CHIMAEEOGALE. 145 



zygapophysis. In the fourteenth, it is nearly as far back as and below the posterior 

 zygapophysis, with the very feebly developed transverse process of that vertebra 

 below it. In the first lumbar, the same process has become, as it were, the upward 

 and backwardly projecting process of the httle-developed transverse process of that 

 vertebra. In the second lumbar it is still more developed, and indicates a decided 

 tendency to separate from the transverse process, which, although far from being 

 strongly pronounced, is directed downwards and forwards. In the third lumbar the 

 transverse process is slightly more marked than in the preceding vertebra, but the 

 process in question is reduced to a mere trace on its posterior margin, and in the 

 following lumbar vertebrae, in which the transverse processes increase in size from 

 before backwards, with, however, only a rod-like antero-posterior expansion, all trace 

 of it is lost. It would thus appear, from the twelfth dorsal to the third lumbar, to be 

 entitled to be regarded as an anapophysial process, but its gradual transition from a 

 metapophysial to an anapophysial position is very suggestive. A true metapophysis 

 may be detected in the second lumbar, and the same process occurs more intensified 

 in the third and fourth, but is reduced in the fifth. A well-developed, posteriorly 

 bifurcate hypapophysis occurs on the third and fourth cervical, and a ridge-like 

 hypapophysis on the vertebrse behind them to the second dorsal. Two rather widely 

 separated hypapophysial processes occur on the thirteenth and fourteenth dorsal and 

 first and second lumbar vertebrae, the remaining lumbar vertebrae having a ventral 

 longitudinal ridge on their centra. 



The dorsal spines of the four sacral vertebrae are united into a rather prominent 

 ridge with a thickened margin, but the spine of the fourth is rather deeply separated 

 from the others. All the sacral vertebrae are firmly united together, but only the first 

 and one-half of the second are applied to the ihum. The first vertebra developes 

 on the sides of the sacral crest what appears to be a trace of united zygapophysial 

 and hyperapophysial elements, and a well-developed, bifurcate hypapophysis on the 

 anterior margin of its centrum : the remainder of the sacral vertebrae being ventrally 

 ridged, the third and fourth developing a feeble transverse process. The spinous 

 processes of the first and second caudals are well developed and halbert-shaped, and 

 backwardly projecting, while only a trace of the process can be observed on the third 

 caudal, whereas it cannot be detected on the following vertebrge. The processes 

 which support the anterior zygapophyses of the first caudal are very long and twice 

 as much so as those of the second caudal. Both processes arise from the anterior 

 surface of the base of the spinous process, and the articular surfaces which they 

 carry on the inner aspect of their extremities are applied, as it were, to each side of the 

 hinder end of the spinous process in front, so sessile are the posterior zygapophyses. 

 On the second caudal there are no traces of these last-mentioned processes, but the 

 last remnant of a neural arch can be detected passing below the spinous process. 

 All the caudal vertebrae, from the third inclusive, are distinguished by the presence 

 of two rod-like processes on the upper surface of their anterior and posterior ends, 

 with another process in close apposition to each externally ; the superior processes 

 doubtless are probably serially homologous with the anterior and posterior zygapo- 



T 



