198 
TEETlAllY YEKTEBKATA OF THE EATOM. 
that Dr. Elliot Sinitli is of opinion that tin ir generic separation would be justified by 
this difference alone; (3) the os innominatum is much less reduced and has a com- 
])l('tely closed obturator foramen and a well-defined acetabulum with which a functional 
femur probably articnlated, whereas in Eosiren the reduction of the pelvis has proceeded 
almost as fixr as in Jlalitherium. Those particulars are taken partly from Abel’s paper 
above referred to, and j^artly from a skull and mandible of Eotherium recently acquired 
by the Geological Museum, Cairo : to this specimen further reference, with figures, is 
made below. 
Erorastomus'^ , another primitive Sirenian from the Lower Tertiary beds of Jamaica, 
differs from the present form both in the general shape of the skull, the slight deflection 
of the snout, the absence of enlarged anterior incisors, and in the form of the 
mandibnlar symphysis. 
Ilalitherium {P7'ototliermm)veronense,7A^xiO’\, approaches this genus very nearly, but 
in it the dentition seems to have already undergone considerable reduction. 
Eosiren libyca, Andrews. 
[Plate XX. figs. 1-7 ; text-figs. 64, 65, 68 B.] 
1902. Eosiren liJo/ca, C. W. Andrews, Geol. Mag. [4] vol. ix. pp. 293-4, figs. 1-3. 
1904. „ ,, 0. Abel, Abhandl. k.-k. geol. lleichsanst. vol. xix. pt. '2, passim. 
Type Specime^i . — A nearly complete skull described and figured, loo. cit. ; Geological 
Museum, Cairo. 
This is the type and only species of the genus at present known. 
Form. (& Loe. — Qasr-el-Sagha beds (Middle Eocene) : north of Birket-el-Qurun. 
S/iitll (PI. XX. figs. 1, 1 A, 1 B ; text-fig. 64). — The type skull is not quite complete, 
the zygomatic arches and a portion of the basal region of the cranium being lost. The 
occipital surface is high and broadens out from above downwards. The occipital 
condyles {cond.) are larger and more prominent than in any other Sirenian skull 
with which comparison has been made, except Eother'nmi \ they are almost pcduncidate 
and their articular surface is strongly convex from above downwards and extends 
considerably on tlieir ventral face. In the mid-ventral line they are sc])arated by a 
deep rounded notch ; but it cannot be determined what share, if any, the basioccipital 
took in their formation, the sutures being obliterated. 
Above the foramen magnum the exoccipitals[exo.) meet in a suture about 2'3 cm. in 
length, thus completely excluding the supraoccipital from the opening; their upper 
■* (^uart. Joiirn. Geol. Soc. vol. xi. (185")) p. 541. 
t /igiio, “Sireiiii fossil! trovati nel Veneto,” jMcid. 1st. Vencto (1. Sci. vol. xviii. (1875) ]>. 415; also by 
the same author, “ Quehjues Observations sur les 8irdiiiens i’ossiles,” Bull. Soe. geol. Franco, [3J vol. xv. 
(1887) p. 728, 
