ADAMS, PHYLOGENY OF THE JAW MUSCLES 123 



feeders and either took slow-moving animals that they could engulf or 

 fed on vegetation. Patten says that they were herbivorous, for he found 

 carbonaceous residue in their bodies. 



The right combination of cartilage jaws covered with bony dermal 

 plates was first worked out by the ganoid fishes, ancestors of the modern 

 fishes on the one hand and of the land-living vertebrates on the other. 



Dinichthys 

 Plate XI, Figs. 1, 2 



The great arthrodires of the Devonian made the best pregnathostome 

 attempt to form a mouth with skeletal supports, but they made the fatal 

 error of trying to form those supports solely from the bony plates of the 

 skin rather than from the branchial cartilages. Coccosteus and Dinich- 

 thys both show in the skull a fine apparatus that serves for seizing their 

 prey. The arthrodiran jaw must have been a formidable weapon, as it 

 was armed with great sharp bony projections analogous to teeth. That 

 these great jaws were much used for biting and shearing is shown by 

 their worn shearing surfaces. Hussakof (1906) says: "The deep scars 

 found on the outer side of certain Dinichthys plates also bear testimony 

 to the savage attacks of their fellows." 



The jaws seem to have worked on the principle of the joints of the 

 insect leg, with most of the muscles attached chiefly to the plates of the 

 body and head. The Arthrodira made another fatal mistake in that they 

 moved the head perhaps more than the mandible. There is a joint be- 

 tween the head and the shoulder plates (Fig. 2), with a good peg-and- 

 socket articulation, so that while the mandibles remained more or less sta- 

 tionary the great head with its dermal plates moved up and down against 

 them ; thus we have the anomaly in the vertebrates of a relatively station- 

 ary jaw and a movable head, just as though the mandible of the gnatho- 

 stomes was attached immovably to the sternum, while the maxillae and 

 skull moved against it. 



It seems surprising that in the discussions as to the relationships of 

 the Arthrodira so little importance should have been given to the peculiar 

 motion of the head upon the thoracic shield, which is unknown among 

 true Pisces. The movement of the mandible of Dinichthys has been very 

 carefully studied by Dr. Hussakof (1906), to whose kindness I owe the 

 opportunity of studying the great collection of arthodiran fossils in the 

 American Museum of Natural History. The movement of the head upon 

 the mandible appeared, however, to have received too little attention, and 

 >his I have accordingly studied with great care. 



