The maxilla named by T. H. Huxley Dasygnathus longidens, from the Trias of Findrassie near Elgin, is re-described. A pterygoid from the same locality is referred to this species and described for the first time. These two bones indicate a large, carnivorous pseudosuchian apparently allied to Erythrosuchus. A detailed description is given of the osteology of the pseudosuchian Stagonolepis, amplified by a large number of hitherto undescribed specimens. The material (considered to represent at least twenty-one individuals) shows an almost complete segregation into two size-groups, distinguished by only a few minor morphological differences of the postcranial skeleton. The larger and small individuals are considered to be males and females respectively, of the one species S. robertsoni Agassiz. Study of Stagonolepis has demonstrated its close relationship to Aetosaurus from the Stubensandstein (Keuper) of Stuttgart. The latter genus, however, was in urgent need of revision. Accordingly a brief account is included of the principal respects in which previous descriptions of this form require modification. Specimens referred by von Huene (1921) to A. crassicauda are shown to have been misidentified in many cases; this material is a composite of a small coelurosaur and a true aetosaurid. Some modifications are also suggested to previous accounts of Typothorax and Desmatosuchus from the Trias of North America, and new restorations are given of the skulls of these forms. The extremely close relationship between Stagonolepis and Aetosaurus, perhaps even indicating generic identity, makes unavoidable the merging of the two families previously founded on these genera. On grounds of priority the name Aetosauridae is retained for the taxonomic unit which includes Aetosaurus, Stagonolepis, Typothorax and Desmatosuchus as principal members. A restricted diagnosis of the family Aetosauridae is given, based on the above four forms, and this is followed by a review of the genera which have from time to time been included in the former families Stagonolepidae and Aetosauridae. The great majority of these genera are excluded from the group as now defined. Possible evolutionary trends within the family are briefly outlined; the sequence of increase of specialization appears to be Aetosaurus, Stagonolepis, Typothorax, Desmatosuchus. The conclusions of Dollo (1884), Adams (1919) and others concerning the function of the preorbital fossa in archosaurian reptiles are endorsed, and it is suggested that a trend towards the reduction of the anterior pterygoid muscle took place in aetosaurids, in parallel with a similar trend in ornithischian dinosaurs. Many features of the skeleton of aetosaurids recall the Ornithischia. These include the elongate naris, reduced dentition, vertical or forwardly inclined quadrate, slipper-shaped jaw, small skull and well-developed dermal armour. However, a direct ancestor-descendant relationship appears to be ruled out by the position of the supratemporal fossa, reduction of the infratemporal opening, probable loss of the coronoid and typically pseudosuchian pubis, although the aetosaurids may well lie close to the root-stock of the Ornithischia. The mode of life of the aetosaurids is considered, and it is concluded that these animals were herbivorous, or possibly feeders upon invertebrates obtained, in the case of Stagonolepis at least, by digging with the peculiar expanded snout-tip and dentary rostrum. A simple muscular mechanism is postulated whereby this could be effected. The stratigraphical implications of these studies are briefly examined and the suggestion, based primarily on the close relationship between Stagonolepis and Aetosaurus, is put forward that the Triassic sandstone of Elgin occupies a higher horizon than has previously been considered.