310 Mr. F. E. Beddard on the 



wanting in A^. ci'istata and in its immediate ally. A^. eduardi. 

 Except for the absence of this foramen the manubrium 

 sterni of Gallus comes nearest to that of N. vulturina, &c. 

 Another feature which seems to distinguisli the sternum of 

 Gallus from that of Numida is in the attachment of the 

 sternal ribs. In all the species of Numida that I have exa- 

 mined, which comprise all those mentioned in the course of 

 this paper (with the exception of N. meleagris) , there are, as 

 in Gallus, four sternal ribs with attachments to the sternum. 

 Of these the last three are crowded together, the first rib 

 being at some little distance in front of this group of three. 

 InGallus, on the other hand, there is a progressive diminution 

 in the distances between the sternal articular surfaces of the 

 ribs from before backward. 



A final characteristic which differentiates the sterna of 

 Gallus and Numida is the breadth of the anterior region, 

 which is much wider in Gallus. The following measurements 

 are taken at the insertion of the second pair of ribs : — 



Leugth. Breadth, 



millini. millini. 



Gallns banhiva 117 35 



„ H2 33 



Nnmidn mitrata 11 o 25 



„ vuHurina* 108 28 



,, cristata 121 27 



„ eduardi 130 28 



„ meleagris 113 25 



„ 2itilorhyncha 105 25 



§ The Scapula and the Clavicle. 

 Prof. Huxley has directed attention to the fact that the 

 acromion is recurved in a hook-like fashion in Numida, and 

 that it is not in Gallus. The existence in Numida cristata 

 and N. eduardi of a hollow enlargement of the conjoined 

 ends of the clavicles is well known. This, of course, does 

 not exist in Gallus. I may point out that in those species 

 oi Numida with an inflated box the two clavicles are disposed 

 in a direction more parallel to each other, and are not bowed 



* Slightly diptorted at the extremity. 



