Summary
The rodent-like tritylodonts, long allied to the multituberculate mammals, were transferred to the Ictidosauria when it was found that they had retained the reptilian quadrate-articular jaw hinge. When it was realized that this supposed order was a mere dumping ground for miscellaneous late mammal-like reptiles, Romer (1966) classed the tritylodonts as an Infraorder of the Suborder Theriodontia. Recently it has become popular to downgrade them to a mere family of the Cynodontia. Nine characters are listed to show that Romer's view is correct. Comments on Kühne's views on the relationships of the monotremes to the marsupials are added since he had strongly reversed the opinion he expressed when he described the skeleton of Oligokyphus.