46 A HISTORY OF BIRDS 



characters, while in the matter of the hind-limbs and hip-girdle 

 it has undergone extreme specialisation (p. 391). In the 

 pterylosis of the wing it shows more primitive characters, 

 probably, than any other group. By many, the extinct 

 /Epyornis of Madagascar is regarded as a close ally of StnitJiio : 

 but until more is known of the palate this must remain open 

 to question ; especially since the form of the hip-girdle, and 

 of the nature of the hind-limbs differs so conspicuously from that 

 of Struthio, and agrees rather with that which obtains in the 

 New Zealand Di norm's . 



The Rheas in the nature of the palate bear a striking similarity 

 to Dinornis ; but in the matter of the hip-girdle the two types 

 differ fundamentally, Rhea, indeed, exhibiting characters in 

 this region of the body that are absolutely unique (p. 392). 

 By many, including the present writer, the Tinamous of the 

 South American Continent are also to be included among the 

 Palaeognathae, being possibly allied to the Rheas. 



The remarkable Apteryx of New Zealand must, it would 

 seem, be regarded as one of the most aberrant of the Pateognathae, 

 though, by some, it is regarded as allied to the Moas (Dinornis). 



The discovery of further and more complete remains, especi- 

 ally of the skull, of an extinct type from the Lower Eocene of 

 Western Europe, now known as Gastornis, promises to throw 

 a flood of light on the subject of the evolution of the Palseognathae 

 (Ratitai), though according to some these remains are those of 

 an Anserine bird, a conclusion which is probably wrong. Only 

 a part of the skull is known, but this shows that the cranial sutures 

 were persistent, or at least slow to close, and further, that the 

 post-orbital region of the skull, as in the Cassowaries, was of 

 considerable length. 



The birds which exhibit the arrangement of palatine bones, 

 here designated Palceognathine^ should probably be regarded 

 as representing some six or seven sub-orders (including Gast- 

 ornis) of an Order Palaeognathae. Time may show that these 

 sub-orders must be further reduced, since Rhea, the Tinamous, 

 yEpyornis and Dinornis may prove to be descended from a 

 common stock. 



With the possible exception of Gastornis, the Palaeognaths, 

 both fossil and recent, are represented only by flightless forms, 

 excepting only the Tinamous, which still retain the power of 



