620 Development of the Pectoral Girdle in the Pipefish. 



With tlie reduction of the scapula the intermediate carti- 

 lage is correspond iugly increased and carries the four 

 proximal radials, which, as in the case of the scapula, become 

 distinct from the remainder of the cartilage by their ossifi- 

 catiou. The transfer of the direct support of the pectoral 

 fin from the internal to the external skeleton renders a largely 

 ossified, coraco-scapular region nnnecessary. 



The coracoid becomes defined by its growth and by the 

 constriction of its connexion with the remainder of the 

 cartilage. The precoracoid process is distinct at first, but 

 eventually forms part of the mass of the coracoid. A poste- 

 rior process is also developed which, although not actually 

 preformed in cartilage, but arising rather as an extension 

 of the ossifying coracoid, occupies the position and relations 

 usually associated with the postcoracoid process in teleosteans. 

 The postcoi'acoid process becomes attached to the jugular 

 plate, although it does not fuse with it. Its relations thus 

 present an interesting resemblance to those of the post- 

 coracoid process and infraclei thrum in Gasterosteus as 

 described by Swinnerton ^. Whether this can be regarded 

 as more than an analogy is extremely doubtful. After 

 ossification has commenced the vertical limb arises as an 

 extension of the coracoid and restricts the movements at the 

 coraco-clavicular joint. The coracoid thus acts as a bracket 

 for the rigid support of the jugular plate at its attachment 

 to the clavicle. 



Jungersen was the first to recognize this bone as the 

 coracoid, previous writers having usually regarded it as an 

 interclavicle. The facts of its development as here shown 

 leave no doubt that Jungersen's interpretation is correct. 



I have to express my gratitude to Professor M'Intosh for 

 continual stimulus and encouragement, and Dr. G. A. 

 Boulenger for his kindly advice and assistance ; also to 

 Professor Sutherland, of University College, Dundee, for 

 the freedom of his laboratory during part of the research. 

 To Dr. Williamson, of Nigg Bay, Aberdeen, I am much 

 indebted for assistance in obtaining material for the work. 



* H. H. Swinnerton, " A Contribution to tlie Morphology and Develop- 

 ment of the Pectoral Skeleton of Teleosteans," Q. J. M. S. xlix. ii. (1905). 



Swinnerton found no postcoracrid process in SipJwnostoma, a closely 

 allied genus, but it is possible that the process had been overlooked or had 

 not yet appeared. 



