On the classification and evolution of geckos

S ummary.1  It is pointed out that the classification of geckos has remained unstable during the last eighty years. 2  The eyelids and spectacles of geckos have been misunderstood; the eyelids are primitive, the spectacle is secondary. 3  The majority of geckos have a pure rod retina in association with nocturnal habits, the rods derived by transmutation of cones. 4  Two types of vertical pupil are distinguished, that with a straight edge and the Gekko ‐type with three lobes on anterior and posterior margins. 5   Diplodactylus has a straight vertical pupil, Phyllodactylus a Gekko ‐type pupil, several species are reallocated. 6   Narudasia is revived and distinguished from Quedenfeldtia on pupil shape. 7   Gymnodactylus is redefined and restricted to South American forms with straight vertical pupil. 8   Phyllurus is used for Australasian “ Gymnodactylus ” with straight vertical pupil. 9   Cyrtodactylus is revived for Old World “ Gymnodactylus ” with Gekko ‐type pupil. 10   Wallsaurus gen. nov. is proposed for Gymnodactylus horridus with straight padless digits and Gekko ‐type pupil. 11  Arrangement of geckos as a superfamily Gekkonoidea with three families is proposed. 12  Boulenger's Eublepharidae is revived for geckos with true eyelids with the addition of Holodactylus and Aeluroscalabates . 13  Sphaerodactylidae fam. nov. is proposed for procoelous geckos with spectacle including Coleodactylus, Gonatodes, Lepidoblepharis , and Pseudogonatodes . 14  The family Gekkonidae is restricted to amphicoelous geckos with spectacle and is divided into two subfamilies. 15  Diplodactylinae subfam. nov. is proposed for Gekkonids with straight vertical pupil, twenty‐two genera are placed here. 16  Subfamily Gekkoninae is restricted to geckos with Gekko ‐type pupil, thirty‐eight genera are placed here. Nine genera remain unplaced. 17  Amphicoelous geckos appear also to have the two elements of the neural arch separate. 18  The position of Uroplatus is discussed. 19  The Eublepharidae are relicts of an archaic specialized group in old stable land areas. 20  The Sphaerodactylidae are a strictly New World group centred on the Caribbean. 21  The Diplodactylinae have a wide broken distribution and appear largely to have been supplanted by the Gekkoninae. 22   Aristelliger is a relict of primitive archaic type in the Antilles. Bavaya and Rhacodactylus , also of peripheral distribution (New Caledonia), may be related to Aristelliger . 23  The Gekkoninae have a wide range and it is suggested that there was an original dispersal of arboreal forms and that there has been independent reversal to terrestrial habits in several regions. 24  The desert region of N. Africa and S.W. Asia contains a wealth of padless terrestrial geckos. 25  Madagascar has eleven genera of geckos all with pads and none archaic. 26  The Oriental Region has a wealth of Gekkoninae, no Diplodactylinae. 27  South Africa and Australia both have notable groups of Diplodactylinae. 28  The problem of the occurrence of a Tarentola in Cuba is discussed. 29  The view that the procoelous condition is primitive and the amphicoelous condition secondary in geckos is upheld. 30  It is argued that geckos whilst ancient are not primitive. 31  A tentative outline of the evolutionary history of geckos is put forward. 32  Evidences for parallel evolution in geckos are discussed in respect of the adaptation of the eye to dim lighting conditions and the secondary reversion to terrestrial habits with loss of digital pads. 33  It is argued that the cases of parallel evolution are much better explained in terms of evolutionary opportunity as a relationship between organism and environment than in terms of any innate tendencies. 34  In several stocks there is an increase over the “standard” number of 14 scleral ossicles, these parallel evolutions are not readily explicable.