( 1T2 ) 



iinnilier known in an}- bird) ; ])nt tlio nlna is sliorteneil to aliont onp-tliird the 

 length of the hnnuTUS. In the flightless A/rn iiniifiiiii.i tlie hniuerns is long 

 and strong ; the hand is larger than the ulna, which is very nmcii shortened. 

 The nlna of normal parrots is abont ono-thinl longer than the humerns : in 

 Stririqojis it only equals it in length ; the hand is also much reduced, although less 

 so in proportion. In the Dodo and in the Solitaire the humerus is still a very strong 

 bone and scarcely reduced at all, while the forearm is much shorter, instead of 

 lunger, than the humerus, and the hand has shrunk extremely. The same applies 

 to Cai<//ar//i.i, Ih-omarns and Ajitrri/.r. Penguins, in s]iite of the greatly reduced 

 length of the forearm, belong to a category by themselves, special conditions 

 ])revailing in conformity with the transformation of the wings into efi'ective 

 screwing jiaddles. 



( 'onelusions: ( 'hange in the power of flight is above all correlated with 

 increased length or shortening of the forearm. Next, the hand is affected, last of 

 all the hnmerns. A much degenerated hand-skeb'ton is a sign of extreme n'duction. 

 This is cpiite understan<lable. The bird's hand is already a much reduced and 

 consoliilateil structure, shajie and size of which are jdedged by the closely packed 

 primarv ijuills ; and their nnndier is almost constant and can never more bo increased. 

 Not so with the ulna. If this is shortened, the cubital fpiills, instead of becoming 

 crowded within the series, dwindle away near the elbow-joint, and they gradually 

 lose their character of reniiges and are either crowded on to the humerus, or they 

 are lost. The reverse process takes ]daee when the nlna is mneli elongated, as for 

 instance in Tnbinares. 



The Sternion is not mncli smaller than it is in a normal cormorant of the size 

 of P. /larr/.v, but the keel is very ranch reduced, only the rostral portion remaining, 

 just sufficient to snjiport the fnrcula,* with which it is not anchylosed, and the short 

 snpracoracoidal muscles. The spina externa is lost, the anterior margin of the 

 rostrnm of the keel doing its duty. A remarkable feature is the wide separation 

 (18 mm.) of the two coracoids, the inner ends of which overlap, or at least touch 

 each other, in all other Cormorauts. 



The Sliouliler-girdle itself exhibits several other interesting points. The furcula 

 is still strong — scarcely, if at all, diminished in size. The coracoids are much 

 reduced in length and in strength. Their feet have remained as broad as they were 

 originallv, but they have crept asunder to the extent of IS mm., and the shafts have 

 been diminished at their median sides. The reduction of the scaj)ulae is extreme. 

 The shortening of the sca])nlae and coracoids, while thefurcnia remained unaffected, 

 and of course never gave up its attachments to the anterior end of the keel, has 

 brought about a peculiar readjustment in the relative positi<uis of the bones. The 

 shoulder-joint has, so to speak, sunk down, and has been shifted tailwards, with the 

 following results: First, the shafts of the coracoids and clavicles are parallel instead 

 of distally diverging : secondly, the scapula and coracoid form a right angle instead 

 of one of (i.j degrees only. 



( 'onsequently, Pli. Iiiirrixi shows five features which occur in the UnlHnc, only 

 that they are intensified in the latter. Hednction of the keel, widening of the space 

 between the coracoids, parallelism of coracoids and clavicles, shortening of the 

 coracoids, and widening of the coraco-scapular angh; to, and beyond, '.Mi degrees. 



* The two ol.iviriilaf fusing tojjether at their ventr.il cnils form llio furrnhi, :inil this wortl is .1 

 diminutive of furra, a forl<. /uirrnhnii. occa-'ionallv si-rMi in En«lisli ornithoIoL'ii-al paper-, is ic^ithfi 

 anatomical nor ciassic. 



