REPTILES. 141 



Plate VI, figs. 10 12) in which an Anouran has the prse-coracoid severed from the coracoid 

 by a notch ; in the other instances there is a fenestra in that region ; this, however, which is 

 exceptional in the tailless Amphibia, is the unvarying rule in the tail-bearing forms (see Plates 

 III and IV). Again, in the Lacertilia, the "prae-" and " epi-coracoids" are continuous, whilst 

 in the Crocodiles the prae- coracoid is aborted all but its upper root : in the Chelonian there is no 

 instance, as far as I know, of a closed inter-coracoid space. The form of the scapula (sc.) is 

 very remarkable (Plate XII, figs. 1 3 ), it is an almost cylindrical rod, becoming flat at the 

 top, where its unossified part is the supra-scapula (s. sc.) ; the position of this part, in relation 

 to the last certical vertebra is essentially the same as that of the Ray, in that creature related to 

 several cervical vertebrae. There is a difference, however, namely, in the Ray the edges of the 

 supra-scapulse (see Plate I, figs. 1 4) come close up to the vertebral spine, whilst in the 

 Chelonian they are attached to the spine by a ligament. Their total independence of the 

 small cervical rib (figs. 3 and 4) is shown in the figures, and a very powerful imagination is 

 required to suppose that the scapular rod is here merely a segment of the rib ; the dislocation 

 is too violent, and the hypothesis cannot be digested. There is nothing that can be called 

 " prse-" or " meso-scapula," save the swollen part in front, which passes interruptedly into 

 the prse-coracoid (p. cr.) ; this front fork forms, with the scapula, a gentle arc (see fig. 3) ; it 

 is of the same thickness, nearly of the same length, and has no separate osseous centre, the two 

 bars being hardened by one ectosteal sheath. This anterior bar has been supposed to be the 

 " acromion process ;" it is that at its root, but is much more, and in the Anourous Amphibia, 

 and often in the Urodela, has its own osseous sheath. It is still soft below in this young 

 specimen, and the relation of this lower end to the dermal bones is shown in figs. 1, 3, and 7 ; the 

 more enlarged figure (ten diameters) shows how this soft end of the pras-coracoid is brought 

 into relation with the clavicles and inter-clavicle (cl., i. cl.) exactly as in the Lizards (see Plates 

 IX XI) ; it is bound to them by a ligament, but there is nothing except fibrous tissue between 

 these parts. In the Mammal the distal end of the pras-coracoid is developed autogenously in 

 relation with the sternal end of the clavicle, and undergoes a peculiar segmentation, as I shall 

 show in due time. The coracoid has its own bony centre (figs. 1 and 2) ; it is the longest rod of 

 the three, and passes backwards and inwards, ending in a hooked, soft epicoracoid (e. cr.). This 

 part (see fig. 1) reaches to within a short distance of the umbilicus (u.), in the interspace between 

 the post-thoracic and pre-abdomiual plates. In some of the Chelonians, especially Testudo, the 

 coracoid is a triangular plate, and not rod-like, as in CJielone mydas ; in the latter type there is a 

 remarkable agreement as to outer form with that of the genus Dactylethra (Plate VI, figs. 10 12) 

 amongst the Frogs, and also in the shape of the prae-coracoid, but nothing could be more unlike 

 than the scapula and supra-scapula of these two animals. But the scapulas of the Systonus (Plate 

 VII, figs. 9, 10), are nearly as narrow as in the Chelonia ; the whole Shoulder-girdle of the 

 Chelonia is a forked ray, the anterior fork being ossified continuously with the primary ray 

 (scapula), and the posterior ray having its own ectosteal sheath ; this is exactly repeated in the 

 African Ostrich, Struthio camelus, see Plate XVII, figs. 5, 6. 



