833

The Journal of Experimental Biology 215, 833-844 © 2012. Published by The Company of Biologists Ltd doi:10.1242/jeb.065979

RESEARCH ARTICLE Is solid always best? Cranial performance in solid and fenestrated skulls

Thomas Kleinteich1,2,*, Hillary C. Maddin3, Julia Herzen4, Felix Beckmann4 and Adam P. Summers2 1Christian-Albrechts-Universität Kiel, Department of Zoology – Functional Morphology and Biomechanics, Am Botanischen Garten 1-9, 24098 Kiel, Germany, 2University of Washington, Friday Harbor Laboratories, 620 University Road, Friday Harbor, WA 98250, USA, 3University of Calgary, 2500 University Drive, Calgary, Alberta, T2N 1N4, Canada and 4Helmholtz Zentrum Geesthacht, Institute of Materials Research, Max-Planck-Straße 1, 21502 Geesthacht, Germany *Author for correspondence ([email protected])

Accepted 23 November 2011

SUMMARY (Lissamphibia: ) are characterized by a fossorial lifestyle that appears to play a role in the many anatomical specializations in the group. The skull, in particular, has been the focus of previous studies because it is driven into the substrate for burrowing. There are two different types of skulls in caecilians: (1) stegokrotaphic, where the squamosal completely covers the temporal region and the jaw closing muscles, and (2) zygokrotaphic, with incomplete coverage of the temporal region by the squamosal. We used 3-D imaging and modeling techniques to explore the functional consequences of these skull types in an evolutionary context. We digitally converted stegokrotaphic skulls into zygokrotaphic skulls and vice versa. We also generated a third, akinetic skull type that was presumably present in extinct caecilian ancestors. We explored the benefits and costs of the different skull types under frontal loading at different head angles with finite element analysis (FEA). Surprisingly, the differences in stress distributions and bending between the three tested skull types were minimal and not significant. This suggests that the open temporal region in zygokrotaphic skulls does not lead to poorer performance during burrowing. However, the results of the FEA suggest a strong relationship between the head angle and skull performance, implying there is an optimal head angle during burrowing. Supplementary material available online at http://jeb.biologists.org/cgi/content/full/215/5/833/DC1 Key words: caecilian, burrowing, skull evolution, 3-D surface manipulation, origin.

INTRODUCTION Caecilians are a monophyletic group of with skulls that Marcus et al., 1933; Carroll and Currie, 1975). This view is are thought to be highly specialized for their burrowing lifestyle; supported by the absence of a real suture between the squamosal they are wedge shaped, compact and robust (Wake, 1993). Many and the parietal in the stegokrotaphic skull. However, the discovery bones of the skull have been fused into larger compound elements, of a stem group caecilian with a stegokrotaphic skull, Eocaecilia e.g. the os basale of caecilians comprises all of the bones of the micropodia, has challenged this view and suggests that the skull base [i.e. the parasphenoid, the exoccipitals, and the caudal zygokrotaphic skull in the Rhinatrematidae is a derived condition parts of the neurocranium, including the otic capsules (Wake, 2003) and that stegokrotaphy is ancestral for all caecilians (Jenkins and (Fig.1)]. There are two distinct skull types in caecilians: (1) Walsh, 1993; Carroll, 2000; Jenkins et al., 2007). zygokrotaphic, in which the skull is fenestrated between the Despite coverage of the temporal region by the squamosal in squamosal and the parietal, and (2) stegokrotaphic, in which the the stegokrotaphic caecilian skull type, there is always a narrow skull is completely roofed. gap between the squamosal and the parietal instead of a tight Based on recently published caecilian phylogenies (Roelants et suture (Wiedersheim, 1879; Sarasin and Sarasin, 1887–1890; al., 2007; Zhang and Wake, 2009; Pyron and Wiens, 2011; Peter, 1898; Versluys, 1912; Abel, 1919; Marcus et al., 1933; Wilkinson et al., 2011) (Fig.2), the zygokrotaphic skull has evolved Goodrich, 1958; Lawson, 1963; Taylor, 1969; Nussbaum, 1977; independently several times in caecilians, in the Scolecomorphidae, Nussbaum, 1983; Wake and Hanken, 1982). This gap is supposed Typhlonectidae and (Brand, 1956; Taylor, 1969; to allow movement of the squamosal and the attached quadrate Nussbaum, 1977; Nussbaum, 1985; Wilkinson and Nussbaum, 1997; and thus plays a role in a complex cranial kinesis (Versluys, 1912; Müller et al., 2009). Zygokrotaphy in the Rhinatrematidae, however, Marcus et al., 1933; Iordansky, 1990; Iordansky, 2000). has usually been considered to be the ancestral condition for the Movement of the squamosal is related to the unique caecilian dual Gymnophiona (Nussbaum, 1977; Nussbaum, 1983; Wake, 2003; jaw closing mechanism that involves an accessory ventral jaw Müller, 2007), and the reduction of bone coverage in the temporal closing muscle (m. interhyoideus posterior) acting simultaneously skull region was considered to be homologous among caecilians, with the primary jaw closing muscles (mm. levatores mandibulae) frogs and salamanders. Thus, the completely roofed stegokrotaphic (Bemis et al., 1983; Nussbaum, 1983). The movement of the skulls of some caecilians were considered to be secondarily derived squamosal during feeding, as revealed by three-dimensional (3- in association with their burrowing lifestyle (Peter, 1898; Goodrich, D) modeling, is a small mediolateral rocking that does not expose 1958; Parsons and Williams, 1963; Nussbaum, 1983; but see a substantial gap in the skull (Kleinteich et al., 2008; Kleinteich,

THE JOURNAL OF EXPERIMENTAL BIOLOGY 834 T. Kleinteich and others

Fig.1. Caecilian cranial anatomy and regions that were Premaxillary constrained in the finite element model applied herein. Nasal Surface renderings in dorsal (left), ventral (right) and Prefrontal lateral (bottom) view based on the HRCT data for Ichthyophis cf. kohtaoensis. Caecilian skulls are compact Frontal Vomer and wedge shaped. The maxillopalatine and the os basale are compound bones that can be ontogenetically Maxillopalatine derived from separate ossification centers. For finite element analysis, we applied point loads to the area Squamosal around the nasal capsules (highlighted in pink) and Pterygoid prevented the joint areas of the occipital condyles Parietal (highlighted in green) from translation or rotation in either direction. The terms for cranial bones follow Wake (Wake, Os basale 2003).

Occipital condyle

Parietal 2 mm Point loads Squamosal Os basale Fixed displacements Prefrontal Nasal

Premaxillary Occipital condyle Maxillopalatine

2010). The stem group caecilian E. micropodia, however, differs zygokrotaphic dermophiid caecilian seraphini shows from extant caecilians by having additional bones in the temporal similar burrowing performance to other stegokrotaphic caecilians region (e.g. a presumed tabular) and a different jaw joint that is (Herrel and Measey, 2010). flat instead of a deep groove (Jenkins et al., 2007), which suggests Here, we present an experimental study where we modified the the presence of a different jaw closing mechanism, possibly 3-D geometry of zygokrotaphic and stegokrotaphic caecilian skulls without movements of the squamosal. to artificially render zygokrotaphic skulls stegokrotaphic and vice Understanding the costs and benefits of one skull type over the versa. We further modified the stegokrotaphic skull models to test other is crucial for evaluating amphibian skull evolution because it a third, akinetic skull type. The original and modified skull gives insight into how strongly the biology of a caecilian is affected geometries were tested using finite element analysis (FEA) for their by its skull. The stegokrotaphic skull would, on casual inspection, performance under the frontal loading regime that caecilians appear better suited to burrowing than the zygokrotaphic skull with encounter during burrowing. The skulls were oriented at different the large unroofed region and, for two caecilian groups with angles relative to their anterior–posterior axis to simulate varying zygokrotaphic skulls, there might actually be an ecological link to directions of the frontal loads that caecilians are likely to encounter poorer burrowing performance; rhinatrematids are considered to be by dorso-ventral movements of the head to manipulate the substrate poor burrowers that can be readily trapped during surface activity (Wake, 1993). The aims of this study were: (1) to evaluate the (Nussbaum, 1983; Gower et al., 2010); typhlonectids are mainly sensitivity of the strain distribution in caecilian skulls as the head aquatic and may therefore be relieved of the constraints of moves through different angles during burrowing; (2) to quantify burrowing (Wilkinson and Nussbaum, 1997). However, the the difference in deformation of zygokrotaphic and stegokrotaphic zygokrotaphic scolecomorphids are considered to be specialized caecilian skulls under frontal loading; (3) to determine whether burrowers (Nussbaum, 1977; Nussbaum, 1983); furthermore, the deflection during burrowing is affected by the state of the skull

Table 1. Specimens used in this study, and parameters for HRμCT imaging Total length Preparation for Energy for HRμCT Resolution of HRμCT I.D. Species (mm) Original skull type HRμCT imaging imaging (keV) dataset (μm) MNHN 1999.8360 Rhinatrema bivittatum 164 Zygokrotaphic 70% ethanol N/A 20.0 (Guérin-Méneville 1838) ZMH A08981 Ichthyophis cf. Kohtaoensis 265 Stegokrotaphic Freeze drying 30 6.8 Taylor 1960 MW 0031 taitana 212 Stegokrotaphic 70% ethanol 20 6.8 Loveridge 1935 ZMH A08984 Typhlonectes natans 330 Zygokrotaphic Freeze drying 30 6.8 (Fischer 1880) ZMH A00235 annulatus 280 Stegokrotaphic Freeze drying 30 9.2 (Mikan 1820) AK 01018 Geotrypetes seraphini 258 Zygokrotaphic Freeze drying 23 3.9 (Duméril 1859) Taxonomic authority is shown below each species name.

THE JOURNAL OF EXPERIMENTAL BIOLOGY Caecilian cranial performance 835

Stegokrotaphy Zygokrotaphy

Fig.2. Distribution of stegokrotaphic (black branches) and zygokrotaphic (white branches) skulls over caecilian phylogeny. (A)The cladogram is derived from the study by Zhang and Wake (Zhang and Wake, 2009); taxon names follow Wilkinson et al. (Wilkinson et al., 2011). *Note that Zhang and Wake (Zhang and Wake, 2009) did not include Crotaphatrema in their analysis. However, Crotaphatrema and Scolecomorphus are the only two genera within the monophyletic Scolecomorphidae (Nussbaum, 1985; Wilkinson and Nussbaum, 2006; Wilkinson et al., 2011) and thus the position of the genus Crotaphatrema in this phylogeny can be assigned with confidence. (B–G) Skulls of the specimens that we examined herein are shown as surface renderings in dorsal view based on HRCT data, and the genera they belong to are highlighted in bold in the cladogram (A). The arrows indicate the temporal fenestration in zygokrotaphic skulls. (B)Rhinatrema bivittatum; (C) Ichthyophis cf. kohtaoensis; (D) ; (E) Typhlonectes natans; (F) ; (G) Geotrypetes seraphini. roofing by generating and testing hypothetical morphologies in species were sampled broadly across caecilian phylogeny and which the skull roofing state is modified relative to the condition represent six of the nine families that are recognized by Wilkinson present in the actual species; and (4) to test whether deformation et al. (Wilkinson et al., 2011) (Fig.2). during burrowing in an akinetic skull representative of a putative The species Ichthyophis kohtaoensis was originally described as caecilian ancestor is affected by the restricted movements of the endemic to Koh Tao Island in Thailand (Taylor, 1960). It is currently squamosal. unclear if I. kohtaoensis can also be found on the mainland of Thailand or if the mainland populations belong to a different species. MATERIALS AND METHODS The specimen examined herein was derived from the pet trade and Specimens it is not possible to assign it to either the island or the mainland Six adult caecilian specimens from six different species were populations of the genus Ichthyophis of Thailand with certainty and available for this study (Table1). The specimens were made thus we prefer to refer to this specimen as Ichthyophis cf. available by the Zoological Museum Hamburg, Germany (ZMH), kohtaoensis. Mark Wilkinson (MW; Natural History Museum London, UK), Alexander Kupfer (AK; University of Siegen, Germany), and the CT imaging Muséum National d’Histoire Naturelle, Paris, France (MNHN). Our We obtained 3-D volume datasets with high-resolution micro- sampling comprised three species with zygokrotaphic skulls and computed tomography (HRCT). Except for the Rhinatrema three species with stegokrotaphic skulls (Table1). Further, the bivittatum specimen (see below), HRCT scanning was performed

THE JOURNAL OF EXPERIMENTAL BIOLOGY 836 T. Kleinteich and others

Zygokrotaphic Stegokrotaphic Akinetic Fig.3. Preparation of alternative skull architectures in caecilians. Skulls are shown as surface renderings that are based on HRCT data in dorsal view. In species that have stegokrotaphic skulls (e.g. Ichthyophis cf. kohtaoensis), we deleted voxels in the dorsal region of the squamosal to generate zygokrotaphic skulls. Vice versa, in

cf. kohtaeoensis species with zygokrotaphic skulls, we added voxels to the dorsal edge of the squamosal to convert them to stegokrotaphic skulls. Further, based on the stegokrotaphic condition, we Ichthyophis 2 mm merged the squamosal with the parietal and parts of the otic capsule to Original architecture generate akinetic caecilian skulls. Geotrypetes seraphini

Original architecture with synchrotron-based x-ray radiation at the beamline W2 of the limitations in the computer hardware for visualization of synchrotron DORIS III accelerator ring of the German Electron Synchrotron HRCT datasets, we pairwise combined neighboring voxels in all (DESY) in Hamburg, Germany. This beamline is operated by the three dimensions of the original synchrotron HTCT datasets (two- Helmholtz Center Geesthacht, Germany. For synchrotron-based x- fold binning), which decreased the resolution of the datasets by a ray radiation HRCT, we used monochromatic x-rays and adjusted factor of two. the energy of the x-ray beam individually to the samples (Table1). The HRCT datasets were then loaded into Amira 5.2 (Visage Details of the setup for HRCT imaging at beamline W2 of the Imaging GmbH, Berlin, Germany). We segmented the cranium German Electron Synchrotron have been published previously except for the lower jaw and the stapes from the volumetric dataset (Beckmann et al., 2006; Kleinteich et al., 2008). The R. bivittatum by using the LabelField function in Amira. Segmentation of HRCT specimen was HRCT-scanned at the 3-D Morphometrics data with the LabelField function results in a so-called Labels dataset Laboratory facility at the University of Calgary, Canada. The scan that contains the assignments of voxels in the HRCT data to was performed on a Scanco CT35 scanner (Scanco Medical AG, specified Materials, in our case the only Material in the Labels Brüttisellen, Switzerland) with the beam energy set at 55kVp and dataset was the segmented cranium. Based on the Labels dataset 72A, and a voxel size of 20m. derived from the original anatomy, we generated two artificial The Boulengerula taitana and R. bivittatum specimens were anatomies for each specimen: (1) by removing voxels in the Labels HRCT scanned in 70% ethanol; the remainder specimens were dataset from the dorsal region of the squamosal in stegokrotaphic decapitated, the heads were exposed to –80°C for 3h and then freeze species (i.e. converting them to a zygokrotaphic skull type) or by dried with a Lyovac GT2 freeze drying system (Leybold-Heraeus adding voxels to the dorsal edge of the squamosal in zygokrotaphic GmbH, Hanau, Germany) for 24h according to the procedure species (i.e. converting them to stegokrotaphic skulls), and (2) by described previously (Meryman, 1960; Meryman, 1961) prior to filling the gap between the squamosal and the parietal with voxels HRCT imaging. Freeze drying is known to increase the contrast to make a tight connection of the squamosal to the remainder in x-ray images (Follett, 1968). Compared with other drying cranium (i.e. converting them to akinetic skulls) (Fig.3). Other than methods, freeze-drying results in less volume shrinkage (Boyde, in previously published studies on manipulations of 3-D datasets 1978) and, especially for hard tissues, shrinkage effects due to freeze (Stayton, 2009; Stayton, 2011), our approach is not based on drying are considered to be negligible. landmark data and the geometric morphometric method. The Labels datasets were then used to calculate surface models Processing of the HRCT data and finite element modeling of the skulls. Each species was represented by three different surface HRCT imaging at DESY resulted in volume datasets with models based on the original and modified Labels datasets (Fig.3). resolutions from 3.9 to 9.2m; the dataset of the R. bivittatum Surfaces that are generated with Amira contain an unnecessarily specimen had a resolution of 20.0m (Table1). Because of high number of polygons, and a reduction in the number of

THE JOURNAL OF EXPERIMENTAL BIOLOGY Caecilian cranial performance 837 polygons usually has no or only little effect on the appearance of the surface model. We reduced the number of polygons until we could visually detect a decrease in the quality of the surface representation of our HRCT datasets. Surfaces were edited in Amira to remove triangular geometries with low aspect ratios and intersecting polygons. We scaled all surface models to identical surface areas to allow for direct comparisons of calculated stresses between the different skulls (Dumont et al., 2009). Original surface areas and linear scaling factors that we applied in the x, y and z directions are shown in supplementary material TableS1. FEA was performed with Marc Mentat 2005 R3 (MSC Software Corporation, Santa Ana, CA, USA) and the surface models were Fig.4. Predicted deformation of the skull in Ichthyophis cf. kohtaoensis under a frontal load when the skull is oriented parallel to the load (head imported with a 3D solid geometry from Amira. The models angle of 0deg herein) in lateral view. The skull shape before application of consisted of approximately 2.3million to 3.2million tetrahedral the load is shown in transparency in the background. This deformation is elements (supplementary material TableS1). We treated the elements upscaled by a factor of 10 to increase the visibility of the deformation. The in the models as an isotropic material with a Young’s modulus of rostral parts of the skull are moved ventrally, which causes a slight ventral 10GPa and a Poisson’s ratio of 0.3. These values have been used bending of the skull under load. in previous studies that involved FEA to assess patterns in non- mammalian tetrapod skull evolution (e.g. Moazen et al., 2008; Moazen et al., 2009). Further, Dumont et al. (Dumont et al., 2009) maximum Von Mises stresses in the regions of the constraints, which demonstrated that FEA can be a useful tool for comparisons of can confound the interpretation of maximum Von Mises stress values shapes, even in the absence of knowledge on the exact material (Dumont et al., 2009). Further, because maximum Von Mises stress attributes in vivo. only represents one node in the model, the key value can refer to We applied two constraints (in Marc Mentat Boundary different regions in different specimens or even different regions in Conditions) to the models (Fig.1): (1) the nodes that define the joint the same specimen under different loading conditions. area of the occipital condyles were prevented from movements by We calculated mean Von Mises stresses under the assumption applying a Fixed Displacement of zero for translations and rotations that skulls that perform better during burrowing will show a along each axis to them (Fig.1, green areas); (2) nodes in the area decrease in Von Mises stress over all elements. Mean Von Mises around the nasal capsule that define the rostral surface of the skull stress is supposed to be less affected by the constraints of the finite were selected and we applied Point Loads to them that summed up element models and accounts for all nodes in the model. to a force of 10N (Fig.1, pink areas). A burrowing force of 10N While Von Mises stress scales linearly with the force per surface lies well within the range of forward pushing forces that were area ratio in finite element models, total strain energy scales with reported for caecilians; in vivo measurements showed that peak the force per volume ratio (Dumont et al., 2009). Although the finite forward pushing forces of caecilians can be up to 20N (O’Reilly element models herein had similar force per surface area ratios, their et al., 1997). Supplementary material TableS1 shows the numbers volumes differed slightly due to different cranial shapes. We of constrained nodes; the total number of constrained nodes is less calculated a force scale factor (supplementary material TableS1) than 1% of the available nodes in the models, and small variations that is based on the differences in the volume of the models. We in the number of constrained nodes between different models are then multiplied the total strain energies that resulted from the FEA assumed to be negligible. To account for different angles of the with the squared force scale factors [because total strain energy head relative to the axis along which caecilians are moving forward, scales with the square of the force (Dumont et al., 2009)] to calculate we incrementally rotated the skulls in five-degree steps for –15deg, total strain energies for equally scaled force per volume ratios. –10deg, –5deg, 0deg, 5deg, 10deg, 15deg relative to the neutral Total strain energy and Von Mises stresses are a standard output axis of the skull (as defined by the occipital condyles and the rostral for FEA with Marc Mentat. Values for Von Mises stresses for every tip of the nasal capsule). node were imported to the statistical computing environment R We used three different measurements to describe the 2.13.1 (http://www.r-project.org) to calculate mean Von Mises performance of the different skull models under the loading regime stress. We fitted the distribution of total strain energy and mean we applied herein: (1) total strain energy, (2) maximum Von Mises Von Mises stress to a raw polynomial function of second order over stress and (3) mean Von Mises stress. Total strain energy equals the different head angles by using a linear model in R. From the the sum of all deformations of the single elements in a model, resulting regression equation, we calculated the angles at which both multiplied by the applied force. This measurement reflects the performance measurements were minimal, as well as the angles amount of work that is put into elastic deformation of the skull. We where total strain energy and mean Von Mises stress were increased can assume that the energy that caecilians invest to load their skulls by 10, 20 and 30% compared with the minimum. For maximum is meant to push the skulls forward into the substrate and not to Von Mises stress we did not calculate a polynomial regression store elastic energy by bending the skulls. Thus, caecilians are because values for maximum Von Mises stress at different head expected to minimize total strain energy on their skulls, i.e. to be angles actually refer to different nodes and thus to different regions more energy efficient during burrowing. in the models. Maximum Von Mises stress is a performance measurement that The effects of skull type and head angle on total strain energy, indicates how close the models are to failure. Structures that maximum Von Mises stress and mean Von Mises stress were encounter higher Von Mises stress are closer to failure than evaluated by calculating a two-way analysis of variance (ANOVA) structures with lower values for Von Mises stress. However, it is for each performance measurement. Further, we grouped known that the constraints on finite element models (i.e. Fixed stegokrotaphic and zygokrotaphic skulls as kinetic skulls and Displacements and Point Loads) can cause artificially high compared them with akinetic skulls by using three-way ANOVAs

THE JOURNAL OF EXPERIMENTAL BIOLOGY 838 T. Kleinteich and others

1.80E+07 Rhinatrema bivittatum Ichthyophis cf. kohtaoensis Fig.5. The effect of head angle during 1.60E+07 burrowing on mean Von Mises stress. 1.40E+07 Stegokrotaphy For finite element analysis, we first 1.20E+07 Zygokrotaphy rotated the models so that in lateral view the rostral tip of the nasal 1.00E+07 Akinetic capsule was aligned with the caudal- 8.00E+06 most tip of the occipital condyles. This 6.00E+06 was defined as the neutral axis (i.e. 4.00E+06 0deg). We then rotated the models to 2.00E+06 simulate different head angles ranging from –15deg to +15deg during burrowing. In all species, we 1.80E+07 Boulengerula taitana Typhlonectes natans observed an optimal head angle 1.60E+07 where mean Von Mises stress 1.40E+07 becomes minimal. This angle lies 1.20E+07 between 0deg and –6deg in all specimens examined herein, 1.00E+07 independent of the species and the 8.00E+06 skull architecture. 6.00E+06 4.00E+06 2.00E+06 Mean Von Mises stress (Pa) Mean Von

1.80E+07 Siphonops annulatus Geotrypetes seraphini 1.60E+07 1.40E+07 1.20E+07 1.00E+07 8.00E+06 6.00E+06 4.00E+06 2.00E+06

–15 –10 –5 0 5 10 15 –15 –10 –5 0 5 10 15

Skull angle (deg) that considered total strain energy, mean Von Mises stress and correlation coefficients (R2>0.97) and a significant relationship maximum Von Mises stress as dependent on the presence of cranial between mean Von Mises stress and head angle and between total kinesis, the species and the head angle. All statistical calculations strain energy and head angle (P<0.00045) for all models tested were performed within the statistical computing environment R 2.13.1. herein (supplementary material TableS4). This shows that the correlation between head angle and skull performance during RESULTS burrowing can be described as a parabola. The effect of head angles during burrowing Fig.7 shows the distributions of Von Mises stress over the skulls Application of a frontal load that is parallel to the neutral axis of of caecilians in dorsal and ventral view at two different head angles the skull of caecilians (i.e. at a 0deg head angle) causes a slight relative to the neutral axis of the skulls. At –10deg (i.e. the nose is posteroventral bending of the nasal capsule region (Fig.4). Variation elevated relative to the occipital condyles) and +10deg (i.e. the nose in the head angle will alter this pattern and either increase this effect is lowered relative to the occipital condyles), the highest Von Mises if the nose is lowered relative to the occipital condyles (i.e. positive stresses are concentrated in the caudal parts of the skulls. High head angles herein) or cause the nasal capsule region to be pushed stresses occur around the occipital condyles and the caudal parts of dorsally instead of ventrally if the nose is elevated compared with the os basale that comprise the otic capsules. At +10deg, Von Mises the occipital condyles (i.e. negative head angles herein). stresses are much higher on the ventral surface of the skulls than Starting from the neutral axis (0deg), mean Von Mises stress at –10deg. The area of high Von Mises stress that is concentrated (Fig.5, supplementary material TableS2) and total strain energy around the premaxillaries and the vomers at –10deg extends all over (Fig.6, supplementary material TableS3) will increase as either the the ventral surface of the skull from the premaxillary to the caudal nose is lifted relative to the occipital condyles or vice versa. Mean parts of the os basale at +10deg. Further, compared with the situation Von Mises stress and total strain energy differ significantly at at –10deg, the dorsal skull bones encounter higher Von Mises different head angles (mean Von Mises stress, d.f.6, F318.23, stresses at +10deg, except for in Typhlonectes natans (Fig.7). P<0.0001; total strain energy, d.f.6, F152.96, P<0.0001). A linear Based on the results of the polynomial regression, we calculated model fit of the observed distributions of mean Von Mises stress the head angles that were optimal for burrowing as well as the angles (Fig.5) and total strain energy (Fig.6) over the different head angles for which mean Von Mises stress and total strain energy increased onto a polynomial function of the second order results in high by 10, 20 and 30% (Table2). In all models that we tested, the optimal

THE JOURNAL OF EXPERIMENTAL BIOLOGY Caecilian cranial performance 839

1.20E–003 Rhinatrema bivittatum Ichthyophis cf. kohtaoensis Fig.6. The effect of head angle during burrowing on total strain energy. The neutral axis (0 deg) 1.00E–003  Stegokrotaphy corresponds to a line from the rostral tip of the 8.00E–004 Zygokrotaphy nasal capsule to the caudal-most tip of the occipital Akinetic condyles. Rotation in the negative direction was 6.00E–004 defined as elevating the nose relative to the 4.00E–004 occipital condyles; lowering of the nose was considered a rotation in the positive direction. 2.00E–004 Similar to the effect of head angles on mean Von Mises stress (Fig.5), all species examined herein 1.20E–003 Boulengerula taitana Typhlonectes natans showed an optimal head angle where total strain energy becomes minimal. The optimal head angle 1.00E–003 lies between 0deg and –5deg in all specimens examined herein, independent of the species and 8.00E–004 the skull architecture. 6.00E–004

4.00E–004

Total strain energy (J) Total 2.00E–004

1.20E–003 Siphonops annulatus Geotrypetes seraphini

1.00E–003

8.00E–004

6.00E–004

4.00E–004

2.00E–004

–15 –10 –5 0 5 10 15 –15 –10 –5 0 5 10 15

Skull angle (deg) head angle for burrowing was slightly off the neutral axis in a (stegokrotaphic and zygokrotaphic T. natans); for total strain negative direction, i.e. with a slightly elevated nose. Optimal head energy, the optimal angles were found to range from –4.1deg angles to minimize mean Von Mises stress were calculated to lie (stegokrotaphic I. kohtaoensis) to –1.7deg (akinetic B. taitana) between –6.1deg (stegokrotaphic R. bivittatum) and –2.5deg (Table2). The offset angles from the optimum that result in an

Table 2. Total strain energy and mean Von Mises stress at optimal head angles and offset angles from the optimal angle for a 10%, 20% and 30% increase in performance parameters Performance parameter: total strain energy Performance parameter: mean Von Mises stress Offset from Offset from Offset from Offset from Offset from Offset from optimal optimal optimal optimal optimal optimal angle at angle at angle at angle at angle at angle at Parameter 10% 20% 30% Optimal Parameter 10% 20% 30% Optimal at optimal increase increase increase angle at optimal increase increase increase Species Skull type angle (deg) angle (J) (deg) (deg) (deg) (deg) angle (Pa) (deg) (deg) (deg) Rhinatrema Zygokrotaphic –3.49 6.71E–05 2.06 2.91 3.57 –5.70 5.95E+06 5.07 7.17 8.79 bivittatum Stegokrotaphic –3.58 7.28E–05 2.06 2.91 3.57 –6.14 6.25E+06 5.20 7.35 9.00 Akinetic –3.44 6.63E–05 2.06 2.91 3.57 –6.07 6.09E+06 5.23 7.39 9.06 Ichthyophis cf. Stegokrotaphic –4.09 9.06E–05 2.21 3.13 3.84 –5.01 5.91E+06 5.32 7.52 9.21 kohtaoensis Zygokrotaphic –3.98 8.81E–05 2.20 3.11 3.81 –4.56 5.65E+06 5.22 7.39 9.05 Akinetic –3.60 7.53E–05 1.97 2.79 3.42 –4.38 6.09E+06 5.06 7.16 8.76 Boulengerula Stegokrotaphic –1.82 7.27E–05 1.56 2.21 2.71 –3.31 6.23E+06 4.63 6.55 8.03 taitana Zygokrotaphic –1.97 7.35E–05 1.56 2.21 2.70 –3.35 6.32E+06 4.60 6.51 7.97 Akinetic –1.74 6.10E–05 1.43 2.03 2.48 –3.26 6.09E+06 4.50 6.36 7.80 Typhlonectes Zygokrotaphic –2.23 8.40E–05 1.63 2.30 2.82 –2.45 5.67E+06 4.51 6.38 7.82 natans Stegokrotaphic –2.20 8.78E–05 1.62 2.30 2.81 –2.49 5.75E+06 4.54 6.42 7.86 Akinetic –2.14 7.71E–05 1.55 2.20 2.69 –2.70 5.77E+06 4.53 6.41 7.85 Siphonops Stegokrotaphic –2.99 1.32E–04 2.14 3.03 3.71 –4.47 6.70E+06 5.13 7.26 8.89 annulatus Zygokrotaphic –3.03 1.33E–04 2.17 3.07 3.76 –4.45 6.64E+06 5.09 7.20 8.81 Akinetic –2.84 1.13E–04 2.00 2.83 3.46 –4.64 6.76E+06 5.11 7.23 8.85 Geotrypetes Zygokrotaphic –2.21 1.13E–04 2.16 3.05 3.74 –3.53 6.16E+06 5.10 7.21 8.83 seraphini Stegokrotaphic –2.13 1.19E–04 2.15 3.05 3.73 –3.58 6.37E+06 5.20 7.35 9.00 Akinetic –2.20 1.13E–04 2.14 3.03 3.72 –3.88 6.44E+06 5.27 7.45 9.12

THE JOURNAL OF EXPERIMENTAL BIOLOGY 840 T. Kleinteich and others

Fig.8 shows the distribution of Von Mises stress on all the –10 deg +10 deg different models we tested herein at the optimal angle (based on Dorsal Ventral Dorsal Ventral mean Von Mises stress; Table2, Fig.5). Comparable to the situation at –10deg and +10deg head angle, the highest Von Mises stresses occur at the premaxillaries and vomers, and close to the occipital condyles. However, at the optimal angle, Von Mises stress in the caudal parts and on the dorsal surface of the skull is lower compared

bivittatum with at –10deg and +10deg (Fig.7). Von Mises stress on the ventral Rhinatrema surface of the skulls at the optimal angle (Fig.8) is higher compared 2 mm with those at a head angle of –10deg (Fig.7) and extends over a wider area; compared with the situation at+10deg (Fig.7), the ventral surface of the skulls experiences lower Von Mises stresses at the optimal angle (Fig.8). For maximum Von Mises stress, the observed values actually Ichthyophis refer to different nodes in different regions of the same model. Thus, cf. kohtaoensis the collected data on maximum Von Mises stress at different head angles of the same model is not strictly continuous and we did not calculate regression equations for maximum Von Mises stress. However, Fig.9 and supplementary material TableS5 show the values for maximum Von Mises stress for the models we tested taitana herein at varying head angles. Comparable to mean Von Mises stress Boulengerula and total strain energy, maximum Von Mises stress tends to increase if the nose is lifted beyond an optimal angle that appears to lie between –10deg and 0deg and also increases if the nose is lowered above the optimal angle. The three tested models that were derived from the R. bivittatum HRCT dataset, however, show a steady decrease in maximum Von Mises stress with increasing head angle, natans which is likely to be an artifact that is caused by a high point load Typhlonectes around a few nodes in these models. Exclusion of these nodes results in the same pattern that we observed in the other caecilian species.

Performance of the three different skull types The three skull types showed very similar behaviors under the frontal loading regime that we applied herein. Values for total strain energy annulatus Siphonops (d.f.2, F0.51, P0.60), maximum Von Mises stress (d.f.2, F0.37, P0.69) and mean Von Mises stress (d.f.2, F0.49, P0.62) were all found to not be significantly different for the three skull types tested. Fig.8 compares the distributions of Von Mises stress on the dorsal and the ventral surfaces of the different skull models for each species

seraphini with each other. In none of the six caecilian species do the dorsal Geotrypetes parts of the squamosal experience notable Von Mises stresses under a frontal loading regime. Thus, varying the shape of the squamosal by removing or adding material does not have an effect on the 0 4E+07 Von Mises stress (Pa) distribution of Von Mises stress over the skull. Further, because the squamosal is not under stress, the shape of the squamosal, and thus Fig.7. Distribution of Von Mises stress over caecilian skulls at two different the skull type, has no effect on the response of the skull to different head angles for burrowing relative to the neutral axis (i.e. from the rostral head angles (Figs5,6,9). The interactions of skull type with head tip of the nasal capsule to the caudal-most tip of the occipital condyle). In angle were found to be not statistically significant for total strain all species and for both angles shown here, the higher stresses are found on the ventral surface of the skulls. Rotation of the skulls in either direction energy (d.f.12, F0.03, P>0.99), maximum Von Mises stress causes an increase in Von Mises stress in the caudal parts of the skull, (d.f.12, F0.01, P>0.99), and mean Von Mises stress (d.f.12, around the region of the occipital condyles and the caudal parts of the os F0.03, P>0.99). basale. Elevation of the nose during burrowing (i.e. negative angles herein) However, akinetic skulls experience slightly higher Von Mises reduces Von Mises stress on the ventral surface of the skull, while lowering stresses at the tooth-bearing bones that define the outline of the skulls of the nose (i.e. positive angles herein) results in higher Von Mises stress in ventral view (i.e. the lateral regions of the premaxillary and the along the ventral surface of the skulls. maxillopalatine) and slightly lower Von Mises stresses at the central ventral bones compared with stegokrotaphic and increase of mean Von Mises stress by 30% were calculated to lie zygokrotaphic skulls (Fig.8). By combining zygokrotaphic and between 7.8deg and 9.2deg; for a 30% increase in total strain energy stegokrotaphic skulls as kinetic skull morphologies and comparing the offset angles were even smaller and ranged from 2.5deg to them to the akinetic skull models in a three-way ANOVA that 3.8deg (Table2). In both performance parameters, B. taitana and accounts for presence of skull kinesis, head angle and species, we T. natans show the smallest offset angles. found that total strain energy was significantly different between

THE JOURNAL OF EXPERIMENTAL BIOLOGY Caecilian cranial performance 841

Fig.8. Distribution of Von Mises stress Zygokrotaphic Stegokrotaphic Akinetic over the models of caecilian skulls that are based on the three different skull Dorsal Ventral Dorsal Ventral Dorsal Ventral architectures (i.e. zygokrotaphic, stegokrotaphic, akinetic) that are considered herein at optimal head angle for burrowing (mean Von Mises stress minimal). In all species, independent of the

bivittatum skull types, Von Mises stresses are Rhinatrema notably higher on the ventral surfaces, 2 mm with a concentration of stress on the bones that form the palate (i.e. premaxillary, vomer, parasphenoid region of the os basale). The different skull architectures only have minor effects on the distribution of Von Mises stress over

Ichthyophis the skulls. cf. kohtaoensis taitana Boulengerula natans Typhlonectes annulatus Siphonops seraphini Geotrypetes

0 Von Mises stress (Pa) 4E+07 kinetic and akinetic skulls (d.f.1, F27.6, P<0.0001). In all the been present in Paleozoic caecilian and tetrapod ancestors. However, species we investigated herein, we found consistently lower (by we found that akinetic skulls have consistently lower values for total approximately 5–15%) values for total strain energy in the akinetic strain energy. We further present evidence that there is an optimal skulls compared with kinetic skulls (supplementary material head angle for burrowing in caecilians around which total strain TableS3). Mean Von Mises stress (d.f.1, F1.53, P0.22) and energy and mean Von Mises stress are minimal. At the optimal head maximum Von Mises stress (d.f.1, F0.01, P0.91) were found angle, stress close to the occipital condyles is minimized, presumably to not significantly differ between kinetic and akinetic skulls. due to a reduction in torque around the occipital joint. We also show that most of the load that caecilian skulls encounter during burrowing DISCUSSION is transmitted along the ventral surface of the skull. Thus, the ventral Our results demonstrate that the two different caecilian skull skull bones are more likely to be shaped by demands of fossoriality architectures (i.e. stegokrotaphy and zygokrotaphy) do not have an than the bones on the dorsal surface of the skull. effect on skull performance under a frontal loading regime, which Traditionally, there has been an emphasis on the different caecilians encounter while pushing the skull into the substrate during caecilian skull types as they relate to the question of whether the burrowing. Further, the kinetic caecilian skulls show similar patterns solid skull roof (stegokrotaphy) is the ancestral condition (Marcus of Von Mises stress distribution to akinetic skulls, which may have et al., 1933; Carroll and Currie, 1975) or whether it is derived as

THE JOURNAL OF EXPERIMENTAL BIOLOGY 842 T. Kleinteich and others

1.20E+009 Rhinatrema bivittatum Ichthyophis cf. kohtaoensis Fig.9. The effect of head angle during burrowing on maximum Von Mises 1.00E+009 stress. The neutral axis (0deg) corresponds to a line from the rostral 8.00E+008 tip of the nasal capsule to the caudal- 6.00E+008 most tip of the occipital condyles. Rotation in the negative direction was 4.00E+008 Stegokrotaphy defined as elevating the nose relative Zygokrotaphy to the occipital condyles; lowering of 2.00E+008 Akinetic the nose was considered a rotation in the positive direction. Similar to the effect of head angles on mean Von 1.20E+009 Boulengerula taitana Typhlonectes natans Mises stress (Fig.5) and total strain energy (Fig.6), the distribution of 1.00E+009 maximum Von Mises stress over 8.00E+008 different head angles suggests the presence of an optimal head angle 6.00E+008 where peak stresses become minimal. The derivation from this pattern that is 4.00E+008 seen in Rhinatrema bivitattum is likely caused by high point loads. 2.00E+008 Maximum Von Mises stress (Pa) Maximum Von 1.20E+009 Siphonops annulatus Geotrypetes seraphini 1.00E+009

8.00E+008

6.00E+008

4.00E+008

2.00E+008

–15 –10 –5 0 5 10 15 –15 –10 –5 0 5 10 15

Skull angle (deg) an adaptation to the burrowing lifestyle of caecilians (Peter, 1898; burrowing performance. In the zygokrotaphic skulls of the Abel, 1919; Goodrich, 1958; Parsons and Williams, 1963). There Rhinatrematidae, the jaw closing muscles pass through the temporal is no strong consensus among authors as to which of the two fenestrae and reach towards the dorsal midline of the skull roof; in caecilian skull types is ancestral (Wake, 2003; Carroll, 2007; all other caecilians with zygokrotaphic skulls, the jaw closing Carroll, 2009; Jenkins et al., 2007). Although we did not gather muscles do not extend beyond the dorsal edge of the squamosal data that directly bears on the question of caecilian ancestry [e.g. (Nussbaum, 1977; Nussbaum, 1983; Wake, 2003). However, we temnospondyl versus lepospondyl origins (see Anderson, 2008)], would like to point out that although the jaw closing muscles in we have shown that with respect to the current utility of the skull caecilians with secondarily derived zygokrotaphic skulls do not during burrowing, a stegokrotaphic skull does not result in improved extend as far dorsal as in species of the Rhinatrematidae, performance over the zygokrotaphic skull. This suggests that if zygokrotaphy does leave more space for the jaw closing muscles. stegokrotaphy is the derived condition within caecilians, it did not Kleinteich et al. showed that the jaw closing muscles that act on evolve primarily as an adaptation to burrowing. Additionally, if the zygokrotaphic skull of T. natans have a higher contribution to stegokrotaphy is the ancestral condition, the evolution of temporal total bite force than the jaw closing muscles in the stegokrotaphic fenestration (zygokrotaphy) does not imply a coupled decrease in species B. taitana, Siphonops annulatus and I. cf. Kohtaoensis cranial performance during burrowing. No matter whether (Kleinteich et al., 2008; Kleinteich, 2010) . In the zygokrotaphic G. stegokrotaphy or zygokrotaphy is ancestral for caecilians, seraphini, the jaw closing muscles extend onto the dorsal surface zygokrotaphy evolved at least three times independently, i.e. within of the parietal (T.K., personal observation), while in stegokrotaphic the Scolecomorphidae, Typhlonectidae and Dermophiidae. If caecilians, the jaw closing muscles originate from the lateral edge zygokrotaphy in the Rhinatrematidae is the derived condition, it may of the parietal (Edgeworth, 1935; Lawson, 1965; Iordansky, 1996; have even evolved four times (Fig.2). The lack of an effect on the Iordansky, 2010). most constraining function of the skull, its use for burrowing, may In all species that we tested, independent of the skull type, we have facilitated the multiple evolution of zygokrotaphy. found that Von Mises stress was higher on the ventral surface of Zygokrotaphy very likely results in more space for the jaw closing the skull than on the dorsal surface. This is because at an optimal muscles (mm. levatores mandibulae) and thus might be related to head angle for burrowing, the sequence of the ventral bones (i.e. differences in feeding biomechanics rather than to differences in premaxillary, vomer and parasphenoid portion of the os basale) is

THE JOURNAL OF EXPERIMENTAL BIOLOGY Caecilian cranial performance 843 almost in line with an axis that directly connects the nasal capsule for this study. We appreciate the help of Tamer Fawzy (Hamburg, Germany), with the occipital condyles. High Von Mises stresses on the ventral Thomas Dejaco (Innsbruck, Austria) and Susanne Kühnel (Jena, Germany) as co- workers during CT imaging sessions. We thank Benedikt Hallgrímsson (University surface of the skull are caused by a slight downward bending of the of Calgary, AB, Canada) for providing time on the Scanco CT35. T.K. expresses skull under load (Fig.4), causing the ventral surface of the skull to his gratitude to Tom Daniel (University of Washington, Seattle, WA, USA), who gave a helpful introduction into finite element analysis and provided access to the be under compression. The ventral bending of the skull follows the finite element software. Two anonymous reviewers provided valuable comments pattern of cranial kinesis that was previously predicted for caecilians that helped to improve a previous version of this paper. by Iordansky (Iordansky, 1990; Iordansky, 2000). However, although much attention has been paid to the skull roof as an FUNDING adaptation for burrowing (Peter, 1898; Nussbaum, 1977; Wake and This research project was funded by the Volkswagenstiftung under the funding Hanken, 1982; Bemis et al., 1983; Nussbaum, 1983; Wake, 1993), initiative ʻEvolutionary Biologyʼ [I/84 206 to T.K.]. the role of the palate during burrowing has been neglected. Preliminary examination of the nature of the sutures between palatal REFERENCES Abel, O. (1919). Die Stämme der Wirbeltiere. Berlin, Leipzig: De Gruyter. bones reveals broadly overlapping joints filled by a dense network Anderson, J. S. (2008). Focal review: the origin(s) of modern amphibians. Evol. Biol. of collagen fibers, suggesting at least minimal kinesis in this region 35, 231-247. Beckmann, F., Donath, T., Fischer, J., Dose, T., Lippmann, T., Lottermoser, L., (H.C.M., personal observation). The anatomy of these sutures is Martins, R. V. and Schreyer, A. (2006). New developments for synchrotron- consistent with a compression-sink model that would alleviate stress radiation-based microtomography at DESY. Proc. SPIE 6318. Doi: 631810-631810- caused under a compression scenario (Wake, 1993). This hypothesis 11. Bemis, W., Schwenk, K. and Wake, M. H. (1983). Morphology and function of the has yet to be tested empirically and is currently under investigation. feeding apparatus in Dermophis mexicanus (Amphibia: Gymnophiona). Zool. J. Linn. Our results show that a skull that is akinetic but otherwise similar Soc. 77, 75-96. Boyde, A. (1978). Pros and cons of critical point drying and freeze drying for SEM. to a skull of extant caecilians has no major advantages or Scanning Electron Microsc. 2, 303-314. disadvantages when exposed to the conditions mimicking burrowing. Brand, D. (1956). On the cranial morphology of Scolecomorphus uluguruensis (Barbour and Loveridge). Ann. Univ. Stellenbosch 32A, 1-25. However, the plots of Von Mises stress reveal slightly decreased Carroll, R. L. (2000). Eocaecilia and the origin of caecilians. In Amphibian Biology, Von Mises stress on the vomer and the os basale and a slightly Volume 4, Paleontology (ed. H. Heatwole, and R. L. Carroll), pp. 1402-1411. Chipping Norton, NSW: Surrey Beatty and Sons. increased Von Mises stress on the tooth-bearing bones. Further, Carroll, R. L. (2007). The palaeozoic ancestry of salamanders, frogs and caecilians. akinetic skulls bend less, which is indicated by significantly lower Zool. J. Linn. Soc. 150, 1-140. total strain energy. There might be a trade-off between reduction Carroll, R. L. (2009). The Rise of Amphibians: 365 Million Years of Evolution. Baltimore, MD: The Johns Hopkins University Press. of skull bending and loading of the bones that define the ventral Carroll, R. L. and Currie, P. J. (1975). Microsaurs as possible apodan ancestors. outline of a caecilian skull. The laterally exposed parts of the Zool. J. Linn. Soc. 57, 229-247. Dumont, E. R., Grosse, I. R. and Slater, G. J. (2009). Requirements for comparing premaxillary, the maxillopalatine and also the squamosal tend to be the performance of finite element models of biological structures. J. Theor. Biol. 256, much thinner bones than the central bones, especially the os basale. 96-103. Edgeworth, F. (1935). The Cranial Muscles of Vertebrates. London: Cambridge We hypothesize that, besides the advantages of a kinetic skull during University Press. biting that were discussed in detail by Summers and Wake (Summers Follett, W. (1968). Radiography of freeze-dried fishes. Copeia 1968, 867-869. Goodrich, E. S. (1958). Studies on the Structure and Development of Vertebrates. and Wake, 2005) and Kleinteich et al. (Kleinteich et al., 2008), New York: Dover Publications. detachment of the squamosal from the dorsal dermal bones also Gower, D. J., Wilkinson, M., Sherratt, E. and Kok, P. J. R. (2010). A new species of might be a mechanism to guide loads towards the ventral centre of Rhinatrema Duméril and Bibron (Amphibia: Gymnophiona: Rhinatrematidae) from . Zootaxa 2391, 47-60. the skulls where thicker bones, and extensive overlapping (and Herrel, A. and Measey, G. J. (2010). The kinematics of locomotion in caecilians: possibly weakly kinetic) joints, are present. effects of substrate and body shape. J. Exp. Zool. A 313, 301-309. Iordansky, N. N. (1990). Evolution of cranial kinesis in lower tetrapods. Neth. J. Zool. Stress on the ventral surface in kinetic and akinetic skulls can be 40, 32-54. reduced by raising the nose slightly compared with the occipital Iordansky, N. N. (1996). Evolution of the musculature of the jaw apparatus in the Amphibia. Adv. Amphib. Res. Form. Soviet Union 1, 3-26. condyles. This reduces ventral bending and at some point will push Iordansky, N. N. (2000). Cranial kinesis in the amphibia: a review. Zh. Obshch. Biol. the tip of the nose dorsally rather than ventrally. However, this results 61, 102-118. Iordansky, N. N. (2010). Pterygoideus muscles and other jaw adductors in amphibians in an increase in Von Mises stress on the dorsal surface of the skull and reptiles. Biol. Bull. Russ. Acad. Sci. 37, 905-914. and in the caudal region of the os basale. Raising the nose beyond Jenkins, F. A. and Walsh, D. M. (1993). An early Jurassic caecilian with limbs. Nature an optimal angle results in higher mean Von Mises stresses, higher 365, 246-250. Jenkins, F. A., Walsh, D. M. and Carroll, R. L. (2007). Anatomy of Eocaecilia maximum Von Mises stresses and higher total strain energies. On micropodia, a limbed caecilian of the Early Jurassic. Bull. Mus. Comp. Zool. 158, the other hand, lowering the nose beyond an optimal angle during 285-365. Kleinteich, T. (2010). Ontogenetic differences in the feeding biomechanics of burrowing results in higher Von Mises stresses on the ventral and oviparous and viviparous caecilians (Lissamphibia: Gymnophiona). Zoology (Jena) dorsal surface of the skulls and at the occipital condyles and also 113, 283-294. Kleinteich, T., Haas, A. and Summers, A. P. (2008). Caecilian jaw-closing decreases overall performance. Even slight deviations from the mechanics: integrating two muscle systems. J. R. Soc. Interface 5, 1491-1504. optimal head angle in either direction might be of biological Lawson, R. (1963). The anatomy of Hypogeophis rostratus Cuvier (Amphibia: Apoda or Gymnophiona). Part I, The skin and skeleton. Proc. Univ. Durham Phil. Soc. 13, significance. This leads us to predict that dorso-ventral movements 254-273. of the skull during burrowing will be restricted to a narrow range Lawson, R. (1965). The anatomy of Hypogeophis rostratus Cuvier (Amphibia: Apoda near the zero angle of incidence. Although the concept of optimality or Gymnophiona). Part II, The musculature. Proc. Univ. Newcastle Phil. Soc. 1, 52- 63. in head angles during burrowing to reduce the stress and bending Marcus, H., Winsauer, O., Hueber, A. (1933). Der kinetische Schädel von energies on the skulls is intuitive, it has not been tested in vivo. Hypogeophis und die Gehörknöchelchen. Anat. Embryol. 100, 149-193. Meryman, H. T. (1960). The preparation of biological museum specimens by freeze- Future studies on skull loads during burrowing in vertebrates drying. Curator 3, 5-19. should address the sensitivity of the results to slight variations in Meryman, H. T. (1961). The preparation of biological museum specimens by freeze- drying: II. Instrumentation. Curator 4, 153-174. the head angle. Moazen, M., Curtis, N., Evans, S. E., OʼHiggins, P. and Fagan, M. J. (2008). Combined finite element and multibody dynamics analysis of biting in a Uromastyx hardwickii lizard skull. J. Anat. 213, 499-508. ACKNOWLEDGEMENTS Moazen, M., Curtis, N., OʼHiggins, P., Evans, S. E. and Fagan, M. J. (2009). We are grateful to Mark Wilkinson (Natural History Museum, London, UK) and Biomechanical assessment of evolutionary changes in the lepidosaurian skull. Proc. Alexander Kupfer (Universität Siegen, Germany), who kindly donated specimens Natl. Acad. Sci. USA 106, 8273-8277.

THE JOURNAL OF EXPERIMENTAL BIOLOGY 844 T. Kleinteich and others

Müller, H. (2007). Developmental Morphological Diversity in Caecilian Amphibians: Stayton, C. T. (2011). Biomechanics on the half shell: functional performance Systematic and Evolutionary Implications. Leiden: Leiden University Press. influences patterns of morphological variation in the emydid turtle carapace. Zoology Müller, H., Wilkinson, M., Loader, S. P., Wirkner, C. S. and Gower, D. J. (2009). (Jena) 114, 213-223. Morphology and function of the head in foetal and juvenile Scolecomorphus kirkii Summers, A. P. and Wake, M. H. (2005). The retroarticular process, streptostyly and (Amphibia: Gymnophiona: Scolecomorphidae). Biol. J. Linn. Soc. 96, 491-504. the caecilian jaw closing system. Zoology (Jena) 108, 307-315. Nussbaum, R. A. (1977). Rhinatrematidae: a new family of caecilians (Amphibia: Taylor, E. H. (1960). On the caecilian species Ichthyophis glutinosus and Ichthyophis Gymnophiona). Occ. Papers Mus. Zool. Univ. Michigan 682, 1-30. monochrous, with description of related species. Univ. Kansas Sci. Bull. 40, 37-120. Nussbaum, R. A. (1983). The evolution of a unique dual jaw-closing mechanism in Taylor, E. H. (1969). Skulls of Gymnophiona and their significance in the of caecilians: (Amphibia: Gymnophiona) and its bearing on caecilian ancestry. J. Zool. the group. Univ. Kansas Sci. Bull. 48, 585-687. 199, 545-554. Versluys, J. (1912). Das Streptosylie-Problem. Zool. Jb. 15 Suppl., 545-716. Nussbaum, R. A. (1985). Systematics of caecilians (Amphibia: Gymnophiona) of the Wake, M. H. (1993). The skull as a locomotor organ. In The skull: Functional and family Scolecomorphidae. Occ. Papers Mus. Zool. Univ. Michigan 713, 1-49. Evolutionary Mechanisms (ed. J. Hanken and B. K. Hall), pp. 197-240. Chicago: OʼReilly, J. C., Ritter, D. A. and Carrier, D. R. (1997). Hydrostatic locomotion in a University of Chicago Press. limbless tetrapod. Nature 386, 269-272. Wake, M. H. (2003). The osteology of caecilians. In Amphibian Biology. Volume 5. Parsons, T. S. and Williams, E. E. (1963). The relationships of the modern Amphibia: Osteology (ed. H. Heatwole and M. Davies), pp. 1811-1878. Chipping Norton, NSW: a re-examination. Quart. Rev. Biol. 38, 26-53. Surrey Beatty and Sons. Peter, K. (1898). Die Entwicklung und funktionelle Gestaltung des Schädels von Wake, M. H. and Hanken, J. (1982). Development of the skull of Dermophis Ichthyophis glutinosus. Morphol. Jb. 25, 555-627. mexicanus (Amphibia, Gymnophiona), with comments on skull kinesis and Pyron, A. R. and Wiens, J. J. (2011). A large-scale phylogeny of Amphibia including amphibian relationships. J. Morphol. 173, 203-223. over 2800 species, and a revised classification of extant frogs, salamanders, and Wiedersheim, R. (1879). Die Anatomie der Gymnophionen. Jena: Gustav Fischer. caecilians. Mol. Phylogenet. Evol. 61, 543-583. Wilkinson, M. and Nussbaum, R. A. (1997). Comparative morphology and evolution Roelants, K., Gower, D. J., Wilkinson, M., Loader, S. P., Biju, S. D., Guillaume, K., of the lungless caecilian Atretochoana eiselti (Taylor) (Amphibia: Gymnophiona: Moriau, L. and Bossuyt, F. (2007). Global patterns of diversification in the history of Typhlonectidae). Biol. J. Linn. Soc. 62, 39-109. modern amphibians. Proc. Natl. Acad. Sci. USA 104, 887-892. Wilkinson, M. and Nussbaum, R. A. (2006). Caecilian phylogeny and classification. Sarasin, P. and Sarasin, F. (1887–1890). Zur Entwicklungsgeschichte und Anatomie In Reproductive Biology and Phylogeny of Gymnophiona (caecilians) (ed. J. M. der ceylonesischen Blindwühle Ichthyophis glutinosus. Band II. Ergebnisse Exbrayat), pp. 39-78. Enfield: Science Publishers. Naturwissenschaftlicher Forschungen auf Ceylon in den Jahren 1884-1886. Wilkinson, M., San Mauro, D., Sherratt, E. and Gower, D. J. (2011). A nine-family Wiesbaden: C. W. Kreidel. classification of caecilians (Amphibia: Gymnophiona). Zootaxa 2874, 41-64. Stayton, C. T. (2009). Application of thin-plate spline transformations to finite element Zhang, P. and Wake, M. H. (2009). A mitogenomic perspective on the phylogeny and models, or, how to turn a bog turtle into a spotted turtle to analyze both. Evolution biogeography of living caecilians (Amphibia: Gymnophiona). Mol. Phylogenet. Evol. 63, 1348-1355. 53, 479-491.

THE JOURNAL OF EXPERIMENTAL BIOLOGY