1S99.J TROM THE LOUDON CLAY OS SHEPPET. 783 



no important information as to the affinities of the bird. It can, 

 however, be seen that the bone is relatively rather short and stout, 

 and that the outer surface of the trochanter is broad and flat and 

 projects forward considerably in advance of the shaft. The 

 proximal end of the tibia (t.) is also preserved, but is too imperfect 

 for description. 



The cervical and dorsal vertebrae are represented by mere 

 fragments, and the only other bone of the skeleton at all well 

 preserved is the scapula (sc), the blade of which is nearly perfect. 

 It hes in approximately its natural position nearly parallel to the 

 vertebral column ; its tip just overlaps the front of the pelvis, and 

 its upper edge for a short distance conceals the lower border of 

 the mandible. The portion preserved is slender, but less so than 

 is the corresponding part of the scapula of Phaethon ; its distal 

 end is slightly expanded. Prom this bone as here preserved no 

 information of importaiice as to the affinities of the bird can be 

 derived. 



Comparison of the pelvis of Prophaethon with those of other 

 Steganopodes shows that in its general form it resembles that of 

 Sula most nearly. The chief differences are that in the fossil th-i 

 interosseous foramina are less distinct, the upper surface of the 

 postacetabular region of the ilia more convex from side to side, 

 and the pelvic escutcheon narrows less towards the hinder end. 

 The pelves of Phalacrocorax and Plotus somewhat resemble the 

 fossil in the expansion of the anterior end of the preacetabular 

 ilia, but differ from it in the large size and number of the inter- 

 osseous foramina which commence opposite the acetabulum, in 

 the general form of the pelvic escutcheon, and in the presence of a 

 sharp ridge (most prominent in Phtus) near the inner border of 

 the postacetabular region of the ilium. 



From the pelves of Fregnta and Phaethon the fossil differs greatly. 

 In both these genera the pelvis is very wide and siiallow, and the 

 ilia are widely separated throughout their length by the synsacral 

 vertebrae, the transverse processes of which are exposed, or at 

 least covered only with ossified fascia {e. r/., in part of the pre- 

 acetabular region oi Phaethon). In fact the fossil pelvis differs 

 much more from those of Phaethon and Firgata than from that of 

 any other of the Stegauopodes ; but since the skull shows con- 

 clusively that Prophaethon is by far most closely related to Phaethon 

 some explanation of this difference is necessary. If the pelves 

 and hind limbs of Freijata and Phaethon be examined, it will be 

 found that, in proportion to the size of the body, they are very 

 small and clearly in a degenerate condition. The explanation of 

 this seems to be that neither of these birds make use of their hind 

 limbs nearly so much as the other Steganopodes, for although no 

 member of the group employs its hind limbs to any great extent, 

 all except Phaethon and Fregata use them in swimming both on 

 and under the surface of the water. I have lately had an excellent 

 opportunity of observing the habits of both Frigate and Tropic 

 birds, and I believe that they subsist entirely on surface-fish and 



