SCANDINAVIAN CETACEA. 259 



a hollow, in which the tuherciilum atlantis anticum is visible. The transverse processes are single, 

 rather long, and obtusely conical ; their height, however, is greater than their width ; their length is 

 not half the height of the atlas ; the foramen spinale is widest above, and its lower, nauch narrower 

 part, is on each side separated from the upper wider part by a projecting process.^ The axis is larger 

 than the atlas, in consequence of its strongly developed lateral processes. Its corpus, 4|" in 

 length, is somewhat longer than that of the atlas. It has a low elevation on its anterior articular 

 surface, answering to the processus odontoideus, and on its posterior articular surface a groove, 

 running inwards from the middle of the upper and lower edges, indicating its division into two 

 lateral parts, but not extending to the middle of the articular surface. Each lateral process, 

 which is longer than the width of the corpus, and not much lower than the height of the axis, has, 

 near the corpus, a large oval perforation, which is directed obliquely upwards. Their ends 

 point upwards, and there is a considerable notch above to their inner side. The lateral pro- 

 cesses are directed somewhat backwards. The processus spinosus is bipartite, and consists of two 

 low, rather dissimilar tubercles, which have their largest measurement in their length. This 

 vertebra, by the dijBFerent direction of the points and the perforations of the processus 

 transversi, differs very much from the same vertebra of Companyo's, according to his figure. 

 This deviation may, however, be a consequence of the different age, as Companyo's whale 

 was young, and this was an old one ; they show also a considerable difference in the form of 

 the atlas, as previously stated. The processus transversi of the 3rd to the 6th cervical vertebræ 

 are ring-shaped, and even the exterior part of this ring is ossified. The lateral process of the 7th 

 cervical vertebra is without the lower ramus or parapophysis ; the corpus of this vertebra is 3|" 

 long. The corpora of the dorsal vertebræ gradually increase in length in passing backwards, as do the 

 processus spinosi. Corpus of 2nd dorsal vertebra 5i", and that of 3rd dorsal vertebra 6" in length. 

 Even the most posterior dorsal vertebra has scarcely any ridge on the lower side of the corpus. The 

 corpus of the lumbosacral vertebræ also increases in length backwards, and has on all a longitudinal 

 ridge on the low^er side, although this ridge is hardly perceptible on the 1st. All the vertebral 

 epiphyses are united on these, as well as on the other vertebræ. The caudal vertebræ have not 

 any ridge on the lower side of the corpus ; those carrying processus spinosi inferiores are, on 

 the contrary, longitudinally concave. Processus spinosi inferiores 18, the posterior ones very 

 small. The sternum resembles in its form very much that of the rorqual from the northern 

 coast of Seeland, which is figured by Eschricht in his 5th memoir,^ and which figure has been 

 copied in ' Skand. Eauna.'^ It differs from this only in the concavity in front being smaller, 

 the lateral processes wider, and the posterior process or point wider and more obtuse. Its width 

 is more than twice its length ; its anterior middle part (manubrium) projects rather strongly, and 



^ It should here be stated that the atlas figured (No. 7) in ' Skand. Fauna Daggdjuren,' 

 p. 644, as belonging to Balæna prisca, and which more recently, in the ' Ofversigten af K. Vetensk. 

 Akadem. Forhandlinger,^ 1860, p. 105, was mentioned as belonging to Balænopt. musculus, is not from 

 this species, but from a Megaptera, probably M. hoops (Fabricius). It agrees in its form very nearly 

 with that which G. Cuvier, in ' Recherches sur les Ossemens Fossiles,' t. v, 1, pi. xxvi, fig. 19, 

 has figured as belonging to the ' Rorqual du Cap,' which, according to Eschricht, is the same species as 

 Megaptera hoops vel longimana. 



^ L. c, p. 364. 



2 ' Daggdj.,' p. 640, fig. 3. 



