- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
Lepidosaurs are frequently described as having highly kinetic skulls, and different forms of cranial kinesis have been described as being characteristic of their radiation. The model of amphikinesis proposed by Frazzetta, J Morphol 1962; 111:287–319, which was long considered a synapomorphy of the large suborder Sauria, is now much debated given its uncertain distribution among the various lizard taxa and the lack of data about its morphological correlates. In this article, we analyze the anatomical correlates of different forms of cranial kinesis, with particular regard to the putative saurian amphikinesis, describing the possible diverse skull movements of several species of European gekkotans (Hemidactylus turcicus, Mediodactylus kotschyi, and Tarentola mauritanica) and lacertids (Lacerta agilis, L. bilineata, Podarcis muralis, P. siculus, and Teira dugesii). Using serial and whole-mount histology, we found clear differences between gekkotans and lacertids in the structure of several cranial joints underlining the existence of two degrees of intracranial mobility. The lacertid species possess the anatomical features for streptostyly (quadrate joints) and metakinesis (parietal-supraoccipital and parabasisphenoid-pterygoid joints) and lack the anatomical correlates for mesokinesis (mobility of frontal-parietal and palatine-pterygoid joints) and amphikinesis (coupled mesokinesis, metakinesis, and streptostyly). In contrast, geckos present all the anatomical correlates for amphikinesis as described by the traditional quadratic crank model. Finally, we present a comprehensive summary of the different forms of squamate cranial kinesis, advancing two alternative hypotheses about the evolutionary origin of amphikinesis. Anat Rec, 297:463–472, 2014. © 2013 Wiley Periodicals, Inc.
Cranial kinesis can be generally defined as the relative movements between particular skull bones, excluding movements of or within the mandible (Arnold, 1998; Metzger, 2002). It has been described in numerous taxa of extant vertebrates, including ray-finned fishes (Westneat, 2004), squamates (Frazzetta, 1962; Metzger, 2002), and birds (Zusi, 1984). Among the derived diapsid reptiles, cranial kinesis is generally considered a plesiomorphic state since skull mobility is ubiquitous in archosaurs and lepidosaurs (Holliday and Witmer, 2008), but particular forms of cranial kinesis are thought to be synapomorphies of recent phylogenetic radiations, particularly in lizards and snakes (Kardong et al., 1997; Evans, 2003, 2008).
The origin, morpho-functional aspects, and evolutionary significance of cranial kinesis within Lepidosauria have been widely discussed for more than a century. In particular, various forms of cranial kinesis were initially defined on the basis of the localization of the relative joints (Versluys, 1910, 1912, 1936; Hofer, 1949). Subsequently, the model of the quadratic crank proposed by Frazzetta (1962) represented for many decades a new starting point in understanding the mechanisms involved in the skull mobility of Lepidosauria and their evolution. This model has long been considered a shared characteristic of the large suborder Sauria, with the highly kinetic snake skull viewed as progressive evolution (see Simonetta, 1963; Iordansky, 1989; Kardong, 1994; Frazzetta, 1999; Lee et al., 1999; Cundall and Greene, 2000).
According to the modern definitions of the various saurian kinetic forms given by Metzger (2002) and Evans (2003, 2008) herein adopted, mesokinesis is the dorsoventral flexion and extension of the skull around an axis that runs transversely through the frontal-parietal joint; hypokinesis is the flexion around an axis running transversely to the palatal zone of maximum flexibility, usually corresponding to the palatine-pterygoid articulation; metakinesis is the relative movement between the dermal skull roof and the braincase by means of a sliding between the parietal and the supraoccipital bones; streptostyly is the movement of the quadrate in relation to the dermatocranium, usually in the form of an anteroposterior rotation at its dorsal end. Frazzetta (1962) used the term amphikinesis to describe the lizard's distinctive cranial kinesis, which amounts to simultaneous metakinesis and mesokinesis coupled with functional streptostyly.
The mechanism proposed by Frazzetta is simple: anterodorsal rotation of the quadrate elevates the muzzle unit through dorsoventral flexion at the mesokinetic joint and, coupled with a metakinetic rotation, results in a depression of the parietal unit (Frazzetta, 1962; Metzger, 2002). It is worth noting that to have a functional mesokinesis (and amphikinesis) a permissive hypokinetic joint must be present ventrally in the palate (usually through the contact between the palatine and the pterygoid or alternatively between the vomer and the palatine bone) and laterally (between the palatine, pterygoid and ectopterygoid) (Metzger, 2002). However, the applicability of Frazzetta's amphikinesis described by the quadratic crank model has been found only in a limited number of species (see Metzger, 2002; Evans, 2003, 2008 for details). In particular, a fully functional amphikinesis appears to be restricted to varanids and gekkotans (Rieppel, 1979; Patchell and Shine, 1986; Condon, 1987; Herrel et al., 1999, 2000; Metzger 2002; Evans, 2003, 2008). However, various studies on the genus Varanus have reported diverse patterns of cranial mobility (Rieppel, 1979; Smith, 1982; Smith and Hylander, 1985; Condon, 1987-1989).
Furthermore, although the presence of skull kinesis in lizards has been successfully demonstrated (e.g., Smith, 1980; Condon, 1987; Herrel et al., 1999, 2000), little is known about the morphological features correlated to these movements (see Meztger, 2002) with few exceptions (e.g. Iordansky, 1966 [see also translation by Kelso, 1968], 2011; Condon, 1988, 1998; Arnold, 1998; Schwenk, 2000; Payne et al., 2011). In particular, while some skull articulations were extensively investigated, such as the various quadrate joints, there are almost no data on the morphological and histological features of the hypokinetic and the metakinetic joints, which are indeed essential for mesokinesis and amphikinesis (see Metzger, 2002; Evans, 2008).
We focused our study in representative members of European lacertids and gekkotans. Members of these different evolutionary lineages demonstrate structurally distinct skull morphologies (e.g., the postorbital bar is complete in lacertids and incomplete in gekkonids) and possibly different degrees of intracranial mobility (Frazzetta, 1962; Herrel et al., 1999; Metzger, 2002; Evans, 2008). Using serial and whole-mount histology, we compared the joint morphology and connective tissue structure of many of the articulations reportedly involved in cranial kinesis. Finally, adding our data to published ones we presented a comprehensive summary of the presence of the different forms of cranial kinesis in the Squamata, advancing two alternative hypotheses on the evolutionary origin of amphikinesis.
- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
We found clear differences between the studied gekkotans and lacertids in some cranial joints (frontal-parietal and pterygoid-palatine), that are in line with two different degrees of intracranial mobility. These differences are highly conserved in all the gekkotans and lacertids analyzed, which share a uniform, lineage-specific organization of the cranial kinetic joints. From comparison of the same cranial joints, we found that there is a different morphology but not a different histology, which allows or constrains certain kinetic skull movements. In fact, although the joint structure does not prove the joint function, it may suggest the potential and/or the exclusion of different forms of kinesis (Metzger, 2002). In addition, the anatomical correlates of the different forms of cranial kinesis found in the studied species can be compared to the same structures found in species in which the kinetic movements were described experimentally, thus allowing a plausible correlation between joint structures and associated movements. For example, the gekkotans here studied share the same general skull joints structures of Gekko gecko and Phelsuma madagascariensis, in which amphikinesis was measured by cineradiography (Herrel et al., 1999, 2000).
In spite of this clear distinctness, a certain number of anatomical similarities are shared by the two different lineages, probably as features of the whole squamate radiation. In general, the skull of the lacertid species examined here is deep, robust, and lacking some of the anatomical correlates of amphikinesis described by Frazzetta's quadratic crank model. However, all the species studied exhibited the potential for a certain degree of streptostylic and metakinetic movements. Their quadrate joints show the main anatomical correlates of streptostyly reported for various different squamate taxa (see Throckmorton, 1976; Rieppel, 1978; Metzger, 2002).
To date, little attention has been given to the histological and morphological features of the metakinetic joints of squamate reptiles, and as a result metakinesis has been considered the most enigmatic form of squamate cranial kinesis (see Metzger, 2002; Evans, 2008). Indeed, our results suggest at least some degree of metakinetic movement can be present in the Lacertidae, in relation to the potential sliding at the syndesmosis between the parietal and the supraoccipital and at the synovial joint between the pterygoid and the basipterygoid. In addition, the epipterygoid joints, which are thought to be important in stabilizing metakinetic movements, show the mobile primitive configuration of the Squamata (see Evans, 2008).
In contrast, both functional mesokinesis and hypokinesis are probably absent in all the lacertid species studied. Indeed, both the frontal-parietal and the palatine-pterygoid joint of the lacertid species are highly interdigitated, probably preventing any flexion or extension movements of a certain magnitude. However, computational model studies (see Moazen et al., 2013) are expected to better elucidate the stiffness degree of various interdigitated sutures. Interestingly, a similar interdigitation at the frontal-parietal joint was found also in Cryptolacerta hassiaca, a lacertid-like lizard fossil from Germany which provided a first morphological evidence for lacertid-amphisbaenians monophyly on the basis of a shared reinforced and akinetic skull (Müller et al., 2011). In addition, in Lacerta bilineata and L. agilis, but not in Podarcis muralis, P. siculus or Teira dugesii, the presence of temporal osteoderms, covering much of the lateral and upper surfaces of the head, further limits skull kineticism.
In general, the gekkotan skull is lightly built, dorsoventrally depressed, and highly kinetic, presenting all the anatomic correlates of quadratic crank model for functional amphikinesis (Frazzetta, 1962; Metzger, 2002; Evans, 2008). The dorsal quadrate articulation and a permissive quadrate-pterygoid syndesmosis allow the anteroposterior rotation of the quadrate bone, similarly to what was observed in lacertids.
The histological structure of the metakinetic joints of the gecko species is similar to those found in all the lacertids studied, both allowing a relative sliding movement between the parietal and the supraoccipital. However, even if the histological structure is conserved between the two families the different osteological relation between the dermatocranium and the braincase suggests a greater possibility of metakinetic movements in geckos.
Unlike lacertids, in the gekkotans the osteological and histological structure of the frontal-parietal joint is consistent with a high degree of mesokinetic movements. In addition, the most notable difference found between the two families lies in the structure of the hypokinetic joint. The contact between the palatine and the pterygoid found in gekkotans involves a small area of juxtaposition and a large gap, mainly filled with fibrous connective tissue. The particular morphology of this joint guarantees the high degree of palatal mobility required by the quadratic crank model (see Frazzetta, 1962; Metzger, 2002). In addition, the movements at the mesokinetic and hypokinetic articulations are facilitated by the reduction of lateral elements to the skull roof and the palate, with particular regard, respectively, to the postfrontal and the ectopterygoid.
Our results are congruent with observations in other gekkotans (Patchell and Shine, 1986; Herrel et al., 1999; 2000; Payne et al., 2011). Our findings highlight the presence of all the anatomical correlates for functional amphikinesis in T. mauritanica, H. turcicus, and in M. kotschyi. In contrast, in the lacertid species examined the osteological and histological features suggest the absence of functional amphikinesis and mesokinesis but not the potential for unlinked streptostyly and metakinesis. A functional metakinesis may be associated with amphikinesis and the quadratic crank model (see Frazzetta 1962; Metzger, 2002; Evans, 2003, 2008), but a form of passive metakinesis can also be related to shock absorption and to the protection of the brain against feeding strains (Rieppel, 1978; De Vree and Gans, 1987, 1989; Moazen et al., 2009).
Many questions about the origin and evolutionary significance of different forms of cranial kinesis in lepidosaurs are still largely unsolved. Adding our results to the data from literature, and mapping the different forms of kinesis into two largely accepted phylogenies of squamates (Conrad, 2008; Pyron et al., 2013), we here present a summary about cranial kinesis, with a particular regard to amphikinesis and its evolutionary framework (Fig. 8). Traditionally, the morphological classifications of squamate reptiles have proposed a phylogenetic tree with the Iguania as the sister group of a large clade, the Scleroglossa, including Gekkota, Scincomorpha and Anguimorpha (see Estes et al., 1988; Lee, 1998; Evans, 2003, 2008; Conrad, 2008). Recently, Pyron et al. (2013) have revised the evolutionary relationships of Squamata using molecular markers. These authors have proposed a phylogenetic tree with the Dibamidae as the sister taxon of a monophyletic group composed of two main evolutionary lineages: Gekkota and a larger clade including two main groups, Scincomorpha and Episquamata: the latter in turn includes Lacertoidea, Anguimorpha, Iguania and Serpentes.
Figure 8. Type of cranial kinesis and phylogenetic relationships of squamates. (A) Cladogram based on morphological characters, modified from Conrad (2008). (B) Cladogram based on molecular analysis, modified from Pyron et al (2013). Figure symbols: □, akinesis; ▪, streptostyly; ○, mesokinesis; •, metakinesis; Δ, amphikinesis; NC, not considered in this study. The question mark indicates lack of information. Asterisk indicates uncertain information. Numbers indicate the two alternative hypotheses about origin of amphikinesis: (1) single origin; (2) multiple origin.
Download figure to PowerPoint
Total akinesis characterizes the skull of Sphenodon (Gorniak et al., 1982; Jones et al., 2011) and various Iguanidae, Agamidae, Scincidae, and Helodermatidae, but the majority of the squamate taxa possess at least some form of cranial kinesis (Metzger, 2002; Evans, 2003, 2008). In particular, streptostyly, found in many different evolutionary lineages, seems to be the more widespread form of cranial kinesis in squamates, and relatively few taxa are known to be mesokinetic or metakinetic. In addition, to date, among the whole squamate radiation, the potential for a functional amphikinesis is known only in Gekkota and Varanidae (Rieppel, 1979; Condon, 1987; Herrel et al., 1999, 2000; Metzger, 2002; Evans, 2003, 2008; this work), which represent two phylogenetically distinct lineages according to both the morphological and the molecular phylogenies (see Conrad, 2008; Pyron et al., 2013). Thus, in the light of available data on the squamate evolutionary history and cranial kinesis, the following alternative hypothesis about the evolution of amphikinesis can be advanced (Fig. 8): (1) amphikinesis evolved only once, in the most recent common ancestor of Gekkota and Varanidae; (2) amphikinesis evolved at least twice independently, in Gekkota and Varanidae.
To date, neither hypothesis can be excluded, but a single evolutionary origin of amphikinesis seems less supported by parsimony as it requires a number of independent events of secondary loss in many squamate lineages. On the other hand, data on cranial kinesis are lacking for many squamate lineages and further studies, particularly on Carphodactylidae, Shinisauridae, and Lanthanotidae are needed to allaying the doubts about the origin of amphikinesis.
A comprehensive and detailed discussion about the functions of the different forms of squamate kinesis is beyond the aim of this study. However, the two different configurations of cranial joints found among lacertids and gekkotans and the relative kinetic implications are expected to be correlated to different functional and structural responses of the skull, with particular regard to feeding performance (Schwenk, 2000, 2001; Wagner and Schwenk, 2000; Evans, 2003). For example, it has been noted that in geckos a highly kinetic skull may result in bite force reduction and in decreased feeding efficiency due to the longer intra-oral transport cycles. However, at the same time there occurs an increase in the velocity of jaw opening and closing, facilitating the capture of elusive prey (Herrel et al., 2007).