1909.] IN THE SOUTHERN SEA-ELEPHANT. 803 



elephant now in the British Museum. These include two males 

 from Macquarie Island presented by the Hon. Walter Rothschild, 

 a male from Chatham Island obtained by Professor H. 0. Forbes, 

 a female from the " Antarctic Seas " obtained during the voyage 

 of the ' Erebus ' and the ' Terror,' and a male from the Crozet 

 group. With these, by the courtesy of Dr. A. Keith, I have been 

 able to compare an old male skull from the Falklands presei-ved 

 in the Museum of the Royal College of Surgeons, being the one 

 on which the above-mentioned paper by Sir W. H. Flower is 

 based. 



In comparing these skulls I find that the most satisfactory 

 distinctive characters are afibrded by the palatal surface, and it 

 is to this aspect that attention will be chiefly restricted. 



Great difference obtains in regard to the proportions of the 

 length to the width of the skull in the different local forms, as is 

 shown in the following table : — 

 Falkland. 



Basal length 20 ins. 



Maximum width ... 15 



Length of palate ... 11 



Width of do. ... 7-3 



Taking first the Falkland race, which, as already mentioned, 

 may be identical with M. I. ti/picus, but which it will be convenient 

 to call 31. l.falclandicus, the skull (text-fig. 183) is characterised by 

 its relative length and narrowness. The palate is also long and 

 narrow, nearly flat in the palatine region, but becoming suddenly 

 hollowed on the line between the 4th and oth cheek-teeth. The 

 palatines themselves form a long median suture ; the process of 

 the pterygoid is small ; and the premaxillse are long, and 

 V-shaped. As additional features may be mentioned the relative 

 narrowness of the condyles, and the circumstance that the lower 

 border of the anterior zygomatic root projects considerably behind 

 the posterior aperture of the maxillary foramen. 



A second skull in the Museum of the Royal College of Surgeons 

 from the Falklands agrees in essential characters with the above. 



Turning to the two Macquarie skulls, which are practically 

 identical, we find the general proportions not very different from 

 those of the Falkland specimen, but the palate (text-fig. 184, 

 p. 604) is much more hollowed*, and this throughout its whole 

 extent. Then, again, the inter-palatine suture is shorter and the 

 pterygoid process much larger ; while the condyles are wider, and 

 the lower border of the anterior zygomatic roof does not project 

 behind the hind aperture of the maxillary foramen. 



For the Macquarie race, as typified by the figured skull (B. M. 

 No. 1.6.22.1), I propose the name M. I. macquariensis. 



The natural supposition would be that the Sea-elephant from 

 the Chatham Islands would be identical with the one inhabiting 

 the Macquaries ; and this appears to be borne out by a skull of 



* I use the term hollowed in place of vaulted as being; more convenient. 



