NOTES ON THE OHIO SHALES AND THEIR FAUNAS 31 



same plane as the top of the skull. The left clavicular lies on 

 the right side on top of the right clavicular which has lost its 

 lower end. They are fitted together in such a way that the 

 left occupies the exact position that the right would occupy if 

 present. The photograph shows the correctness of the writer's 

 statement in the Ohio Naturalist where he said: "Every 

 bone is represented, though in some of the paired bones one of 

 the pair may be missing and the one preserved not be perfect. 

 The bones missing are the right mandible, the right postero- 

 supero-gnathal, the left antero-supero-gnathal, and the left 

 postero-lateral." The left antero-lateral is not shown in the 

 photograph on account of its being greatly distorted. The 

 left mandible is shown on the right side in order to compare it 

 with the length of the skull. The antero-supero-gnathal was 

 omitted from the composition. 



The main fault with Dean's restoration* is the size and 

 position of the clavicular. As shown in the photograph of the 

 Oberlin College specimen the clavicular comes against the 

 posterior part of the suborbital probably fitting against it, 

 though the imperfection of the posterior edge of the latter pre- 

 vents a positive determination of that point. The inner arm 

 of the clavicular must have come inside the suborbital and 

 prevented the mandible resting against it as shown by Dean. 

 As compared with the Oberlin College specimen the clavicular 

 figured by Dean is almost two diameters too large for the 

 specimen that it accompanies. On that account the lower end 

 of the bone is thrown too far back and its relationships to sub- 

 orbital and supero-laterals are incorrectly shown. Such a 

 restoration also makes the animal much thicker dorso-ventrally 

 than it should be. 



The Oberlin specimen shows a distinct overlap of the 

 clavicular on the marginal of the skull and Dean's restoration 

 shows no such overlap. In Dean's restoration the anterior 

 part of the suborbital is not shown. In an earlier paper I called 



'Mem. Amer. Mus. Nat. Hist., IX, PartV, Plate XXXVIII. 



