GREGORY, PRESENT STATUS OF ORIGIN OF TETRAPODA 373 



A, Sphenodon, right shoulder-girdle, outer view ; B, ditto, ventral view. The coracoid 

 is single. The metacoracoid may have disappeared. The epicoracoid (ep.co.) as in the 

 lizard (C,D) is bounded by the scapula, coracoid, interclavicle and clavicle. Beneath the 

 T-shaped interclavicle is the intercoracoid, or anterior sternal region. 



C, Varanus, left shoulder-girdle, outer view ; D, ditto, ventral view. 



In Sphenodon and Varanus the so-called "suprascapula" (f cl) may possibly represent 

 the cleithrum of primitive Tetrapoda (cf. Fig. 14 D). The "suprascapula," like the 

 cleithrum, is connected below with the clavicle ; to it is attached the trapezius muscle. 

 The coracoid is single ; anteriorly it is fenestrated for the pectoral muscles. 



E, Shoulder-girdle of a Cotylosaur, Latidosaurus. Ventral view, after Wiillston. The 

 coracoid and metacoid are distinct ; the cleithrum is absent. 



F, Shoulder-girdle of a Permocarboniferous reptile, Edaphosaurus novomexicanuSj 

 after Williston and Case. Right side. The expanded coracoid region, short scapula and 

 large clavicle are perhaps inherited from primitive amphibians. The metacoracoid is 

 present. 



and with the glenoid; its wide distribution as a vestigial or reduced 

 ossific center in various orders of mammals suggests that it is an ancient 

 reptilian inheritance. 



In reference to the shoulder-girdle of monotremes we would apply the 

 name coracoid to the posterior element^ which forms part of the glenoid 

 (Fig. 13), since this part closely resembles the true coracoid of embryo 

 marsupials. The anterior element (Fig. 13, ep. co.) may correspond 

 with the epicoracoid (Fig. 14) of lizards and Sphenodon, as suggested 

 by Wiedersheim (1909, p. 190), which never participates in the glenoid 

 and always lies beneath the interclavicle and clavicle. Thus the true 

 coracoid of monotremes may either have resulted from the loss of the 

 suture between the coracoid and metacoracoid or the metacoracoid may 

 have disappeared, while the epicoracoid has been developed, perhaps in 

 adaptation to fossorial habits. 



In the lizards, Sphenodon and other reptiles with a "single" coracoid 

 (Fig. 14), this singleness may also have resulted either from the loss of 

 the suture separating the coracoid from the metacoracoid, or from the 

 disappearance of their coracoid. Their epicoracoid is also well developed 

 and ossified. 



In the lower Tetrapoda (Fig. 15), including various Permian orders, 

 the coracoid and metacoracoid are often divided by suture, but the epi- 

 coracoid is not ossified. 



In Ceratodus (Fig. 5), the sturgeons (Fig. 4) and the sharks, which 

 have the best developed and largest coracoscapula cartilages among recent 

 fishes, there is no sutural separation of parts and it seems likely that the 

 segregation of ossific centres corresponding to the coracoid and metacora- 

 coid is a later advance, perhaps correlated with the great expansion of the 

 coracoscapular mass and the higher differentiation of the limb muscles 

 in tetrapods. 



