THE DEVELOPMENT OF THE MARSUPIAL SHOULDER GIRDLE. 763 



The Morphology of the Coracoid. 



Until comparatively recently there seemed little doubt in the minds of the majority 

 of morphologists that the rudimentary coracoid process of the higher mammals was 

 the true homologue of the larger and posterior of the two coracoidal elements in the 

 monotremes, and of the well-developed bone which unites the scapula with the sternum 

 in birds. 



In 1887, Howes (10) published a paper on the morphology of the mammalian 

 coracoid, in which he very ably discusses the whole question, and gives reasons for 

 doubting the correctness of the currently accepted views. The coracoid of the rabbit 

 and some other mammals he has shown to be developed from two distinct ossifications, 

 which he holds to be probably homologous with the two coracoidal elements of the 

 monotreme. The smaller of the two ossifications which forms part of the articular 

 surface for the head of the humerus he regards as the homologue of the posterior 

 element in the monotreme usually regarded as the coracoid, while the larger ossification 

 which forms the coracoid process he believes to be the representative of the ' epicoracoid ' 

 of the monotremes. His chief reasons for this determination are that in the rabbit the 

 ossification which gives rise to the coracoid process lies anterior to that which forms 

 part of the glenoid cavity, and also that this larger ossification, like the ' epicoracoid ' in 

 the monotremes, takes no part in the formation of the glenoid surface. The ' epi- 

 coracoid' of the monotremes he regards as the homologue of the anterior coracoidal 

 element in Dicynodon, but does not regard it as homologous with the precoracoid of the 

 Amphibia. His reason for this latter opinion is that the true precoracoid is believed to 

 become absorbed by the clavicle in the mammalia and most of the higher vertebrates. 



In 1893, he published (14) a second paper, giving the results of his further researches 

 in the development of the mammalian coracoid process, and also dealing with the 

 terminology. As Howes holds that the two coracoidal elements of the monotreme are 

 together homologous to the single coracoid of the bird or the true coracoid of the 

 Amphibian, he proposes calling the posterior coracoidal element the ' metacoracoid,' as 

 suggested by Lydekker, and to retain the name ' epicoracoid ' for the anterior element. 



In this same year, Lydekker (15), by a different argument, came to a conclusion some- 

 what similar to, though not identical with, that of Howes. By comparing the shoulder 

 girdle of the sloth with that of Dicynodon he comes to the conclusion that the coracoid 

 of the sloth is the homologue of the anterior coracoidal element in the Anomodont, from 

 their having somewhat similar relations ; and consequently, that the mammalian coracoid 

 process is homologous with the ' epicoracoid ' of the monotremes. 



If we compare the early condition of the marsupial coracoid with that of the 

 Monotreme or Theromorph there seems no difficulty in homologising the cartilaginous 

 element with the posterior of the two coracoidal elements in the lower forms. In the 

 monotreme it is the posterior element which forms with the scapula the g]enoid cavity, 



