32 SHOULDER- GIRDLE AND BREAST-BONE. 



rest completely on the first abdominal-line scale; and are wholly behind the clavicle. The presence of 

 the first of these splints would have been no difficulty ; the presence of a second need be none, 

 when " Nature herself teaches us " how she has subdivided the clavicle in Coitus. The foremost 

 bone (a. i. cl.) is like a hatchet, set on end ; the pointed haft running up behind the descending 

 cms of the clavicle. The next bone (p. i. cl.), or " posterior inter-clavicle," has a blade below, 

 and a haft, set on obliquely, above : this haft is flat, and broad at the top ; it is turned somewhat 

 outwards and backwards. The meaning of this broad top will be seen when we come to Ostracion ; 

 its double lamina and open end caused me trouble at first ; but high magnifying powers show 

 nothing more than fine fibrous tissue in the interspace between the outer and inner plates ; so 

 that it is a purely aponeuorotic bone. 



It is well that there is always a good landmark in the main Shoulder-splint, namely, the 

 clavicle, for, whilst one of the typical upper splints is wanting here, there is the apparent con- 

 fusion of two atypical bones below. 1 



The true Shoulder-girdle of Syngnathus is of extreme interest, as it is persistently soft, and 

 has undergone no segmentation. 



This curious thin flap of hyaline cartilage (see Woodcut, fig. 3 C, where the right moiety 

 is shown from the inside, magnified 10J diameters) increases in width in passing from the 

 scapular region (sc.) to the coracoid (cr.). The scapula is somewhat emarginate above, and the 

 coracoid below ; and the posterior margin is crenate, and interdigitates with the brachial rays 

 (b. 1 4). The outline of the entire cartilage reminds one of those Caterpillars that lift their head 

 whilst crawling over the plant on which they feed. The gently sinuous, scooped front margin of 

 the cartilage can scarcely be said to be " notched," and there is no " fenestral" rudiment of a 

 cleft. The small " feet" which insinuate themselves between the massive, bony brachials have 

 a clean margin, towards which, as towards the whole circumference, the cartilage cells become 

 flat and parallel with each other. 



The brachials (fig. 3 C, b. 1 4) are deeply notched both above and below, the hinder 

 parts being a partly separated second series. The anterior brachial moieties are almost semi- 

 circular, and lie in the moon-shaped " glenoid" spaces of the coraco-scapular cartilage. They 

 have a core of cartilage in their inner part ; but their outer halves, or post-brachial moieties, are 

 intensely ossified, and converted by periosteal growths into extraordinary stag's-horn appendages. 

 On the outer side, some of these become upturned hooks of great length, and flattened like 

 anchor-flukes ; so that the rays of the fin-root are in the strongest histological contrast to the 

 coraco-scapular foundation of the fin. 



In the still more bizarre forms seen in the genus Hippocampus the abdominal-line series are 

 much more carinate than in Syngnathus; and the whole of its structure would well repay the morpho- 

 logist who would work it well out. But the most unlooked-for modification of the ichthyic verte- 

 brate is to be seen in the genus Phyllopteryx, Kaup. I refer the reader to the exquisite plates 

 (by Mr. G. H. Ford) in the 'Transactions of the Zoological Society' for 18G5 (Part I, plates 

 14 and 15, p. 327). A knowledge of the structure of the common kinds of Syngnathus 

 makes this strangest of types quite intelligible. 



1 The relation of these Fishes to the generalised Ganoids is well shown by their possession of 

 " malars" or " jugals," and still more by the presence of the "jugulars," so rare in the Teleostei, namely 

 in Slops and Megalops (Cuv., ' Reg. Anim.'). These are well developed in Syngnathus and Hippocampus ; 

 although I am not aware that any other Ichthyotomist has observed this fact. 



