Abstract

Kluge (1967, 1983) proposed that the four subfamilies within the Gekkonidae were monophyletic assemblages, but that the Gekkoninae could be divided into two tribes on the basis of hyoid apparatus structure. Whilst agreeing that four subfamilies were present in the Gekkonidae, Moffatt (1973) argued that those groups of non-Eublepharine gekkos which remained after the differentiation of the Diplo- dactylinae and Sphaerodactylinae, and which had not become sufficiently distinct to be classified into separate subfamilies, had been lumped together as the Gekkoninae. Subsequently, Russell (1976, 1979) found that at least seven distinct groups could be defined within the Gekkoninae on the basis of toe structure. In the present paper I compare chromosomal evolution in the monophyletic Diplodactylinae and that in the possibly polyphyletic Gekkoninae, to test whether the tribal subdivision made by Kluge (1983) is valid, or whether this is a far more heterogeneous group as Russell and Moffatt proposed. The chromosomal data from 47 of the 92 species show that the Diplodactylinae have evolved from a 2n = 38 all acrocentric ancestral karyotype by the processes of pericentric inversion and centric fusion. In contrast, an analysis of 74 species from the Gekkoninae shows that eight distinct putative ancestral karyomorphs are present, 2n=32, 34, 36, 38, 40, 42, 44 and 46, each of which is acrocentric or telocentric. Numerous fusions, inversions, additions and tandem fusions have occurred within each of these categories. These data suggest that the Gekkoninae are a polyphyletic assemblage, and group comparisons indicate that there is some agreement with the morphogroups proposed by Russell (1976).