18()8.] PROF. HUXLKY ON THE ALECTOROMORPBiE. 299 



In the Alectoropodes — 



1. The osseous junction of the lophosteon and metosteon is nar- 

 row, the inner notch being always more than half as long as the 

 sternum (fig. 2, p. 297). 



2. The costal processes of the pleurostea (c p, fig. 2) are more 

 prolonged and more nearly parallel with the axis of the sternum 

 than in the preceding case. 



3. The hallux is raised above the level of the other toes, and its 

 basal phalanx is much shorter than that of the third toe. 



With a single exception, the second metacarpal always has a 

 backward process*. A tubercle is very commonly present upon the 

 posterior edge of the base of the phalanx of the third digit ; and 

 the ilio-pectineal processes are generally very well developed. The 

 vomer, wherever I have been able to observe it, has been weak and 

 flattened from above downwards. 



Three groups are readily distinguishable by osteological charac- 

 ters among the Alectoropodes. 



The Numididen-f differ from the other members of this division in 

 the absence of any backward process of the second metacarpal, 

 and in the obtuseness and somewhat outward inclination of the 

 costal processes. The acromial process of the scapula is also sin- 

 gularly recurved. 



In all the rest the backward process of the second metacarpal is 

 distinctly developed, and the costal processes are more acute (gene- 

 rally very much so) and pass more directly forwards. Among these 

 the Meleagridce are peculiar in three respects. 



1 . The length of the ilium from tlie centre of the acetabulum to 

 its posterior margin (which may be called the postacetabular length) 

 is greater than the distance from the same point to the anterior 

 margin of the ilium {ox prceacetahidar length). 



2. Viewing the pelvis from above, the postacetabular area is 

 longer than it is broad (fig. 5, p. 300). 



3. The furcula is singularly weak and straight (viewed laterally), 

 and has a straight rod-like hypocleidium. 



In all the other genera which I have examined, the prseacetabular 

 length is greater than, or, in the solitary case of Tetrao cupido, equal 

 to, the postacetabular. The postacetabular area is broader than it 

 is long ; the lateral contour of the furcula more curved ; and the 

 hypocleidium expanded antero-posteriorly. 



The great series of Galline, Pavonine, Phasianine, and Tetraonine 

 birds included under the title of Phasianidce, which offer these 

 characters, present two types of structure, the one of which may be 

 termed Galline, and the other Tetraonine, and which are well defined 

 and contrasted in their extreme forms, though I am by no means 

 clear that they do not graduate into one another. 



* M. Blanchard (/. c. p. 99) gives the pi-esence of this process as a univei'sal 

 character of the " Gallinaces," merely mentioning that "dans les IIoccos et Ics 

 Penelopes elle s'affaiblit beaucoup." 



t That is to say, the species of the genus Nitmida. I have seen no skeletons 

 of Agelastus ov Phasiduf:. 



