222 ME, W. P. PTCEAFT ON THE MOEPHOLOGT AND 



sterni, near the middle of its extent (fig. 4 c), is an area of bone conspicuous on 

 account of its thinness. Now it seems more than probable that these thin areas are 

 the last traces of a once deep fissure which has gradually closed up, thus uniting to 

 the body of the sternum a pair of posterior lateral processes. Parker has noticed a 

 similar indication of the closui'e of these fissures in Apteryx. 



Fusion probably first took place, in the case of Rhea, between the free ends of the 

 processes and the mesosternum by syndesmosis of the membrane, by which the fissure is 

 invariably closed. A slight expansion of the free ends of the processes and raetasternum, 

 such as is indicated in Tiiiamidoe, would materially hasten this process of closing. It 

 is also to be noted that the membrane closing this fissure is generally thicker posteriorly 

 than anteriorly ; this also would contribute towards the work of obliteration. If, once 

 again, we imagine these fissures open, the resemblance between the sternum oi Rhea 

 and Cryptnri will be found to be significantly striking. 



It has been alieady suggested by Gadow that the protuberantia sterni may represent 

 a degenerate keel. This seems quite probable. Seen in profile its keel-like resemblance 

 is quite as close as that claimed for Aptertjx by Parker (fig. 4 d). The sternal callosity 

 so conspicuous in the living bird is represented amongst the Neognathw in Opistho- 

 comus and is a quite secondary character. 



b. The Pectoral Arch. 



The shoulder-girdle of the flightless Pala'ognathce has undergone considerable 

 modification and bears evidence throughout of retrogression. The coracoid and scapula 

 are no longer free, except in the very young bird, but completely anchylosed, forming, 

 the one with the other, a very obtuse angle, instead of an acute angle as in modern 

 birds. 



In Droriutus the sternal (epi-coracoid) border of the coracoid, though wide, is less than 

 the length of the whole coracoid. Only the inner half of this border is received within 

 the coracoid grooves ; its outer portion is free and underlies the anterior-lateral process. 

 The pro-coracoid is ligamentous. The supra-coracoid foramen is large and receives 

 several pneumatic apertures. 



The scapula is slender, and looks like an outward, upward, and backward outgrowth 

 of the coracoid. It bears a strongly marked articular f;xcet for the vestigial furcula. 



The pectoral arch of Dromceus ater does not appear to differ materially from that of 

 D. novm-hollandioe. I gather this from a sketch by Dr. Henri P. Gervais, kindly 

 made for me at the request of Prof. Oustalet, to whom I am much indebted, as 

 well as to Dr. Gervais. 



The following measurements (in millim.) accompanied the sketch of Dr. Gervais : — 



Length from pro-coracoid process to tip of scapula 0"127 



Greatest length of coracoid 0-062 



Width, epi-coracoid region 0-047 



