428 Mr. H. Seebohm on the 



Alectorides of Sclater, which are synonymous with the Gera- 

 nomorphffi of Huxley. I see no reason why Opisthocomus 

 and Cariama should not also be included in the Grallae. The 

 former has been regarded as closely allied to the Cracidse, 

 and the latter to the Serpentariidae; but in each case there 

 are at least six facts which are opposed to their supposed 

 affinity, but are consistent with their close relationship to the 

 Grallse. They appear to be like Mesites, Eurypygus, Ehino- 

 chetus, and Psophia, the scattered remnants of a once large 

 and widespread group, from which the Gruidae and Rallidse 

 are also descended. 



Opisthocomus differs from the Gallinae in the following 

 particulars : — 



1. It has no basipterygoid processes. 



2. It has subclavicular processes. 



3. The posterior margin of the sternum is not very deeply 

 cleft. 



4. It has very few ventral processes on the dorsal vertebra. 



5. The angle of the mandible is not prolonged and recurved. 



6. The episternum is not pierced so as to allow the cora- 

 coids to meet at their bases. 



In each of these six characters in which it differs from the 

 Gallinse it agrees with the Otididse. 



Cariama differs from Serpentarius in the following par- 

 ticulars : — 



1. It is schizognathous. 



2. It has no basipterygoid processes. 



3. It has subclavicular processes. 



4. It has no median process to the furculum. 



5. It has lateral processes to the sternum. 



6. It has very few ventral processes on the dorsal vertebra. 

 In each of these six characters in which it differs from 



Serpentarius it agrees with the Rallida^. 



The following families appear to be sufficiently allied to 

 be associated together in a suborder, to which the name of 

 Grallae may be applied : — 



