170 



E.N. ARNOLD 



Among the same species, the African forms assigned to Chamaeleo 

 Laurenti, 1763 have small but well developed buccal openings into 

 the middle ear, but these are less obvious although present in the two 

 large Madagascan forms examined. In Bradypodion ventrale there 

 are indentations where the openings would normally be but the 

 openings themselves are absent. This also true of the Rhampholeon 

 GLinther, 1 874 and Brookesia Gray 1 865 studied in which there are 

 not even indentations. 



Ear reduction in the Phrynosomatidae 



(Fig. 4) 



Relationships within the Phrynosomatidae are discussed by De 

 Queiroz ( 1 992) and ear structure of various members of the group is 

 described by Earle (1961a, b, c; 1962), Wever( 1978) and Montanucci 

 (1987). 



In Phrynosoma there is stiff skin over the tympanum and some- 

 times this is very like that surrounding it, but middle ear structure is 

 basically normal. Uma has an essentially unmodified ear, but in 

 Callisaurus the columella is more robust and contacts the inner edge 

 of the quadrate bone, while the extracolumella is more heavily built, 

 directed backwards and has much stronger dorsal and internal 

 processes; the quadrate is also somewhat modified. In Holbrookia 

 and Cophosaurus, there is no ear opening, the tympanum is absent 

 and the columella is even more robust attaching to the quadrate via 

 a short, broad internal process. Cophosaurus has the tympanic area 

 partly covered by the m. depressor mandibulae and the quadrate is 

 more or less straight instead of cup-shaped. In Holbrookia, the 

 extracolumella is very reduced, covering of the tympanum by 

 muscles is greater than in Cophosaurus but modification of the 

 quadrate rather less. 



PATTERNS OF MODIFICATION 



In summary, a limited degree of external and middle ear reduction, 

 including covering of the tympanum by unmodified skin, has prob- 

 ably occurred at least fourteen times in the Iguania. More extensive 

 modification has taken place in Tympanocryptis, the Cophotis- 

 Lyriocephalus-Ceratophora clade, Phrynocephalus, the 

 Chamaeleonidae and the Callisaurus-Holbrookia-Cophosaurus 

 clade. In all these groups, different species exhibit markedly differ- 

 ent degrees of modification. When these varied conditions are 

 plotted on phylogenies of the groups concerned (Figs 2-4), it is 

 possible to get some idea of the order in which particular features of 

 the modified ears appeared. To be able to reconstruct a complete 

 sequence on a lineage, the origin of each new feature must be 

 separated from those of others by side-branches, otherwise, recon- 

 struction will be impossible or incomplete (Arnold, 1994). In the 

 Chamaeleonidae only two basic degrees of modification exist and 

 only three in Tympanocryptis, in Aphionotis-Cophotis- 

 Lyriocephalus-Ceratophora, and in Callisaurus-Holbrookia- 

 Cophosaurus. In the clade containing Phrynocephalus, on the other 

 hand, there are at least six successive conditions. 



It is apparent that the sequence of ear modification has been 

 different in some groups even though the end results have substan- 

 tial similarity. Thus, In the Oriental agamids, and Ctenophorus and 

 Tympanocryptis, the tympanum was first obscured by becoming 

 scaly (Fig. 5) and only then did other changes take place, such as the 

 m. depressor mandibulae moving forwards to cover the tympanic 

 area and the columella becoming more robust. In contrast, more 



basal members of the clade containing Phrynocephalus show that 

 movement of the depressor mandibulae muscle took place first, 

 creating and then narrowing a meatus (Fig. 6) and it was only when 

 this process was complete that the tympanum was entirely cut off 

 from external contact. In phrynosomatids, the earliest change in- 

 volved the columella and extracolumella becoming more robust, 

 rather than the tympanum becoming obscured. Complete closure of 

 the buccal opening is the final stage in the groups where it occurs 

 but, in the Callisaurus-Holbrookia-Cophosaurus clade, has not 

 taken place, even though the ear is otherwise highly modified. 



The reasons for the extensive ear modification found in some 

 iguanians are uncertain. The ground-dwelling forms that exhibit 

 these conditions occur in arid areas and many of them burrow 

 directly into loose sand or earth. In this situation, covering or 

 modifying the primitively delicate tympanum and associated middle 

 ear structures may protect them from damage. Again, a more robust 

 columella with direct attachment to the quadrate may be more 

 efficient at transmitting low amplitude vibrations generated on or in 

 the substrate by predators, prey or conspecifics. Ability to detect 

 such vibrations sometimes has clear benefits, for instance in the 

 sand-burrowing Scincus scincus (Hetherington, 1989). However 

 these possible performance advantages seem unlikely to apply to the 

 various tree dwelling forms that exhibit reduction of the external and 

 middle ears. It might be expected that the differences, in the order in 

 which specific modifications of the ear appear, result from different 

 selective regimes or different sequences of these. However, differ- 

 ences in order of modification even occur between ground-dwelling 

 forms, in which selective regimes are likely to be similar. 



USE OF THE OUTER AND MIDDLE EARS IN 

 THE SYSTEMATICS OF THE IGUANIA 



Although loss and reduction features are often said to be of low 

 value in phylogeny reconstruction (see for instance Hecht and 

 Edwards, 1977), changes in the outer and middle ear of the Iguania 

 do not always fit this preconception. Obscuring of the tympanum 

 has occurred many times, but loss of more structure is often congru- 

 ent with changes in other characters within groups. Thus ear 

 alterations do not conflict with phylogenies based on other features 

 in Tympanocryptis, the Aphaniotis-Cophotis-Lyriocepalus- 

 Ceratophora clade and in the phrynosomatid sand-lizard group. 

 Similarly, there is some congruence with classifications of chamele- 

 ons based on lung and hemipenial structure (Klaver & Bohme, 

 1986). For instance within Chamaeleo the development of a ptery- 

 goid connection to the columella is confined to the subgenus 

 Chamaeleo. However, loss of a buccal opening to the middle ear 

 occurs in the Brooke sia-Rhampholeon group but is also found in 

 Bradypodion which may be more closely related to other larger 

 chameleons (Klaver & Bohme, 1986). Although many changes in 

 the ear on the Phrynocephalus lineage are congruent with the 

 phylogeny, this group is exceptional in showing reversal in some ear 

 features. 



REFERENCES 



Arnold, E. N. 1 992. The Rajasthan toad-headed lizard, Phrynocephalus laungwalaensis 

 (Reptilia: Agamidae), represents a new genus. Journal ofHerpetology 26: 467^72. 



, 1994. Do ecological analogues assemble their common features in the same 



order? An investigation of regularities in evolution using sand-dwelling lizards as 

 examples. Philosophical Transactions of the Royal Society of London B. 344: 277- 

 290. 



