23O NEW YORK STATE MUSEUM 



The large, straight and stout coracoids of Crax are also pneu- 

 matic, as are also the powerfully developed scapulae. In the former, 

 very extensive apertures are found at their bases on the posterior 

 aspects, while in the latter, they occur as small foramina on the 

 under sides of the anterior ends. The os furcula is of the U-pat- 

 tern, and has a remarkably long and somewhat slender hypo- 

 cleidium of nearly three centimeters in length. Above, the moder- 

 ately dilated free end of either clavicular end makes an extensive 

 articulation with the corresponding scapula, but does not, in the 

 specimen at hand, come in contact with the coracoid of the same 

 side. Below, the coracoids fit in deep grooves on the sternum, and 

 apparently come in contact in the median line through a foramen 

 that pierces the manubrium sterni, as seen in so many of the 

 Gallinae. 



In P h a s i a 11 u s c o 1 c h i u s [pi. 1, fig. 6 ; pi. 2, fig. 18 ; pi. 6, 

 fig. 28] the skull reminds me much of the skull in Centrocercus 

 [pi. 2, fig. 16], and in reality this part of the skeleton is consider- 

 ably alike in the two forms. In Phasianus, however, the frontal 

 processes of the premaxillary not only persist throughout life as in 

 the Sage cock, 'but they are produced backward farther between the 

 nasals. The external narial apertures are both really and compara- 

 tively smaller in Phasianus, and in its skull, too, the pars planae 

 do not ossify as they conspicuously do in Centrocercus. A con- 

 spicuous vomer is present in the pheasant, whereas in the Sage 

 cock if it ossifies at all it is in the most rudimentary manner. For 

 the lacrymals and the balance of the cranium, these two birds are 

 markedly alike. 



Xearly the entire skeleton in Phasianus is highly pneumatic, but 

 this condition is not enjoyed by the bones of the leg (except the 

 femur), and those of the antibrachium and pinion. The pattern 

 of the sternum agrees with that of Gallus and the grouse, but the 

 pelvis is entirely different from that bone in Centrocercus, coming 

 nearer as it does to the pelvis in Gallus [compare the figures in the 

 text and plates, especially pi. 6, fig. 28, and pi. 7, fig. 31]. Four of 

 the dorsal vertebrae in Phasianus coossify to form one bone, their 

 common neural plate being perfectly smooth, and showing no traces 

 of the original separate neural spines. The skeleton of the limbs 

 reminds one of G. bankiva in no small degree, and, in reality 

 the various bones composing them are typically galline. 



Argus giganteus has an interesting skeleton in many re- 

 spects, and it is not only thoroughly after the gallinaceous order, 



