ORCELLA. 411 



outwardly directed until they are nearly at right angles to the long axes of the bodies, 

 but yet with a slight backward tendency. Measured from the centre of the body 

 of the 5th vertebra posteriorly to its tip, the length of this process is only slightly 

 in excess of the interval between the first mentioned point and the apex of the 

 spinous process, which is exactly the reverse of the proportion between the transverse 

 and spinous processes of Globicephalus, from which the skeleton of Orcella is 

 remarkably different, in such details. In the 6th, 6th, and 7th lumbar vertebrEe the 

 lengths of the transverse processes and their proportions to the spinous processes 

 are much the same as in the immediately preceding vertebrae, but in the 8th the 

 transverse process becomes slightly shorter, but hardly perceptibly so. In the 10th 

 the diminution ia length is more visible, and with this diminution the process tends 

 to curve forwards. Decreasing in length, the processes increase in antero-posterior 

 breadth towards their tips ia the lumbar vertebrge. 



In the two sacral vertebrae there is a considerable alteration in the character of 

 the transverse processes as compared with the 8th lumbar, but the transition is 

 gradual. In the sacral, the process is contracted at its base and expanded towards 

 its tipa and the forward curve of the external half of the process is more marked 

 than in the preceding vertebrae. The transverse process is nearly equal to the height 

 of the spinous process measured from the centre of the body. The height of the 

 neural canal is about one-third, and its greatest breadth is about one-eighth the 

 height of the neural arch and its process combined. 



The four succeeding caudal vertebrae have much the same character as the 

 sacral vertebrae, but they differ from them and the four preceding lumbar vertebrae 

 ia the greater development of the processes occupying the position of zygapophyses 

 and in the appearance of a ridge which passes backwards from the front margins of 

 the latter ia the locality of a mammillary process. The length of the spinous and 

 transverse processes gradually diminishes, and the breadth of the latter increases with 

 their forward curve, and the spinous processes become nearly vertical. 



In the succeeding caudals the bifurcate process in the position of a zygapo- 

 physis is even better marked, as it is traced to the 41st vertebra, in which the height 

 of the spinous process measured from the middle of the body is less than the length 

 of the transverse process taken from the same point. In the 40th vertebra, the free 

 end of the transverse process is obHquely shortened, the direction of the process 

 being much forwards. These two last-mentioned characters become more pro- 

 nounced in the next five vertebrae, associated at the same time with a reduction in 

 the length of the transverse process which in the 44th vertebra is reduced to a hook- 

 like process. The spinous process suiiers considerable diminution in length with a 

 gradual downward subsidence, towards the bodies of the vertebrae, of the processes 

 in the position of the anterior zygapoj)hyses, and the ridge connected with them 

 becomes more marked as it is traced backwards. In the 46th vertebra the transverse 

 process becomes reduced to a lateral ridge on the front half of the body, but the 

 spiaous process is well marked, also the neural canal. The former of these processes 

 disappears on the 47th and the latter on the 51st segment. 



