAM Female Molossidae Yangochiroptera Insectivore Carollia perspicillata* AMNH 266126 AMNH Male Phyllostomidae Yangochiroptera Frugivore Chalinolobus gouldii M11635 WAM Female Vespertilionidae Yangochiroptera Insectivore Chalinolobus gouldii M11911 WAM Male Vespertilionidae Yangochiroptera Insectivore Chalinolobus gouldii M11964 WAM Female Vespertilionidae Yangochiroptera Insectivore Chalinolobus gouldii M12177 WAM Male Vespertilionidae Yangochiroptera Insectivore Craseonycteris thonglonyai* USNM 528306 NMNH Male Craseonycteridae Yinpterochiroptera Insectivore Cynopterus brachyotis* AMNH 232506 AMNH Female Pteropodidae Yinpterochiroptera Frugivore Desmodus rotundus NA UNSW Male Phyllostomidae Yangochiroptera Sanguivore Eptesicus fuscus* UW238 UW Female Vespertilionidae Yangochiroptera Insectivore Furipterus horrens* AMNH 244493 AMNH Female Furipteridae Yangochiroptera Insectivore Hipposideros ater M26060 WAM Male Hipposideridae Yinpterochiroptera Insectivore Hipposideros ater M26063 WAM Male Hipposideridae Yinpterochiroptera Insectivore Hipposideros ater M40102 WAM Female Hipposideridae Yinpterochiroptera Insectivore Hipposideros ater M42453 WAM Female Hipposideridae Yinpterochiroptera Insectivore Hipposideros commersoni* AMNH 244493 AMNH Female Hipposideridae Yinpterochiroptera Insectivore Hipposideros pomona NA CUHK Female Hipposideridae Yinpterochiroptera Insectivore Lavia frons* AMNH 49384 AMNH Male Megadermatidae Yinpterochiroptera Insectivore Macroglossus minimus M15729 WAM Female Pteropodidae Yinpterochiroptera Frugivore Macroglossus minimus M38808 WAM Male Pteropodidae Yinpterochiroptera Frugivore Macroglossus minimus M40890 WAM Male Pteropodidae Yinpterochiroptera Frugivore

282

Macroglossus minimus M40891 WAM Female Pteropodidae Yinpterochiroptera Frugivore Megaderma spasma NA CUHK Male Megadermatidae Yinpterochiroptera Carnivore Miniopterus schreibersii M42841 WAM Male Miniopteridae Yangochiroptera Insectivore Miniopterus schreibersii M44582 WAM Female Miniopteridae Yangochiroptera Insectivore Miniopterus schreibersii M44647 WAM Male Miniopteridae Yangochiroptera Insectivore Miniopterus schreibersii M62231 WAM Female Miniopteridae Yangochiroptera Insectivore Molossus molossus* AMNH 149261 AMNH Female Molossidae Yangochiroptera Insectivore Myotis daubentoni* AMNH 218932 AMNH Female Vespertilionidae Yangochiroptera Insectivore Mystacina tuberculata NA UNSW Female Mystacinidae Yangochiroptera Omnivore Myzopoda aurita* USNM 449282 NMNH Female Myzopodidae Yangochiroptera Insectivore Natalus stramineus* AMNH 206695 AMNH Male Natalidae Yangochiroptera Insectivore Noctilio albiventris* AMNH 243904 AMNH Female Noctilionidae Yangochiroptera Piscivore Nycteris grandis* AMNH 268369 AMNH Female Nycteridae Yangochiroptera Carnivore Phyllostomus hastatus* AMNH 239876 AMNH Male Phyllostomidae Yangochiroptera Omnivore Pipistrellus abramus NA CUHK Female Vespertilionidae Yangochiroptera Insectivore Pteronotus parnelli* AMNH 269063 AMNH Female Mormoopidae Yangochiroptera Insectivore Rhinolophus ferrumequinum NA CUHK Female Rhinolophidae Yinpterochiroptera Insectivore Rhinolophus ferrumequinum* AMNH 245591 AMNH Female Rhinolophidae Yinpterochiroptera Insectivore Rhinonycteris aurantius NA WAM Rhinonycteridae Yinpterochiroptera Insectivore Rhinopoma hardwickei* AMNH 208125 AMNH Female Rhinopomatidae Yinpterochiroptera Insectivore Rousettus bidens* ZMA 23.100 ZMA Male Pteropodidae Yinpterochiroptera Frugivore Saccopteryx bilineata* AMNH 265962 AMNH Male Emballonuridae Yangochiroptera Insectivore Scotoropens sp. NA UNSW Male Vespertilionidae Yangochiroptera Insectivore Tadarida brasiliensis* NA NA Male Molossidae Yangochiroptera Insectivore Taphozous georgianus M15709 WAM Female Emballonuridae Yangochiroptera Insectivore Taphozous georgianus M26959 WAM Male Emballonuridae Yangochiroptera Insectivore Taphozous georgianus M30914 WAM Female Emballonuridae Yangochiroptera Insectivore Taphozous georgianus M62505 WAM Male Emballonuridae Yangochiroptera Insectivore

283

Taphozous mauritianus* AMNH 49366 AMNH Male Emballonuridae Yangochiroptera Insectivore Thyroptera tricolor AMNH 246236 AMNH Female Thyropteridae Yangochiroptera Insectivore Vespertilio sinensis NA CUHK Female Vespertilionidae Yangochiroptera Insectivore

284

Appendix Table 8 P values of Pairwise comparisons of differences in scaling of shape and biomechanical properties of the humerus in bats between foraging and roosting guilds: Carnivore (C), Frugivore (F), Gleaning (G), Hawking (H), Trawling (T), Terrestrial Locomoting (TL) and Upstand Roosting (UR). CircMaxR J Iy Ix Imax Imin Shape Foraging guilds only C:F 0.7507 0.5001 0.4706 0.471 0.4715 0.4696 0.1906 C:G 0.5767 0.4911 0.4452 0.4364 0.43705 0.4387 0.109 C:H 0.9062 0.20315 0.6599 0.2534 0.27985 0.188 0.681 C:T 0.7507 0.5001 0.4706 0.471 0.4715 0.4696 0.18885 C:TL 0.7507 0.5001 0.4706 0.471 0.4715 0.4696 0.18885 C:UR 0.7507 0.5001 0.4706 0.471 0.4715 0.4696 0.18885 F:G 0.6740 0.5094 0.5447 0.544 0.5444 0.54305 0.88715 F:H 0.6722 0.40405 0.734 0.32195 0.3526 0.2805 0.6436 F:T 0.5001 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 F:TL 0.5001 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 F:UR 0.5001 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 G:H 0.7051 0.405 0.7378 0.33805 0.37275 0.28395 0.9346 G:T 0.6740 0.5094 0.5447 0.544 0.5444 0.54305 0.76335 G:TL 0.6740 0.5094 0.5447 0.544 0.5444 0.54305 0.76335 G:UR 0.6740 0.5094 0.5447 0.544 0.5444 0.54305 0.76335 H:T 0.1187 0.8006 0.4083 0.67635 0.67755 0.71955 0.1925 H:TL 0.1187 0.8006 0.4083 0.67635 0.67755 0.71955 0.1925 H:UR 0.1187 0.8006 0.4083 0.67635 0.67755 0.71955 0.1925 T:TL 0.5001 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 T:UR 0.5001 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 TL:UR 0.5001 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 Foraging guilds + upstand roosting C:F 0.22445 0.5001 0.4706 0.471 0.4715 0.4696 0.11275

285

C:G 0.3617 0.89275 0.20375 0.5807 0.58135 0.89785 0.07665 C:H 0.33195 0.56365 0.4052 0.9649 0.0473 0.39705 0.96735 C:T 0.22445 0.5001 0.4706 0.471 0.4715 0.4696 0.11915 C:TL 0.22445 0.5001 0.4706 0.471 0.4715 0.4696 0.11915 F:G 0.78185 0.3091 0.3272 0.81375 0.82 0.3093 0.4896 F:H 0.4177 0.5637 0.40555 0.96545 0.0479 0.41645 0.9871 F:T 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 F:TL 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 G:H 0.45225 0.26535 0.49555 0.9496 0.06005 0.1761 0.9754 G:T 0.40345 0.107 0.7698 0.39085 0.3908 0.0915 0.89135 G:TL 0.40345 0.107 0.7698 0.39085 0.3908 0.0915 0.89135 H:T 0.829 0.58925 0.40555 0.74205 0.0479 0.41645 0.97225 H:TL 0.829 0.58925 0.40555 0.74205 0.0479 0.41645 0.97225 T:TL 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005 0.50005

286

Appendix Table 9 P values of Pairwise comparisons of differences in shape and biomechanical properties of the humerus in bats between foraging and roosting guilds: Carnivore (C), Frugivore (F), Gleaning (G), Hawking (H), Trawling (T), Terrestrial Locomoting (TL) and Upstand Roosting (UR). CircMaxR J Iy Ix Imax Imin Shape C:F 0.8674 0.691 0.7244 0.6661 0.8156 0.5854 0.6972 C:G 0.9065 0.0405 0.0506 0.0358 0.0954 0.018 0.0193 C:H 0.5271 0.0366 0.0229 0.0413 0.0242 0.0428 0.0139 C:T 0.2527 0.0154 0.0131 0.0159 0.0156 0.019 0.3406 C:TL 0.0994 0.1159 0.1445 0.1029 0.1526 0.1236 0.0551 C:UR 0.3738 0.4192 0.6153 0.2202 0.6872 0.2051 0.1254 F:G 0.7653 0.0786 0.084 0.0757 0.121 0.0591 0.0309 F:H 0.6319 0.0774 0.0402 0.0934 0.0241 0.1353 0.0149 F:T 0.2996 0.027 0.0217 0.0301 0.0187 0.0459 0.4608 F:TL 0.1143 0.1932 0.2195 0.1796 0.1861 0.261 0.0732 F:UR 0.2675 0.6268 0.8189 0.3762 0.8129 0.4022 0.1837 G:H 0.3686 0.6029 0.8348 0.523 0.8755 0.3505 0.7747 G:T 0.1863 0.4476 0.4001 0.4942 0.3059 0.6669 0.2904 G:TL 0.063 0.916 0.8833 0.9182 0.9597 0.7026 0.9705 G:UR 0.399 0.3543 0.2484 0.5855 0.3095 0.4928 0.6321 H:T 0.405 0.2023 0.2549 0.2025 0.286 0.2326 0.3212 H:TL 0.1416 0.8085 0.9779 0.7477 0.9587 0.8192 0.8344 H:UR 0.0912 0.4905 0.2301 0.8772 0.1822 0.92 0.6931 T:TL 0.6537 0.4526 0.3897 0.4961 0.404 0.4877 0.3494 T:UR 0.0557 0.1419 0.0826 0.2868 0.0755 0.3359 0.5414 TL:UR 0.0148 0.488 0.3914 0.7035 0.3537 0.8029 0.6565

287

Appendix Table 10 Procrustes ANOVA and pairwise comparisons of differences in humeral cross-sectional shape between foraging and roosting guilds: Carnivore (C), Frugivore (F), Gleaning (G), Hawking (H), Trawling (T), Terrestrial Locomoting (TL) and Upstand Roosting (UR). Procrustes variances for defined groups C F G H T TL UR 0.2947812 0.51333958 0.05286328 0.19851127 0.73764189 0.04543731 0.02153752 P values / Pairwise absolute differences between variances C F G H T TL UR C 0.4036 0.3609 0.7051 0.2097 0.3764 0.3382 F 0.21855838 0.0893 0.116 0.4016 0.1677 0.1567 G 0.24191792 0.4604763 0.5074 0.0645 0.9716 0.9366 H 0.09626993 0.3148283 0.14564799 0.095 0.5775 0.465 T 0.44286069 0.2243023 0.68477861 0.53913062 0.1377 0.1185 TL 0.24934388 0.4679023 0.00742597 0.15307396 0.6922046 0.8834 UR 0.2732437 0.4918021 0.0170959 0.1842164 0.7161044 0.02389979

288

Appendix Table 11 Description of landmarking protocol used to capture humeral morphological shape Landmark Description Label 1 Proximal end of the spina entepicondyli 0 2 Base of the spina entepicondyli 1 3 Inner most depression of the medial process 2 4 Inner most depression between the spina entepicondyli and the trochlea 3 5 Distal end of the lateral capitullum 4 6 Proximal end of the lateral capitullum 5 7 Inner most point of the lateral fossa 6 8 Distal end of the trochlea 7 9 Proximal most point of the depression between the trochlea and central capitulum 8 10 Distal end of the central capitulum 9 11 Anterior end of the lateral capitullum 10 12 Posterior end of the lateral capitullum 11 13 Proximal end of the trochlea 12 14 Proximal end of the central trochlea 13 15 Inner most point of the posterior depression of the distal epiphysis 14 16 Proximal end of the taberculum majus 15 17 Proximal end of the caput humeri in posterior view 16 18 Proximal end of the taberculum minus in posterior view 17 19 Medial end of the taberculum minus in posterior view 18 20 Lateral end of the taberculum majus in posterior view 19 21 Posterior end of the caput humeri in lateral view 20 22 Inner most point of the supraglenoid fossa 21 23 Inner most point of the medial fossa 22 24 Anterior end of the trochlea 23

289

25 Medial end of the caput humeri 24 26 Lateral end of the caput humeri 25 27 Distal end of the caput humeri in posterior view 26 28 Proximal end of the taberculum minus in anterior view 27 29 Anterior end of the central capitulum 28 30 Inner most depression between the lateral and central capitulum 29 31 Inner most point of the radial fossa 30 32-71 Curve along the anterior face of the diaphysis between the medial base of the trochlea and the lateral base of the pectoral ridge C0 72-111 Curve along the medial face of the diaphysis between the base of the trochlea and the base of the taberculum minus C1 112-151 Curve along the posterior face of the diaphysis between the base of the caput humeri and landmark 14 C2 152-191 Curve along the lateral face of the diaphysis between the base of the taberculum majus and the base of the supraepicondylar groov C3 192-201 Curve along the medial edge of the pectoral ridge C4 202-211 Curve along the edge of the medial ridge C5 212-221 Curve along the lateral edge of the pectoral ridge C6

290

Appendix Table 12 Species list with information on familial and subordinal taxonomic arrangement, diet, humeral length, wing loading, aspect ratio and foraging guild ID Species Family Suborder Diet HL HLCat WL AS FG P_Ase_dong_hum_l.75.Per Aselliscus dongbacana Hipposideridae Yinpterochiroptera Insectivore 23.673 Small 4.9 5.7 G P_Aus_aus_hum_l.75.Per Austronomus australis Molossidae Yangochiroptera Insectivore 34.88575 Medium 13.4 8.2 H P_Car_per_hum_l.75.Per Carollia perspicillata Phyllostomidae Yangochiroptera Frugivore 25.09 Small 11.4 6.1 F P_Cha_gou_hum_l.75.Per Chalinolobus gouldii Vespertilionidae Yangochiroptera Insectivore 26.38625 Small 10.7 7.8 H P_Cra_thong_hum_l.75.Per Craseonycteris thonglonyai Craseonycteridae Yinpterochiroptera Insectivore 16.631 Small 5.2 7.1 G P_Cyn_brac_hum_l.75.Per Cynopterus brachyotis Pteropodidae Yinpterochiroptera Frugivore 32.992 Medium 13.1 7.7 F P_Des_rot_hum_l.75.Per Desmodus rotundus Phyllostomidae Yangochiroptera Sanguivore 36.689 Medium 14 6.7 TL P_Ept_fus_hum_l.75.Per Eptesicus fuscus Vespertilionidae Yangochiroptera Insectivore 28.497 Medium 9.4 6.4 H P_Fur_hor_hum_l.75.Per Furipterus horrens Furipteridae Yangochiroptera Insectivore 28.497 Medium G P_Hip_ate_hum_l.75.Per Hipposideros ater Hipposideridae Yinpterochiroptera Insectivore 24.29525 Small 6.3 5.8 H P_Hip_gig_hum_l.75.Per Hipposideros gigas Hipposideridae Yinpterochiroptera Insectivore 19.408 Small 15.7 7.7 H P_Hip_pom_hum_l.75.Per Hipposideros pomona Hipposideridae Yinpterochiroptera Insectivore 23.8 Small 5.4 4.8 H P_Mac_min_hum_l.75.Per Macroglossus minimus Pteropodidae Yinpterochiroptera Frugivore 26.3475 Small 12.3 6.5 F P_Meg_spa_hum_l.75.Per Megaderma spasma Megadermatidae Yinpterochiroptera Carnivore 26.718 Small 11.8 6.2 C P_Min_sch_hum_l.75.Per Miniopterus schreibersii Miniopteridae Yangochiroptera Insectivore 27.43525 Medium 10.2 7 H P_Mol_mol_hum_l.75.Per Molossus molossus Molossidae Yangochiroptera Insectivore 8.771 Small 14.1 9.2 H P_Myo_dau_hum_l.75.Per Myotis daubentoni Vespertilionidae Yangochiroptera Insectivore 22.799 Small 7 6.3 T P_Mys_tub_hum_l.75.Per Mystacina tuberculata Mystacinidae Yangochiroptera Omnivore 29.068 Medium 12.3 7 TL P_Myzo_aur_hum_l.75.Per Myzopoda aurita Myzopodidae Yangochiroptera Insectivore 24.411 Small H P_Nat_stra_hum_l.75.Per Natalus stramineus Natalidae Yangochiroptera Insectivore 21.539 Small 3.9 5.8 H P_Noc_alb_hum_l.75.Per Noctilio albiventris Noctilionidae Yangochiroptera Piscivore 8.64 Small 13.9 7.8 T P_Nyc_gran_hum_l.75.Per Nycteris grandis Nycteridae Yangochiroptera Carnivore 32.147 Medium 11.4 5.2 C P_Phy_has_hum_l.75.Per Phyllostomus hastatus Phyllostomidae Yangochiroptera Carnivore 52.697 Large 25.2 7.6 C

291

P_Pte_par_hum_l.75.Per Pteronotus parnelli Mormoopidae Yangochiroptera Insectivore 31.132 Medium 7.9 6.7 H P_Rhi_aur_hum_l.75.Per Rhinonycteris aurita Rhinonycteridae Yinpterochiroptera Insectivore 28.904 Medium H P_Rhi_fer_hum_l.75.Per Rhinolophus ferrumequinum Rhinolophidae Yinpterochiroptera Insectivore 35.44 Medium 12.2 6.1 H P_Rhi_hard_hum_l.75.Per Rhinopoma hardwickii Rhinopomatidae Yinpterochiroptera Insectivore 30.427 Medium 14 6.9 H P_Rou_bid_hum_l.75.Per Rousettus bidens Pteropodidae Yinpterochiroptera Frugivore 65.189 Large 24.6 5.9 F P_Sac_bil_hum_l.75.Per Saccopteryx bilineata Emballonuridae Yangochiroptera Insectivore 24.605 Small 5.9 6.1 G P_Sco_sp._hum_l.75.Per Scotoropens sp. Vespertilionidae Yangochiroptera Insectivore 16.96 Small 7.85 6.16 H P_Tad_bras_hum_l.75.Per Tadarida brasiliensis Molossidae Yangochiroptera Insectivore 23.166 Small 11.5 8.2 H P_Tap_geo_hum_l.75.Per Taphozous georgianus Emballonuridae Yangochiroptera Insectivore 37.546 Medium 22.4 9.5 H P_Taph_mau_hum_l.75.Per Taphozous mauritianus Emballonuridae Yangochiroptera Insectivore 37.71 Medium 25.9 10 H P_Thy_tri_hum_l.75.Per Thyroptera tricolor Thyropteridae Yangochiroptera Insectivore 17.716 Small 4.1 6 G

292

Appendix Table 13 Pairwise comparisons of magnitudes of humeral morphological disparity between species of different foraging guilds WHOLE BONE C F G H T TL C 0 0.01038068 0.0054748 0.00386693 0.00196047 0.00333931 F 0.081 0 0.00490588 0.00651375 0.01234115 0.00704137 G 0.335 0.321 0 0.00160787 0.00743527 0.00213549 H 0.383 0.105 0.674 0 0.0058274 0.00052762 T 0.701 0.097 0.237 0.243 0 0.00529978 TL 0.544 0.263 0.692 0.919 0.376 0 DIAPHYSIS C F G H T TL C 0 0.01478315 0.00690507 0.0051916 0.00297628 0.00471345 F 0.073 0 0.00787808 0.00959155 0.01775943 0.01006969 G 0.394 0.259 0 0.00171347 0.00988135 0.00219162 H 0.425 0.081 0.755 0 0.00816788 0.00047815 T 0.676 0.086 0.271 0.247 0 0.00768973 TL 0.543 0.259 0.758 0.948 0.368 0 EPIPHYSIS C F G H T TL C 0.00E+00 4.72E-05 1.81E-04 2.61E-04 5.79E-05 4.79E-05 F 0.766 0 2.28E-04 3.08E-04 1.07E-05 9.51E-05 G 0.252 0.128 0.00E+00 7.95E-05 2.39E-04 1.33E-04 H 0.044 0.013 0.48 0 3.19E-04 2.13E-04 T 0.763 0.952 0.18 0.057 0.00E+00 1.06E-04 TL 0.802 0.633 0.467 0.195 0.601 0.00E+00

293

Appendix peer-reviewed publication 1 Peer-reviewed published version of Chapter 2

López-Aguirre, C., Hand, S.J., Koyabu, D. et al. Postcranial heterochrony, modularity, integration and disparity in the prenatal ossification in bats (Chiroptera). BMC Evol Biol 19: 75 (2019). https://doi.org/10.1186/s12862-019-1396-1

Appendix peer-reviewed publication 2 Peer-reviewed published version of Chapter 3

López‐Aguirre, C, Hand, SJ, Koyabu, D, Son, NT, Wilson, LAB. Prenatal allometric trajectories and the developmental basis of postcranial phenotypic diversity in bats (Chiroptera). J Exp Zool (Mol Dev Evol) 332: 36– 49 (2019). https://doi.org/10.1002/jez.b.22846

294