22 SHOULDER-GIRDLE AND BREAST-BONE. 



what overlapped by the large clavicle below it, but the overlapping of splint-bones is very variable 

 (witness the interdigitation of the teeth of sutures) ; and as this bone is very thin and small, 

 the bone below very large and thick, and as they both are in close contact with the outer face of 

 the cartilage, which they invest, it is evident that this is merely a matter of overshadowing growth, 

 having no more morphological value 'than the variations in the teeth of a suture. Using some 

 degree of forecast in my comparisons, I may remark that if the outer lamina of the clavicle of 

 Coitus (Plate II, fig. 12, cl., s. cl.) had grown forwards, and the inner plate had been arrested, then 

 the supra-clavicle would have been overlapped by the clavicle, instead of the contrary, which 

 actually obtains. Again, in the Herring (Plate II, fig. 6, cl., s. cl., p. cl.) the largely- developed 

 posterior margin of the clavicle creeps within the post-clavicle as well as within the supra-clavicle. 

 Below, in front of, and a little outside the little thin supra-clavicle, we see the large thick clavicle 

 (fig. 1, cl.), which is sub-falcate in general form, scooped on, its outer and convex on its inner 

 face, and having its lower two fifths semicylindrical, the scooping being behind. This is a 

 strong green bone, very typically ichthyic, and, like all the splint-bones of the Lepidosiren, wholly 

 devoid of ganoid growth. Here we have the Teleostean Fish (excluding the aberrant Siluroids 

 and Hippocampoids) closely approached by this most generalised type. Amongst the Ganoids 

 it comes nearest to what obtains in Lepidosteus and Amia, for it has no " interclavicles," and 

 the feebleness of the post-temporals reminds one of the Muraenoids, which are actually devoid 

 of these bones (see Plate II, fig. 9). 



Not only did Professor Owen fail to find the post-temporal and supra-clavicles, but also 

 the whole of the cartilaginous true Shoulder-girdle. As to this soft condition of the endo-skeletal 

 part of the cincture, we have an interesting correspondence with those living Ganoids, 

 CalamoicMhys, Amia, and Sturio ; and undoubtedly also with many of the forms that are 

 buried in the primary rocks ; and as we shall soon see, with the existing Hippocampoids. 

 The coraco-scapular plate (fig. 1, sc. cr.) is scarcely half the size of the clavicle a thoroughly 

 ichthyic character; it is not unlike that bone in shape, being narrow above and in front, pro- 

 jecting backwards as a broad plate in the middle, and then narrowing to a blunt point below. 

 It cleaves to the inner face of the clavicle and supra-clavicle, growing a little in front of the 

 former, but not so high as the latter ; there is a thin stratum of fibrous web between them. 

 Behind, the upper moiety of the broad part is formed into a pedicellated cup the true " glenoid 

 cavity" (gl.) ; below the glenoid part there is a thin broader projection, which ends in a rather 

 sharp angle, where the cartilage becomes suddenly cut away, in a crescentic manner, behind ; the 

 anterior margin is gently sinuous. A transverse line passing across the glenoid pedicel would be 

 the base of the scapula ; the rest belongs to the coracoid ; this region has scarcely any projection 

 forwards that can be called " pra?coracoid." Now, whilst the Lepidosiren agrees with the 

 Ganoid genera Amia and Sturio in having an unossified Shoulder-girdle, it also equally 

 agrees with the Placoids in having a well-developed epicoracoid belt (figs. 1, 2, 3, e. cr.). 

 Professor Owen, speaking of the Sturgeon, expressly says ('Lectures,' vol. ii, p. 133) that " the 

 scapulo-coracoid arch is completed below, as in Lepidosiren, by ligamentous union, not, as in 

 Sharks, by cartilaginous confluence." True, indeed, as to the Sturgeon, but erroneous as to 

 the Lepidosiren. Originally, the epicoracoid mass must have been double; and perhaps in a 

 very early stage each moiety was continuous with the coracoid proper, but a wide transverse 

 cleft ie soon formed, such as is seen between the scapula and the coracoid in Coitus (Plate II, fig. 

 12, sc. p. cr.), and in other Fishes with large brachials, e.g. Anarrhichas, Triyla, &c. Above, this 



