48 CARINAT.T?. 



The specific name emidnus, although the earlier, is rejected as 

 being inappropriate. The type of Alegalornis was ideutified by its 

 describer with LifJiornis. 



Hah. Europe (England). 



The following specimens, or their originals, were obtained from, the 

 London Clay (Lower Eocene) of the Isle of Sheppetj. 



A. 1. The imperfect cranium, probably belonging to this species. 



{Fi<l.) Erom its somewhat small size as compared with the limb- 

 bones, this specimen may indicate a female. Figured by 

 Owen in the ' Quart. Journ. Geol. Soc' vol. xxxvi. pi. ii. 

 In the description this specimen was compared with the 

 skull of Diomcdca, from which it is at once distinguished 

 by the presence of the transverse hinge between the 

 frontals and the rostrum, and also by the absence of 

 supraorbital grooves. In these respects, as well as in 

 general contour (allowance being made for the depression 

 of the frontal region through crushing), it accords very 

 closely with the skull of Sida. This is well shown by the 

 form of the rostrum, in which the lateral groove formed 

 by the junction of the two portions of the horny sheath 

 has precisely the same position. Again, the lachrymals 

 have a similar depression on the external surface between 

 the expanded extremities, and are totally different from 

 those of Diomedea. An equally close resemblance to Sula 

 and difference from Diomedea is displayed by the form 

 and position of the posterior nares. Purchased, 1880, 



A. 5-9. Fragments of the two associated humeri. The types; 



(Fig.) figured by Owen in the ' Quart. Journ. Geol. Soc' vol, 

 xxxiv. pi, vi. A. 5 is the proximal part of the right 

 humerus, represented in fig. 3 of the plate ; in the form 

 and position of the head, subtrochanteric fossa, and delto- 

 pectoral crest, it has all the characters of Sula, and is 

 totally unlike Diomedea. A. 6 are two portions from the 

 distal end of the left humerus, with the adjacent ex- 

 tremities cut and polished, of which the distal fragment 

 is represented in figs. 11, 12 of the plate. In these 

 figures the specimen is assumed to comprise the whole of 

 the distal end, whereas the condylar region is really 

 wanting ; the surface marked l.m. in fig. 12, and corre- 

 lated with the whole of the surface for the brachiaHs 



