24 THE VOYAGE OF H.M.S. CHALLENGER. 



surfaces corresponded to the direction of the surfaces of their transverse processes, which 

 surfaces were flattened and not rounded. The eighth and ninth dorsal vertebrae had 

 no transverse process projecting from the side of the neural arch, but instead a massive 

 process was directed horizontally outwards from the side of the body nearer the anterior 

 than the posterior surface. This process was nearly three times as big in the ninth as 

 in the eighth vertebra, and in both it had a large articular surface at its outer end for 

 the head of the corresponding rib. Zygapophyses were present as far back as the anterior 

 pair of the sixth dorsal vertebra, but behind that they had disappeared, and a pair of 

 well-developed metapophyses projected forwards, from the laminae of the seventh, 

 eighth, and ninth dorsal, to overlap and articulate with the laminae of the vertebra 

 immediately in front. 



Compared with Drvon Haast's specimen this animal has one vertebra less in the dorsal 

 series. This does not invalidate the opinion that they are of the same species, as it is 

 well known that in the Cetacea variations to the extent of a vertebra and a pair of ribs 

 may take place in the thoracic region. In Mesoplodon grayi and Mesoiolodon soiuerhyi 

 there are also ten dorsal vertebras, whilst in Mesoplodon austrcdis the number is only nine. 

 The transverse processes in my sjDecimen were not rounded as in Dr von Haast's animal, 

 and the articular surfaces for the heads of the ribs did not appear to rise quite so far 

 above the base of the neural arch as he describes. 



The lumbar vertehrce were almost uniform in shape, but increased in size from before 

 backwards. Tlie bodies were keeled on their inferior surface. At the anterior and 

 posterior ends of the series the longitudinal and transverse diameters of the body were 

 almost equal, but in the intermediate vertebrae the longitudinal was greater than the 

 transverse. The transverse processes were not so long as the width of the body except in 

 the anterior vertebrae. The base of the process sprang from the anterior half of the side of 

 the body in series with the transverse processes of the eighth and ninth dorsal vertebrae 

 — the processes projected forwards and outwards, and the free end was convex. The 

 spines were long, laterally compressed, and sloped slightly backwards : the length of the 

 eighth lumbar spine was 6 inches. A pair of broad lamelliform metapophyses projected 

 forwards from the anterior border of the laminae close to the root of the spine, but did not" 

 articulate with the vertebra in front, from the posterior edge of the laminas of which a 

 pair of much smaller processes projected backwards. As in Dr von Haast's specimen, 

 the neural arches sprang from the centre of the bodies. 



The caudal vertehrce diminished in size from before backwards. The first was 9-| inches 

 high and 7 inches between the tips of its transverse processes. The last was 9-lOths of 

 an inch in greatest breadth, and 7-lOths of an inch in height. In the anterior four 

 vertebrae the spines were massive, and these processes were present as far back as the tenth 

 caudal, in which the neural arch and spine formed a slight ridge, and the spinal canal was 

 diminished to the diameter of a goose-quill. The transverse processes were strong in the 



