THE SHOULDER-GIRDLE IN FISHES. 13 



Fishes) has been supposed (by Cuvier and others) to be the scapula. As there are still two 

 more dermal bones in each semi-girdle below the one in question, I was perplexed for some time 

 as to which of the outer shoulder-plates I must call the scapular splint : it can, however, all be 

 explained. Figs. 6 and 8 show this bone, which may be called the " supra-clavicle " (s. cl.) ; it 

 protects the posterior face of the scapular cartilage, just down to the great scapular fenestra ; and 

 in this half-grown Sturgeon it is 20 lines long and 4 lines in breadth. This scale has a small 

 ganoid tract along its middle, an easy step towards its entirely subcutaneous condition in 

 the Teleostei ; it is overlapped by the post-temporal above, whilst the lower half of its anterior 

 margin articulates with the great clavicle (figs. G and 8, p. t., s. cl., cl.). Its counterpart in other 

 types will be easily recognised in Plate II ; but it is not clearly shown in Huxley and Hawkins's 

 ' Atlas' (see plate 5, fig. 3a), for there are minor and less specialised bone-grains running 

 behind the great shoulder-plates, and the small amount of ganoid surface which this bone pos- 

 sesses makes it difficult of discovery at first. In some instances I shall show that this bone is 

 not always ossified separately from the true clavicle ; but whether as a region or as a distinct 

 dermal plate, its identification is of the utmost importance. 



The large plate next below the " supra-clavicle " at one time beguiled me into supposing 

 that it answered to the so-called " scapula" of authors ; for it is not the lowermost bone in this 

 series. As a large number of Mammals including Man possess the counterpart of the bone 

 now under consideration, its real and relative morphology must be thoroughly mastered : it is 

 the true " clavicle," and a knowledge of it must be made the foundation-stone of all our scientific 

 knowledge of the structures and relations of the outer and inner Shoulder- girdle. This large ganoid 

 plate is rough externally, where it applies itself to the outer part of the scapular and post-scapular 

 bars ; and further downwards to where it articulates to the plate beneath it (Plate I, figs. G 

 and 8, cl.). It is smooth, thin, and radiating as it passes inwards, behind the operculum (see 

 Huxley and Hawkins's 'Atlas,' plate 5, fig. Ba; and Plate I, figs. G 8, cl., of this work) ; it is 

 this inner plate to the convex hinder surface of which the prse-coracoid is attached, snail-like 

 (fig. 7, cl., p. cr.) : this point of relation will serve as a most capital landmark all the way along 

 our devious journey to the shoulder of the youngest-born of the Vertebrates. Another waymark 

 is to be seen below ; for where the spreading, outer plate of this bone begins, below, to lose its 

 ganoid character, and to be overlapped by the bottom plate there, behind, the " epicoracoid " is 

 naked, and on the inner side of the clavicle apart of its snail-footed lamina is spread, the rest lying 

 on the lower bone (see figs. 6 8, cl., i. cl., p. cr.). The eye will rest on this landmark when the 

 complexities of the Reptilian Shoulder-girdle are being sighted. 



To an anatomist familiar only with the skeleton of ordinary Osseous Fishes, the bone now 

 to be described would be an anomaly ; it is the " interclavicle" in the nomenclature of this paper ; 

 and in one form or another it will meet the eye until we are actually within the Mammalian circle. 

 This bone (figs. 6 8, i. cl.) is equal in size to the clavicle, but its ganoid part is much broader ; 

 this part, as in the clavicle, is bent at an acute angle on the inner plate; it is oblong, and is 

 notched in front and at its lower angle. The inner plate is, like that of the clavicle, convex 

 behind and concave towards the operculum ; it is very smooth and thin, so thin as to be fenestrate, 

 and its outline is four-lobed. In the Sturgeon there is this instructive condition, namely, the thin, 

 outspread lamina of the epicoracoid rests upon the upper part of this inner plate, as well as upon 

 that of the clavicle ; and this notwithstanding the overlapping of the interclavicle (figs. 6 S) : 

 I must recur to this when I come to speak of Polyptervs and its fossil congeners. The pras- 



