13 



He would divide the Ratitse into six Orders : — 



1. ^PYORNITHES Fam. ^pyornithid^. 



2. AFTER YGES Fam. Apxeeygid^. 



3. IMJMANES Fam. 1. Dinoenithid^. 



2. Palapterygid^. 



4. MEGISTANES Fam, 1. Casuariid.=e. 



2. Drom^id^. 



5. RHE^ Fam. Rheid^. 



6. STRUTHIONES Fam. Struthionid^. 



In discussing the Carinata, Professor Newton exhibits the utmost 

 caution, and is unwilling to commit himself to a new arrangement of 

 the Subclass, but he points out that the Crypturi must be considered as 

 the first Order of the CarinatcB, as proposed by Professor Huxley, in 

 whose classification they follow the Ratita. He next insists on the 

 distinctness of the Impennes. He doubts the afiinity of the Podici- 

 pedidce with the Pygopodes, and he would, associate the latter with the 

 Gavioe and Limicolce as one Order, the Tubinares being distinct and 

 forming a separate Order by themselves. 



The Grallce would contain the Fulicaria and Grues, the contents of 

 the latter being the Gruidce, Psophiidce, and Aramidce, along with Eury- 

 pyga and Rhinochetus. 



The Gallince should rank as an Order, the Hemipodii, " standing 

 somewhat apart,^^ the Alectoropodes (normal Gallince), and the Peri- 

 steropodes [Megapodiidce and Cracidcs) forming the three component 

 Suborders. Opisthocomus is the single representative of a separate 

 Order. 



The Pteroclidce are an intermediate group between the Game-Birds 

 and Pigeons, and must either be raised to the rank of an Order, as 

 proposed by Huxley and Gadow, or they should belong to one large 

 group, which should contain the Gallince and Columbce. The author 

 does not fail, however, to emphasize the fact of the Dasypsedine con- 

 dition of nestling Sand-Grouse, a fact he has since confirmed in an 

 interesting memoir in 'The Ibis' for 1890 (pp. 207-214, pi. vii.). 

 The Columbcs would possess two Suborders, the Didine birds and the 

 true PigeonSj with which latter group Didunculus would rank, being 

 probably connected with the more typical Columbidcs by the Papuan 

 genus Otidiphaps. 



The GavicB are closely allied to the Limicolce ; and the Grallce, through 

 Eurypyga and Rhinochetus, allied to Limicolce in some respects, also 

 trend toM^ards the Herodiones. The Gavice are further allied through 

 Phaeton to the Steganopodes. 



Cariama is an indefinite form, possessing some Accipitrine characters. 



