22 PEOF. T. JEFFEEY PAEKEE ON THE 



increased, and a haemal process is given off from it. The latter increases in length, 

 the haemal plate continues to be divided into a larger anterior (fig. 22, h.pl), a smaller 

 posterior (h.pl 1 ) portion, and the zygapophyses become practically obsolete, until, in the 

 thirty-second vertebra, the characters shown in fig. 22 are assumed. In this there are 

 seen to be only three strengthening ridges on each side, making with the neural and 

 haemal plates five pairs of radiating lamellae. 



Between the thirty-sixth and the forty-fifth vertebrae, the increase in length of the 

 centrum is small ; the neural and haemal plates diminish in height, the length of the 

 haemal processes increases, and the lateral ridges are reduced from three to two. In the 

 fiftieth a small bony lump or knot appears in the middle of the lateral surface of the 

 centrum : this occurs in all succeeding vertebrae, attaining its maximum in the ninety- 

 second (fig. 25, c. 92) : the bone of which it is composed is the hardest in the body. 



Between the fiftieth and the sixtieth vertebra, the two strengthening ridges become 

 gradually fused into one ; this is retained throughout the rest of the vertebral column 

 (figs. 24 and 25) : the neural and haemal plates also undergo further reduction. In 

 foe fifty-second, the haemal processes attain a length of a little more than 1^ inch, and 

 from about the seventieth to the ninetieth this is increased to the maximum length of 

 nearly 2 inches : they continue to be slightly inclined backwards up to the eighty-eighth, 

 in which the inclination becomes more marked. In succeeding vertebrae they become 

 more and more tilted upwards, until in the ninety-second (fig. 25) they are nearly 

 parallel to the centrum. 



It is somewhat remarkable that in no case do the haemal processes unite distally to 

 form a haemal arch, and that consequently there are no haemal spines in any part 

 of the vertebral column. 



As mentioned above, the neural spines from about the seventh vertebra onwards are 

 very slender, not more than ^ inch in diameter. They gradually increase in length as 

 far as about the thirtieth, where the maximum length of 4J inches is reached, and is 

 retained to about the sixty-sixth vertebra, from which point they begin gradually to 

 shorten, being reduced to 3 inches in the eightieth, to 1^ inch in the ninetieth, and to 

 i|inch in the ninety-second. 



The vertebral column is terminated by a bone which I have called, for convenience, 

 the ninety-third vertebra. This (fig. 25, c. 93) is a subconical bone nearly 2-| inches in 

 length, having low neural and haemal plates, but neither neural, anal, nor haemal 

 processes, and marked externally by a lateral ridge which undergoes a great thickening 

 at the posterior end of the bone. In longitudinal section (fig. 26) it is seen to have 

 a single subconical cavity ending bluntly near the hinder end, and, like the cavities of 

 the other vertebrae, filled with the gelatinous remains of the notochord. The bone thus 

 has all the characters of the anterior half of a vertebra, and may be conveniently called 

 a demivertebra. Its size distinctly tends to confirm this interpretation of its nature ; 

 if completed it would be nearly 5 inches long, that is about an inch longer than its 



