Abstract
The isolated skull of a Lower Carboniferous anthracosaur labyrinthodont from the Midlothian coalfield was formerly but erroneously attributed to Pholidogaster pisciformis. It is now redescribed after 'Airbrasive' development as the holotype of a new genus and species. The anthracosaur agrees in a number of diagnostic cranial characters with the embolomerous forms of the Coal Measures but other characters, particularly those of the braincase and lower jaw, suggest relationship to the Seymouriamorpha. Its cranial anatomy could well be antecedent to that of both these groups and the sparse evidence available suggests close relationship to the American Mississippian anthracosaur Proterogyrinus. A new classification of the Batrachosauria is presented with a major subdivision into Anthracosauria and Seymouriamorpha. The former group comprises the newly proposed infraorder Herpetospondyli to include the Lower Carboniferous forms, together with the infraorders Embolomeri and Gephyrostegoidea. It is concluded that while no batrachosaur with an otic notch could be a reptile ancestor a primitive 'notchless' anthracosaur could be. The cranial anatomy of Pholidogaster pisciformis is then redescribed, after complete 'Airbrasive' preparation of the holotype skull. The holotype, like the anthracosaur, comes from the Gilmerton Ironstone of the Edinburgh region, but the earlier holotype skull of Otocratia modesta, long considered to be an ichthyostegid, is also attributed to Pholidogaster. After redescription of 'Otocratia' a composite restoration of the Pholidogaster skull is presented. The nature of the dermal ornament, the post-parietal-supratemporal suture, the absence of tabular horns and the nature of the post-cranial skeleton all demonstrate that Pholidogaster is an early temnospondyl labyrinthodont in no way closely related to anthracosaurs. It is closely similar to the American Mississippian form Greererpeton burkemorani Romer, notably in the presence of large premaxillary fangs and the absence of otic notch and intertemporal. Both may be attributed to the family Colosteidae. It is concluded that there is no strong evidence of close relationship between the colosteids and the Devonian ichthyostegids.

This publication has 26 references indexed in Scilit: