1896.] ZOOLOGICAL EXPEDITION TO MADAGASCAR. 977 
developed in JH. liberiensis, whilst in H. amphibius and the 
H. major of the Upper Pliocene the cranial portion is much 
reduced, the facial portion on the contrary enormously produced. 
In connection with this is the great elongation of the frontal bones 
of H. liberiensis, whilst they are broad and short in H. amphibwus 
and H. major. H. sivalensis is still very near H. liberiensis in this 
respect, the antero-posterior extension of the frontal being, as was 
shown by Falconer and Cautley, twice as great as in H. amphibius. 
An expression of the relative proportion between the anterior and 
posterior portions of the cranium is given by the position of the 
orbits. The various Hippopotamus crania from Madagascar have, 
in this respect, much resemblance with H. sivalensis, the cranial 
portion being, however, somewhat more shortened, the facial 
portion somewhat more lengthened; so that the orbit occupies a 
less central position than in H. liberiensis, and, as a matter of 
course, still less so than in H. swalensis. The Malagasy forms 
thus constitute a step farther in the direction of H. amphibius, 
the breadth of the intraorbital region being much less than in the 
African species and the same as in H. sivalensis. 
These changes are reflected by the position which the lachrymal 
occupies. In H. liberiensis, as shown by Leidy, who had at his 
disposal the skull of a younger animal, exhibiting distinctly all the 
sutures, the lachrymal is entirely separated from the nasals by the 
anterior prolongation of the frontal, which last thus comes in 
contact with the maxillary. This is, with the exception of the 
Ruminants, almost the rule in Ungulates. As to H. sivalensis, in 
six out of seven skulls figured in the ‘ Fauna Antiqua Sivalensis’ 
the sutures are distinctly to be seen; and we find here again the 
lachrymal excluded by the frontal from contact with the nasal and 
joining the maxillary. The originals of most of the skulls figured 
being in the National Museum, I have had an opportunity of 
verifying the accurateness of the drawings, so that we may fairly 
conclude that H. sivalensis had, as a rule, the character mentioned 
above in common with H. liberiensis. The same is the case with 
regard to H. paleindicus, as shown in the F. A.S8., with the shght 
difference that the anterior tongue of the frontal is. somewhat 
shortened. 
In the Malagasy Hippopotami we find, as a rule, the following 
relations in this part of the skull. The lachrymal departs from 
the orbital margin in an inward direction and reaches the nasal, 
with which it unites, thus shutting out the frontal from a connection 
with the maxillary. Anteriorly to the lachrymal, exactly corre- 
sponding to the place which in H. liberiensis and H. sivalensis is 
occupied by the foremost tongue of the frontal, we find here a 
separate bone of various dimensions, interposed between the uasal 
and lachrymal, and touching the maxillary in front and sometimes 
the malar bone as well. In H. amphibius the lachrymal is usually 
broadly interposed between the frontal and maxillary; but in 
ycung specimens we meet occasionally with the same supra- 
numerary bone; sometimes, as in H. hberiensis and H. sivalensis, 
