436 MR. F. E. BEDDARD ON THE [ JuDC 3, 



the keel is not deep as in many purely flying birds. Eurypyga lias 

 a very deep keel to the sternum. 



The clavicles have a large interclavicular piece which is prolonged 

 in front as well as behind. 



Here, again, Podica appears to be peculiar, or rather to resemble 

 Ardea, which has also an anterior and posterior interclavicular 

 process. 



Among the supposed near allies of Podica, Coh/mbus and Podiceps 

 have both a posterior interclavicular process only, which is present, 

 though extremely small, in Eails. 



The pelvis of Podica is in some respects like that of many Rails ; 

 but in other particulars again it is Grebe- or Diver-like, The ilia in 

 front (see fig. 5, p. 437, and fig. 6, p. 438) do not reach up to the 

 top of the spines of the dorsal vertebras ; each ilium also is deflected 

 away from the vertebral column and overlaps several ribs. In most 

 Rails that I have examined, the iUa completely cover the vertebra), 

 being fused with the summit of their neural spines ; this is the case 

 at any rate with Ocydromus and Aramides. In Trihonyx and 

 Ftilica the ilia do not completely cover up the dorsal vertebrae which 

 they overlap ; but in both these forms the ilia approach each other at 

 their anterior extremity, and are not deflected away from each other 

 as they are in Podica. In this particular the pelvis of Podica is 

 decidedly Colymbine. 



The general outline of the pelvis is as decidedly that of a Rail, 

 being wider behind than in front ; but in all the Rails which I 

 have examined the pelvic bones are perfectly free from the ischia, 

 though they do not extend very far behind the termination of the 

 latter. 



In Podica there is in places a close union between the pubis and 

 ischium of each side, amounting to a synostosis, while the pubic 

 bones themselves extend for about an inch in length beyond the 

 ischia, and are curved inwards towards each other. 



As regards the non-fusion of the pubis and ischia, the Divers are 

 at one extreme and Podica at the other — the Rails occupying an 

 intermediate position ; so that, although Podica is in some respects 

 peculiar, it must be considered as coming nearer to the Rails than 

 the Grebes. The backward extension of the pubes is, however, a 

 Grebe-like character, though it is after all rather slight. The ilia of 

 Podica. end posteriorly in a strong blunt point on each side, about 

 as long as the first two vertebrae, and there is no strong ridge, such 

 as we meet with in the Rails ; the hinder part of the pelvis is 

 smoothed and rounded. Although there are differences here from 

 the Rails, it cannot be said that there are any marked resemblances 

 to the peculiar pelvis of the Diver and of the Grebe. 



The anterior parts of the ilia together with the intervening 

 vertebrae are narrower than the posfcacetabular portion of the pelvis ; 

 but the disproportion is certainly not so great as in most of the Rails : 

 the long and narrow pelvis of Fidica ardesiaca, though it does not 

 approach in shape that of Podica, does so in the proportion of the 

 anterior and posterior regions ; so also that of Aramides, with which 



