Evolution of Selected Characters
The new phylogeny has interesting implications for the evolution of four characters frequently used in considerations of amniote relationships.
Tabular-Parietal Contact
The tabular-parietal contact has usually been considered to unite embolomeres and seymouriamorphs to diadectomorphs and amniotes (Gauthier et al., 1988), but this character is also present in all lepospondyl groups that have a tabular (Figs. 2 and 3). Therefore, this character is diagnostic of node J (embolomeres and batrachomorphs). Of course, the identity of the temporal element in most lepospondyls is difficult to determine and it has been assumed to be a tabular in this analysis, but even if this element were a supratemporal the presence of a tabular-parietal contact would be an autapomorphy of node J. Some nectrideans also have a supratemporal and the tabular-parietal contact is definitely present in this group (Fig. 3).
Transverse Flange of the Pterygoid
The transverse flange of the pterygoid has been usually considered to be diagnostic of amniotes and diadectomorphs (Gauthier et al., 1988; Panchen and Smithson, 1988), but it is a synapomorphy of seymouriamorphs, gephyrostegids, diadectomorphs, and amniotes and also appeared convergently in dissorophoids. The transverse flange was lost in lepospondyls (see Appendix 3).
Supraoccipital
The supraoccipital has often been considered to be a synapomorphy of amniotes and diadectomorphs (Gauthier et al., 1988) and its presence in some lepospondyls (Fig. 4) was often assumed to result from convergence, but it seems to be a synapomorphy of all these taxa. Among the lepospondyls included in this analysis, a supraoccipital is present in Pantylus, Rhynchonkos, brachystelechids, and lysorophians. Nectrideans and lissamphibians appear to have lost this element (this character could not be coded in aistopods and in adelogyrinids).
Parasternal Process of the Interclavicle
The parasternal process (long posterior stem) of the interclavicle has been thought to be a synapomorphy of amniotes, diadectomorphs, and seymouriamorphs (Carroll, 1968), but this character seems to be an autapomorphy of the clade that includes all these taxa and lepospondyls. A parasternal process may have been present primitively in lepospondyls because it is present in Pantylus, brachystelechids, and some lysorophians. However, the optimization of this character at the base of Lepospondyli is ambiguous because the interclavicle of adelogyrinids and nectrideans lacks a parasternal process (this character could not be coded in aistopods). Lissamphibians have lost the interclavicle.
Post a comment