PEOFESSOE FLOWEE ON THE EECENT ZIPHIOID WHALES. 227 



and posterior edges of the arch until between the eighth and ninth dorsal vertebrae 

 inclusive, but not developed between the ninth and tenth. 



Metapophyses first appear as distinct tubercles on the transverse processes of the 

 third, and gradually increase in size and become more compressed, pointing forwards 

 and slightly upwards. 



Articular surfaces for the heads of the ribs are developed only on the hinder edges of 

 the bodies, without any corresponding surface on the anterior edge of the next vertebra, 

 so that the head of the rib appears not to articulate directly with the body of the same 

 vertebra to which the tubercle is attached, but only to the one in front of it. In the 

 first vertebra this surface is entirely on the side of the body, in the second at the 

 junction of the body and the arch, from the third to the seventh at the root of the 

 pedicle of the arch ; on and after the eighth it is absent altogether, and the rib is attached 

 only to the transverse process. 



The transverse processes, in a line with the upper transverse processes (diwpophyses) of 

 the cervical region, are short and thick, with large rounded articular extremities for the 

 tubercles of the ribs. In the seventh vertebra this process is small, and in the eighth 

 reduced to a mere low longitudinal ridge on the outside of the metapophysis, which 

 has here acquired a considerable size. In the ninth vertebra a large and massive process 

 springs from the upper part of the side of the body near the anterior edge, in a situation 

 corresponding to which no trace of a process exists on any of the vertebras in front. It 

 has a large articular surface at its extremity, looking somewhat backwards, for the 

 ninth rib. The tenth vertebra bears a corresponding process, but rather longer, more 

 depressed, wider from before backwards, situated rather lower on the side of the body, 

 and not quite so near its anterior edge. Its articular surface (for the tenth rib), also 

 directed obliquely backwards, is not so large as that of the ninth. This process corre- 

 sponds serially with the transverse processes of the lumbar vertebra?. 



Berardius thus conforms to the type of the Physeteridae in the transverse processes of 

 the dorsal vertebrae not gradually sinking from the arch to the body, as in the true 

 Dolphins, but disappearing near the end of the series, and being replaced by a new 

 process; but it differs from Physeter, and exactly agrees with Mesoplodon 1 , in not 

 having both upper and lower processes developed simultaneously on several of the 

 vertebrae. Hyperoodon approaches nearer to Physeter in this characteristic feature, as 

 its seventh thoracic vertebra has distinct upper and lower transverse processes, which in 

 some specimens completely unite at their extremities, so as to form a ring, to the outer 

 edge of which the rib is attached. 



The twelve lumbar vertebrae are very much alike. Their bodies increase in size 

 towards the hinder end of the series, where they are remarkably elongated. Below 



1 In the skeleton of Sowerby's Mesophdon in the Brussels Museum, the upper process continues as far as 

 the seventh vertebra, and the lower process commences abruptly on the eighth. 



vol. viii. — paet in. September, 1872. 2 l 



